HOME HELP ORNL-TM- 3039
Contract No. W-7405-eng-26 Reactor Division
MSRE SYSTEMS AND COMPONENTS PERFORMANCE by Molten-Salt Reactor Experiment S t a f f
--Edited and Compiled by R. H. Guymon
JUNE 1973
This document t a i n s information of a preliminary nature and was prepared primarily f o r i n t e r n a l use a t t h e Oak R i d g e National Laboratory. It i s subject t o r e v i s i o n or correction and t h e r e f o r e does not represent a f i n a l report. NOTICE:
i
OAK RIDGE NATIONAL LABORATORY Oak Ridge, Tennessee 37830 operated by U. S. ATOMIC ENERGY COMMISSION NOTICE
L
T h b report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the United States Atomic Energy Commission, nor any of their employees, nor any of their contractors, subcontractors, or their employees, m k e a any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights.
DISTRtdUTION OF THIS DOCUMENT IS UNUM -
- -
., ' - - -
I
.
.. . ,
.-
.
.'
. . .
i
i
iii
G
CONTENTS Page 1 3
..,..... ..............-..... .... ................ 2. DESCRIPTION OF THE PLANT . . . . . . . . . . . . 4 3. CHRONOLOGY OF OPERATION AND MAINTENANCE . . . . . 9 4. PLANT PERFORMANCE AND STATISTICS. . . . . . . . . . 15 4.1 Cumulative Statistics. . . . . . . . . . . . . . . . 15 4-2 Availability during Various Periods. . . . . . . . 18 4.3 Interruptions of Operations. . . . . . . . . . . . . 22 4.4 Time Required for Operational Tasks. . . . . . . . 48 4.5 Changes Made in the Plant. , . . . . . . . . . . . . 59 4.6 Tabulation of Recorded Variables at Full Power . . . . 63 5. FUELSYSTEM.. . . . . . . . . . . . . . . . . . . . . . . 89 5.1 Description. . . . . . . . . . . . . . . . . 89 5.2 Purging of Moisture and Oxygen from the System . . . . 89 5.3 Fuel-Circulating-System Volume Calibration . . . . . 91 . 5.4 Drain Times. .-.. . . . . . . . . . . . . . . . . . 91 5.5 Mixing of Fuel and Flush Salts . . . . . . . . . . . 93 5.6 Primary System Leak. . . . . . . . . -.. . . . . . . 93 5.7 Operation . . . . . . . . . . . . . . . . . . . . . 94 5.8 Fuel Pump and Overflow Tank. .. .. . .. .. .. .. . . 110 94 5.9 Primary Heat Exchanger . . 5.10 Reactor Vessel and Reactor Access Nozzle . .. . 112 5.11 Fuel and Flush Salt Drain Tanks. .. 129 . . . . 137 6. COOLANTSYSTEM . . . . . . . . . . . . . . . 6.1 Description . . . . . . . . . . . . . . . . 137 6.2 Purging Moisture and Oxygen from the System. . . . . . 137 6.3 Coolant Circulating System Calibration and Drain The. 137 6.4 Operation. . . . . . . . . . . . 139 6.5 Coolant Salt circulating Pump. . . . 139 6.6 Radiator . . . . . . . . . . . . . . 140 6.7 Performance of Main Blowers, MB-1 and MB-3 . 147 6.8 Coolant Drain Tank .... . .. . 159 ABSTRACT1. INTRODUCTION
1
e
2
7.
COVER-GAS SYSTEM 7.1 7.2 7.3
e................161
. . . . . . . .. .. . . . . .. . .. .
Initial Testing Normal Operation Conclusions and Recommendations.
e
161 162 164
- .
. iv
.
........................ 165 8.1 Description ...................... 165 8.2 Experience with Coolant Salt Off-gas System . . . . . . 167 . 8.3 Experience with the Fuel Off-Gas System . . . . . . . . 173 8.4 Difficulties with the Fuel Off-gas System . . . . . . . 177 8.5 Subsequent Operating Experience with Fuel Off-gas System ....................... 201 8.6 Discussion and Conclusions . . . . . . . . . . . . . . 212 9. FUEL ANDCOOLANTPUMPLUBEOILSYSTEMS . . . . . . . . . . . . 215 9.1 Description ...................... 215 9.2 Installation and Early Problems. . . . . . . . . . . . 215 Addition of Syphon Tanks to the Oil Catch Tanks . . . . 216 9.3 9.4 OilLeakage . . . . . . . . . . . . . . . . . . . . . . 217 9.5 Change-out of Oil Pumps . . . . . . . . . . . . . . . . 218 9.6 Test Check of One'Oil System Supplying Both the Fuel and Coolant Salt Pumps . . . . . . . . . . . . . . . 218 Oil Temperature Problems . . . . . . . . . . . . . . . 218 . 9.7 Increase in Radiation Levels the Lube Oil Packages. 219 9.8 Analysis of Oil .................... 220 9.9 Replacement of Oil . . . . . . . . . . . . . . . . . . 220 9.10 9.11 Discussion agd Recommendations . . . . . . . . . . . . 222 10. COMPONENT COOLING . . . . . . . . . . . . . . . . . . 223 10.1 Primary System .................... 223 10.2 Secondary System . . . . . . . . . . . . . . . . . . . 231 232 11. VENTILATIONSYSTEM ..................... 11.1 Description. . . . . . . . . . . . . . . . . . . . . . 232 11.2 Operating Experience . . . . . . . . . . . . . . . . . 232 11.3 Conclusions and RecommendatZons. . . . . . . . . . . . 235 12. WATER ........................ 237 12.1 Potable Water System . . . . . . . . . . . . . . . . . 237 12.2 Process Water System . . . . . . . . . . . . . . . . . 238 12.3 Cooling Tower Water System . . . . . . . . . . . . . . 239 12.4 Treated Water System . . . . . . . . . . . . . . . . . 240 12.5 Condensate System . . . . . . . . . . . . . . . . . . 245 12.6 Nuclear Instrument PenetratLon. . . . . . . . . . . . .245 12.7 General Water Systems Conclusions. . . . . . . . . . . 246 247 13 . LIQUID WASTE SYSTEM ..................... 13.1 Description. . . . . . . . . . . . . . . . . . . . . . 247 13.2 Preliminary Testing . . . . . . . . . . . . . . . . . . .247 13.3 Operating Experience . . . . . . . . . . . . . . . . . 248 14. HELIUM DETECTOR SYSTEM . . . . . . . . . . . . . . . . . 251 14.1 Description and Method of Operation. . . . . . . . . . . .251 14.2 Calibration . . . . . . . . . . . . . . . . . . . . . .251 14.3 Operating Experience . . . . . . . . . . . . . . . . . .255 14.4 Discussion and Recommendations . . . . . . . . . . . . 256 8
OFF-GAS SYSTEM
at
SYSTEMS
SYSTEMS
LEAK
bd
V
Page
. . . . . . . . . . . . . . . . . . . . 259 15.1 Description . . . . . . . . . . . . . . . . . . . . . 259 15.2 Operating Experience . . . . . . . . . . . . . . . . . 259 15.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . 260 261 16 . ELECTRICAL SYSTEM . . . . . . . . . . . . . . . . . . . . . . 16.1 Description. . . . . . . . . . . . . . . . . . . . . . 261 16.2 Alternating Current System . . . . . . . . . . . . . . 262 16.3 250-V dc System . . . . . . . . . . . . . . . . . . . . 265 16.4 Reliable Power System . . . . . . . . . . . . . . . . . 265 16.5 48-V dc System . . . . . . . . . . . . . . . . . . . . 266 16.6 Diesel Generators. . . . . . . . . . . . . . . . . . . 267 16.7 100-kVa Variable Frequency Motor-Generator Set . . . . 267 268 16.8 Conclusions . . . . . . . . . . . . . . . . . . . . . . 17 . HEATERS . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 17.1 Description . . . . . . . . . . . . . . . . . . . . . . 269 17.2 Preoperational Checkout . . . . . . . . . . . . . . . . 273 17.3 System Heatup . . . . . . . . . . . . . . . . . . . . . 273 17.4 Heater Performance . . . . . . . . . . . . . . . . . . 278 17.5 Systems Cooldown Rate . . . . . . . . . . . . . . . . . 287 17.6 Discussion . . . . . . . . . . . . . . . . . . . . . . 291 18. SAMPLERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 18.1 Fuel Sampler-Enricher . . . . . . . . . . . . . . . . . 292 18.2 Coolant Sampler. . . . . . . . . . . . . . . . . . . . 320 19 . CONTROLRODS . . . . . . . . . . . . . . . . . . . . . . . . . 323 . 323 -19.1 Description. . . . . . . . . . . . . . . . . . . . . . 19.2 Initial Testing . . . . . . . . . . . . . . . . . . . . 324 19.3 Periodic Testing . . . . . . . . . . . . . . . . . . . 325 19.4 . . . . . . . . . . . . . . . . . 326 19.5 Improved Rod Scram Testing . . . . . . . . . . . . . . 336 19.6 Maintenance Experience . . . . . . . . . . . . . . . . 338 19.7 Discussion and Recommendations . . . . . . . . . . . . 339 20. FREEZE VALVES . . . . . . . . . . . . . . . . . . . . . . . . 341 20.1 Introduction . . . . . . . . . . . . . . . . . . . . . 341 20.2 Description of the Design of the Freeze Valves . . . . 341 20.3 Description of the Operation of the Freeze Valves . . . 344 20.4 Operating Experience . . . . . . . . . . . . . . . . . 346 20.5 Recommendations. . . . . . . . . . . . . . . . . . . . 353 21 . FREEZE FLANGES . . . . . . . . . . . . . . . . . . . . . . . . 356 21.1 Description . . . . . . . . . . . . . . . . . . . . . . . 356 21.2 Operation . . . . . . . . . . . . . . . . . . . . . . . 356 21.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . 358 15
.
INSTRUMENT AIR SYSTEM
Operating Experience
.
..
....
.......
......
...
.
. .
~
vi Page
. . . . . . . . . . . . . . . . . . . . . . . . . 360 22.1 Description and Criteria . . . . . . . . . . . . . . . 360 22.2 Methods Used to Assure Adequate Containment and Results . . . . . . . . . . . . . . . . . . . . . . . 360 22.3 Discussion of Cell Leak Rate Determination . . . . . . 370 22.4 Vapor Condensing System . . . . . . . . . . . . . . . 376 22.5 Recommendations .................... 377 23 . BIOLOGICAL SHIELDING AND RADIATION LEVELS . . . . . . . . . . 380 23.1 Radiation Surveys -Approach to Power . . . . . . . . . 380 23.2 Radiation Levels During Operation. . . . . . . . . . . 383 23.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . 386 24. INSTRUMENTATION . . . . . . . . . . . . . . . . . . . . . . . 389 24.1 Introduction . . . . . . . . . . . . . . . . . . . . . 389 24.2 Description . . . . . . . . . . . . . . . . . . . . . . 389 24.3 Initial Checkout and Startup Tests . . . . . . . . . . 393 24.4 Periodic Testing . . . . . . . . . . . . . . . . . . .394 24.5 Performance of the Nuclear Safety Instrumentation. . . 397 24.6 Performance of the Wide-Range Counting Channels. . . . 400 24.7 Performance of the Linear Power Channels . . . . . . . 402 BF. Nuclear Instrumentation. . . . . . . . . . . . . . 403 24.8 24.9 Nuclear Instrument Penetration . . . . . . . . . . . . 403 24.+0 Performance of the Process Radiation Monitors . . . . . 403 24.11 Performance of the Personnel Radiation Monitoring and Building Evacuation System . . . . . . . . . . . . . 404 24.12 Performance of the Stack Monitoring System . . . . . . 404 24.13 Performance of Thermocouples and the Temperature Readout and Control Instrumentation. . . . . . . . . 404 24.14 Performance of Pressure Detectors . . . . . . . . . . . 408 24.15 Performance of Level Indicators . . . . . . . . . . . 408 24.16 Performance of the Drain Tank Weighing Systems . 409 24.17 Performance of the Coolant Salt Flowmeters . . . . . . 410 24.18 Performance of Relays . . . . . . . . . . . . . . . . . 411 24.19 Training Simulation . . . . . . . . . . . . . . . . . 411 24.20 Miscellaneous . . . . . . . . . . . . . . . . . . . . 412 24.21 Conclusions and Recommendations . . . . . . . . . . . . 412 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 22
.
CONTAINMENT
bi
MSRE SYSTEMS AND COMPONENTS PERFORMANCE
ABSTRACT When t h e MSRE was shut down i n December 1969, i t had accumulated 13,172 full-power hours of operation,
S a l t had been c i r c u l a t e d i n t h e
f u e l system f o r 21,788 hours and i n t h e coolant system f o r 26,076 hours. E s s e n t i a l l y no d i f f i c u l t y w a s encountered with t h e primary system during operation.
After t h e r e a c t o r was shut down t h e r e w a s an i n d i c a t i o n
of a l e a k i n t h e drain-tank piping a t / o r near a f r e e z e valve.
Further in-
v e s t i g a t i o n w i l l be made later as t o t h e nature and cause of t h i s leak. There w a s a small continuous leakage of l u b r i c a t i n g o i l i n t o t h e f u e l pump throughout t h e operation.
This, togdther with s a l t m i s t , caused peri-
odic plugging i n t h e off-gas system which was designed f o r clean helium. F i l t e r s i n s t a l l e d i n t h e main l i n e s proved very e f f e c t i v e . I n e a r l y operation, d i f f i c u l t y w a s encountered with t h e coolant radiator.
The doors would not seal, t h e r e were too many a i r leaks, and thermal
i n s u l a t i o n w a s inade+ate.
After t h e s e were repaired, t h e system operated
f i n e except f o r a f a i l u r e of one of t h e main blowers and some trouble with t h e blower bearings. Only r e l a t i v e l y minor d i f f i c u l t i e s were encountered with t h e containment and o t h e r systems.
2
td
1.
INTRODUCTION P. N. Haubenreich Operation of t h e MSRE constituted a major s t e p toward the objectives of t h e Molten-Salt Reactor Program.
1
The goal of t h i s program i s t h e de-
velopment of l a r g e , fluid-fuel reactors having good neutron econorqy and producing law-cost e l e c t r i c i t y . '
The MSRE w a s b u i l t t o demonstrate t h e
p r a c t i c a l i t y of t h e molten-salt r e a c t o r concept, with emphasis on t h e compatibility of t h e materials (fluoride salts , graphite , and container a l l o y ) , the performance of key components , and t h e r e l i a b i l i t y and'maint a i n a b i l i t y of t h e plant. I n t h e course of 5 years of t e s t i n g and operation of t h e MSRE (1964
-
1969) t h e operators accumulated considerable experience w i t h t h e various components and systems i n t h e r e a c t o r plant.
This experience, properly
disseminated, should be valuable i n t h e continuing development of moltensalt reactors.
Much has already been published i n t h e Molten S a l t Reactor
Program semiannual progress reports ( R e f s . 2 t o 14) but such reporting is pie'cemeal and sometimes r a t h e r condensed. much very d e t a i l e d information i n t e s t
On t h e other hand, t h e r e i s
reports and operations and mainte-
nance f i l e s , but these are r e l a t i v e l y inaccessible and s p e c i f i c informat i o n i s tedious t o e x t r a c t .
It w a s considered worthwhile, t h e r e f o r e , t o
e x t r a c t , organize, evaluate and report t h e experience with MSRE systems and components. The purpose of t h i s report is t o present a convenient, comprehensive
description of t h e MSRE experience.
The first chapters b r i e f l y describe
t h e p l a n t and o u t l i n e t h e chronology of i t s operation.
N e x t there is a
chapter on t h e o v e r a l l p l a n t performance, includihg s t a t i s t i c s r e l a t i v e t o r e l i a b i l i t y and maintainability.
The chapters which follow a r e each
devoted t o one system o r component.
Finally, t h e r e i s a chapter of dis-
cussion and conclusions.
LiJ
3 2.
DESCRIPTION OF THE PLANT
R. H, Guymon 5-
The MSRE was a single-region, c i r c u l a t i n g molten-salt fueled, thermal r e a c t o r which produced heat at t h e r a t e of about 8 Mw.
The f u e l was UF4
i n a c a r r i e r salt of LiF-BeF$ZrFt.,.’ A t t h e operating temperature of 1200°F, t h i s salt i s a l i q u i d which has very good physical properties:
viscosity
about 8 centipoise, density about 135 l b / f t 3 , and’vapor pressure l e s s than 0.1 mm Hg. The design conditions a r e shown i n t h e flow diagram (Fig. 2-11,
general arrangement of t h e plant i s shown i n Fig. 2-2. ~
The
The salt-containing
piping and equipment was made of Hastelloy-N, a nickel-molybdenum-ironchromium a l l o y with exceptional r e s i s t a n c e t o corrosion by molten f l u o r i d e s and with high s t r e n g t h at high temperature. I n t h e r e a c t o r primary system, t h e f u e l salt w a s r e c i r c u l a t e d by t h e sump-type c e n t r i f u g a l pump through t h e shell-and-tube h e a t exchanger and
W
t h e r e a c t o r vessel. i n Figure 2-3.
The 5-ft diam. by 8-ft high r e a c t o r v e s s e l i s shown
It w a s f i l l e d with 2-in. by 2-in.
graphite moderator s t r i n g e r s
which had grooves machined i n t h e s i d e s t o form flow channels f o r t h e f u e l
salt.
Since t h e graphite i s compatible with t h e molten salt, it was possible
t o use unclad graphite which i s d e s i r a b l e t o obtain good neutron economy. The heat generated i n t h e f u e l salt as it passed through t h e r e a c t o r w a s t r a n s f e r r e d i n t h e heat exchanger t o a molten LIF-BeF2 coolant s a l t .
The
coolant salt w a s c i r c u l a t e d by means of a second sump-type pump through t h e h e a t exchanger and through t h e r a d i a t o r . flow blowers p a s t t h e r a d i a t o
A i r was blown by two a x i a l
tubes t o remove t h e heat which w a s s e n t
up t h e coolant s t a c k where it w a s d i s s i p a t e d t o t h e atmosphere.
Drain tanks were provided f o r s t o r i n g t h e f u e l and coolant salts at high temperature when t h e r e a c t o r was not operating.
LiF-BeF2 f l u s h salt
used f o r flushing t h e f u e l system before and a f t e r maintenance was s t o r e d i n the fuel flush tank.
The salts were drained by gravity.
They were
t r a n s f e r r e d back t o t h e c i r c u l a t i n g systems by pressurizing t h e tanks with helium.
L,
i c\
J
FILTEl
STrn
FbN
I-= I I
1-
i i
Fig. 2.1
Design Flowsheet of the MSRE
I
C
c
c
c
c
C
OWL-DWG 634209R
REACTORCONTROL
6. COOLANT PUMP
Fig. 2.2
Layout of the MSRE
12. CONTAINMENT VESSEL 13. FREEZE VALVE
6 OR&-LR-OW
GRAPHITE SAMPLE ACCESS PORT
61097RqA
FLEXIBLE CONDUIT TO CONTROL ROD DRIVES
COOLING AIR LINES
ACCESS PORT COOLING JACKETS
FUEL OUTLET CORE ROD THIMBLE LARGE GRAPHITE SAMPLES
REACTOR ACCESS PORT SMALL GRAPHITE SAMPLES HOLD-DOWN ROD OUTLET STRAINER
CORE CENTERING GRID
FLOW DISTRIBUTOR VOLUTE GRAPHITE -MODE
CORE WALL COOLING ANNUI REACTOR CORE CAN REACTOR VESSEL
ANTI-SWIRL VANES VESSEL DRAIN LINE
Fig. 2.3
SUPPORT GRID
Details of the MSRE Core and Reactor Vessel
'
7
The f i s s i o n product gases, krypton and xenon, were removed continuously from t h e c i r c u l a t i n g f u e l salt by spraying salt at a r a t e of 50 gpm i n t o t h e cover gas above t h e l i q u i d i n t h e f u e l pump tank.
There they trans-
f e r r e d from t h e l i q u i d t o t h e gas phase and were swept out of t h e tank by a small purge of helium.
After a delay of about 1-1/2 h r i n t h e piping,
t h i s gas passed through water-cooled beds of activated charcoal.
The
krypton and xenon were delayed u n t i l all but t h e 8 5 K r decayed and then were d i l u t e d w i t h a i r and discharged t o t h e atmosphere. The f u e l and coolant systems were provided with equipment f o r taking samples of t h e molten salt while t h e reactor w a s operating at power.
The
fuel sampler was a l s o used f o r adding s m a l l amounts of f u e l t o t h e reactor
while at power t o compensate f o r burnup. The negative temperature c o e f f i c i e n t of r e a c t i v i t y of t h e f u e l and graphite moderator made nuclear control of t h e system very simple.
However,
t h r e e control rods were provided f o r adjusting temperature, compensating f o r buildup of f i s s i o n products, and f o r shutdown. The p l a n t w a s provided with a simple processing f a c i l i t y f o r t r e a t i n g
f'ull 75-ft,3 batches of h e 1 salt with hydrogen f l u o r i d e o r f l u o r i n e gases. The hydrogen f l u o r i d e treatment was f o r removing oxide contamination from
t h e salt as H20.
The fluorine treatment u t i l i z e d t h e f l u o r i d e v o l a t i l i t y
process f o r removing t h e uranium as uF6. Auxiliary systems included:
(1)a cover-gas system w i t h t r e a t i n g
s t a t i o n s f o r removing oxygen and moisture from t h e helium cover gas; (2) two closed-loop o i l systems f o r cooling t h e f u e l andd.coolant pumps
and providing l u b r i c a t i o n t o t h e bearings; ( 3 ) a closed loop component coolant system f o r cooling t h e control rods and other i n - c e l l components;
( 4 ) s e v e r a l cooling water systems including a closed-loop t r e a t e d water system f o r cooling c e r t a i n i n - c e l l equipment; (5) a v e n t i l a t i o n system f o r contamination control; and (6) an instrument a i r system. A l l of t h e primary salt system w a s located i n t h e r e a c t o r and drain
tank c e l l s .
These s e a l e d pressure vessels provided secondary containment.
For a fuller description of t h e p l a n t , see Reference 15.
9
t,
3.
CHRONOLOGY OF OPERATION AND MAINTENANCE . R. H. Guymon
Design of t h e MSRE began i n t h e summer of 1960 and by August 1964, i n s t a l l a t i o n w a s f a r enough along t o permit t h e planned non-nuclear t e s t i n g 1 6 t o begin.
Milestones t h a t were passed i n t h e years which fol-
lowed a r e l i s t e d i n Table 3-1. Table 3-1.
Milestones i n MSRE Operation
October 24, 1964
S a l t f i r s t loaded i n t o tanks S a l t first c i r c u l a t e d through core First criticality (
'U)
F i r s t operation i n megawatt range
~
Full power reached
Nuclear operation w i t h
' U concluded
S t r i p uranium from f u e l s a l t F i r s t c r i t i c a l with 233U Reach f u l l power with 2 3 3 U Nuclear operation concluded
1965 June 1, 1965 January 24, 1966 May 23, 1966 March 26, 1968 August 23-29, 1968 October 2 , 1968 January 28, 1969 December 12, 1969 J ~ ~ U ~ W 12,Y
The a c t i v i t i e s during t h e p e r i o d of non-nuclear t e s t i n g , zero-power
experiments, and preparation f o r power operation a r e outlined i n Fig. 3-1. During t h e prenuclear t e s t i n g , except f o r t h e usual s t a r t u p troubles due t o e a r l y f a i l u r e s and instrument malfunctions, all systems operated w e l l and t h e r e were no unanticipated problems i n handling t h e molten salt.
Some plugging of t h e offgas system valves and f i l t e r s w a s encoun-
tered. b u t t h i s did not seem serious. C r i t i c a l i t y w a s a t t a i n e d by adding UFb-LiF enriching salt t o t h e carrier salt.
More uranium w a s added t o b r i n g t h e f u e l gradually t o t h e
desired operating concentration while t h e control rods were c a l i b r a t e d and r e a c t i v i t y c o e f f i c i e n t s were measured.
kid
A t t h e conclusion of t h e zero-
power nuclear experiments t h e r e a c t o r w a s shut down t o f i n i s h construction of t h e vapor-condensing system, t o carry out some inspection, and t o
M S R E ACTlVlTl ES JULY 4964
INSPECTION AND
-
DECEMBER 1965
P R E N U C L E A R TESTS C O M P L E T E SYSTEM
PRELlM IN A R Y T E S T I N G
F I N A L PREPARATIONS FOR POWER OPERATION
OF
-- -- - r
I
3
FINISH INSTALLATION OF SALT SYSTEMS
9
INSTALL CORE SAMPLES INSPECT FUEL PUMP HEAT-TREAT CORE VESSEL
INSTALL CONTROL RODS SAMPLER-ENRICHER
LEAKTEST, PURGE Et HEAT SALT SYSTEMS
Ezzzzzm
TEST SECONDARY
TEST AUX. SYSTEMS
TEST TRANSFER, FILL Et DRAIN OPS.
OPERATOR TRAINING
LOAD SALT INTO DRAIN TANKS COOLANT SALT
FLUSH SALT
m
1 I
I
I
J
A
1 I
S
= a -
I I
O
Fig. 3.1
c
I I
I
N
D
LOAD Et CIRCULATE CARRIER
CIRCULATE C R FL SALTS
I
J
CONTAINMENT ADJUST Et MODIFY RADIATOR ENCLOSURE
LOAD U - 2 3 5 INZERO-POWER NUCLEAR EXPERIMENTS
I
I
I
1 I
I
I
M
A
M
I
,
I
J
J
w
0
LOW- POWER (0-50 kw) EX ezzl PTS.
mezzZ2za
1
F
FINISH VAPOR-CO~D. SYSTEM MODIFY CELL PENETRATIONS REPLACE RADIATOR DOORS
OPERATOR TRAINING
1
I, I
I I
I
A
S
I
O
'
I
N
I
0
'
MSRE Activities July 1964 -December 1965
c
c
t
11
’
make some repairs and modifications, including replacement of t h e heatwarped radiator doors.
The first t e s t of t h e secondary containment w a s
made during t h i s shutdown period. After tests i n t h e kilowatt range showed t h a t t h e dynamics of the system were as expected, t h e approach t o f u l l power w a s s t a r t e d i n January 1966. Plugging i n t h e f u e l offgas system occurred immediately
af%erincreasing t h e power t o one megawatt.
Almost three months were
spent investigating and remedying t h e offgas problem by i n s t a l l a t i o n of
a large, efficient f i l t e r .
As shown i n Fig. 3-2, t h e power ascension w a s
resumed i n April and fulLpower, which w a s limited by t h e capability of t h e heat-removal system, w a s attained i n M a y . High-power operation w a s abruptly h a l t e d i n July when t h e hub and blades of one of t h e main blowers i n t h e heat-removal system broke up. While t h e reactor w a s down, t h e array of graphite and metal specimens was removed f r o m t h e core and a new array i n s t a l l e d .
During the flushing
operations before t h e specimen removal, t h e fuel-pump bowl w a s accidentally over-filled and some flush salt froze i n t h e attached gas l i n e s .
Tempo-
rary heaters were i n s t a l l e d remotely t o c l e a r t h e l i n e s and i n November t h e offgas l i n e at t h e pump bowl w a s cleared by running a t o o l through it. Most of t h e remaining operation with 235Uw a s at power.
Restrictions
continued t o develop i n t h e offgas system but caused l i t t l e delay i n t h e program.
Shutdowns were necessary t o make repairs on t h e f u e l sampler-
enricher, t h e component coolant pumps, and in-cell air l i n e disconnects of high c e l l leak r a t e . The core whose leakage caused an specimens were replace and various experiments conducted.
l
nated on March 26, 1968 s o
ly, annual containment t e s t s were made, Operation with 235U f u e l w a s termie uranium could be s
ed from t h e
f i e 1 salt and replaced w i t During t h e shutdown.
-
and t h e f u e l piping and vesse troscopy t o determine t h e dis
lowed, t h e core specimens were replaced,
e surveyed using remote ganrma ray specion of f i s s i o n products. Necessary and
preventive maintenance w a s a l s o done i n preparation f o r operation w i t h 2 3 3 ~fuel.
b.
A f t e r t e s t i n g and extensive modification of t h e excess fluorine d i s posal system, t h e flush salt and f u e l s a l t were processed t o s t r i p the
12
SALT
w
SALT N NEL UMI
Nn Loop UfNAMKS TESPS
A
Y
J
866 J-
A
S 0
N
N V E m P T E GAS NFUELLDU'
J (967
J
A
:c
_] FUEL
-
0
2 4 6 8 WWER (Mw)
0
INMSTIGATE COMR GAS. XENON. AND FWON PROWCT BEHAVlOA ADD PLUTONIUM R R A D I A E MCdPSUITED U
MAP EP. aposlm WlTH W M P , SPECTROMETER
MEASURE TRITIUM, SAMPLE FUEL REMOVE CORE WRAY PUTREACTORINSTANDBY
-0
Fig. 3 . 2
Outline of the Four Years of MSRE Power Operation
dd
13
uranium and remove t h e corrosion products produced duri tion."
the fluorina-
Loading of 233Ui n t o t h e c a r r i e r salt, throu& s p e c i a l equipment
attached t o a drain tank began immediately thereafter.
The c e i l s were
closed and leak-tested before t h e reactor w a s made c r i t i c a l by s m a l l additions of 233U through t h e sampler-enricher.
On October 8, 1968, Glenn
Seaborg manipulated t h e controls t o r a i s e t h e power t o 100 kw, making the MSRE t h e world's first reactor t o operate on 233U. E a r l y i n t h e zero-power physics experiments, t h e amount of gas en-
t r a i n e d i n t h e c i r c u l a t i n g f u e l had increased from c0.l t o
- 0.5 uol %.
Investigation of this, as w e l l as d i f f i c u l t i e s w i t h t h e sampler-enricher and plugging i n t h e offgas system, delayed the approach t o f u l l power ~
u n t i l January 1969.
As the power w a s increased i n t o t h e megawatt range, there were observed f o r t h e first time sporadic s m a l l increases ( - 5 t o 10%) i n nuclear power f o r a few seconds, occurring w i t h a varying frequency somewhere around 10/hr.
The characteristics of t h e t r a n s i e n t s pointed t o changes
i n gas volume i n t h e f u e l loop and it appeared t h a t they were most l i k-. ely caused by occasi9nal release of some gas t h a t collected i n t h e core. This hypothesis w a s -supported when, l a t e i n February, a variable frequency generator w a s used t o operate t h e f u e l pump at reduced speed.
The gas
entrainment i n t h e f u e l circulatfng loop decreased sharply (from 0.7 t o
cO.1 vol $) snd t h e p e r t y b a t i o n s ceased e n t i r e l y . 1 8 Operatibns continued at various nuclear powers and fuel-pump speeds u n t i l June 1when t h e reactor w a s shut down'to replace t h e core specimens, investigate the d i s t r i b u t i o n of f i s s i o n products i n the primary loop by gamma scanning, remove t h e offgas r e s t r i c t i o m , and t e s t the secondary containment.
For t h e first time one of t h e control rods did not scram
d it w a s replaced during t h e shutdown.
The. remaining months of operation were spent i n various studies of '
t h e behavior of tritium, xenon, and c e r t a i n other f i s s i o n products i n the reactor,
.In November it became evident t h a t s u f f i c i e n t funds would.'not
14 t h e c e l l atmosphere, indicating a leak i n t h e primary system apparently near a freeze valve. After t h e f i n a l shutdown, t h e f a c i l i t y w a s placed i n a standby cond i t i o n t o await post-operation examinations planned f o r early i n t h e next f i s c a l year.lg
The operating crews were disbanded and over t h e next s i x
months most of t h e engineers were reassigned as they completed analyses and reporting of t h e reactor experience.
As of t h i s writing, t h e post-
operation examination has not been accomplished. In t h i s report and elsewhere MSRE run numbers a r e often used t o i d e n t i f y t h e period of operation.
Generally a new run number was assigned
at t h e end of a major shutdown o r when there were s u b s t a n t i a l changes i n t h e purpose o r type of operation.
The s t a r t i n g dates f o r each of the 20
runs are l i s t e d i n Table 3-2.
Table 3-2.
Run No.
.
S t a r t i n g Datea
Dates of MSRE Runs
Run No.
S t a r t i n g Datea
1
January 9 , 1965
11
January 24, 1967
2
M s y 11, 1965
12
June 8, 1967
3
M a y 31, 1965
13
September 3, 1967
4 5
December 5, 1965
14
September 19, 1967
February 7, 1966
August 1 4 , 1968
6
March 26, 1966
15 16
7 8 .9
June 11, 1966
10
December 10, 1968 January 10, 1969
September 14, 1966
17 18
Movember 6, 1966
19
July 31, 1969
December 6 , 1966
20
April 11, 1969 November 20, 1969
%ese are t h e dates on which samples , logs , e t c . , began t o receive new nmibers, and a r e not generally t h e dates of reactor s t a r t u p .
15
bi
4.
PLANT PERFORMANCE AND STATISTICS
R. H. Guymon
P. N. Haubenreich -"
"They've kept the darned thing running and when they shut down they can get back on the line; you can't knock that!" An unidentified "AEC spokesman", as quoted i n October 1 2 , 1967 Nucleonics Week. '
.
I
"So far the.MoZten S a l t Reactor Experiment has operated successfully and has earned a reputation for r e l i u b i t i t y . ' I USAEC &'airman Glenn T. Seaborg a t the ceremony marking the f i r s t operation of a reactor fueled w i t h 233V, October 8 , 1968. The IdSFU3 ran long and it ran w e l l .
It ran long enough with 235U I
.
f u e l t o a t t a i n t h e o r i g i n a l goals of t h e experiment, then continued through more than a year of added experimenis with 233U f'uel.
These
statements a r e supported by t h e s t a t i s t i c s on t h e o v e r a l l plant performance presented i n t h i s chapter.
Later chapters describe the performance
of individual components and systems which made up t h e plant. ..
4.1
Cumulative S t a t i s t i c s
. t h e MSRE ran are presented Several d i f f e r e n t indications of how .. long i n Table 4-1. The time c r i t i c a l i s a commonly used index f o r r e a c t o r ex.I
periments.
Another b a s i s of comparison with other reactors i s t h e number of equivalent full-power hours, Integrated power (Mw-hrs) is closely related.
(The figures quoted h e r e are based on a Full power of 8.0 Mw,
which i s t h e value i n d i c a t e d by heat balances.20 Analyses of i s o t o p i c ., changes14 i n d i c a t e t h a t f u l l power w a s 7.34 Mw, i n which case t h e tabul a t e d Mw-hrs should be
l i e d by 0.92.)
I
, w e amounts of time t h a t
salt c i r c u l a t e d i n t h e loops
of i n t e r e s t from t h e standpoint of t h e
demonstration of m a t e r i
i b i l i t y . ' Each pump operated a length of
t i m e equal t o t h e salt
n plus helium c i r c u l a t i o n i n that loop. of various kinds on d i f f e r e n t p a r t s of
.'
The number of t h e r
t h e salt system are l i s t e d i n Table 4-2.
The numbers a r e small i n com-
parison with t h e permissible numbers of cycles except i n t h e case of t h e
,
i
16
Table
I
4-1.
W
Accumulated Operating Data
I
i
' U Operation
i
3U Operation
Total
Time C r i t i c a l ( h r s )
11,515
6,140
17,655
Integrated Power (Mw-hrs)&
72 ,441
33,296
105,737
Equivalent Nl-Power Hours
9,005
4,167
13,172
15,042 16,906
6,746 9,170
21,788 26,026
4,046 3,172
3 ,384 1,535
7,430 4,707
20,789 17,444
10,059 9,994
30,848
37
14 6
4
!
S a l t Circulating Time ( h r s )
Fuel Loop
Coolant Loop i
H e l i u m Circulating Time (hrs )
,
Fuel Loop
1
Coolant Loop
1
Time Above 900'F
(hrs)
Fuel Loop
Coolant 'Loop
27,438
F i l l and Drain Cycles Fuel Loop Coolant Loop
13
.
, !
$ased
on heat balances w h i c h i n d i c a t e d full power w a s 8.0 Mw.
51 19
(c3
c Table
4.2
MBRE Cumulative
Cycle History
Coolwlt PlmlR 12 ; 19 971 I 1 101 ! Freeze Flames 100. 101. 102' 13 ' 51 I 18 ;QTi Freeze Flames 200. 201 12 : ., 18 ; 97' Penetrations 200. 201 12 Freeze Vdilve 101 13 Freeze Valve 104 '21 Freeze Valve 105 21 23 Freeze Valve ’ 106 15 Freeze Valve 107 Freeze Valve 108 16 15 Freeze Valve 109 Freeze Valve 110 8 Freeze Valve 111 6 2 ,FYeeee Valve 112 Freeze Valve 204 12 peeze Valve 206 12 1 4 hrs FD-1 Cooler 7. pp-4hr :_ * These figures are based on the original calculations, If 'they were cycle tests, the usage factors would be 23.04s and 13.37s.
: ,
156
L 1 ! 1 I !,
: 1f 1 1
* 57 . 50" 29 12 20 14 14 li 21 4 4
i
1
1
15 13
62 14 57 44 22 28 30
10 6 2 42 41
based on freeze
flangd
thermal
18 freeze flanges.
Here t h e f u e l flanges experienced 99% of t h e permissible
number of cycles based on early calculations o r 23% of the number permissi-
tad
b l e on t h e b a s i s of extended tests of a prototype flange. In order t o put t h e MSRF: record i n proper perspective, it i s necess a r y t o measure it against those
development.
of other reactors i n a similar stage of
A comparison on t h e b a s i s of equivalent full-power hours is
made i n Table 4-3.
It is not invidious t o s a y t h a t t h e MSRE compares w e l l
with other reactors having t h e same purpose; t h a t i s , t o demonstrate t h e p r a c t i c a l i t y of a reactor concept. 4.2
Availability during Various Periods
The b e s t index of the r e l i a b i l i t y of a plant should be t h e f r a c t i o n of t i m e (over some extended period) that it i s available f o r i t s intended service.
Perhaps inevitably, however, i n experimental plants w h e r e t h e
objectives include more than simply generating power, t h e d e f i n i t i o n of a v a i l a b i l i t y i s not always so clearcut.
Therefore we have presented in.
Table 4-4 an index which, i n p r i n c i p l e , i s ,less s i g n i f i c a n t b u t whose d e f i n i t i o n is q u i t e clear; namely, t h e time t h a t t h e r e a c t o r w a s actually
u
c r i t i c a l i n each 3-month period during t h e 4 years of parer operation. Percentages of elapsed t i m e are shuwn f o r s e l e c t e d i n t e r v a l s , but these must not be regarded as a measure of r e l i a b i l i t y since t h e r e a c t o r w a s s u b c r i t i c a l much of t h e t i m e because the planned program ,required it. (During shutdowns f o r core specimen removal o r t h e s u b s t i t u t i o n of 2 3 3 ~ f, o r
exmple. )
I n t h e MSRE t e s t program'6 it w a s planned t h a t t h e r e be a period of operation f o r t h e primary purpose of demonstrating plant r e l i a b i l i t y . This phase of t h e program covered t h e last 1 5 months of operation w i t h
235U. Table 4-5 gives a breakdown of t h e time during t h i s period. The reactor w a s c r i t i c a l 80% of the time and t h e a v a i l a b i l i t y of t h e p l a n t , as defined i n t h i s t a b l e , w a s 86%. This amply j u s t i f i e d D r . Seaborg's statement quoted at t h e head of t h i s chapter. Another indication of r e l i a b i l i t y is haw long a plant can be kept continuously on l i n e .
In t h e f i n a l run with 235U, t h e f u e l salt w a s i n
t h e loop continuously f o r j u s t over 6 months, before it w a s drained f o r
u
19
Table 4.3
Equivalent N l - P o w e r Hornsa
'
Produced by Early U. S. Reactors of Several Types
me Aqueous Homogeneous Organic-Cooled
Reactor
First Critical
m-2
12/57
OMRE Piqua
9/57
LMFBR
PWR BWR MSR
4/63 1/66
Interval (years)
EFPH
3,100 ,-
5.6
5,934 5,642
2.6
2/64
6.8
8/62
9/64
2.1
2,661
Peach Bottom
3/66
b
3.8
10 ,836b
8/51
SRE
4/57
'
3.4
Hallam
Sodium-Gr aph it e
I-FI'GR
6/63
Operation Terminated
8,140
EBR-1 EBR-2 Fermi
12/62 b b
11.3
11/63 8/63
Shippingport (Core #I)
12/57
2/64
6.2
27,781'
10.5
6.1 6.3
EBWR
12/56
6/67
VBWR
8/57
12/63
6.3
11,164 11,814
MSRE
6/65
12/69
-4.5
13,172
.
a
Calculated from thermal Mw-hrs and i n s t a l l e d capacities reported i n USAEC-DRIYT booklet "Operating History of' U. S. Nuclear Power Reactors d a t a from April 1964 Nucleonics. 1969" except Shippingport Co
-
EFPH quoted i s through 1969.
boperation i s not
ed.
C
two cores) through 1969 i s 43,400 EFPH.
Total f o r Shippingport
20
Table 4-4
Time C r i i c a l Each Quarter dur,.ig t h e
Four Years o f Power Operation (1966
Year
Quarter 1 2
1966
3 4
1
62 1070 413 1221
57%
(235u)
73.9% (1967 1
2045
2
0
3
0
.4
735 1800
1969
C r i t i c a l Time Elapsed Time
1852 1186 1292 2144
1967
1968
Hours Critical
- 1969)
1375 1054 1176
(d i
I
21
-
Table 4-5
Breakdown of Time during Spstained Operation Phase-
O f MSRE Program (December
1 4 , 1966
Activity o r Condition
Available
i
Maintenance
Critical Changing specimens Annual tests
Time
26 d 5 d
Sampler l a t c h (8/67)
35 d 3 d 3 d
V i e w i n r e a c t o r c e l l (6/67)
1.1
0.6
0.6
6 d 0 -9
Miscellaneous
Total elapsed
Per centage
8934 h r
13 d
Component cooling pump (9/67 )
r 1
26, 1968)
Air-line disconnects (1/67)
Sampler wiring (12/67)
Other
- March
468 d
100.0
2.2
22
t h e planned processing.
During these 188 days the r e a c t o r w a s counted as
hd
unavailable only 63 h (1.4% of t h e time) while r e p a i r s were being made t o t h e fuel-sampler drive.
It w a s a c t u a l l y c r i t i c a l f o r 97.8% o f t h e time.
The f i n a l phase of t h e MSRE operation, w i t h 233U f u e l , was not aimed
primarily at demonstrating r e l i a b i l i t y , 2 1 but it turned out t h a t t h e availab i l i t y of t h e p l a n t w a s s t i l l remarkably high.
This i s shown i n Table
4-6.
Here available t i m e includes ( i n addition t o t h e c r i t i c a l time) s u b c r i t i c a l i n t e r v a l s during the zero-power experiments, time spent i n charging the pump power supply during variable-speed t e s t s , changing t h e core specimens, gamma-scanning t h e drained salt systems, and experiments on behavior of gas i n f l u s h salt.
4.3
Interruptions of Operations
Figure 4-1 provides a broad v i e w of t h e MSRE operation and m a d o r i n t e r ruptions.
This figure w a s prepared f o r inclusion (together w i t h similar
graphs from many other r e a c t o r s ) i n t h e USAEC's annual presentation t o t h e JCAE and, according t o t h e rules, assigns a b r i e f , descriptive reason f o r each shutdown of 5 days o r more.
A great deal more d e t a i l as t o t h e nature
and cause of these and s h o r t e r i n t e r r u p t i o n s i n operation i s given i n t h e
4 figures and
11 t a b l e s w h i c h follow.
Figure 4-2 covers 1966, the f i r s t y e a r of power operation.
It shows
when salt w a s i n the f u e l loop, when t h e r e a c t o r w a s at power, and assigns
a nwtiber t o each i n t e r r u p t i o n i n e i t h e r .
A number i n a c i r c l e means t h e
i n t e r r u p t i o n w a s due t o some experiment.
A number i n a square means t h a t
t h e i n t e r r u p t i o n w a s necessary because of trouble w i t h some system o r com-
ponent.
Figures 4-3, 4-4, and 4-5 display t h e same kind of information
f o r 1967, 1968, and 1969 respectively. I n Tables 4-7 through 4-10, t h e i n t e r r u p t i o n s i n operation during each year are categorized as t o t h e type of i n t e r r u p t i o n , t h e cause, and other work t h a t w a s accomplished during t h e interruption.
Each interrup-
t i o n l i s t e d i n these four tables i s described more fully i n Tables 4-11 through
4-14.
kd
23
Table 4-6
Time C r i t i c a l and Time Available During 233U Phase
O f MSRE Program (September 10,
Elapsed
Period
.
Zero-power expts (9/10/68 - 11/28/68) 1895 Interim (11/28/68 - 1/12/69) 1092 F i r s t Power R u n s (1/12/69
1968
- 6/1/69)
3355 1642
- December 1 2 ,
Hours Crit.
Avail.
1969)
Percent age C r i t . Avail.
649 86
1852
34.2
97.7
110
7.9
10.1
3175
3268 408
94.6
97.4 24.8
- 8/9/69) 0 0 Later Parer R u n s (8/10/69 - 12/12/69) 2974 2230 2971 75.0 99.9 - - Overall (9/10/68 - 12/12/69) 10959 6140 8609 56.0 78.6 S h u t d m (6/1/69)
t
r
. .
..
..
...
...
. ..
..,
..
.
.
.
_ _ . . .... ..
...._.____......I.
.
.
.
.
~
.... .. .
_..._..._.._..-.I.
..
.
.. .
.,
.
.
i .
._... ...... ~
120
110
100
MAJOR SHUTDOWN PERIODS 1. Lowpowertosting. 2. Remove plugged capillary, check valve and filter in offgas system. 3. Identify plugging material md install larger valves and newly designed filter in offplr system. 4. Repair elactrial wiring short in fuel sample enricher. 6. Leak-ten m t w cell. 6. Removeand replace coMample array; replace ona
MSRE Initial Criticality
6/1/85
Critical on 2 3 3 ~
10/2/88
Installedcapacity Reactor Thermal Mw
8
/
Power Generation 1-
Thermal Mwh
33,283 I
blower. 7. Clear BM line plugcsursd by salt ovarfill; replace second blown. 8. Ctar salt plug from gas line; repair air nhn in
TOTAL TMwh
,L/
105,737
reactor cell.
9. Replam leaking alr-line d i m n e c n in reactor cell. 10. Remove and replace eocbumple army; annual tom of containmentand controls. 11. Replace fuel sampler drivo; rnriwe sample latch. 12. Remove and replace corr-umple array. Annual tests. Shake down q g a r i n g plant. Psalt torem0v.U. Load U. 13. Mix salt. Rsprir component cooling blown. 14. Repair fud sampler drive. Sewice control-rod drive. 15. Remornand replace cornample array. Replace control rod and d r h . Clar offgm lina. Annual
........
12 1 1 1 . 1 . .
133
'
taa. 18. Meaure fission product deposition.
17. Conclude nuclear operations 12/12/89.
I I
20
yj
0
10
1
..A. I 1
0
.J
2 3 ,
1965
Fig. 4 . 1
c'
Integrated Power and Major Shutdowns of the MSRE
c
c
,-
tl3MOd
do01 1 3 n d NI 1 l V S
8
ID
4
I
E
F 0
rt
' d .
c
w
.b
0
m
0
<
2 0
Ga
C
D
c C
z
c C
z D
W
n m
z
D
c
FUEL
0
FLUSH
r[
"1
SALT IN FUEL LOOP Loop
Q,
POWER (MW)
P
POWER
3
m
r
'p
z3
zW
Dm
'v, 3
-I
2
...
w
tD
VI
I
4 0
I
r
3 2
0
I
d
I
0, u)
0 yc
D I A
a
z
0
w
z K
I
0.
a
3 I
v) v)
3 LL
P
W
(MW) ki3MOd
tf3MOd
HSfllJ
d
do01 13nd NI I l V S
0
n
w
0
W
I
> 0 z
m w
LL
U
,
ID
0
rn
0 rn 0
z 0 <
8-I
0
0
-.
0
I
FUEL
-@
/
U
v)
Z
...
c)
9
r I
z
=xI
0
.I__
-
--
29 l b b b 4-7.
IWlWWFTIOm W OPBRLTIOR DURIRG 1W
C ~ U Mand Related Actlvlties
i!
i!
8
I I -
-
lo.
rite
1
r5-11%
2 3 4 5 6 7
1/21
A*
1/22
Y A*
I
-
I
I
-
IS 66 -
1
3-
B -
f:
P
P Y
P
9
1/23 1/23 1/23 1/24 1/25 kyi/:
P-Y
H
lo
2/16
Y
Y
11 l2
416 4/12 4/14 4/16 4/20 4/20 4/21 4/22 4/25 4/25 4/28 4/28 4/29 4129 51IJ
P-Y
13 14 15 l6
17
M 19 2O 21 22
23 24 25 26
bj
?.
B
8
a
n 28 29 9 31 32
33 34 35
36
37 38 39
ho bl
b2 L3
P P U
A'
P P
A Y*
Y
P
Y
Y
P P
A* Y
P Y
P
A*
P P-H
A*
A
P-Y
A
P
A'
A* A*
P-1
A
A*
A
Y
Y
U PH
51l2 5/16
A*
A
5/19 5/23 5/25 5/26 5/26 518 5126 5/28 6/14 6/14 6119 6/26 6/n 7/14 7/15 7/17
A*
A
P
A*
P
P P Yff
P-H
Y
Y P-H
A*
A*
A A
A*
A
A*
A
P
Y
om
15
0/10
h6
0112
A*
b7
A*
ba
0/16 0131
19
a/ P
YJ
1f.U 1/12
P P A
P-Y
P-Y
1/15 1/17
!2
i3 54 55 56
P-Y
Y
1/
2/23 2/24
A
'1 2/25
-.
P
8
P
bb
51
P
P P -
P -
*%portable Unscbedulsd Rod &m
A X
P 8 H
--
-
Autcoultic Yanrvll primary &Cotrdary
Worked on huing period Follovlng ihe Interruption; Includes Preventiw Walntenancs
P-H
30
-I
Utut H a p p e d
L
k
w
a
I
z
E
w
1
1
2
1
1
1
Period Contaimlent Testa
Samplers
l m e z e Valves
IControl Rods
1nstrumantati01
Air System
Instrument
Electrical and Hcaters
TVA Power outage
Contalmnt & Ventilation
Ccmponent Coolant System
oiigas System 1
Fuel & Coolant S a l t system
~ u m a nError
Check L i s t s
Unplanned, but due to lkpt.
Coolant Drain
Fuel Drain
LcaB &ran
R o d Scram
Radiator or Blowers
X
x x
1
s x s s
i
1
r
= x
> >
5
32
5imt % p p n d
-
2
-1
i
I -i -
-
?! 88
a: -
I
m
F
II
F
Y
P P
Y
Y
P P
Y Y Y
P
Y Y
P P
Y
P
Y
P
P
8
P
8
PY P
Y Y Y
Y
P
Y
P
Y Y
P P P
Y
8 8
P ll
P P P
Y Y
8
8 8
ll
P
Y II
P P
8 8 8
K
P
8
P P
8 8
1
P
I 1 1
I 1
I
I
Y
, 1 _1
33
bd
Table 4.11
Description of Interruptions of Operation .
bi
No.
Date
1
1/54/20
Low-power t e s t s t o c a l i b r a t e nuclear instruments, check r a d i a t i o n l e v e l s , c o l l e c t data f o r CS and noise analysis, obtain r e a c t i v i t y b a l and check out t h e control c i r c u i t s .
2'
1/21
Rod scram due t o accidental shorting of t h e 32-v dc power supply while investigating a No. 3 s a f e t y channel t r i p .
3
1/22
Rod and load scram t o t e s t c i r c u i t r y .
4+
1/23
Rod scram due t o a voltage sag.
5
1/23
Power reduction t o attempt t o blow out plugs A offgas l i n e , equalizer l i n e s , and main charcoal' beds.
'6
1/23
7
1/24
Load scram due t o f a l s e low coolant flow indication.
8+
1/25
Rod scram due t o inadvertent actuation of t h e manual s crgm- switch.
Description and Related A c t i v i t i e s
I
I
Rod scram due t o instrument malfunction when switching l i n e a r channel ranges.
Fuel and coolant drain due t o plugging i n t h e offgas system and t o i n s u l a t e t h e coolant flow t r a n s m i t t e r l i n e s . While shut down, t h e r e s t r i c t o r i n t h e equal i z e r l i n e (521) and t h e check valve i n t h e vent l i n e (533) t o t h e a u x i l i a r y charcoal bed were removed. The f i l t e r i n t h e main offgas l i n e w a s replaced a i N t h e pressure control'valve (PCV-522) w a s replaced w i t h a hand valve w i t h l a r g e r p o r t s . Work was done on No. 3 control rod d r i v e , component coolant pump No. 1, r a d i a t o r door seals, and motor generator No. 4. The thermal s h i e l d air lock problem w a s investigated.
9
2/16
10,
u
During 1966
n
Fuel drain due t o a f a i l u r e o f t h e drain tank c e l l space cooler motor. While shut down, t h e i n l e t valves t o t h e charcoal beds were replaced and t h e offgas l i n e at t h e f i e 1 pump w a s reamed out. Rep a i r s were made on a damaged e l e c t r i c a l penetration on t h e sampler-enricher and a le& on r e a c t o r c e l l
. Reportable unscheduled rod scram.
34
4 .ll (continued)
Table
No.
Date
10 ( c o n ' t )
Description and Related A c t i v i t i e s space cooler No. 2 w a s fixed. A s t a t i c i n v e r t e r w a s i n s t a l l e d t o replace motor generator No. 4. The t r e a t e d water corrosion i n h i b i t o r was changed from potassium t o lithium n i t r i t e - t e t r a b o r a t e .
11
416
Fuel drain due t o a plug at t h e a u x i l i a r y charcoal bed which turned out t o be a poppet from a drain tank vent check valve.
12
4/12
Rod scram and load scram t o t e s t performance of t h e rods and r a d i a t o r doors.
13
4/14
Power reduction t o c o l l e c t data f o r noise analysis.
14% 4/16
Rod scram due t o t h e f u e l pump stopping which w a s caused by a f a l s e low salt l e v e l indication.
15
4/20
Manual rod scram and load scram t o end t h e 2.5-Mw operation.
16*
4/20
Rod scram due t o a f a l s e s i g n a l on s a f e t y channel No. 1 while s a f e t y channel No. 2 w a s being t e s t e d .
17
4/21
Power reduction due t o charcoal beds plugging.
18% 4/22
Rod scram, load scram, f u e l drain and coolant drain i n i t i a t e d by a f a u l t y pump pressure r e l a y which t r i p p e d instrument power breakers. Some salt froze i n r a d i a t o r because t h e coolant pump could not be s t a r t e d due t o a l u b r i c a t i n g o i l flow switch failure ( FS-754).
19*
4/25
Rod scram due t o i n v e r t e r f a i l u r e .
20*
4/25
Rod scram and load scram due t o f a l s e low fuel-pump l e v e l i n d i c a t i o n while doing periodic instrument check l i s t s .
21
4/28
Rod scram and load scram due t o TVA power outage caused by a storm.
22%
4/28
Rod scram and load scram due t o spurious s a f e t y chann e l t r i p s possibly caused by an e l e c t r i c a l storm.
23
4/29
Manual rod scram and load scram t o check e f f e c t of cold doors on r a d i a t o r temperatures. Shutdown w a s necessary t o r e p a i r t h e l a t c h on t h e samplerenricher.
*
*
Reportable unscheduled rod scram.
.
,
35
bd
Table 4 .ll (continued) No.
Date
24
4/29
Description and Related A c t i v i t i e s
Fuel drain necessitated by a f a i l u r e of t h e samplerenricher cable drive motor. While shut down, t h e b e l t s were tightened on component cooling pump No. 1 and attempts were made t o blow out t h e plugs i n the offgas l i n e s .
25
5/11
Power reduction due t o offgas l i n e f i l t e r plugging.
26*
5112
Rod scram and load scram due t o operator f a i l i n g t o ins e r t a control c i r c u i t jumper w h i l e doing periodical instrument check l i s t s .
27
5/16
Power reduction t o increase p i t c h on main bZower ,blades t o increase possible maximum power.
28*
5/19
Rod scram and load scram due t o a TVA power outage. While shut down, measured offgas pressure drop and attempted t o blow out plugs.
29*
5/23
Rod scram caused by instrument malfunction and operator e r r o r . Operator had switched t o servo with a 40째F difference between a c t u a l temperature and temperature demand setpoint.
30
5/25
Power reduction t o check t h e power c o e f f i c i e n t of reactivity.
31
5/26
Power reduction t o obtain r e a c t i v i t y balance data.
32
5 126
Power reduction due t o main blower No. 1 stopping and f a i l i n g t o start.
33
5/26
Power reduction t o c o l l e c t data f o r anoise analysis.
34%
5/26
35
5/28
ai
?
%
scram and load scram t o i n v e s t i g a t e a false f i r e alarm. Fuel drain due t o a f a i l u r e of component coolant pump No. l w h i l e component coolant pump No. 2 w a s valved o f f . A planned drain had been scheduled due t o t h e high indicated c e l l leak r a t e . This w a s found t o be leak from t h e thermocouple headers i n t o t h e c e l l . While shut down, a 20-psig leak t e s t w a s made
Reportable unscheduled rod scram.
36 Table 4.11 ( continued)
No.
Date
35 (can't)
Description and Related A c t i v i t i e s on t h e c e l l s , repairs were made on both component coolant pumps, t h e e l e c t r i c a l load w a s r e d i s t r i b u t e d and attempts were made t o locate t h e water leak i n t o the cells.
36
6/14
Power reduction due t o a f a i l u r e i n t h e e l e c t r i c a l supply t o t h e nuclear instruments.
37
6/14
Power reduction due t o failure of a f l e x i b l e coupling on main blower No. 1.
38
6/19
Power reduction t o empty t h e overflow tank.
39*
6/20
Rod scram and load scram due t o t h e operator inadvert e n t l y i n s e r t i n g a t e s t source at t h e wrong process radiation detector (RE-528 instead of RE-565) while doing periodic instrument check l i s t s .
40*
6/27
Rod scram, load scram, f u e l drain, and coolant drain due t o t h e e l e c t r i c a l supply t o component coolant pump No. 1 shorting out at t h e penetration.
4 1 '
7/14
Rod scram and load scram due t o a TVA power outage.
42*
7/15
Rod scram and load scram due t o a TVA power outage caused by a storm.
43
7/17
Power reduction and coolant drain due t o a catastrophic failure of main blower No. 1 hub. The f u e l s a l t w a s drained on 7/24/66. While f i l l i n g t h e f u e l system w i t h flush s a l t , t h e fuel-pump bowl w a s accidentally o v e r f i l l e d and some salt froze i n t h e sampler and offgas l i n e s . Heat had t o be applied remotely t o remove t h e plugs. While shut down, t h e core specimens were replaced, and a new p a r t i c l e t r a p w a s ins t a l l e d i n t h e offgas l i n e . Leaks i n reactor c e l l space cooler No. 1 and i n t h e t r e a t e d water heat exchanger were repaired and t h e thermal s h i e l d degass i n g tank w a s i n s t a l l e d . Both component coolant pumps were repaired and piping w a s i n s t a l l e d t o drain t h e condensedwdter from them. Repairs were made on t h e r a d i a t o r door s e a l s , r a d i a t o r heaters, and t h e main blower motors. A l l t h e e l e c t r i c a l switch gear breakers were calibrated. P r i o r t o s t a r t q p , the f u e l and coolant pumps lube o i l w a s chang&d, t h e periodic check l i s t s were completed and t h e c e l l s were leak-tested a t 10 psig.
IC
Reportable unscheduled rod scram.
gs
37
W
4.11 (continued)
Table
Date
No.
L,
Description and Related A c t i v i t i e s
44
lO/10
Power reduction d u e - t o a false s i g n a l from a process r a d i a t i o n detector (RE-528).
45
10/10
Power reduction due t o an experiment when t h e fuel-pump pressure was rapidly vented from 1 5 t o 3 psig.
46*
10/12
Rod scram due t o 'inverter trouble.
47*
10/16
Rod scram and load scram due t o f a l s e s i g n a l from a process r a d i a t i o n detector ( RE-528).
48
10/31.
Power reduction t o i n s t a l l a new r o t o r on main blower No. 1.
49
10/31
Fuel drain due t o upper offgas l i n e (524) being plugged. While shut down, t h e flow r e s t r i c t o r i n t h i s l i n e w a s replaced and heat w a s applied t o main offgas l i n e at t h e f u e l pump t o unplug it.
50
11/11
Power reduction t o check main blower vibration.
51
11/12
Power'reduction t o empty t h e overflow tank.
52
11/15
Power reduction t o empty t h e overflow tank.
53
11/17
Power reduction t o attempt t o blow t h e plug out of t h e main offgas l i n e at t h e f u e l pump.
54
11/20
Fuel drain due t o a high indicated c e l l leak rate which proved t o be due t o leaking a i r l i n e disconllects on i n - c e l l valve operators. Rotameters were i n s t a l l e d on these air l i n e s . While shut dawn, t h e plug i n t h e offgas l i n e at t h e f u e l pump w a s reamed out and r e p a i r s were made nn t h e component coolant pumps and t h e component coolant system pressure control valve (pdcv-960). Heaters were i n s t a l l e d on t h e i n l e t s of main charcoal beds IA and B. The c e l l s were leakt e s t e d at 10 psig.
55'
12/23.
Rod scram and load scram.by accidentally opening a c i r c u i t while i n s t a l l i n g jumpers f o r an experiment.
*
Reportable unscheduled rod scram.
38
V
Table 4.11 (continued) Date
No.
Description and Related A c t i v i t i e s
56
12/24
Parer reduction caused by a pressure r e l e a s e experiment. The r e a c t i v i t y decreased, t h e regulating rod withdrew t o i t s upper l i m i t and t h e operator f a i l e d t o change the l i m i t .
57
12/26
Power reduction when venting t h e f u e l pump after an attempt t o blow out t h e plug i n t h e main offgas l i n e . The regulating rod reached i t s upper l i m i t and t h e operator f a i l e d t o change t h e l i m i t .
8
,
Reportable unscheduled rod scram.
39
W
Table 4.12
Description of Interruptions of Operation
.
I
During 1967
No.
w
Date
1c
1/12
2
1/14
3
1/16
4
1/28
5
1/30-2/2'
6
2/5
7
2/26
Description and Related A c t i v i t i e s d load scram due t o a freeze valve (FV-108) r e l a y t r i p p i n g t h e r e a c t o r out of RUN during an investigation of t h e p l a s t i c keepers on t h e relays. Power reduction t o observe xenon decay rake and obt a i n zero-power r e a c t i v i t y balance data. due t o leakage from t h e i n - c e l l a i r l i n e disconnects. The measured leakage w a s s o high t h a t reasonable c e l l leak rate calculations were not possible. The disconnects were replaced. While shut down, t h e main offgas f i l t e r and control. valve (PCV-522) were removed and a new f i l t e r w a s i n s t a l l e d . The t r e a t e d water heat exchanger w a s replaced and maintenance w a s done on t h e component coolant pumps. The c e l l s were leak-tested at 10 psig. Power reduction t o t e s t automatic load control. ion t o take pictures of t h e r a d i a t o r and i n v e s t i g a t e he t r a n s f e r of the r a d i a t o r and heat ade dynamic tests and checked t h e failure of a freeze valve pped the r e a c t o r out of on due t o main blower No. 1 vibration. t o obtain heat-transfer data. t o coolant system offgas f i l t e r due t o e l e c t r i c a l power supply t r o u b l e e t y and rod servo.
3/7
ll
t o a bad bearing on main blower at low p y e r t o observe e f f e c t of ' xenon decay on r e a c t i v i t y balance.
*
Report able unscheduled rod scram.
i
40
Table 4.12 ( c o n ' t ) ~~
No.
Date
Description and Related A c t i v i t i e s
Y
3/13
Rod scram and load scram due t o f a l s e s i g n a l on s a f e t y channel 3 while doing periodic instrument, check l i s t s .
13
3/19
Rod reverse and load scram due t o false low fuel-pump speed indication while doing periodic instrument check lists.
14
3/23
Power reduction due t o vibration on main blower No. 3.
15
4/11
Power reduction t o obtain a s p e c i a l fuel-pump gas sample. Also obtained heat-transfer data.
16
4/28-4/30
Power reduction t o c6Xlect data f o r noise analysis.
17
515
Power reduction t o t e s t r e a c t o r response t o load changes without rod changes.
18
518
Rod scram, l o i d scram, f u e l drain and coolant drain t o remove reacto'r core specimens. While shut down, gamma-scan data were taken, a new source w a s ins t a l l e d , and heaters were i n s t a l l e d on t h e i n l e t s t o main charcoal beds 2A and 2B. Reactor c e l l space cooler No. 2 w a s replaced and r e p a i r s were made on control rod No. 2, r a d i a t o r door brakes, component coolant pumps and t h e main blowers. A boot w a s replaced on t h e sampler-enricher. Prior t o startup, t h e periodic check l i s t s were completed, t h e c e l l s w e r e leak-tested at 20 psig.
19
6/21
Power reduction t o obtain heat-transfer data.
20
6/23
Power reduction t o r e p a i r an o i l pressure switch on 3 component coolant pump No. 1.
21
6/25
Rod scram and load scram due t o a TVA power outage caused by a storm. While shut down, r e p a i r s were made on component coolant pump No. 1 o i l pressure switch, a t r e a t e d water rupture d i s c (844) w a s replaced, and a heavier base w a s i n s t a l l e d on main blower No. 3.
22"
6/30
Rod scram and load scram due t o an operator r e s e t t i n g the wrong channel while doing periodic instrument check l i s t s .
*
)t
Reportable unscheduled rod scram.
dj
41
hd
Table 4.12 ( c o n ' t )
No.
Date
Description and Related A c t i v i t i e s
Y
23"
b
7/12
Rod scram and load scram due t o a TVA power outage caused by a storm.
24
Power reduction t o obtain heat-transfer data.
25
Power reduction, f u e l drain and coolant drain due t o t h e sampler-enricher drive cable being severed. During shutdown, t h e l a t c h was r e t r i e v e d but t h e capsule remained i n the f u e l pump. P r i o r t o s t a r t u y , maintenance w a s done on component coolant pump No. 1.
26
Coolant drain due t o t h e r a d i a t o r door binding i n t h e guides.
27
Power reduction due t o an o i l leak i n component coolant pump No. 2.
28
Fuel d r a i n due t o a leak through the discharge valve of component coolant pump No. 2 which prevented repair of t h e o i l leak.
29
10/1
Power reduction t o c o l l e c t data f o r noise analysis.
30*
10120
Rod scram and load scram due t o a power f a i l u r e due t o an a r c between a switch actuator rod and t h e main power l i n e t o t h e building.
31
10/2 3
Power reduction t o observe t h e e f f e c t of xenon decay on t h e reactivity.balance.
32
11/14
Power reduction t o c o l l e c t data f o r noise analysis.
11/16
Power reduction t o r e p a i r sampler-enricher wiring.
34%
11/22
Rod scram while maintenance w a s being performed on t h e t rument at b n .
35
12/13
Power reduction t o observe t h e e f f e c t of xenon poisoni n g on t h e r e a c t i v i t y balance and t o c o l l e c t data f o r noise analysis.
33
,.
*
Reportable unschedule
42
Table 4.13
Description of Interruptions of Operation During 1968
No.
Date
Description and Related A c t i v i t i e s
1
1/22
Power reduction t o observe t h e e f f e c t of xenon poisoning on t h e r e a c t i v i t y balance and t o c o l l e c t data f o r noise analysis.
2
2/21
Power reduction t o t e s t t h e rod servo instrumentation under simulated 3U conditions.
'3
2/29
Power reduction t o observe t h e e f f e c t of xenon poisoning on t h e r e a c t i v i t y balance and t o c o l l e c t data f o r noise analysis.
4
317
5
3/22
Power reduction t o make dynamics t e s t s .
6
3/25
Manual rod scram and load scram t o observe xenon poisoning e f f e c t on r e a c t i v i t y balance.
7
3/25
Manual rod scram t o end 235U operatfon.
8
3/26
Fuel drain t o end 235U operation. Shutdown w a s emminent due t o a t a n g l e d drive cable i n t h e samplerenricher. A f t e r t h e .drain, t h e capsule w a s accidentally dropped i n t o t h e f u e l pump when t r y i n g t o untangle t h e cable. Unsuccessful attempts were made t o recover t h i s . While shut down, gamma-scan data w a s taken, the core specimens were removed, and t h e main offgas l i n e at t h e f u e l pump w a s unplugged. The flush and f u e l salts were t r a n s f e r r e d t o t h e f u e l processing p l a n t where t h e 235Uw a s removed. 233U w a s added t o t h e f u e l salt. Repairs were made on two heat exchanger h e a t e r s , control rod No. 2, both component coolant pumps, t h e i n v e r t e r , and t h e r a d i a t o r doors. Main blowers No. 1 and No. 3 bearings were replaced and t h e coolant offgas f i l t e r w a s replaced. P r i o r t o s t a r t u p , t h e periodic check l i s t s were completed and t h e c e l l s were l e a k - t e s t e d ' at 20 psig.
9
9/14
Fuel drain f o r adding more 233U as p a r t of t h e c r i t i c a l i t y experiment.
.
Power reduction t o observe t h e e f f e c t of xenon poisoni n g on t h e r e a c t i v i t y balance, t o make dynamics tests and take heat-transfer data.
43
&i
Table 4.13 ( c o n ' t ) Date
No.
br
Description and Related A c t i v i t i e s
10
9/17
Fuel drain f o r adding more 233U as p a r t of t h e c r i t i c a l i t y experiment. .
11
9/21
Fuel d r a i n - f o r adding more 233U as p a r t of t h e c r i t i c a l i t y experiment.
12
10121
Fuel drain due t o i n v e r t e r failure. Operator f a i l e d t o r e s e t r e a c t o r c e l l r a d i a t i o n monitor i n time t o prevent d r a i n .
13*
11/27
Rod scram and load scram due t o operator f a i l i n g t o ins e r t a jumper around f u e l pump l e v e l contacts. The fuel-pump pressure was being vented a f t e r attempting t o blow out a plug i n t h e main offgas l i n e .
14
11/28
Fuel d r a i n and coolant drain t o t h e loop and t h e drain tank. f u e l pump w a s reamed o u t , t h e was cleaned, and r e p a i r s were coolant pumps.
15
12/17
Fuel drain due t o d i f f i c u l t y with the l a t c h and cable drive of t h e sampler-enricher. While shut down, control rod drive No. 3 w a s inspected and a weight added t o decrease i t s drop time.
6
Reportable unscheduled rod scram.
mix t h e f u e l salt i n The offgas l i n e a t t h e coolant offgas piping made on t h e component
44 Table
No.
4 .I4 Description of Interruptions of Operations During 1969 .'
Date
Description and Related A c t i v i t i e s
1
1/15
Power reduction t o r e p a i r lower limit switches on r a d i a t o r doors.
2
1/23
Power reduction t o i n v e s t i g a t e e f f e c t of power on "blips".
3
1/26
Power reduction t o i n v e s t i g a t e e f f e c t of stopping t h e f u e l pump on "blips".
4
1/28
Power reduction t o i n v e s t i g a t e "blips".
5
217
Power reduction t o c o l l e c t d a t a f o r noise analysis.
6
2/n
Power reduction t o run t e s t s on n a t u r a l convection.
7
2/16
Power reduction t o run dynamics t e s t s .
8
2/23
Power reduction due t o failure of b e l t s on s t a c k fan No. 1.
9
2/27
Power reduction t o connect t h e f u e l pump t o a variable speed motor generator set.
10
314
Manual rod scram and load scram due t o failure of t h e variable speed motor generator set.
11
318
Power reduction t o - a l l o w xenon t o decay i n preparation t o making void f r a c t i o n t e s t s . While shut down, t h e coolant offgas f i l t e r w a s replaced.
12
3/20
Power reduction t o connect t h e f u e l pump t o t h e normal e l e c t r i c a l supply.
13
3/25
Power reduction t o connect t h e f u e l pump t o t h e variable frequency motor generator set.
14*
3/25
Manual rod scram and load scram due t o failure of t h e variable speed motor generator set.
15
4/3
Power reduction t o connect t h e f u e l pump t o t h e normal e l e c t r i c a l supply.
ic
ic . Reportable
u
unscheduled rod scram.
hd
45
u
Table 4-14 (continued) ~
No.
Date
~
~
~
~
_
_
_
Description and Related A c t i v i t i e s
16*
4/10
Rod scram, load scram, f u e l drain, and coolant d r a i n due t o a burned-out fuse on one phase of t h e main transformer. Drains could have been averted by operating a l t e r n a t e equipment.
17
4/12
Load scram due t o a disturbance i n t h e 48-v dc power supply '
18*
4/15
Rod scram, load scram, and f u e l drain due t o t h e drain freeze valve thawing. The setpoint on a temperature switch had d r i f t e d o f f t h e setpoint and an inadequate plug had been established.
19
514
Power reduction due t o a f a i l u r e of t h e coolant stack b e r y l l i ~monitor.
20
515
Power reduct'ion due t o an e r r o r i n calculating t h e r e a c t i v i t y balance.
21
5/13
Power reduction t o connect t h e f u e l pump t o t h e variable speed motor generator s e t .
22
5/19
Power reduction t o connect t h e f u e l pump t o t h e normal e l e c t r i c a l supply.
23
5/25
Power reduction due t o plug i n overflow tank vent l i n e 1:hich caused d i f f i c u l t y i n emptying t h e overflow
bd
I
tank.
24
5/26
Power reduction t o connect the fuel pump t o t h e variable speed motor generator s e t .
25*
5/26
Marpu
26
5/27
scram and load scram due t o f a i l u r e of t h e variable speed motor generator s e t .
scram and load scram due t o f a i l u r e of t h e variable speed motor generator s e t . scram, load scram, f u e l drain, and coolant drain t o remove core specimens and due t o plugging i n t h e offgas system. Control Rod No. 3 d i d not scram. It w a s replaced during t h e shutdown. Extensive gamma-scan data w a s taken. A h e a t e r w a s installea on t h e main offgas l i n e near t h e f u e l pump which aided i n removing t h e plug. The overflow tank
27
*Reportable
unscheduled rod scram.
_
46 Table 4.14 (con't) Date
NO.
27 (con't)
Description and Related A c t i v i t i e s offgas l i n e w a s replaced and t h e coolant offgas syst e m w a s unplugged. Repairs were made on t h e samplerenricher, and on a leaky in-cell air l i n e disconnect. Before s t a r t u p , t h e periodic check lists (modified) were completed and t h e c e l l s were leak-tested at 20 psig. N l power w a s delayed while t e s t s were made at various fuel-pump speeds t o i n v e s t i g a t e bubbles i n t h e loop.
28
8/20
Power reduction due t o f a i l u r e of t h e variable speed motor generator s e t .
29
8/20
Power reduction due t o f a i l u r e of t h e variable speed motor generator s e t .
30
8/20
Power reduction due t o f a i l u r e of t h e variable speed motor generator s e t .
31
8/20
Power reduction due t o f a i l u r e of t h e variable speed motor generator s e t .
32
8/20
Power reduction due t o failure of t h e variable speed motor generator set^.
33
8/21
Power reduction due t o f a i l u r e of t h e variable speed motor generator s e t .
34
8/21
Power reduction due t o f a i l k e of t h e variable speed motor generator set.
35
8/28
Manual rod scram and load scram due t o f a i l u r e of t h e variable speed motor generator s e t .
36'
8/29
Manual rod scram and load scram due t o f a i l u r e of t h e variable speed motor generator s e t .
37
8/29
Power reduction f o r xenon decay i n preparation f o r argon cover-gas experiments.
38
9/7
Power reduction due t o an e r r o r i n c a l c u l a t i n g t h e r e a c t i v i t y balance.
39
919
Power reduction t o c o l l e c t data f o r noise analysis, t o make gamma scans and run rod drop t e s t s .
40
9/16
Power reduction t o connect t h e f u e l pump t o t h e normal e l e c t r i c a l supply.
*
Reportable unscheduled rod scram.
LJ
47
Li
Table 4.14 ( c o n ' t ) No.
Date
Description and Related A c t i v i t i e s
41
9/18
Power reduction t o replace a bearing on Main Blower No. 1.
42
9/23
Power reduction t o replace a bearing on Main Blower No. 1. While shut down, connected t h e f u e l pump t o t h e variable speed motor generator s e t .
43
9/30
Load scram due t o d i r t y contacts on relays.
44
9/30
Load scram due t o d i r t y contacts on relays.
45
9/30
Load scram due t o d i r t y contacts on relays.
46
9/30
Load scram due t o di'rty contacts on relays.
47
9/30
Load scram due t o d i r t y contacts on relays.
48
10/1
Power reduction t o connect t h e f u e l pump t o t h e normal e l e c t r i c a l supply.
49
10/3
Power Yeduction due t o door being inadvertently l e f t open i n an exclusion area which caused inadequate containment v e n t i l a t i o n .
LJ
Power reduction t o take a s p e c i a l fuel-pump gas sample. Power reduction t o take s p e c i a l fuel-pump offgas samples and t o run t e s t s with t h e f u e l pump o f f .
51 52
10/23
Load scram due t o a f a l s e t r i p of a coolant s a l t flow relay while doing periodic instrument check l i s t s .
53
10/24
Load scram w h i l e t e s t i n g relays a f t e r they were repaired. Operator f a i l e d t o r e s e t . t h e relays.
54
11/2
Manual rod scram, l o a d scram, f u e l drain and coolant drain t o take gamma-scan data,
55
12/3
Power reduction t o take gamma-scan data.
56
12/12
Manual rod scram, load scram, f u e l drain, and coolant drain t o end operation of t h e reactor.
48 The next t h r e e t a b l e s summarize t h e foregoing m a s s of information on
W
Table 4-15 is a summary by type; Table 4-16, by cause and
interruptions.
work done during t h e i n t e r r u p t i o n .
Because unscheduled scrams of t h e con-
t r o l rods are of s p e c i a l i n t e r e s t , a separate breakdown of these i s given
i n Table 4-17.
It should be noted t h a t never were t h e rods scrammed because
of t h e r e a c t o r power, t h e r e a c t o r period, o r t h e f u e l temperature going out of limits.
4.4
Time Required f o r Operational Tasks
One aspect of t h e o p e r a b i l i t y of a plant (and one which a f f e c t s plant a v a i l a b i l i t y ) i s haw long it takes t o do various tasks required i n t h e operaThe purpose of t h i s s e c t i o n i s t o summarize t h e MSRE experience i n
tion.
t h i s regard. Manpower and t h e approximate number of working hours required f o r various t a s k s are l i s t e d i n Table 4-18.
One must recognize t h a t some of
these figures are subject t o considerable variation.
*
The time required t o
do a p a r t i c u l a r t a s k depended on chance d i f f i c u l t i e s t h a t were encountered,
LiJ
t h e experience of t h e crew, o t h e r jobs being done concurrently and t h e urgency of speedily c o ~ l e t i n gt h e t a s k , i . e . , whether o r not it w a s given a high p r i o r i t y .
Furthermore, sometimes extra requirements such as s p e c i a l
data-taking caused t h e job t o take longer than usual.
The figures i n Table
4-18 apply t o f a i r l y normal s i t u a t i o n s ; not t h e b e s t nor t h e worst.
Very
b r i e f descriptions of the t a s k s a r e given below, w i t h references t o t h e s e c t i o n of t h e MSRE Operating ProceduresP2 followed i n each case, i f applicable.
4.4.1
Auxiliary Systems S t a r t u p Check L i s t s During a shutdown, much of t h e equipment remained i n operation.
However,
a f t e r long shutdowns, t h e a u x i l i a r y systems s t a r t u p check lists (Section 4) were completed t o assure that nothing had been overlooked.
A l l valves and
switches were checked, equipment w a s put i n t o operation and standby equipment
w a s tested.
4A
These were done as l a t e i n t h e shutdown as p r a c t i c a l .
- Electrical
Startup Check L i s t assured t h a t all main breakers were
closed so t h a t equipment and heaters could be started from t h e control boards.
tr
49
Table 4-15
Summary of Interruptions by Type
Number of Interruptions Type of Interruption Power Reduction: Rod Scram: Load Scram:
Fuel Drain Coolant Drain
Total
1966
Automatic Manual
3 23
Aut omat ic Manual
20
Automatic Manual
13
6
1967” ’
1968
.
1969
Total
0
0 23
39
7 1
8
3 :.
90
7 3
12
11
6
1
14 3
3
35
36
Aut omat ic Manual
0
2
6
7
4
3
20
Aut omati c Manual
1 3
0
0
1
3
1
3
2 10
-
-
-
-
-
85
47
18
75
225
50
*
Table 4-16.
Summary of Interruptions by Cause and
Work Done During Interruption
P
sw
Number of InterruDtions- Total 1969 .1967 1968 P S W P S W P S W P sw
8
o o
14 0'
1966 Category Planned
~
o
12 0'
o
24
o o 58
0' 0 .
Unplanned
3
0 0
0 0 0
o
5,0
1 2 0 ' 0 0 0
0
Human Error
7
i o
2
2 3 0 1 1 6 3
.
1
1
1
0 0 0 1 3 0 0 1 6 0 0 ~
0
1 1 0
1
0
, 1 8 0
Fuel and Coolant S a l t Systems
a
0 0
0
0
0 0 0
0
Radiator o r Blowers
6
1 5
4 0 5
0 0 1
2 0 3 1 2 1 1 3
10
0*12
1
0
2
0
0
2
1 1 2 1 2 1 1 8
o
1
3
0
0
0
0
0
0
2
o o o
~
0
0
Offgas Bsytem
,
i
But Due t o an Ekperimen t Check L i s t s
Water System Component Coolant Syst e m
,
1
0
4
l
.
I
.
0
2
0 6
2
0
2
Containment and Ventilation
1
1
2
0
0
0
TVA Power Outage
I 3
1
0
2
0
0
I
01
0
0
0
1
0
1
I
E l e c t r i c a l and Heaters
4
0 . 4 .
Instrument A i r System
1
O
Instrumentation
11
0 0 0
0
1
1 1 7
4 0 1 0 ' 2
1
3
2
0
3
0 0 0 I
1
0 1
o o
1
0
0 0 0 ; o 0
1
1
g 30 0 19
1 , 2 1 3 1
I l
i
1
Control Rods
0
0 31 7 0 7 0 2 ' 0 0 2
Freeze Valves
0
0
0
i
~
0
0
0 0111 0
2io
0O 0 0 O 0
I
3 ' 2.0 3
0
0 1 O 0
Samplers
2
0
Periodic Containmen t Tests
0
0 3 f O O 2
0 0
1 l O
Core Specimens
0
O
0 0
NOTE:
3
1 1 2 1 0
0
'
7
0 1 0
0
1
5 1 9
0
1
0
0
7
l i l 0
1
2
0
4
!
l
i
1
0
1
= Primary cause of interruption. S = Secondary cause of interruption.
P W
*No
= Worked on during period following the i n t e r r u p t i o n (includes preventive maintenance).
interruption of operations during this e n t i r e period was caused by the cover-gas system, lube-oil system, o r leak-detector system.
c
C
Table 4.17
MERE Cumulative Cycle History
I
ul w
*These
f.igures are based on the original calculations. cycle t e s t s , the usage factors would be 23.04s and 13.37s.
If they were based on freeze flange thermal
..
. .
. ..
..
...
...
..
....
.
__...
..............I
... .... ...
..
-
I
Table 4-18.
Operating Procedures Section Number
-~ -~~
" . . . I
. _ ._ ..
l_l
Time t o Complete (hrs)
Operation
io
- 15
1-T
Instrument a i r system s t a r t u p check l i s t
4c 4D 43
4H
Water system s t a r t u p check l i s t Component cooling system s t a r t u p check l i s t Shield and Containment s t a r t u p check l i s t s Leak test of containment valves Pressurize the c e l l s ( t o 20 p s i g ) Cell l e a k rate a t pressure Evacuate c e l l s and purge with nitrogen F i r s t c e l l leak r a t e a t -2 p s i g V e n t i l a t i o n system s t a r t u p check l i s t h a k d e t e c t o r system s t a r t u p check l i s t Instrumentation s t a r t u p check l i s t
5A
Purging oxygen and moisture from the f u e l c i r c u l a t i o n system
5B
Startup of cover-gas and offgas systems
2 - 4
1-T*
5E
Startup of lube o i l systems
2 - 4
1-T*
5F
Heatup of fuel' and coolant systems
40
%
Routine pressure test
20
4G
4-8 4-8 1 - 2
- 160 15 - 18 >60
130
-
18 24 >I70 4 - 8
2 - 4 160 - 200 30
- 40 - 50 - 30
1-T* 1-T* H
',
1-E, 1-T, 1-P *H
w iwt
vl
ru
M
1-T* 1-T*
I
1-E, 1-T, 1-1 )Hm
w w
on s h i f t . Most of t h i s *as done on day s h i f t w i t h some a s s i s t a n c e from the s h i f t s ,
Wt
M
~.
1-T*
*Done w
-~
.
*
4B
4F
.
MANPOWER E= Engineer T= Technician P= Pipef itter I= Instrument Mechanic
E l e c t r i c a l s t a r t u p check l i s t
(5
..,.
APPROXIMATE TlMES RE&UIRFD TO PERFORM VARIOUS OPERATIONS
4A
I:\ 81
. " _ ~.I
This d i d n o t r e q u i r e much a t t e n t i o n . This required the a t t e n t i o n of most of the s h i f t operating crew.
e
c
c
c Table 4-18.
APPROXIMATE TIMES REQUIRED TO PERFORM VARIOUS OPERATIONS (continued)
Operating Procedures Section Number
MANPOWER Enaineer T= Technician P= Pipef itter I= Instrument Mechanic
E=
Operation
Time t o Complete (hrs)
F i l l i n g the f u e l and coolant systems
g
F i l l i n g the c o o l a n t system F i l l i n g the fuel system w i t h flush s a l t Drain of flush s a l t and secure Prepare f o r fuel f i l l
g
Fuel s a l t sampling o r enriching Coolant salt sampling
8
12 12
-x-w M
3-5
-x-w
8 - 9
M
1 - 2 1
"t
3 - 4
2-T* 1-T*
10
Preparation for c r i t i c a l i t y and power operation S u b c r i t i c a l t o f u l l power
-
- 12
1
**
*
1-T
P e r i o d i c instrument checks
8A 8B 8D
5 - 7
Neu+xon l e v e l instruments P r o c e s s . r a d i a t i o n monitors Safety ' c i r c u i t checks Emptying the overflow tank
91 12A 12A-2A
m3s F i r s t c o n t r o l room log on each s h i f t
*Done
1
2 - 4
on s h i f t . Most of this was done on day s h i f t with some a s s i s t a n c e from the shifts.
w
*This
.x-x-H-
d i d n o t r e q u i r e much a t t e n t i o n .
This required the a t t e n t i o n of most of the s h i f t operating crew.
-1
2-T*
- 1-1/4
1-T
112
3/4
2-I* 3-T* 2 t o 3-T
*
vl
w
... .
.
. . ..
Table 4-18.
Operating Procedure s Section Number
.
.
.. .
. .
.. .
.
I
_,____I
....
_...__I__...____-...
~
...
APPROXDlA!l’E TlMES ReQUIRED TO PERFORM VARIOUS OPERATIONS (continued)
MANPOWER Engine e r TI Technician PI Pipef itter I= Instrument Mechanic
E= Time t o Complete (hrs)
Operation
~
12A
L
\
Lags (continued)
12A-2A 12A-2B
12A-2B m-2B
116
Second c o n t r o l room l o g on each s h i f t F i r s t b u i l d i n g log each day F i r s t b u i l d i n g l o g on o t h e r s h i f t s Second b u i l d i n g log on each s h i f t
-- 21
0 - 1
1-T, 1-T* 1-T
1-112 112
113
F u l l power t o s u b c r i t i c a l Fuel s a l t d r a i n and secure Coolant salt d r a i n and secure Cool down t o 400’F
3 - 5
2 - 4 48 60
-
R e s t a r t i n g equipment after an e l e c t r i c a l power outage
Return to power after a d r a i n Drain to both d r a i n tanks ( t r a n s f e r through t r a n s f e r l i n e s ) Drain t o both d r a i n tanks ( t r a n s f e r through f i l l l i n e s ) Drain t o one d r a i n tank
*Done
I
- 60 - 30 io - 1 5
50 20
on s h i f t .
4w
Most of this was done on day s h i f t w i t h some a s s i s t a n c e from t h e s h i f t s . w This d i d n o t r e q u i r e much a t t e n t i o n .
**
This required the a t t e n t i o n of the s h i f t operating crew.
c
*
*
116
R e t u r n t o power after a load and rod scram
C
- 112
Shutdown of t h e Reactor
10
*
l-T* 1-T* l-T* 1-T
1-T*
VI
c
55
u
- Instrument
4B
A i r System Startup Check L i s t assured t h a t t h e corn-
pressors and d r i e r s were operating properly, t h a t t h e standby compressor would automatically s t a r t i f needed, and t h a t emergency nitrogen cylinders were i n s t a l l e d .
- Water
4C
System Startup Check L i s t assured t h a t all pumps and tower
fans were operable and t h a t proper water flow w a s established on all equipment. 4D
- Component
Cooling Systems Startup Check L i s t assured t h a t a l l
valving w a s s e t properly and t h a t all blowers were operable. 4E
- Shield
5 parts.
and Containment Startup Check L i s t i s divided i n t o
P a r t (1)checked t h a t t h e primary and secondary block and check
valves, containment enclosures, and l i n e s were leak-tight t o assure adequate containment.
Removable shielding .was a l s o checked t o assure t h a t it
had been r e i n s t a l l e d .
+
,
This check l i s t was assigned t o an operating engi-
neer on day s h i f t with technician and p i p e f i t t e r help as needed. Part ( 2 )
b.i
- this
i s t h e time required t o pressurize t h e c e l l s t o 20 psig
by adding instrument a i r .
Part ( 3 ) involved taking pressure and tempera-
t u r e d a t a periodically t o e s t a b l i s h t h e c e l l leak r a t e . valves and l i n e s were checked while at pressure.
Some containment
The time required f o r
checking these i s included i n P a r t (1). P a r t ( 4 ) i s t h e time required t o vent t h e c e l l s , evacuate t o -2 p s i g , and purge w i t h nitrogen.
a l s o involved taking periodic data.
Part ( 5 )
Since t h e r e w a s a s m a l l water leak i n
the c e l l , it took about 7 days f o r t h e atmosphere t o reach equilibrium. 4F
- Ventilation
Syste
t a r t u p Check L i s t assured t h a t a s t a c k fan
w a s i n operation, t h a t t h e other one would automatically s t a r t , and t h a t
v e n t i l a t i o n w a s adequate i n a l l areas. 4G
- Leak
Detector Systems S t a r t u p Check L i s t assured t h a t a l l valves
were open t o leak-detected flanges and that t h e o v e r a l l leak-rate w a s a t i s f actory
4H
.
- Instrumentation
S t a r t u p Check L i s t assured t h a t each instrument
and c i r c u i t functioned properly.
Abnormal conditions were simulated and
where possible t h e control action w a s allowed t o occur.
One operating
engineer on day s h i f t was assigned r e s p o n s i b i l i t y f o r completing t h i s check l i s t with technician and instrument mechanic assistance as required.
Ld
Some of t h e t e s t s were done on s h i f t . \
56
4.4.2
Reactor Startup The manipulations involved i n a c t u a l s t a r t u p of t h e r e a c t o r
(Section 5) are described below.
5A -Purging Oxygen and Moisture f r o m t h e Fuel Circulating System involved s e t t i n g t h e helium purge at' maximum rate and operating t h e f u e l pump t o c i r c u l a t e it. The t i m e of purge w a s a calculated number. 5B - S t a r t u p of Cover Gas and Offgas System involved s e t t i n g t h e valves and checking t h e helium t r e a t i n g s t a t i o n .
-
'
'5E S t a r t of Lube O i l Systems involved checking t h e supply tank, draining t h e o i l catch tank, s e t t i n g valves and checking t h a t t h e standby pumps would start automatically. 5F
- Heatup of
Fuei and Coolant Systems involved r a i s i n g t h e heater
s e t t i n g s and maintaining a reasonable temperature d i s t r i b u t i o n . and coolant systems could be heated si&taneously.
The f u e l
The t i m e given i s
f o r heating t h e coolant system from room temperature and the r e a c t o r syst e m from 400째F (normal shutdown condition) t o 1200째F and reaching
equilibrium.
5H -Routine Pressure T e s t involved r a i s i n g t h e f u e l system pressure t o 60 p s i g (usually w i t h f l u s h salt c i r c u l a t i n g ) , then lowering t h e f u e l system pressure and increasing t h e coolant system pressure t o 60 psig. A l l pressure switches and t h e i r c o n t r o l o r alarm actions w e r e checked.
51 - F i l l i n g t h e Fuel and Coolant Systems is divided i n t o 4 p a r t s . P a r t (1)involved thawing t h e freeze valves , f i l l i n g t h e coolant loop, checking thaw times of t h e freeze valves, r e f i l l i n g and s t a r t i n g t h e coolant pump.
P a r t ( 2 ) involved thairing t h e freeze valve, f i l l i n g t h e
f u e l loop w i t h f l u s h salt and s t a r t i n g t h e f u e l pump.
P a r t ( 3 ) involved
draining t h e flush s a l t , emptying t h e overflow tank and freezing t h e freeze valve.
P a r t ' , ( 4 ) involved thawing t h e freeze valve, adjusting t e m -
peratures, running rod drop times and various other s a f e t y checks and taking base l i n e countrate data.
P a r t (5) involved f i l l i n g t h e f u e l
loop w i t h f u e l salt, stopping at s i x l e v e l s t o take count r a t e data, freezing t h e drain freeze valve and s t a r t i n g t h e f u e l pump.
5J - C r i t i c a l i t y
and Power Operation i s divided i n t o 2 p a r t s .
Part (1)involved running rod drop and o t h e r tests.
The time does not
W
57
include check l i s t s 8A, 8B, o r 8D which sometimes had t o be done before c r i t i c a l i t y (see 4.4.3).
P a r t ( 2 ) involved manipulating t h e rods and
heat-removal system components t o reach f u l l power.
4.4.3
Operation Various jobs were done while t h e r e a c t o r w a s operating.
These are
described below.
-
6A3 o r 6 ~ 4 Fuel System Sampling o r Addition of Enriching Capsules involved a series of manipulations t o i n s e r t and w i t h d r a w a capsule w h i l e maintaining containment. '
Two samples could be delivered t o the laboratory
per s h i f t i f no difficulties were encountered.
This rate could not be
sustained due t o decontamination of c a r r i e r s , etc.
6B
- Coolant
S a l t Sampling involved a less complicated series of
manipulations t o i n s e r t and w i t h d r a w a sample from t h e coolant pump.
8 - P e r i o d i c Instrument Checks involved t e s t i n g as much of t h e c r i t i c a l equipment as possible without i n t e r r u p t i n g operation of the reactor.
9I - m p t y i n g t h e Overflow Tank involved pressurizing it w i t h helium and venting when it w a s nearly empty.
This took about twice as long when
t h e FP offgas l i n e w a s plugged.
12A -Control Room and
l d i n g Logs were taken twice p e r shift.
Some adjustments were usually required, water treatment w a s added and other odd jobs were done as part of t h e logs.
4.4.4
Shutdown of t h e Reactor ibed below.
P a r t (1)involved reducing t h e power.
It i s divided i n t o
4 parts.
This could be done by scramming
t h e rods and t h e load o r by i n s e r t i n g t h e rods and lowering t h e r a d i a t o r
doors. valves.
P a r t (2) involved draining the fuel system and freezing t h e freeze P a r t ( 3 ) involved
freeze valves.
4.4.5
Part
ning t h e coolant system and freezing t h e
( 4 ) involved codling
t h e systems t o 400'F.
Recovery from Unplanned Shutdowns A nuuiber of times during operation, power outages and o t h e r unex-
pected events caused shutdowns. operation as soon as possible.
Often there w a s a desire t o r e t u r n t o Some of these are described below.
.
..
...~
.
-.
. ..
....
. .~~,~
~
-
. . . ...
.
.
- -1
ORNL-DWG 73-596 DAYS
,
30 ,-25 ,-M ,-15 ,-la -5 I
I
I
I
I
1
I
I
2
I
I
3
I
I
4
I
I
6
, I
DAYS 8
I
I
7
I
1
B
I
1
9
I
10
I
Repin. modifiutiom. prvsntatiw maintlruncs a d w m specimen mplawmant
Instrumentationstartup c h k list. Contdnmat N r m p duck M8t.
Other sysmm sunup check Iim
Lac ctoum of prinwy systm.
Ruor of Pimvy rymm. Weld nnmbmrn and install Mocks on d k Rarurizr mlh to 20 pig.
--art
d h n 20 pdg.
EVUUM d h md pu01with nitrogen. H n i up fuel md ~001mtlwph
MO~S~UN ruchingquilibium in mllr
Fim (411 I n k rata I( VIEUU~. Fill molmt s m . Fill fuel system with flush salt
& m y e d r n rah
srmpk flush salt Dmin fhnh salt R.pmf a fuel Ylt fill.
Fill fuel system with fuel ok. &mph fuel salt Prepmtion f a criticality and power opsntion. Critiulity md taking tho ramor to full power.
Fig
c
4.6
Return to Full-Power Operation (Typical Schedule)
I
I
11
1 1 2
1 1 3
I
I
I
I
15
14
1
59
Lli4
9A-1 - R e s t a r t
Equipment After a Power Outage involved s t a r t i n g t h e
d i e s e l s , r e s t a r t i n g equipment and r e s e t t i n g t r i p p e d instruments.
55-1-
Return t o Power After a Load and Rod Scram involved manipu-
l a t i n g t h e rods and heat-removal system components t o return t o power.
If
no instrument o r equipment t e s t s were necessary, t h i s could be done i n l e s s than an hour. l l A , 51, and 5J
3 parts.
- Return t o Power After
a Drain i s divided i n t o
Part (1)assumes a drain t o both drain tanks followed by a
t r a n s f e r through t h e t r a n s f e r l i n e s , r e f i l l and return t o power ( t h i s is t h e design method of t r a n s f e r ) .
P a r t ( 2 ) assumes a drain t o both drain
tanks followed by a t r a n s f e r through t h e f i l l l i n e s , r e f i l l , and r e t u r n t o power.
(This method of t r a n s f e r w a s used t o save time.
interlocks w a s necessary.)
Jumpering of
Part ( 3 ) assumes a drain t o only one drain
tank followed by a f i l l and r e t u r n t o power.
4.4.6
LJ
Reactor Startup After a Major Shutdown Each s t a r t u p w a s d i f f e r e n t
requiring scheduling between t h e com-
p l e t i o n of maintenance and start of operation.
The c e l l membrane w a s
usually i n s t a l l e d s h o r t l y a f t e r t h e f i n a l closure of t h e primary system. Assuming t h i s , and t h a t maintenance on out-of-cell
components w a s com-
pleted by t h e time they were needed, a somewhat t y p i c a l s t a r t u p would be
as shown i n Fig. 4-6.
4.5
Changes Made i n t h e Plant
The number and kinds of changes made i n a p l a n t a f t e r it begins operation a r e influenced by two d i f f e r e n t things: changing a c t i v i t i e s .
o r i g i n a l design and
I n some single-purpose systems t h e number of changes
may simply r e f l e c t how well t h e o r i g i n a l design met i t s goals.
I n an ex-
perimental plant such as t h e MSRE, most of t h e changes may be required f o r experimental purposes o r changes i n t h e mode of operation.
I n any event,
t i m e spent i n making changes a f f e c t s t h e a v a i l a b i l i t y of t h e plant. During t h e course of t h e MSRF: operation many changes were made, but only a few caused s i g n i f i c a n t delay i n t h e program.
60
A f t e r t h e end of construction and t h e non-nuclear checkout of t h e
U
MSRE, it w a s required t h a t a change request be formally i n i t i a t e d , reviewed and approved before any modification w a s made.
(Section 13 of
Reference 22, which prescribes t h e procedure, defines a modification as tta change i n t h e physical plant which produces a s i g n i f i c a n t l y d i f f e r e n t
c h a r a c t e r i s t i c o r function i n any component o r system.")
I n t h e 55 months
from June 1965 through December 1969, a t o t a l of 633 requests f o r changes i n t h e reactor system were i n i t i a t e d , of which 512 were approved.
For t h e
chemical processing p l a n t , 113 change requests were i n i t i a t e d of w h i c h 87
were approved. Table 4-19 summarizes f o r each 6-month period t h e number of requests i n i t i a t e d and approved f o r t h e reactor.
The table a l s o categorizes t h e
approved changes as t o t h e reason f o r making t h e change and t h e type of
As indicated, most of t h e changes were aimed a t f i l l i n g t h e needs of normal. operation and over h a l f were changes i n e i t h e r instruchange t h a t i t - w a s .
mentation o r setpoints. Table 4-20 summarizes t h e change requests f o r t h e chemical processing
plant. Although w e know how many changes w e r e made, we cannot accurately sum up t h e times required f o r making t h e changes.
We can s a y , however, t h a t i n
t h e case of t h e r e a c t o r change requests, t h e vast majority e i t h e r took
very l i t t l e t i m e t o execute o r w e r e made while other work w a s going on. Exceptions include t h e work on the f u e l offgas system i n the spring of
k966.
I n the-processing plant, t h e summer of 1968 w a s spent i n t e s t i n g
and modifications, p a r t i c u l a r l y of the fluorine disposal system.
u
c
c Table 4-19
Number of Requests
Period
1966
1967
1968
1969
Summary of Reactor
Reason for
System Requests
Change Requests
12 $: d
3 2 1965
c
ti c 0’
lsta
34
'27
1
6
13
5
0
0’
4
0
6
3
2
10
2
4
2nd
126
115
5
9
86
9
1
1
4
2
11
17
8
61
12
5
1st
173
J-37
2
12
104
7
2
0
9
2
11
29
60
14
10
2nd
103
76
1
7
59
5
2'
l-
1
1
11
14
34
12
4
lst
69
55
3
1
47
1
1
1
1
0
3
18
1
25
10
2
2nd
34
30
0
1
25
1
0
2'
-0
0
3
4
1
13
7
3
1st
32
23
1
1
11
1
3
6
0
0
3
5
2
ll
1
1
2nd
18
17
0
0
12
0
5
0
0
0
0
2
6
8
0
lst
24
21
0
0
12
0
8
1
0
3
5
3
8
6
0
2nd
20
ll
1
0
4
0
4
2
0
0
3
2
6
3
0
14
37
373
26
14
51
100
46
234
75
29
Total
633 'After
institution
512
of change request
22 procedure
i9
on June 21, 1965.
.l
Suminaryof ChemPlant Change Requests
Table 4-20 Number c Request
Period
iP) 2 g u8
9 8 1965
1966
1967
1968
1969
'otal
d
3
5
1 0
1
4
0 2
0 2
0 0
0 0
0
5
0 0
0 0
53
9
5 0
1st 2nd
56
44
30
'25
0 0
0 0
1st 2nd
1
0 0
0 0
0
0
1 0
113
94
1
1
1st
5
4
2nd
5:
1st 2nd 1st 2nd
2 3 8 II
b g 1
1
0 ,O
4
0 0
0 0
0 0
0 0
0 0
.O 2
0 0
0 0
1 1
4
0 2
1 3
0 0
0 0
3
17 10
4
1.6
5
2
7
6
1 0
0
0 0
0 0
0 0
1 0
0 0
6
37
8
33
12
1
0 0
0 0
0 0
0 0
0 1
0
3
0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0
' 0
0 0
0 0
1 0
0 0
1 0
0 0
1 0
0 0
0 0
0
0
0 0
89
0
0
2
0
2
2
0 i;!
25
t c_.
Type Change
Reason for Change Requests
c
1
2
63
4.6
Tabulation of Recorded Variables a t F u l l power
It seemed d e s i r a b l e t o record a set of readings of a l l of t h e various
r e a c t o r v a r i a b l e s a t normal conditions.
However, even though t h e MSRE r a n
a t f u l l power much of t h e t i m e , conditions were constantly changing due mainly t o t h e various-experiments which were performed. Also some of
t h e v a r i a b l e s were not recor
select t h e b e s t t i m e .
r e g u l a r l y thus making i t d i f f i c u l t t o
After consideration of these f a c t o r s , t h e 12 t o 8
s h i f t on 10/12/69 was selected as a period a t f u l l power when conditions were f a i r l y normal and considerable information w a s available. Values
were taken from the various recorder charts, r o u t i n e c o n t r o l room and building logs and t h e weekly, monthly, and o t h e r check lists. These are tabulated i n Tables 4.21 and 4.22.
A s n a p s h o t . l i s t i n g a l l of t h e computer
inputs w a s also r e t r i e v e d from the computer tapes f o r 0400 on 10/12/69. The l o c a t i o n of t h e sensing element These values are l i s t e d i n Table 4.23. is described b r i e f l y i n t h e d e s c r i p t i o n column.
The number and l e t t e r s i n
t h e i d e n t i f i c a t i o n column correspond t o those used i n t h e design drawings and o t h e r design documents. They can be used t o f u r t h e r - i d e n t i f y the variable.
64 Table 4.21.
Identification
W
Tabulation of recorded variables . On 10/12/69 't-
Description
Reading 10/12/69
Line 100 (Reactor butlet) temperature
1205'F 1169'F
TE-101-2A
Line 100 temperature Line 101 temperature
TE-101-3A
Line 101 temperature
1215'F
TE-102-3A
Line 102 temperature
1160'F
TE-102-B4A
Line 102 temperature
1152'F
TE-102-5A
Line 102 (Reactor inlet) temperature
1173'F
TE-103-AlB "E-103-B1
Line 103 temperature Line 103 temperature
908'F 798'F
TE-103-6
Line 103 temperature
1150'F
TE-103-8
Line 103 temperature
TE-103-Bll
Line 103 temperature
1223'F 983'F
TE-103-14B
Line 103 temperature
1200'F
TE-104-5B
Line 104 temperature
918'F
TE-106-5B
Line 106 temperature &ine 108 temperature
2080'F 222'F
TE-109-7 TE-200-B7B
Line 109 temperature
227'F
Line 200 temperature
1018'F
TE-200-D7B
Line 200 temperature
1025'F
TE-200-A8B
Line 200 temperature
1035'F
TE-200-BSB
Line 200 temperature
lOlO'F
TE-200-C8B
Line 200 temperature
1015'F
TE-200-AgB
Line 200 temperature
1010' F
TE-200-BgB
Line 200 temperature
TE-200-14A TE-200-16B
Line 200 temperature Line 200 temperature
1025'F 1018'F
TE-200-19A
Line 200 temperature
1023'F
TE-200-AS-AlB
Line 200 penetration temperature
965'F
TE-200-AS-BlA
Line 200 penetration temperature
642'F
TE-100-A5 TE-100-A6
TE-108-7
1200'F
1015'F
LJ
.
65
I
Table 4.21.
Description
Identification
Reading 10/12/69
TE-200-AS-CIA
Line 200 penetration temperature
332°F
TE-2 00-AS-2A
Line Line Line Line Line
200 penetration temperature
363°F
201 temperature 201 temperature
1072°F 1062°F
201 temperature
1068°F
201 temperature
1078°F 1045°F
TE-201-B1OB
Line 201 temperature Line 201 temperature Line 201 temperature
TE-201-C1OB
Line 201 temperature
1086°F
TE-201-A11B
Line Line Line Line
201 temperature
1078°F
201 temperature
1083°F
201 temperature
1075°F
201 penetration temperature
706°F
Line 201 penetration temperature Line 201 penetration temperature
772°F
TE-201-2B TE-201-5A TE-201-7B TE-201-AgB TE-201-BgB TE-201-AlOB
TE-201-CIlB
&,
(continued)
TE-201-D11B TE-201-AS-AIB TE-201-AS-BIA TE-201-AS-C1B TE-2 01-AS-2A TdI-201A Xpr 201 FR-201
'
1073°F 1059°F
350°F 350°F
Line 201 penetration temperature Radiator AT (salt)
59°F
Radiator power
7.1 MW
Bolant salt flow
860 gpm
TE-2 03-1
870°F
TE-203-2
107°F
TE-204-1A
Line 204 temperature
1135°F
TE-204-2A
204 temper
1155°F
204 tempe
1205°F
204 temper
1200°F
204 temper
1193°F
TE-204-6A
Line Line Line Line Line
204 temp
1170°F
TE-204-A7B
Line 204 temperature
1178°F
TE-204-B7B
Line 204 temperature
1170°F
TE-204-8B
Line 204 temperature
1130°F
TE-2 04-3A TE-204-4A TE-204-5A
66 Table 4.21
Description
Identification
b,
(continued)
Reading 10/12/69
TE-206-lA
Line 206 temperature
1073°F
TE-206-2A
Line 206 temperature
1124°F
TE-206-3A TE-206-4A
Line 206 temperature Line 206 temperature
1147°F 1163°F
TE-206-5B
Line 206 temperature
1150°F
TE-206-6A
Line 206 temperature
1073°F
PI-407
Leak detector pressure
100 psig
RM-500
Cover-gas supply radiation
0.3 mr/hr
PI-500A
Helium header pressure
223 psig
FIC-500A
H e l i u m f l o w rate
6 litershin.
PI-5OOG2
Helium treating station pressure
250 psig
PI-500M
Reduced helium pressure
35 psig
PI-501A
Reduced helium header pressure
PI-51OA
Oil tank No. 2 pressure
PR-51lD
Coolant drain tank pressure
TE-512-1
Line 512 temperature
34.5 psig 7.5 psig 4.8 psig 96'F
FI-512A PI-5 13A
Coolant pump purge gas flow
0.6 liters/min
Oil tank No. 1 pressure
8 Psig
TE-516- 1
.Line 516 temperature
107°F
PR-516
Fuel pump pressure
FI-516B
Fuel pump purge gas flow
5 Psig 2.4 liters/min.
PR-522A
Fuel pump pressure
5.2 psig
TE-524-2
Line 524 temperature
93°F
FI-524B
Fuel pump upper off-gas flow rate
47%
LI-524C
Oil citch tank No. 1 level
FI-526C
Coolant pump upper-off-gas flow rate
LI-526C
Oil catch tank No. 2 level
RTI-528
Coolant gas supply radiation
PR-528A
Coolant pump pressure
14% 0.04 liters/min. 18% 1.5 mr/hr 5.1 psig
(I'
67 Table 4.21
tJ Identification
Reading 10/12/69
Description
A 021 548
Helium oxygen content
0.4 ppm
A HpO I 548 FI-548A ,
H e l i u m moisture content Helium oxygen analyzer flow
2 PPm 100 cc/min
FI-548BI
H e l i u m moisture analyzer flow
100 cc/miil
PdI-556A
Main charcoal bed AP
3.4 p s i
RBI-557
Charcoal beds off-gas radiation
0.1 mr/hr
RM-565
C e l l a i r radiation
1.5 mr/hr
FI-566
Reactor c e l l a i r oxygen analyzer flow
100 cc/min
A01 I 566
Reactor c e l l a i r oxygen content
2.7%
PR-572B
Fuel drain tank No. 1 pressure
5.9 p s i g
PR-574B
Fuel drain tank No, 2 pressure
5.5 psig
PR-576B
Fuel f l u s h tank pressure
PI-589
Werflow tank pressure
5.1 p s i g
FI-589
Overflow tank bubbler flow rate
27 psig
PI-592
Fuel pump pressure
5.4 p s i g
FI-592
b,
(continued)
.
5.0 psig
25 psig
Fuel pump bubbler flow rate
FI-593
Fuel pump bubbler flow rate
23 p s i g
LR-593C
Fuel pump l e v e l
53%
FI-594
Coolant pump bubbler flow rate
24.8 psig
FI-595
Coolang pump bubbler flow rate
25.5 p s i g
LR-595C
5 7%
FI-596
25.2 p s i g
RM-596
0.1 mr/hr 24.0 p s i g
FI-598
Coolant pump bubbler flow rate
FI-599
Overflow tank
LI-599B
Overflow tank
FI-600
Overflow tank
LI-600B
Overflow tank
24%
PI-7OlA
Fuel 011 pump
10 psig
PI-702A
Fuel o i l pump No. 2 pressure
"E-702-lA
Line 702 temperature
ler flow rate
27.0 p s i g 2 3%
ler flow rate
26.5 psig
-
64 p s i g 135'F
68 Table 4.21
Identification
W
(continued) Reading 10/12/69
Description
FI-703
Fuel pump lube oil flow rate
3.8 gpm
FI-704
Fuel pump coolant oil flow rate
8.2 gpm
PI-751A
Coolant oil pump No. 1 pressure
PI-752A
Coolant oil pump No. 2 pressure
8 P a 56 psig
TE-752-lA
Line 752 temperature
FI-753
Coolant pump lube oil flow rate
FI-754
Coolant pump coolant oil flow rate
LI-806A
Steam dome water level (Pp-1)
LI-807A
Steam dome water level (FD-2)
FI-810 FI-812
40 gpm 40 gpm
FI-817
Condenser water flow rate (-1) Condenser water flow rate (-2) Offgas particle trap water flow rate
TI-820-1
OT-1 water outlet temperature
87°F
TI-821-1
OT-1 water inlet temperature
80°F
FI-821A TI-822-1
OT-1 water flow rate OT-2 water outlet temperature
TI-823-1
OT-2 water inlet temperature
8.4 gpm 81°F 77°F
FI-823A
OT-2 water flow rate
TI-826
Treated water cooler (TW out) temperature
RM-827
Process water radiation
TI-829
Treated water cooler (TW in) temperature
PI- 829A FI-830
Treated water pump pressure
FI-832
Coolant pump motor cooling water flow rate
3 mr/hr 108°F 70 psig 4.6 gpm 4.8 gpm
FI-836A
Drain tank cell space cooler water flow rate
63 gpm
FI-838A
Reactor cell space cooler No. 1 water flow rate
FI-840A
Reactor cell space cooler No. 2 water flow rate
FI-844A
Thermal shield water flow rate
53 gpm 59 gpm 50 gpm
PI-85U
Cooling tower water pump pressure
34 psig
FI-85lC
Cooling tower water to cooler flow rate
273 gpm
Fuel pump motor cooling water flow rate
130°F 4.0 gpm 6.7 gpm 0% 0%
4
mm
8.6 gpm 98°F
gs
69 Table 4.21
Identification
Reading 10/12/69
Treated water cooler (CTW out) temperature Cooling tower water temperature
84°F 80°F
FI-859 FI-862A FI-864A
Thermal shield slide water flow rate Coolant cell space cooler (west) flow rate Coolant cell space cooler (east) flow rate CCP gas cooler water flow rate CCP No. 1 and 2 oil cooler water flow rate Air compresser cooler outlet water temperature Air compresser head outlet water temperature
7 gpm 20 gpm
PI-927C
CCP AP Ventilation filter suction Ventilation stack filters AP Ventilation stack fan suction
PdI-937A PdI-938A
TR to 840 level AP SESA to TR AP
RI4-6000-1 RM-6000-2 RM-6000-3
Reactor cell radiation
PdI-9OOA PI-927A PdI-927B2
RM-6000-4
RM-6000-5 RM-6000-6 RM-6010 RR-8100 RR-8200 FI-9000 PIC-9006-1 FI-9006 PdI-AD2-A2
Lid
Description
TI-854 TI-858
FI-873 FI-875 TI-881-2A TI-881-2B
W
(continued)
Reactor cell radiation Reactor cell radiation Drain tank cell radiation Drain tank cell radiation Drain tank ce Coolant cell r Reactor power Log reactor p Instrument air f Emergency N2 setpoint Instrument air flow (emergency header) Radiator air pressure drop
21-AD2
Bypass damper position
FI-AD3A
Radiator stack flow rate
TI-AD
Radiator air inlet temperature
TI-AD3- 8A
Radiator air outlet temperature
20 gpm 15 gpm 6.3 gpm 86°F 92°F
8 psi 2.2 in. H a 0 3.3 in. H a 0 5.5 in. H a 0 0.05 in. HzO 0.04 in. HzO 50,000 R/hr 30,000 R/hr 50,000 R/hr 10 R/hr 60 R/hr 500 R/hr 100 R/hr 8.5 MW 7Mw
20% 65 psig 44% 67%
10% 70% 71°F 190°F
70 .L
.
W
Tab be 4 21 (continued)
Identification
Description
Reading 10/12/69
PI-AR-1
Instrument air receiver tank pressure
TE-CC-1
Coolant cell ambient temperature
TE-CC-2
Coolant cell ambient temperature
88 psig 88°F 90°F
TE-CC-3
Coolant cell ambient temperature
78°F
TE-CC-4
Coolant cell ambient temperature
80°F
TE-CC-5
Coolant cell ambient temperature
76°F
TE-CC-6
Coolant cell ambient temperature
115°F
TE-CC-7
Coolant cell ambient temperature
105OF
TE-CC-8 TE-CDT-IA
Coolant cell ambient temperature Coolant drain tank top temperature
TE-CDT-3A TE-CDT-5A
Coolant drain tank top temperature Coolant drain tank middle temperature
103°F 1195°F 1182°F 1200°F
WR-CDT-C1
Coolant drain tank weight
3%
EiI COP-1 EiI COP-2
Coolant oil pump No. 1 current Coolant oil pump No. 2 current
0 amps
TE-CP-1B
Coolant pump flange-neck temperature
8.7 amps 245°F
TE-CP-B2A
Coolant pump bowl-neck temperature
786°F
TE-CP-C2A
Coolant pump bowl-neck temperature
795°F
TE-CP- 3B
Coolant pump bowl top temperature
957°F
TE-CP-4A
Coolant pump bowl top temperature
TE-CP-5 A
Coolant pump bowl top temperature
965°F 965°F
TE-CP-6A
Coolant pump bowl bottom temperature
1042°F
TE-CP-7A
Coolant pump bowl bottom temperature
1020°F
TE-CP-8A
Coolant pump bowl flange top temperature
107°F
TE-CPM-1B
Coolant pump motor temperature
100°F
SI-CP
Coolant pump speed
1780 rpm
EiI-CP
Coolant pump current
51 amps
EWI-CP
Coolant pump power
56.5 kW
LR-CPA
Coolant pump level (float)
TE-CPLE-A2
CP level element pot (lower) temperature
4.8 in. 1140°F
hd
71 Table 4.21
Description
Identification
CP level element pot (upper) temperature CP level element pipe (top) temperature
1097°F 1102°F
CP level element sensor (lower) temperature
515°F
TE-CPLE-A7 TE-CR-121 TE-CR-123
CP level element sensor (upper) temperature Coolant radiator inlet pipe temperature Flow venturi pipe temperature
372°F 1060°F 1020°F
LI-DC LI-DTC
Decontamination cell sump level Drain tank cell sump level Fuel drain tank No. 1 (upper) temperature
0 in. 0 in. 1115'F
Fuel drain tank No. 1 (middle) temperature Fuel drain tank No. 1 bayonet (top) temperature
1170°F
TE-FD1-18B TE-FD1-19A TE-FD 1-2OA WR-FD1c TE-FD2-2A TE-FD2-13A
Fuel Fuel Fuel Fuel
drain drain drain drain
tank tank tank tank
No. 1 bayonet (lower) temperature 1161°F No. 1 bayonet (bottom) temperature 1161°F No. 1 weight 0% No. 2 (upper) temperature
Fuel drain tank No. 2 (middle) temperature. No. 2 bayonet (top) temperature
TE-FD 2- 17A TE-FD2-18B
No. 2 bayonet (upper) temperature
TE-FF-102-2 TE-FF-200-2 TE-FF-201-2
1130°F
Fuel drain tank No. 1 bayonet (upper) temperature 1146°F No. 1 bayonet (center) temperature 1158°F Fuel drain t
TE-FD2-16A
TE-FD2-19A TE-FD2-20A WR-FD2c TE-FF-100-2
LJ
Reading 10/12/69
TE-CPLE-A4 TE-CPLE-A5 TE-CPLE-A6
TE-FD-1-2A TE-FD1-13A TE-FD1-16A TE-FD1-17A
u
(continued)
1140°F 1150°F 1115°F 1130°F 1140°F 1140°F 1140°F
Fuel drain tank No. 2 bayonet (center) temperature Fuel drain tank No. 2 bayonet (lower) temperature 0 . 2 bayonet (bottom) temperature Fuel drain tank No. 2 weight 5% 885°F ter temperature 974°F ter temperature ter temperature ter temperature
754°F 806'F
TE-FFT-SA
Fuel flush tank (top) temperature
1132°F
TE-FFT-7B
Fuel flush tank (middle) temperature
1174°F
72 Table 4.21
Identification
,
. .
V
(continued)
Description
Reading 10/12/69
WR-FFT-C
Fuel flush tank weight
69.6%
PdI-FI-AI.
Vent. system roughing filters AP
0.6 in. HzO
PdI-FI-A2
Vent. system absolute filters AP
1.6 in. HzO
EiI FOP-1 EiI FOP-2
Fuel oil pump No. 1 current
0 amps
Fuel oil pump No. 2 current
0.2 amps
TE-FP-1B
Fuel pump neck-flange temperature
305°F
TE-FP-2B
Fuel pump neck temperature
559°F
TE-FP-3B
Fuel pump neck-bowl temperature
TE-FP-4B TE-FP-5B
Fuel pump bowl top temperature Fuel pump neck-bowl temperature
962°F 992'F 990°F
TE-FP-6A TE-FP-9B
Fuel pump bowl top temperature
1030°F
Fuel pump neck-bowl temperature
TE-FP-1OB
Fuel pump bowl top temperature
955'F 1002'F
TE-FP-11B TE-FP-12B
Fuel pump flange top temperature Fuel pump bowl center temperature
"E-FPM- 1B
Fuel pump motor temperature
998°F 120°F
SI-FP
Fuel pump speed
1176 rpm
Ei-FP
Fuel pump current
44.5 amps
Ew-FP
Fuel pump power
34 kW
LI-FSC
Fuel storage cell sump level
1.1 in.
FT-2olA-lA
Flow element 20lA pipe temperature
1130'F
FT-20lA-2A
Flow element 20lA top flange temperature
1210'F
FT-2 0lA-3A FT-2 0lA-4A FT-ZOlA-SA FT-2OlA- 6A FT-201B-lA FT-201B-2A FT-201B-3A FT-201B-4A
Flow element 20l.A bottom flange temperature Flow element 20l.A pipe temperature
1160°F 1130'F
Flow element 20lA top flange temperature
1210°F
Flow element 2OlA bottom flange temperature
1250'F
Flow element 201B pipe temperature
970°F
Flow element 201B top flange temperature
1180°F
Flow element 201B bottom flange temperature
1170°F
Flow element 201B pipe temperature
1130°F
gd
150'F
.
73 Table 4.21
(continued) ,_
.
.
.
Reading 10112169
Flow element 201B top flange temperature
1200°F
FT-201B-6A
Flow element 201B bottom flange temperature
1130°F
TE-FV-103-1B
Freeze valve 103 shoulder temperature
1000°F
TE-FV- 103-2A
Freeze valve 103 center temperature
390°F .
TE-FV-103-3B
540°F
TE-FV-104-3B
Freeze valve 103 shoulder temperature Freeze valve 104 shoulder temperature
TE-FV-104-B4
Freeze valve 104 adjacent pipe temperature
TE-FV-104-5B TE-FV-104-6B
Freeze valve 104 pot temperature Freeze valv pipe temperature
TE-FV-105-2A
Freeze valve
TE-FV-105-A4A
Freeze valve 105 adjacent pipe temperature ' .
1160°F
Freeze valve 105 adjacent pipe temperature
center temperatur e
_,
4500F 450°F 590°F
,
650°F 1250°F
TE-FV-105-5B
Freeze valve 105 pot temperature
1190'F. 108D.">F .
TE-FV-105-6B TE-FV-106-2A
Freeze valve 105 pipe temperature
1120°F
Freeze valve 106 center temperature
1215°F
TE-FV-106-A4A
Freeze valve 106 adjacent pipe temperature
1140°F
TE-FV-106-B4A
Freeze valve 106 adjacent pipe temperature
1190°F
TE-FV-106-5B TE-FV-107-lA TE-FV-107-3B
Freeze valve 106 pot temperature Freeze valve 107 shoulder temperature oulder temperature Freeze valv adjacent pipe temperature
1120°F
TE-FV-107-5B TE-FV-107-6A Freeze valve
TE-FV-108-3B TE-FV-108-5B
._
TE-FV-108-6A
<
500°F 499°F 530°F
ot temperature
610°F
temperature
540'F 440°F
oulder temperature temperature
450°F; 640°F
temperature
543°F
acent pipe temperature
TE-FV-108-B-4
(Ilr
-
FT-201B-5A
TE-FV-J.O7-A4 '
..-
Description
Identification
b,
. .
TE-FV-lOg-lB
Freeze valve 109 shoulder temperature
TE-FV-109-6A
Freeze valve
TE-FV-204-1B
Freeze.valve 204 shoulder temperature
pot temperature
470°F 595°F 790°F
,
.
74 Table 4.21
Identification
W
(continued)
Description
Reading 10/12/69
TE-FV-204-2A TE-FV-204-3B
Freeze valve 204 center temperature Freeze valve 204 shoulder temperature
235°F 830°F
TE-FV-206-1B
Freeze valve 206 shoulder temperature
710'F
TE-FV-206-2A TE-FV-206-3B
Freeze valve 206 center temperature
300°F
Freeze valve 206 shoulder temperature Heater 103 temperature
830°F 908'F
High bay pressure
-0.19 in. HzO
TE-H 103 PI-HB-A
i
TE-HX-lA
Heat exchanger coolant out temperature
1060°F
TE-HX-4A
1180°F
TE-EIX-7A
Heat exchanger coolant !Ln temperature Heat exchanger shell (center) temperature
21-ID-A
Inlet radfator door position
82.5%
TE-OFT-1
Overflow tank pipe temperature Overflow tank top temperature
934'F
TE-OFT-2B
1165°F
Overflow tank side temperature
1202°F 1190°F
LI-OTlA3
Outlet radiator door,position Fuel oil supply tank level
79.5% 64%
RM-OT1
Fuel oil supply tank radiation
1.7 mr/hr
LT-OT2A3
58%
RM-OT2 TIC-02 Rl-1
Coolant oil supply tank level Coolant oil supply tank radiation
0.1 mr/hr
Helium oxygen removal No. 1 temperature
790°F
TIC 02 R2-1
Helium oxygen removal No. 2 temperature
1235'F
TIC
R1-2
Helium oxygen removal No. 1 wall temperature
513°F
TIC 02 R2-2 TIC PH 1
Helium oxygen removal No. 2 wall temperature Helium preheater No. 1 temperature
860°F 790'F
TIC PH 2
Helium preheater No. 2 temperature
800°F
TE PT-1
Particle trap temperature
360°F
TE PT-2
Particle trap temperature
360°F
TE PT-3
Particle trap temperature
360'F
TE R-5A
Reactor top temperature
1206°F
TE R-6A
Reactor top temperature
1210°F
TE-OFT-4 ZI-OD-A
03
hs
75
W
Table 4.21
Identification
(continued)
Description
Reading 10/12/69 .
b,
TE R-7A
Reactor neck temperature
TE R-8A
Reactor neck temperature
799°F 680°F
TE R-9 TE R-10
Reactor neck temperature
600°F
Reactor neck temperature
534°F
TE R-17
Reactor side temperature
1190°F
TE R-23A
Reactor side temperature
1185°F
TE R-32A
Reactor bottom temperature
1170°F
TE R-34
Reactor neck flange temperature
233°F
TE R-35
Reactor neck flange temperature
198°F
TE R-36A
ReacGor contr
rod No. 1 (upper) temperature
449°F
TE R-37A
Reactor control rod No,. 2 (upper) temperature
431°F
TE R-38A
Reactor control.rod No. 3 (upper) temperature
TE R-39A
Reactor control rod No. 1 (lower) temperature
460°F 855°F
TE R-40A
Reactor control rod No. 2 (lower) temperature
786°F
TE R-41A TE R-43B
Reactor control rod No. 3 (lower) temperature Reactor graphite tube (lower) temperature
689°F 1025°F
TE R-44A
Reactor neck (bottom) temperature
1218°F
TE R-46A TE R-47
Reactor neck (upper) temperature Reactor neck (upper) temperature Reactor neck (upper) temperature rmal well temperature
250°F
LI RC-C PI RC-A
L#
-
1 sump level Reactor cell pressure
215°F 212°F 810°F
0 in. -2.5
psig
FI-S1 RM-SlA
75% 100 cpm
RM-SIB RM-s1c LI SC-A
Containment stack iodine
11 cpm 530 cpm 1.0 in. H a 0
LI TC-A
Spare c e l l sump level.
0.4 in. Ha0
PI VT-1
Vapor suppression tank pressure
LI WT-A LI WTC-A
Waste tank level
0 . 4 psig 107 in. H2O
Waste tank cell level
2.6 in. H a 0
76 Table 4.21
Descr3p t ion
Identification
W
(continued)
Reading 10/12/69
Main blower vibration
e1 mil
OACOT
Official average coolant outlet temperature
1011.5," F
OAFOT
Official average fuel outlet temperature
1208.3'F
Control rod No. 1 position
35.5 in.
Control rod No. 2 position
44 in.
Control rod No. 3 position
43.1 in.
Fission chamber No. 1 position
60 in.
Fission chamber No. 2 position
65.7 in.
Fission chamber No. 1 count rate Fission chamber No. 2 count rate
io4 cps
Control rod No. 1 clutch current
143 ma
Control rod No. 2 clutch current
144 ma
Control rod No. 3 clutch current Motor generator 2 current
147 ma 28 amps
Motor generator 3 current Motor generator 2 voltage
32 amps 52 volts
Motor generator 3 voltage
52 volts
Battery voltage Motor generator No. 1 current
50.5 volts 260 volts 130 amps
Inverter voltage
206 V
Inverter current
89 amps 260 amps 280 amps 0 amps 88 amps 12 cfm
Motor generator No. 1 voltag-
Main blower No. 1 current Main blower No. 3 current Component coolant pump No. 1 current Component coolant pump No. 2 current Instrument air dryer purge rate
104cps
u
77
bi
Table 4.22,
'
Reading 10/12/69
.Description
Reading 10/12/69
Description
Reading 10/12/69
H-CR-1
15
H203-2
0
gCH-5
12
H-CR-2
17
H204-lA
3
RCH-6
15
H-CR-3
23
LE-CP-1
4
RCH- 7
8
H-CR-4
18
LE-CP-2
7
H102-2
13
H-CR-5
18
FV204-1
2.5
R- 1
18
H-CR-6
19
Fvj204-2
1
R-2
19
H-CR-7
15
FV206-1
2
R-3
19
H-CR-8
25
FV206-1A
1.5
HX- 1
0
200-13
17
H204-1
15
Hx-2
16
201-12
13
H206-1
11
Hx-3
202-2
14
CDT-1
15
FP-1
6
200-14
6
CDT-2
11
FP-2
6
200-15
10
CDT-3
12
RAN-1
0
201-10
11
CP-1
13
RAN-2
0
201-11
6
CP-2
12
200-16
2
201-13
7
H200-1
10
201-14
1
202-1
7
H200-11
12
102-1
3
204-2
10
H200-12
14
522
0
205-1
7
H201-1
13
102-4
9
204-3
6
H201-2
10
102-5
1
FT20lAl
5
H201-9
16
103
26
FT201A3
5
H100-1
2
FV-103
0
FT20lA2
6
RCH-
16
H-104-1
8
FT201Ar
4
RCH-
13
FV-104-1A
4
FT-201B1
6
RCH-3
19
FV-105-1A
11
FT201B3
5
21
FFT-1
16
m201B2
6
18
FFT-2
15
FT201B4
6
H101-2
11
-1-1
18
H203-1
0
H101-3
13
FD-1-2
17
. Description *
,
h,
Tabulation of recorded h e a t e r current on 10/12/69 (average of 3 phases)
-
3
I
16
*
‘78 Table 4.22
(continued)
Description
Reading 10/12/69
Description
Reading 10/12/69
Description
Reading 10/12/69
FD-2-1
18
FV-104-3
11
FV-108-3
3
-2-2
13
FV-104-4
10
FV-108-1
1
FV-104-1
5
FV-105-2
6
FV-108-3
5
FV-104-3
7
FV-105-3
7
FV-109-1
3
H-104-5
10
FV-104-7
13
FV-109-2
4
H-104-6
8
FV-106-2
8
FV-109-3
2
FV-105-1
12
FV-106-3
10
FV-109-1
1
FV-105-3
11
FV-110-2
0
FV-109-3
5
FV-105-1
7
FV-110-3
0
FV-110-1
0
FV-105-4
11
FV-107-1
2
FV-106-1
11
FV-107-2
3
FV-106-3
9
FV-107-3
3
FV-106-1
6
FV-107-1
1
FV-106-4
3
FV-107-3
5
FV-106-IA
11
FV-108-1
3
FV-104-2
9
FV-108-2
5
sj
79
bj
Table 4.23. Computer snapshot taken at 0400 on 10/12/69
Identification
Scan Seq.
Description
No. EWM-CP-D
256
Coolant pump power
EWM-F'P-D
255 349
Fuel pump power Reactor cell evacuation flow Radiator stack flow
Fq I-569 FT-AD3-A FT-S1-A FT-20 1-A ET-201-B FT-512-A FT-516-B FT-524-B FT-526-C FT-703-A FT-704-A FT-753-A FT-754-A LE-CP-A
40 242 15 28 236 2 34
Containment stack flow Coolant salt flow Coolant salt flow Coolant pump purge gas flow Fuel pump purge gas flow Fuel pump upper offgas flow
235 237 238
Coolant pump upper offgas flow Fuel pump lube oil flow
239 &O
Fuel pump coolant oil flow lube oil flow Coolant pump coolant oil flow
LT-OT-1-A
241 65 248
LT-OT-2-A LT-5244
250 251 252
Coolant pump level Fuel oil tank level Coolant oil tank level Oil catch tank No. 2 level
50
LT-595-C LT-596-B LT-598-C LT-599-B LT-600-B PDT-AD2-A PDT-556-A
61 54 62 57 58 24 228
Coolant pump level
Reading 10/12/69 31.8 kW 34.8 kW 1.27 liter/min 195,000 cfm 20,200 cfm 849 gpm 840 gpm 0.65 liter/min 2.39 liter/min 133 cc/min 83 cc/min 3.75 gpm 8.04 gpm 3.87 gpm 6.54 gpm 4.6 in. salt 12.4 in. oil 11.1 in. oil 11.3 in. oil 16.0 in. oil 6.1 in. salt 5.6 in. salt 5.1 in. salt 5.5 in. salt 5.6 in. salt 6.3 in. salt
9.1 in. H20 Main charcoal bed pressure drop
3.4 psi
80 T a b l e 4.23
U
&continued)
Identification
Scan Seq. No
PDT-960-A
230
Component coolant pump LIP
8.0 p s i
PT-HB-A
233
High bay pressure
-.07
PT-RC-A
348
Reactor c e l l pressure
-1.98 p i g
PT-500-A
223
H e l i u m header pressure
220 p s i g
PT-510
z7
O i l tank No. 2 pressure
6.8 p s i g
PT-511-C
225
Coolant drain tank pressure
-- p s i g
PT-511-D
218
Coolant drain tank pressure
5.7 p s i g
PT-5 13-A
226
O i l tank No. 1 pressure
7.7 p s i g
PT-516
347
Fuel pump pressure
5.0 p s i g
PT-517-A
224
Drain tank supply pressure
PT-522-A
13
Fuel pump pressure
8.5 p s i g 5.0 p s i g
PT-528
66
Coolant pump pressure
Description
.
,
Reading lo/ 12/69
in. HzO
4.6 p s i g
PT-572-B
219
Fuel drain tank No. 1 pressure
5.2 p s i g
PT-574-B
220
Fuel d r a i n tank No. 2 pressure
5.1 p s i g
PT-576-B
221
Fuel f l u s h tank pressure
5.3 p s i g
PT-589-A
53
Overflow tank pressure
7.2 p s i g
PT-592-B
33
Fuel pump pressure
5.6 p s i g
PT-608-B
222
Fuel storage tank pressure
-1.2
RE-NLC1-A
32
Reactor power
8.6 MW
RE-NLC2-A
36
Reactor power
8.6 MW
RE-OT-1-B
262
O i l tank No. 1 r a d i a t i o n
1.8 mr/hr
RE-OT-2-B
263
O i l tank No. 2 r a d i a t i o n
0.09 mr/hr
Reactor Power
8.5 MW
RE-SCl-AI.
9
LiJ
psig
RE-SlA
277
Containment stack alpha
14%scale
RE-S1B
278
Containment stack beta-gamma
7% scale
RE-s1c
270
Containment stack iodine
21% scale
RE-500-D
261
Cover gas supply r a d i a t i o n
0.3 mr/hr
RE-528-B
275
Coolant gas supply r a d i a t i o n
1.5 mr/hr
RE-528-C
276
Coolant gas supply r a d i a t i o n
2.'1 mr/hr
RE-557-A
273
Offgas from charcoal beds r a d i a t i o n
0.1 mr/hr
hd
81 Table 4.23
(continued) ~~
Identification
RE-557-B RE-565-C RE-596-A RE-596-B RE-596-C RE-675-A RE-675-B
283 284 285 286 287
RE-678-C
LJ
1
7
RE-827-C RM-NCC1-A6 RM-NCC2-A6
264 265 266 259 260
Description
Offgas from charcoal beds radiation Cell Cell Fuel Fuel
air radiation air radiation pump gas supply radiation pump gas supply radiation
Fuel pump gas supply radiation Sampler cold offgas Sampler cold offgas Sampler hot offgas Sampler hot offgas Process water radiation Process water radiation Process water radiation Reactor count rate Reactor count rate
RM-NCC1-A7
41
Reactor power
RM-NCC2-A7
42
Reactor power
RM-NCC1-A9
44
RM-NCC2-A9
45
SE-CP-G1-A
52
SE-F"-E l-A TE-AD1-1A TE-AD3-4 TE-AD3-SA TE-AD3-6 TE-AD3-7A
11 184
Reactor period Reactor period speed eed
187 188
TE-AD3-8A
190
TE-CDT-2A
181 182 127
TE-CDT-8
'L,
.
274 271 272 280 282
RE-565-B
RE-678-D RE-827-A RE-827-B
Scan Seq. No
TE-CP-A2B
185 186
Reading 10/12/69
0.1 mr/hr 1.0 mrlhr 1.4 mr/hr 0.1 mr/hr 0.1 mr/hr 0.1 mr/hr 1.1 mr/hr 4.2 mr/hr 2600 mr/hr 4000 mr/hr 25 mr/hr 26 mr/hr 34 mr/hr 10,000 cps 10,000 cps' 9.6 Mw 9.2 MW . ' -300 sec '
-150 sec
Radiator inlet air temperature Radiator outlet duct wall temperature
1775 rpm 1190 rpm 67°F 107°F
Radiator outlet duct wall temperature tlet duct wall temperature tlet duct wall temperature
120°F 129°F 173°F
tlet air temperature Coolant drain tank bottom temperature Coolant drain tank middle temperature
178°F 1210°F 1203°F
Coolant pump bowl-neck temperature
804°F
'
82 Table 4.23
(continued)
Identification
scan Seq. No
TE-CP-lA
126
Coolant pump flange-neck temperature
247°F
TE-CP-3A
128
Coolant pump bawl top temperature
955°F
TE-CP-8B
125
Coolant pump flange top temperature
102°F
TE-CP-SA TE-CPM-IA
129 124
Coolant pump bowl middle temperature
1018°F 96°F
TE-CTS-D
90
Coolant pump motor temperature Surveillance rig top zone temperature
TE-CTS-E
88 87 267 137 138 303 304 305 306 307 309 166 163 164 167 171 168 169 172 106 102 108 109 110
Surveillance rig mid zone temperature Surveillance rig bottom zone temperature
1230°F 1220°F
Diesel house ambient temperature Computer room ambient temperature
76°F 71"F
TE-CTS-F TE-DH-1 TE-DL-1
TE-DL-2 TE-DTC-1 TE-DTC-2 TE-DTC-3 TE-DTC-4 TE-DTC-5 TE-DTC-6 TE-FD1-lA TE-FD1-3A TE-FD1-12A TE-FD1-18A TE-FD2-lA TE-FD2-3A TE-FD2-12A
TE- FD2- 18A TE-FF100-4 TE-FF100-5 TE-FF100-6 TE-FF101-4 TE-FF101-5
.
Description
Reading 10/12/69
1200°F
Computer reference thermal plane temperature 69°F Drain tank cell ambient temperature
148°F
Drain tank cell ambient temperature Drain tank cell ambient temperature
144°F 151°F
Drain tank cell ambient temperature
145°F
Drain tank cell ambient temperature D r a h tank cell ambient temperature
149°F 150°F
Fuel drain tank No. 1 top temperature
1073°F
Fuel drain tank No. 1 bottom temperature
1149°F
Fuel drain tank No. 1 middle temperature
1178" F
Fuel drain tank No. 1 bayonet temperature
1158°F
Fuel drain tank No. 2 top temperature
1062°F
Fuel drain tank No. 2 bottom temperature
1130°F
Fuel drain tank No. 2 middle temperature
1159°F
Fuel drain tank No. 2 bayonet temperature Freeze flange 100 inner temperature
1138°F 904°F
Freeze flange 100 middle temperature
648°F
Freeze flange 100 outer temperature
559°F
Freeze flange 101 inner temperature
824°F
Freeze flange 101 middle temperature
585°F
LJ
83 Table 4.23
(continued)
_. Scan Identification Seq.
Description
Reading 10/12/69
NO TE-FF101-6
111
Freeze flange 101 outer temperature
TE-FF102-4
112
Freeze flange 102 inner temperature
868°F
TE-FF-102-5
113
Freeze.flange 102 middle temperature
631°F
TE-FF-102-6
114
Freeze flange 102 outer temperature
541°F
TE-FF200-4
115
Freeze flange 200 Inner temperature
769°F
TE-FF200-5
116
Freeze flange 200 middle temperature
546°F
TE-FF200-6
117
477°F
TE-FF201-4
118
Freeze flange 200 outer temperature Freeze f4ange 201 inner temperature
TE-FF201-5
120
Freeze flange 201 middle temperature
605°F
TE-FF201-6
121
Freeze flange 201 outer temperature
495°F
TE-FFT-1A
175
Fuel flush tank top temperature
1134°F
TE-FFT-2A
173
Fuel flush tank bottom temperature
1157°F
TE-FFT-10
Fuel flush tank middle temperature
1174°F
TE-FP-1A
174 92
Fuel pump neck-flange temperature
303°F
TE-FP-2A
93
Fuel pump neck temperature
TE-FP-3A
94
Fuel pump neck-bowl temperature
518°F 962°F
TE-FP-4A
98
Fuel pump bowl top temperature
1004°F
TE-FP-SA TE-FP-7B
95
Fuel pump neck-bowl temperature
984°F
er temperature
I
454°F
796°F
.1212"F
bot t-om temperature
1208°F
TE-FP-1OA
bowl temperature top temperature
950°F 992°F
TE-FP-11A
e top temperature
146°F
TE-FP-8B TE-FP-9A
TE-FST-10
117°F 81°F
TE-FV103-ZB
407°F
TE-FV104-1B
462°F
r temperature
TE-FPM-lA
center temperature
TE-FV105-2B
1214°F
TE-~~106-2 B TE-FV107-1B
1236°F
144
Freeze valve 107 shoulder temperature
102°F
84 Table 4.23
V
(continued)
Identification
scan Seq. No.
TE-FV107-2B
145
Freeze valve 107 center temperature
TE-FV107-3B TE-FV108-1B
146
Freeze valve 107 shoulder temperature
490°F 90°F
147 148 149 150 151 152 153 154 155 156 157 158 160 161 162 17 43 176 34 67 91 134 133 132 131 345 105 183
Freeze valve 108 shoulder temperature
77°F
Freeze valve 108 center temperature
446°F
Freeze valve 108 shoulder temperature
76°F
Freeze valve 109 shoulder temperature
103°F
'
478OF
Tdeeze valve 109 shoulder temperature
108°F
Freeze valve 110 shoulder temperature Freeze valve 110 center temperature
106°F 77°F
Freeze valve 110 shoulder temperature Freeze valve 111 shoulder temperature
88°F 88°F
Freeze valve 111 center temperature
80°F
Freeze valve 111 shoulder temperature
88°F
Freeze valve 112 shoulder temperature
89°F
Freeze valve 112 center temperature
83'F
Freeze valve 112 shoulder temperature
89OF
Freeze valve 204 center temperature Freeze valve 206 center temperature
254°F 303OF
High bay ambient temperature
83°F
HX fuel outlet nozzle temperature
1170°F
HX fuel inlet nozzle temperature HX shell center temperature
1207°F 1185°F
Main blower No. 1 bearing temperature
86°F
Main blower No. 1 bearing temperature
68°F
Main blower No. 3 bearing temperature Main blower No. 3 bearing temperature
93OF 68°F
Nuclear instrument penetration temperature
139°F
Overflow tank bottom temperature
1172°F
Particle trap temperature
203°F
TE-FV108-2B TE-FV108-3B TE-FV10 9-18 TE-FV109-2B TE-FV109-3B TE-FV110-1B TE-FV110-2B TE-FV110-3B TE-FV111-1B TE-FV111-ZB TE-FV111-3B TE-FV112-1B TE-FV112-2B TE-FV112-3B TE-FV204-2B TE-FV206-2B TE-HB-1 TE-HX-2A TE-HX-3A TE-HX-9A TE-MB1-1 TEiMB1-2 TE-MB3-1 TE-MB3-2 TE-NIP-2 TE-OFT-6B TE-PT2YM
Description
Freeze valve 109 center temperature
Reading 10/12/69
td
85
hi
able 4.23 Scan Identification Seq. No. TE-PT-2FM
165
TE-PT-2FF
170
.
Description -
I
-
.
Reading -10/12/69
Particle trap temperature
167°F
Particle trap temperature
94°F
Reactor top temperature
1189'F
TE-R-2
26
TE-R-4A
73
TE-R-15A
78
Reactor side temperature
1181°F
TE-R-18A
79
Reactor side temperature
1196°F
TE-R-2OA
80 82
Reactor side temper
1182'F
Reactor side temper
1178°F
Reactor bottom temperature
1182'F
Reactor bottom temperature
1182°F
TE-R-26A
Li
(continued)
1209°F
_*
TE-R-2 7A
55 83
TE-R-28A
84
Reactor bottom temperature
1181°F
TE-R-29A TE-R- 30A
56 85
Reactor bottom temperature
1180°F
TE-R-31A
59
Reactor Bottom temperature
TE-R-42A TE-R-45A
77
Reactor graphite tube temperature Reactor neck upper temperature
431°F 282°F
Reactor top temperature,
1185°F . .
Reactor top temperature
1191°F 1185°F
I
76
Reactor bottom temperature
~
,
1181°F 1180'F
i TE-R-49
8
TE-R-50 TEkR-51
10
12
TE-RC-1
290
ient temperature
139'F
TE-RC-2
291
nt temperature
143'F
TE-RC-3
292 294
ient temperature
149'F 135°F
295
nt temperature
138°F
296
Reactor ce
127°F
TE-RC-7
297
142°F
TE-RC-8
298
Reactor cell ambient temperature Reacto ient temperatur
TE-RC-9
299
Reacto
ient temperature
149°F 139°F
Reactor cell ambient temperature
153°F
Special equipment room ambient temperature
101°F
i
b.
TE-RC-10 TE- SER-1
301 178
-
\
86
Table 4.23 Scan Identification Seq. NO
(continued)
Description
Reading ; 10/12/69
TE-TRM-1
177
Transmitter room ambient temperature
81°F
TE-VH-1
180
Vent house ambient temperature
76°F
TE-VT-1
143
Vapor tank water temperature
63°F
TE-VT-2
289
Vapor tank air temperature
66°F
TE-100-Al
5
Line 100 temperature
1207°F
TE-10042
25
Line 100 temperature
1207°F
TE-100-A3 TE-100-lA
46
Line 100 temperature
1208°F
68
Line 100 temperature
1209°F
TE-100- 3A
70
1208°F
TE-101-SA
60
Line 100 temperature Line 101 temperature
TE-102-A4A
72
Line 102 temperature
1167°F
TE-102-lA
71
Line 102 temperature
1165°F
TE-102-5D
6
Line 102 temperature Line 200 temperature
1167°F
64 22
Line 200 temperature
1022°F
Line 201 temperature
1068°F
Line 201 temperature
1068°F
TE-201-A2B
20 18
Line 202 temperature
i0ll"F
TE-20l-MC
16
Line 202 temperature
i0lO"F
TE-201-B11B
123
Line 201 temperature Line 201 temperature
1068°F
TE-200-C7A TE-200-20A TE-2 01-AlB TE-201-Alc
122
1210°F
1014°F
TE-201-1B
63
TE-202-Al
7
Line 202 (well) temperature
997°F
TE-202-B1
27
Line 202 ( w e l l ) temperature
1007OF
TE-202-D1
48
Line 202 (well) temperature
1007°F
TE-522-1
135
Line 522 temperature
84°F
TE-524-1
136
Line 524 temperature
101°F
TE-556-lA
201
Line 556 temperature
94°F
TE-702-1B
192
Line 702 temperature
132°F
TE-705-lA
193
Line 705 temperature
142°F
TE-707-1A
194
Line 707 temperature
140°F
TE-75 2-1B
195
Line 752 temperature
123°F
1072°F .. '
87 Table 4.23
a-,
(continued)
TE-755-lA
196
Line 755
126°F
TE-575-lA
197
Line 575 temperature
1270F
TE- 791-1
140
Line 791 temperature
102°F
TE-795-1
Line 795 temperature
146°F
TE-804-1
141 215
Line 804 temperature
100°F
TE-805-1
216
Line 805 temperature
104°F
TE-811-1
211
Line 811 temperature
81°F
TE-813-1 TE-826-1
212
79°F
202
Line 813 temperature Line 826 temperature
99°F
TE-831-1
206
Line 831 temperature
102°F
TE-833-1
213
Line 833 temperature
99°F
TE-837-1
203
Line 837 temperature
102°F
TE-841-1 TE-845-1
205
Line 841 temperature
106°F
TE-846-1
207 204
Line 845 temperature Line 846 temperature
124°F 107°F
TE-85 1-1
210
Line 851 temperature
79°F
TE-874-1
208
Line 874 temperature
118°F
TE-876-1
214
Line 87
106OF
TE-916
200
Line
335°F
124°F
TE-917 TE-922
199
Line 922
118°F
WM-CDT WM-FDl WM-FD2
246 243
Coolant d Fuel dra
152 lbs 0 lbs
244
Fuel d
468 lbs
W-FFT
245
Fuel f
8800 lbs
WM-FST
247
Fuel s
XPM-201
268
Reactor
0 lbs 7.6 MW
ZM-FC1
257
Fission ch
ZM-FC2
258
ZM-NCRl
19
Fission chamber No, 2 position Compensated ion chamber No. 1 position
No. 1 position
61 in. 66 in. 36 in.
.
88
Table 4.23 .
Identification
Scan Seq. No.
.'., w
(continued)
*
Description
Reading 10/12/69
Compensated ion chamber No. 2 positian Compensated ion chamber No. 3 position
44 i n .
ZM-NCR3
21 23
ZT-ID
37
Radiator i n l e t door position
89 i n .
ZT-OD
38
Radiator o u t l e t door position
85 i n .
ZM-NCR2
44 i n .
-_
89
5.
FUEL SYSTEM
J. L. Crowley R. H. Guymon
5.1
C. H. Gabbard J. K. Franzreb
. Description
The fuel-circulating loop consisted of a graphite-moderated r e a c t o r ,
a c e n t r i f u g a l type f u e l pump with an overflow tank, and a shell-and-tube h e a t exchanger, all connected by 5-inch Hastelloy-N piping.
The normal
operating temperature was about 1200°F and f u e l o r f l u s h salt w a s c i r c u l a t e d
at 1200 gpm.
When t h e r e a c t o r w a s not i n operation, t h e s a l t was drained
t o one o r both of t h e f u e l drain tanks o r t o t h e f u e l f l u s h tank.
Inter-
connecting salt piping and freeze valves permit f i l l i n g t h e r e a c t o r o r t r a n s f e r r i n g salt between tanks by manipulating valves i n t h e helium supply and vent l i n e s . 5.2
Purging Moisture and Omgen from t h e System
I n t h e f a l l of 1964, before s a l t was charged i n t o t h e d r a i n tanks, oxygen and moisture were purged from t h e f u e l c i r c u l a t i n g system and t h e drain tanks.
This w a s accomplished by pressurizing t h e system with helium
t o assure t h a t it was l e a k - t i & t and purging with dry h e l i
followed by a combination of evacuation
efore and during t h e heatup.
Details of t h e
. heatup a r e given i n Chapter 17. The helium w a s introduced a t t h e f u e l - pump which was i n operation t o provide c i r c u l a t i o n i n t h e loop. The system w a s vented o r evacuated at t h e normal f u e l pump offgas l i n e (518) o r at t h e provided a longer flow path Figure 5-1 gives t h e i n t h e e f f l u e n t gas were The system w a s evacuated i n g t o 1130’F d i d not c
Later analyses of salt s effective.
a more e f f e c t i v e purge. f operations used.
Peaks of moisture
about 250°F and again at about. 650OF.
r each of -these moi.sture peaks.
Further heat-
s i g n i f i c a n t addikional moisture releases. a t e d t h a t t h e purging’had been very
. .
.
.E .-cE c
._ 3
5
s
0
‘
c- 4000 5 5 a
E
750
I
,....
.~
’A
I -
//
-1
REACTORVESSEL TEMPERATURE
j
-
500
/ ’ ,
4
I
250
/
! A
0 20
22
24
26 OCT 1964
Fig. 5 . 1
C
. . ,
10
5
W
.
28
30
I’
I
I
3
I
I
5
I
I
7
I
9 NOV 4964
I
I
I
11
I n i t i a l Removal of Moisture from Fuel System
c
I
I
13
I
I
45
I
.
. _ .
.
1
91
bi
5.3
Fuel-Circulating-System Volume Calibration
After t h e f l u s h s a l t had been added t o t h e d r a i n tanks and t r a n s f e r s made t o f i l l t h e freeze valves (see Section 5.11), system was f i l l e d and operated f o r 8 days.
the f u e l circulating
It was then drained and f i l l e d
s e v e r a l times t o check t h e d r a i n times and c a l i b r a t e the system.
The C a l i b r a t i o n w a s done by increasing t h e d r a i n tank pressure i n increments and recording t h e d r a i n tank weight and t h e d i f f e r e n t i a l pressure between t h e d r a i n tank and t h e f u e l pump.
This i s p l o t t e d i n Fig. 5-2.
The system
w a s purposely o v e r f i l l e d t o determine t h e p o s i t i o n of t h e fuel-pump overflow l i n e and t o test t h e overflow-tank l e v e l indicators.
Overflow oc-
curred a t readings on the two fuel-pump bubblers of 89.5 and 91%. A t o t a l of 105 l b of s a l t t r a n s f e r r e d t o t h e overflow tank produced a reading on t h e l e v e l instrument of 11.5% o r 4.2 inches o f g a l t , i n acceptable agreement with t h e calculated response.
The temperature d i s t r i b u t i o n i n t h e d r a i n valve (FV-103) w a s cont r o l l e d so t h a t it would thaw i n 9 t o l l m i n u t e s a f t e r an emergency d r a i n
signal.
(See Chapter 20.)
The t i m e required t h e r e a f t e r f o r t h e s a l t t o
I
d r a i n from t h e loop depended on which valves were open i n t h e s a l t and gas lines.
Normally the f r e e z e valves t o both d r a i n tanks w e r e kept thawed,
but f o r a while ' a f t e r a s a l t f i l l one valve would s t i l l be frozen. emergency d r a i n s i g n a l acted t
An
t h e f r e e z e valves and t o open valves
i n d r a i n tank ven
equalizer l i n e s between t h e gas i n t h e
f u e l pump and i n t h e d r a i n t
t i t was considered possible t h a t one
o r more of these va
en.
t o measure d r a i n t i
On a normal a f t e r FV-103 thaw
Tests were conducted, therefore,
nceivable combinations ..of valving. loop drained completely i n 9-11 min. zer valves open but one f r e e z e valve
kept frozen, t h e d r a i n t i m
about 30 minutes, regardless of whether o r
not t h e vent valv
With t h e equalizers closed, one f r e e z e
valve frozen and
t i m e s were deemed acceptable.
d r a i n time was 41 min.
These d r a i n
-
92
ORNL-DWG 73-597
24 U v)
n
Y
s
22
e.20
4a.
2
18
a Y
f za
0:
l6
14
0
z
E
12
w
a
IO
2i 8
6 0
2000
4Ooo
6000
8000
10,m
POUNDS OF SALTTRANSFERRED
Fig. 5.2
Fuel Circulating System Calibration
'
c
93
5.5
Mixing of Fuel and Flush S a l t s
During 235U f u e l operation, t h e uranium.concentration of t h e f l u s h
salt increased an average of about 215 ppm each time it was-used after f u e l s a l t operation.
D u r i ~ ~ g operation, - ~ ~ ~ U w i t h a lower uranium concen-
tration i n the fuel, the
39 ppm.
responding increase should have been only
"he a c t u a l increases during t h e t h r e e flushes a f t e r 233U fuel c i r -
culation were 36, 42, and 39 ppm.
1
Using these figures, approximately
1
40 lbs of f u e l salt mixed w i t h t h e flush salt during each flushing opera.tion.
i
I
(For more d e t a i l s on s a l t . a n a l y s i s and i n t e r p r e t a t i o n of r e s u l t s ,
I
see Reference 23.) System Leak During operation t h e c e l l - a i r a c t i v i t y w a s continuously monitored t o detect any leaks from t h e primary system.
No leaks were detected u n t i l after t h e f i n a l fuel salt drain i n December, 1969, A t t h i s time t h e c e l l -
air a c t i v i t y d i d increase which indicated a leak.
Subsequent t e s t s showed
t h a t t h e leak w a s at o r near one of t h e drain-tank freeze valves (FV-105). The a c t i v i t y w a s mostly xenon w i t h some iodine, krypton, and noble metals.
Four days a f t e r t h e first r e l e a s e , t h e r e w a s less than 25 curies of xenon and less than 50 m i l l i c
ne i n t h e c e l l .
t h e atmosphere without s a f e t y limits .24
r e l e a s e rate permitted by t h e MSRE
The p r o b a b i l i t y of
review of t h e operation thermal s t r e s s e s .
u l t i n g from corrosion seems remote.
between t h e air shroud s p e c i f i e d as a f'uI.1 penetr
A
e freeze valve does not i n d i c a t e any excessive
No
x-rays and other inspect
This w a s released t o
eports
.
ere found upon reviewing t h e construction It w a s noted, however, t h a t t h e weld
s a l t piping a t t h e freeze valve w a s not eld.
This l e d t o t h e suspicion that t h e
eak may be a crack t h a t started a t t h i s point and w a s propagated by s t r e s s
cling.
Determinatio
attempted during t h
ocation and nature of examinations.
More i
on t h e preliminary evaluation of t h e leak i s given i n Reference 25.
id
,
;
94
5.7
Oper a t i o n
The operation of t h e f u e l system w a s s a t i s f a c t o r y .
D i f f i c u l t i e s en-
countered with t h e components are described i n the following sections.
The
rate of t r a n s f e r of salt t o the overflow tank w a s higher than expected. Details on t h i s and t h e loop void f r a c t i o n and xenon poisoning are given i n Reference 18.
5.8 5.8.1
Fuel Pump and Overflow Tank
Description The f u e l salt c i r c u l a t i n g pump w a s a centrifugal sump-type pump w i t h
an overhung impeller, developed at ORNL expressly f o r c i r c u l a t i n g molten fluoride salts of t h e type used at t h e MSRE:.26
(See Fig. 5-3 and 5-4.)
At t h e normal operating speed of 1189 rpm, it had an output of about 1250 gpm at a 48.5-ft
salt head.
About 50 gpm of t h e pump output was circulated
i n t e r n a l l y t o t h e pump bowl via a spray r i n g t o promote t h e release of entrained o r dissolved gaseous f i s s i o n products.
The gas space i n
t h e pump bowl w a s purged w i t h helium t o sweep these t o t h e offgas disposal system.
The helium w a s introduced j u s t below t h e s h a f t s e a l i n the bearing
housing.
Most of t h e gas flowed downward through t h e labyrinth between t h e
shaft and the s h i e l d block t o prevent radioactive gas and salt mist from
reaching t h e seal.
The remainder flowed upward t o prevent any o i l which
leaked through t h e seal Prom g e t t i n g i n t o t h e pump bowl.
O i l f o r lubri-
cating t h e bearings and cooling t h e s h i e l d plug w a s recirculated by an external pumping system.
H e l i u m bubbler type instruments w e r e used t o
measure t h e l i q u i d l e v e l as a means of determining t h e inventory of salt i n t h e f u e l system.
Small capsules were lowered i n t o the bowl t o take
samples for analysis o r t o add f u e l salt. The maximum height of t h e l i q u i d i n t h e pump bowl w a s l i m i t e d by the top
of an overflow l i n e (5-1/2 i n . above t h e center l i n e of t h e volute) which connected t o a 5.5-ft3 overflow tank located beneath t h e pump.
H e l i u m bub-
b l e r type instruments were a l s o used f o r measuring t h e l i q u i d l e v e l i n t h e overflow tank.
Since the overflow l i n e extended t o t h e bottom of t h e over-
f l o w tank, closing. a valve i n t h e overflow tank vent l i n e forced t h e salt back t o t h e f u e l pump.
6,
I
Fig. 5.3
ORNL- LR. m0-56043-0
MSRE Fuel Pump
I-
LL
a I v)
96
,
97
5.8.2
Early Operation The f u e l pump had been loop-tested with molten salt f o r 100 h r s at
1200'F before it w a s i n s t a l l e d a t t h e MSRE i n October, 1964.
A f t e r instal-
l a t i o n no s i g n i f i c a n t difficulties were encountered during helium circulat i o n while t h e f u e l system was being purged, during e a r l y flush salt operat i o n which s t a r t e d on January 12, 1965, o r during the c r i t i c a l i t y experiments. A continuous, but very slow, accumulation of salt i n t h e overflow tank w a s
observed throughout t h e early'operation.
5.8.3
(See 5.8.9.)
Examination of Fuel Pump afier C r i t i c a l i t y ExperimentP7 llhe fuel-pump rotary element was removed f o r inspection i n September 1965
at t h e end of Run 3.
The pump had been used f o r c i r c u l a t i n g helium f o r 1410
hours and it -had been f i l l e d with. salt at .the MSRE f o r 2120 hours, 1895 hours of which t h e s e t w a s being t i o n and appeared rea*
The pump was generally f o r Pull-power operation.
t o be used
The only dimensional change w a s a 0.006-in.
gro.cith
of t h e pump tank bore diameter where t h e upper O-ring mated with t h e pump tanlr. The most s i g n i f i c a n t discovery w a s evidence of a small o i l leak through the gasketed Joint at t h e catch basin f o r t h e lower o i l seal. This o i l had
run down t h e surface of t h e s h i e l d plug, where it had become coked by t h e higher temperatures near t h e bottom.
Some of t h e o i l had reached t h e upper h i e l d plug and had become coked i n t h e aked p a s t t h e O-ring during high-
.
il w a s observed below t h e r i n g af'ter the decontamination c e l l , but it w a s
t h e open l i n e s during the t r a n s f e r t o J
ions, considerabl c u l t y w a s enn offgas system, due largely t o decomleaked i n t o t h e pump l e , t h e gasketed Joint' w a s seal-welded ver, t h e spar6 never had t o be i n s t a l l e d
at t h e MSRE.
i
i
/
1
98
A l a y e r of flush salt about 3/8-inch deep containing about 40 i n . 3
w a s trapped and frozen on top of t h e labyrinth flange.
Apparently t h e
salt had drained through t h e l/&inch diameter holes i n t h e flange u n t i l
surface tension e f f e c t s balanced the hydrostatic head.
This l a y e r can be
seen i n Fig. 5-5, w h i c h i s a photograph taken during the inspection.
This
photograph a l s o shows the contrast between t h e surfaces exposed t o t h e
s a t , which'were b r i g h t but not corroded, and those above the salt, which were d i s t i n c t l y darker.
There w a s a coating of f i n e salt mist p a r t i c l e s
on the lower face of t h e s h i e l d plug and i n an "0"-ring
groove around t h e
s h i e l d plug t h e r e were s m a l l amounts of f l u s h and f u e l salt t h a t must have been transported as mist. The pump was r e i n s t a l l e d using remote maintenance techniques s o t h a t
these techniques could be evaluated.
Four universal j o i n t s on t h e flange
b o l t s t h a t had been found broken during the disassembly were repaired p r i o r t o t h e r e i n s t a l l a t i o n of t h e r o t a r y element.
These failures r e s u l t e d from
excessive b o l t torque t h a t had been used e a r l i e r t o obtain an i n i t i a l seal
on t h e flange.
( I t turned out t h a t the jack screws had not been f u l l y
backed o f f before t h e flange b o l t s were tightened.)
5.8.4
Ftmp and Pipe Support Problem
A problem r e l a t e d t o t h e o v e r a l l fuel-pump i n s t a l l a t i o n became evident during the p o s t - c r i t i c a l i t y shutdown i n t h e f a l l of 1965.
The f u e l pump
could move i n t h e horizontal plane, b u t w a s fixed against v e r t i c a l movement. The heat exchanger could move horizontally; t h e heat exchanger support frame
w a s fixed against v e r t i c a l movement at t h e north end, but w a s mounted on spring supports at t h e south end.
The south end of t h e heat exchanger w a s
coupled t o t h e fuel-pump bowl by a s h o r t length of 5-in. Sched.-bO pipe. The heat exchanger w a s supported from below, and t h e pump bawl w a s supported from above.
The connecting piping was supposed t o move upward at t h e heat
exchanger and downward at t h e pump when t h e system w a s heated. Because of the physical arrangement of t h e piping and equipment, s t r e s s ranges i n the piping due t o thermal cycling were calculated t o reach a
maximum of 20,000 p s i , which is acceptable.
Uncertainties e x i s t e d i n
c r i t i c a l p a r t s of t h e heat exchanger, however, p a r t i c u l a r l y i n t h e nozzle
0
66
100
W
where estimates were as high as 125,000 p s i during cycling from150 and The end of t h e exchanger toward t h e pump bowl w a s therefore
130O0F.
mounted on springs.
This should have reduced s t r e s s e s i n t h e nozzle t o t h e
range of 20,000 t o 70,000 p s i . C a r e f u l observations during a heating cycle t o 1200째F, showed t h a t t h e
equipment did not move as expected, however.
Because of t h e complex equip-
ment configuration and t h e inevitable uncertainty of t h e calculations, it
w a s decided t o make s t r a i n gage measurements with t h e equipment cold and moving t h e piping by mechanical means f o r measured distances with measurable loads.
The highest stresses w e r e found t o be i n t h e f i l l e t where t h e
nozzle w a s welded t o the heat exchanger head when a spring force of 2000 pounds w a s exerted t o raise t h e end of t h e heat exchanger 3/16 i n .
The
measured s t r e s s i n t h e f i l l e t w a s 13,000 p s i and w a s a f a c t o r of about
5 greater than t h e stress i n the nearby piping. Dye-check o f t h e nozzle t o t h e head of t h e heat exchanger showed no indication of cracks. The conclusion fromthese tests w a s that t h e mounting arrangement w a s
adequate t o allow t h e system t o go through more than t h e 50 thermal cycles required without a fatigue failure.
The system w a s then put t o use f o r
power operation.28
5.8.5
Effect of Bubbler Flow Rates on Indicated Fuel-Pump Level The fuel-pump l e v e l w a s determined by bubbling helium through t h e salt
and measuring t h e d i f f e r e n t i a l pressure between t h i s l i n e a&
a reference
l i n e w h i c h connected t o t h e gas space of t h e pump (see Fig. 5-6).
The end
of one of t h e bubbler dip tubes (596) w a s 1.874 i n . lower than t h e other
(593). A common reference l i n e (592) w a s used f o r t h e two bubblers.
The
l e v e l readout instrument had a full s c a l e (0-100$) range of 10 inches of The centerline of t h e volute w a s at 35%. Compensation w a s provided
salt.
i n t h e instrumentation f o r changes of density between flush and f u e l salt and f o r differences i n t h e lengths of t h e b d b l e r tubes. Since t h e d/p c e l l s used to indicate l e v e l were located outside t h e reactor c e l l , t h e r e w a s some pressure drop i n t h e l i n e s between these and t h e pump. lines.
I
The amount of pressure drop w a s dependent upon flows through t h e
T e s t s were made e a r l y i n 1965 t o e s t a b l i s h t h e relationship between
these flow rates and t h e indicated levels.
I
1
W
101
ORNL-DWG 70-5193
Fig. 5.6
Schematic Representation of Fuel Pump Bowl Level and Density Indicators
102
Data f o r t h e upper probe are presented i n Fig. 5-7. t h e lower probe were similar.
The curves f o r
It can be seen t h a t t h e indicated l e v e l de-
creased w i t h increasing pump-bowl reference l e g flow and t h a t t h e indicated l e v e l increased w i t h increased f l o w i n t o t h e dip tubes. l e v e l w a s not changed during t h e tests.
The actual s a l t
Normal operating conditions were
s e t at 25-psig forepressure on all three of t h e bubbler f l o w elements,
5 psig i n t h e fuel-pump cover gas. 5.8.6
Fuel-Pump Level Changes and Limitations Based on t h e recommendations of t h e pump development group at ORNL, the
l e v e l alarm and interlocks were o r i g i n a l l y s e t as shown i n Table 5-1. The differences i n t h e l e v e l a t which the f u e l pump could be s t a r t e d
(64%) and
t h e l e v e l at which t h e pump would-automatically s t o p (55%) w a s
necessary because t h e indicated l e v e l decreased 10 t o 12% immediately upon s t a r t i n g t h e pump.
This w a s due t o f i l l i n g t h e spray r i n g and fountain flow
chamber. The narrow differences between the high and low alarm and control setpoints caused considerable operational d i f f i c u l t i e s .
Prior t o s t a r t i n g the
pump after a fill, t h e average loop temperature could not be accurately determined.
Since t h e f'uel-pump l e v e l changed about 12.4% p e r 100째F change
i n loop temperature, the s e l e c t e d fill point w a s not a l w a y s s a t i s f a c t o r y f o r operation.
I n which case t h e freeze valve had t o be thawed and t h e system
l e v e l adjusted.
During operation t h e reactor o u t l e t temperature w a s usually
held constant when t h e power w a s changed.
Therefore, t h e bulk average tem-
perature changed w i t h power changes and t h i s caused changes i n t h e fuel-pump level.
I n addition t o t h i s , experiments were run which required operation
at d i f f e r e n t reactor o u t l e t temperatures. During load and rod s c r a m t h e system cooled rapidly. Sometimes it w a s necessary t o reheat t h e system before r e s t a r t i n g t h e pump.
The operating l e v e l s were f u r t h e r restri'cted
when it appeared t h a t t h e offgas plugged more rapidly when t h e salt l e v e l w a s above 60 t o 65% and gas entrainment i n t h e f u e l loop seemed t o increase below about 50%. The problem w a s f u r t h e r compounded by t h e changing pumpbowl l e v e l due t o salt t r a n s f e r t o t h e overflow tank.18
103
\
c, ORNL-DWG 73-598
74
73
72 71 70 69
68 67 66 65 Fl-593 (pig) NOTES:
FI-592 and FI-593 flow rates were proportional t o the difference between the pressure upstream o f the flow restrictors and the pressure in the fuel pump. The fuel pump pressure was held constant a t 5 psig during these tests. LR-593C was read from the bottom o f the inked space which was about 1.5% i n width.
Fig. 5.7
Effect of Bubbler Flow on Fuel Pump Level Indicators
. .
/
104
Table 5-1.
Level
+78%
' 475%
+64%
Original FP Level Interlocks
Action and Reason f o r Interlock Stops f i l l , gives a temperature setback and rod reverse t o prevent o v e r f i l l i n g of t h e Fuel pump (overflow point i s about 90%). Annunciation hnnp cannot be s t a r t e d b e l o w t h i s l e v e l .
This i s t o
prevent cavitation.
. +55%
Annunciation
+
Pump w i l l s t o p below t h i s level.
+53%
,
prevent cavitation.
Again, t h i s is t o
Q,
105
bi
Tests were therefore performed whereby t h e interlocks were bypassed and t h e pump w a s started and operated at lower l e v e l s .
Since t h e r e appeared t o be no cavitation o r adverse e f f e c t s on t h e pump, t h e interlocks given i n Table 5-1 were changed t o 78%, 75%, 55%, 40%, and 38% l a t e i n 1965. Normal l e v e l during operation w a s s t i l l maintained between 50% and 60% due t o above considerations.
However, recovery after a load and rod scram w a s
much easier. 5.8.7
Coolant Air t o Fuel P m p
I n t h e design of t h e MSRE, it was calculated t h a t t h e upper portion of t h e fuel-pump tank would be subject t o s u b s t a n t i a l heating from f i s s i o n products i n t h e gas space above t h e salt. pump tank would be limited
Since t h e useful l i f e of t h e
thermal-stress considerations at t h e junction
of t h e volute support cylinder w i t h t h e spherical t o p of t h e tank, close control was at first maintained over t h e temperatures i n t h i s region. ,
Design s t u d i e s had indicated t h a t t h e maximum l i f e t i m e would result if the junction temperature were kept about 100째F below t h e temperature on
t h e tank surface 6-in. out from t h e junction.
Component cooling "air"
(95%N 2 ) was provided t o maintain t h i s temperature d i s t r i b u t i o n .
A secondary
consideration i n controlling t h e temperatures w a s a desire t o keep as much of t h e pump tank as PO
l e above t h e liquidus temperature of t h e
salt. I n operating t h e r e a c t o r , it would be i d e a l i f a fixed flow rate of
air over t h e pump
ovide a s a t i s f a c t o r y temperature d i s t r i b u t i o n
f o r all conditions
gn calculations indicated that t h i s condition
could be met with an ai
of 200 c f ~ n . However, ~ ~ temperature measure-
ments on t h e pump-test
d during t h e i n i t i a l heatup of t h e MSRE indi-
cated t h a t only about
ould be required, and t h a t the air would have
t o be turned o f f when t h e
To minimize t h e t e
air f l o w during power
tank w a s empty. e f f e c t s when t h e cooling air w a s turned on,
as t o have been set at t h e m i n i m t h a t
ould give t h e desired temperatures.
It w a s found t h a t an air flow of 30 cfIn
gave a s a t i s f a c t o r y temperature d i s t r i b u t i o n at a l l power l e v e l s .
"
I n order
t o achieve and control t h i s r e l a t i v e l y low flow rate, a new valve, having
106
a lower Cv had t o be s u b s t i t u t e d f o r t h e o r i g i n a l valve.
L+
Figure 5-8
shows t h e temperatures i n the two regions of i n t e r e s t as a function .of reactor power l e v e l with t h a t air flow.
The v a r i a t i o n s i n t h e individual
temperatures were caused by v a r i a t i o n s i n t h e pump-bowl.leve1 salt tempera-
ture and air flow.
Both the individual temperatures and t h e temperature
differences increased l i n e a r l y w i t h power, as expected.
It w a s f e l t during
e a r l y MSRE operation t h a t , although temperature differences would exceed 100째F at f u l l power, t h e r e a c t o r could be so operated w i t h t h e 30-cfm air Mow without s i g n i f i c a n t l y reducing the l i f e of the pump tank.
When t h e r e a c t o r w a s s t a r t e d up f o r R u n 8 i n Septeniber 1966, an unexplained'shif't downward i n these temperatures w a s noted.
Later t h e cooling
air t o t h e pump tank w a s turned o f f during t h e attempts t o m e l t out t h e salt plug i n t h e 522 l i n e , and although the temperatures on t h e pump-tank surface were higher than w i t h t h e cooling air, t h e temperature gradient w a s less. Since t h e temperature d i s t r i b u t i o n w a s as good o r better than w i t h t h e air cooling, t h e use of air cooling w a s discontinued.
5.8.8
S a l t Transfer t o the Overflow Tank E a r l y operation of t h e r e a c t o r showed t h a t by some unexplained mecha-
n i s m , salt gradually accumulated i n t h e f u e l pump overflow tank even when t h e salt l e v e l i n t h e pump bowl w a s w e l l below t h e overflow point.
The
t r a n s f e r rate depended on salt l e v e l , and t h e t r a n s f e r ceased when t h e l e v e l w a s about 3 i n . below the overflow point.
This s i t u a t i o n e x i s t e d
u n t i l about A p r i l 1966, when transfer began t o be observed at lower salt levels.
The rate appeared t o increase gradually as t i m e went on u n t i l
it leveled o f f i n June and J u l y a t 0.57 l b of salt per hour, independent of salt l e v e l as far down as 4.7 i n . below t h e overflow point.
The change
occurred at t h e time of t h e stepwise increase i n power, but no mechanism connecting t h e two has been i d e n t i f i e d .
This rate of t r a n s f e r continued
through t h e 235U operation and required emptying t h e overflow tank about t h r e e times per week. P r i o r t o operation w i t h 233U f u e l salt, t h e flush salt which had been processed t o remove t h e 235U was c i r c u l a t e d f o r 42 hours.
During t h i s and
t h e first 16 hours of f u e l c a r r i e r salt c i r c u l a t i o n , t h e t r a n s f e r rate and
LJ
c
â&#x20AC;&#x2DC;C
ORNL-DWG 66-4 4444
800
_
0
4
Fig.
5.8
2
Variation Power.
3
4 POWER
5
6
7
(Mw)
of Fuel-Pump Tank Temperatures with Cpoling-air flow, 30 cfm.
Reactor
8
108
loop void fraction appeared t o be unchanged from t h e previous periods. hours after t h e start of a =-hour
Two
exposure of a beryllium rod i n t h e pump
bowl, t h e bubble fraction i n t h e system increased from t h e normal 0.1 vol
% and t h e rate of t r a n s f e r t o t h e overflow from about 0.4 lb/hr t o greater than 4 lb/hr.
t o about 0.6 vol
%
tank increased
During t h e remainder of t h e 233U operation t h e t r a n s f e r rate was high
(up t o 72 l b / h r ) and variable.
The results of t h e investigation of t h e
overflow rate, the changing brrbble fraction, and subsequent power perturbations are given i n Referen6e 18.
Burps of t h e Overflow Tank
5.8.9
During early operation of t h e MSRE, when t h e overflow rate was very low and t h e need t o push salt from t h e overflow tank back t o t h e fuel-pump bowl
w a s infrequent, t h e practice w a s t o empty the overflow tank completely.
The
sudden pressurization of t h e f i e 1 pump at t h e end of t h e burp gave false l e v e l indications and stopped t h e pump.
Also when power operation w a s
started, it w a s noted that gaseous f i s s i o n products were being pushed out
the o i l seal l i n e (524) by t h e sudden pressurization.
Therefore, t h e
procedures w e r e changed such t h a t t h e overflow tank w a s not completely emptied of salt during operation.
In February 1969, t h e main offgas l i n e plugged t o t h e point where it presented a 4-psi pressure drop t o t h e normal 3.2 l i t e r / m offgas flow. During a burp of t h e overflow tank, t h i s plug b l e w out w i t h t h e r e a c t o r at fUl power r e s u l t i n g i n a complete burp of t h e overflow tank.
On at least
t h r e e other occasions when t h e overflow tank w a s being emptied, t h e offgas plug b l e w out causing more salt t o be t r a n s f e r r e d than planned. 5.8.10 -,
Variable-Speed Drive f o r t h e Fuel PUQ
P r i o r t o February 1969 t h e f u e l pump w a s a l w a y s operated with the
normal 60-HZ power supply at
- 1189 rpm.
Then, i n order t o investigate the
e f f e c t of fuel circulation rate on system b e h a k o r (bubble ingestion, xenon stripping, t r a n s f e r t o overflow tank, b l i p s ) , l e a variable-speed motorgenerator set w a s brought t o the,:.MSRE t o supply power t o t h e f u e l pump during experiments.
As described i n Chap. 16 ,â&#x20AC;&#x2DC;considerable e f f o r t w a s
u
109
W
expended i n modifying and repairing t h e M-G set t o obtain s a t i s f a c t o r y reliability.
The pump i t s e l f , however, operated with no d i f f i c u l t y f o r
s u b s t a n t i a l periods at speeds between 50% and 105% of normal.
5.8.11
Flooding of t h e Pump Bowl w i t h Flush S a l t
A t t h e end of Run 7 i n July 1966, t h e f u e l loop w a s f i l l e d with f l u s h
salt t o r i n s e out residual pockets of f u e l salt and thus reduce t h e radiation l e v e l s f o r t h e scheduled work i n t h e reactor c e l l .
As t h e salt
w a s overflowing i n t o t h e overflow tank t o r i n s e it out, t h e l e v e l i n t h e f l u s h tank w a s lowered too far which exposed t h e bottom of t h e dip tube.
.
A l a r g e bubble of helium gas at a pressur,e of about 30 psig entered t h e
As t h e bubble rose, i t s volume increased due t o t h e
bottom of t h e loop,.
decreased pressure and t h e fuel-pump l e v e l increased f a s t e r than it could overflow t o the overflow tank.
S a l t flooded t h e reference bubbler, t h e
annulus around t h e s h a f t , t h e sampler tube, and t h e main offgas l i n e . Several factors contrib
4J
vented f i l l i n g t o t h e over flushing-of t h e overflow tank.
d t o t h e accident.
These had been bypassed t o allow
line.
The flush-salt inventory w a s rn-ginal
normal operating temperatures (1200'F). . low (about 114OOF) which mad
Interlocks normally preat
The flush-tank temperatures were
t h e salt more dense and thus there w a s an
i n s u f f i c i e n t volume. 5.8.12
PluRRing of t h e Offgas Line at t h e Fuel Pump
Intermittent problems were encountered w i t h plugging of t h e 1/2-in. point ,lust beyond where it ~ched-40 main offgas l i n e a t
I n t h e e a r l y years of operation, t h i s e i t h e r melted @;ht t o power, o r it w a s periodically i t s e l f f r e e when t h e re shut down, through t h e use of a mechanical, reamed out whe
t h e fuel-pump bawl.
f l e x i b l e r o t a t i n g "snake"
J u l y 1969, a specially b u i l t heater assembly
consisting of two 1000
ds w a s i n s t a l l e d remotely k m d t h e
522 offgas l i n e between t h
e pump bowl ahd t h e gas cooling "shroud" back-blowing with helium, w a s suc-
of t h e fuel pump. cessful i n c l e
This
-
are given i n Chapter 8.
t a i l s on t h e offgas plugging problems
110
5.8.13
Conclusions and Recommendations
The fuel pump w a s used t o c i r c u l a t e salt f o r 21,788 hours at temperat u r e s near 1200째F with no perceptible change i n performance and no failure. Some plugging of t h e offgas l i n e w a s encountered. sidered i n future designs.
This should be con-
Perhaps dual l i n e s could be used w i t h heaters t o
m e l t out plugs i f they develop.
Leakage of o i l (1t o 2 cc/day) i n t o t h e
pump bowl contributed t o t h e offgas plugging problem.
This p o s s i b i l i t y w a s
eliminated i n t h e replacement rotary element by seal-welding a gasketed j o i n t , but because the plugging problem w a s manageable, the spare element
w a s never i n s t a l l e d . The t r a n s f e r of salt t o t h e overflow tank w a s an operating nuisance i n that it had t o be periodically t r a n s f e r r e d back t o t h e pump.
The Mark-I1
pump (which has operated f o r over 13,000 hours i n a t e s t loop) has more height i n t h e pump b o w l , eliminating t h e need f o r an overflow tank.
5.9
Primary He at Exchanger
The primary heat exchanger w a s a horizontal s h e l l and U-tube type.
During e a r l y power operation t h e heat t r a n s f e r c a p a b i l i t y w a s found t o be considerably lower than expected.
A reevaluation of t h e physical properties
showed t h a t t h e t h e m conductivity of both t h e f u e l and coolant salts w a s s u f f i c i e n t l y below t h e value used i n design t o account f o r t h e overestimate * I
of the o v e r a l l c o e f f i c i e n t of heat t r a n s f e r .
Table 5-2 shows a comparison
of t h e physical property data used i n t h e o r i g i n a l design t o t h e current data.
The heat t r a n s f e r coefficients calculated by t h e conventional design
procedures using these two sets of data are a l s o shown. The heat t r a n s f e r d i d not change during subsequent operation of t h e
reactor and no other d i f f i c u l t i e s occurred.
The measured o v e r a l l heat trans-
f e r coefficients ranged from 646 t o 675 w i t h an average of 656 Btu/(hr-ft2-OF) f o r 8 measurements.
These measurements were made on t h e basis of nominal
full power at 8.0 Mw and a coolant salt flow of 850 gpm. If t h e a c t u a l coolant f l o w rate were 770 gpm, which would be t h e flow consistent with a power l e v e l of 7.34 Mw, t h e measured o v e r a l l c o e f f i c i e n t would be 594 as compared t o a calculated value of 599 Btu/(hr-ft2-'F). covered i n d e t a i l i n Referenees 30 and 31.
The performance i s
111
Table 5-2.
Physical Properties of Fuel ark Coolant S a l t s
Used i n MSRE Heat Exchanger Design and Evaluation
Original Fuel
Coolant
Current Fuel
Coolant
Thermal Conductivity, Btu/(hr-ft-OF)
2.75
3.5
o .a32
Viscosity , l b / ( f't-hr )
20 .o
18.7
Density, l b / f t 3
17.9 154.3
0 0659 23.6
120.0
141.2
123.1
Specific Heat , Btu/(lb-OF)
0.46
0.57
0.4735
Film Coefficient , Btu/(hr-ft2-OF)
3523
5643
1497
0.577 1989
Overall Coefficient, Btu/(hr-ft2-OF)
1186
618
112
5.10 5.10.1
Reactor Vessel and Reactor Access Nozzle
General Description
Reactor Vessel and Core -The 8-ft-high
r e a c t o r v e s s e l w a s a 5-fi-diameter by
cylinder which contained a 55-in.-diameter
graphite core.
A
cutaway drawing of t h e r e a c t o r v e s s e l and core i s shown i n Figure 5-9.
The vessel design pressure'and temperature were 50 p s i g and 1300'F with an allowable stress of 2750 p s i . The f u e l salt entered t h e flow d i s t r i b u t o r where it was directed downw a r d around t h e circumference of t h e vessel.
It flowed i n a s p i r a l path
through a 1-in. annulus between t h e v e s s e l w a l l and t h e core can f o r cooling purposes.
Anti-swirl vanes i n t h e lower head of t h e v e s s e l straightened t h e
f l o w path before i t entered t h e graphite moderator core.
Flow passages,
formed by grooves'in t h e s i d e s of t h e graphite moderator bars, d i r e c t e d t h e laminar salt flow t o t h e upper head plenum.
The salt flow then l e f t t h e re-
a c t o r vessel through t h e s i d e o u t l e t of t h e r e a c t o r access nozzle which i s described i n t h e following section. During operation t h e f u e l salt w a s h e l d i n the c i r c u l a t i n g system by
a freeze valve (FV-103) attached t o t h e lower head of t h e r e a c t o r vessel. The drain l i n e and freeze valve w a s a 1-1/2 i n . Sch.-hO pipe which w a s f l a t t e n e d f o r about 2 i n . t o give a flow cross section about 1/2-in. w i d e . Cooling air i n a shroud surrounding the f l a t t e n e d ' s e c t i o n maintained a frozen
salt plug f o r a "closed" valve.
The 1-1/2 i n . pipe extended 2-3/4 i n . i n t o
t h e lower head of t h e v e s s e l and w a s covered with a hood f o r protection against sediments c o l l e c t i n g on t o p of t h e frozen plug making it inoperative. The 1-1/2 i n . hooded drain thus would remove a l l but a small puddle of salt
from t h e lower head even i f there had been heavy sedimentation.
To provide
f o r drainage of t h e remaining puddle, a 1/2-in. tube was mounted through t h e w a l l of t h e portion of t h e drain protruding i n s i d e t h e v e s s e l and extended downward through t h e freeze valve below.
This, i n e f f e c t , formed
p a r a l l e l freeze valves operated by t h e same controls. The r e a c t o r core consisted of 617 f u l l and f r a c t i o n a l s i z e graphite
elements 2 x 2 i n . cross s e c t i o n and about 67 i n . long.
S a l t flow channels
were formed by machined half channels on each of four faces of each element.
113
ORNL-LR-GWG 61097RiA
P
GRAPHITE SAMPLE ACCESS PORT
FLEXIELE CONDUIT TO CONTROL ROD DRIVES
COOLING AIR LINES
.
Fig. 5 . 9
E R E Reactor Vessel
114
When not buoyed by being immersed i n salt, t h e v e r t i c a l graphite elements
were supported by a l a t t i c e of graphite blocks which i n t u r n a r e supported by a Hastelloy-N g r i d fastened t o t h e bottom of t h e core can.
The score can
w a s thus f r e e t o expand downward r e l a t i v e t o i t s t o p attachment t o t h e reactor vessel while t h e graphite was free t o expand upward r e l a t i v e t o i t s support at t h e lower end of t h e core can.
The graphite moderator elements
were restrained from f l o a t i n g out of the core by a Hastelloy-N rod through holes i n t h e dowel section at t h e bottom of each graphite element.
In
addition a wire passing through t h e upper graphite elements prevented t h e upper portion of a broken element from f l o a t i n g out of t h e core. To prevent possible overheating i n an otherwise stagnant region, a
small portion-of s d t enteripg t h e reactor w a s diverted i n t o t h e region j u s t above t h e core can support flange i n t h e annulus between t h e vessel and t h e core can.
Vessel w a l l temperatures i n t h i s region Etnd a l s o t h e lower
head were monitored during operation f o r possible deviations. A t t h e center of t h e core were located sample specimens, t h r e e control
rod thimbles, and f i v e removable graphite elements. w e r e located i n one corner of a 4-in. occupying t h e o*her three corners.
The 'sample specimens
square with the t h r e e control rods
The five.removable+elements then occupied
t h e remaining spaces between these four-positions.
The controls rods are
described elsewhere i n t h i s report (Chapter 19). The sample specimen assemblies mentioned above were made of var.ious com-
binations and arrangements of graphite and m e t a l .
They w e r e exposed t o much
t h e same salt velocity, temperature, and nuclear f l u x as t h e core matrix.
These are described in general elsewhere i n t h i s section.
The reactor vessel w a s supported-from t h e 'top removable cover of t h e thermal s h i e l d by twelve hanger-rodcasseniblies
.
'
These hanger-rod assemblies
were pinned t o lugs welded t o t h e reactor vessel just above t h e flow distributor.
The-support ,arrangement w a s such t h a t t h e reactor vessel could be
considered t o be anchored at t h e support lugs.
The portion of t h e vessel
below t h e lugs w a s free t o expand downward w i t h no r e s t r a i n t . React0.r Access nozzle (RAN) -Attached vessel w a s a 40-in.
long lO-in.-diam
nozzle.
t o t h e upper head of t h e reactor The nozzle had a 5-in.-diam
I
s i d e o u t l e t f o r t h e leaving salt located about 10 i n . above t h e r e a c t o r
b)
115
L.,
v e s s e l upper head.
The remaining extension of t h e lO-in.-diam
nozzle pro-
vided a pocket of trapped gas during t h e f i l l i n g of t h e reactor.
However,
most of t h e volume of t h i s extension w a s occupied by t h e nozzle plug which
i s a removable support f o r t h e t h r e e control rod thimbles, t h e 2-1/2 i n . graphite sample access pipe, and f o r t h e discharge screen. . See Fig. 5-10.. Conventional Leak-detected metal oval-ring- j o i n t flanges were used on both sample access â&#x20AC;&#x2122;openings t o provide t h e necessary containment of t h e primary system.
The radial clearance between t h e removable nozzle plug and t h e
nozzle w a s 1/8-in. ered annulus.
at t h e top and 1/4-in. a t t h e bottom t o provide a tap-
It was intended t h a t salt be frozen i n t h i s tapered annulus
providing a salt s e a l t o prevent molten s a l t from contacting t h e metal seali n g surfaces. possible.
However, maintaining a frozen salt s e a l w a s found t o be not
This is discussed i n more d e t a i l l a t e r .
Cooling a i r w a s pro-
vided on both i n s i d e of t h e plug and outside of t h e 10-in.lnozzle but only t o the i n s i d e of t h e plug of t h e 2-1/2 i n . graphite sample access.
Heaters
were provided t o thaw any frozen salt remaining i n t h e 10-in. annulus after
b,
a salt drain.
For thawing t h e 2-1/2 in. annulus t h e cooling tube w a s re-
moved from t h e plug as p a r t of t h e sampling procedure and temporarily replaced w i t h a metal sheathed heater. Some twenty thermocouples (not including spares) were i n s t a l l e d a t various locations on t h e FWl t o monitor temperatures of t h e nozzle, both plugs and.contro1 rod thimbles.
A thermocouple w e l l attached t o t h e
graphite-sample access plug extended i n t o t h e flowing s a l t stream t o i n d i c a t e r e a c t o r o u t l e t tempe Strainers w e r graphite chips lar lower end of t h e 10-in. n region of t h e r e a c t
i t e sample assembly ve gr,aphite elements were
r e a c t o r o u t l e t t o prevent passage of s t r a i n e r basket w a s attached t o t h e and extended downward i n t o t h e upper e t h r e e control rod thimbles and t h e ugh t h e s t r a i n e r basket.
Since t h e
ed as were t h e remaining moderator ele-
ments, a cross-shaped extension of t h e basket assembly projected beneath t h e basket t o provide a hold-down.
See Fig. 5-11.
The core sample specimens were removed and replaced onlyâ&#x20AC;&#x2122; when t h e re-
bt
a c t o r w a s drained of s a l t and p a r t i a l l y cooled.
The specimens were removed
116 i ORNL-DWG 68-5507 /COOLING
COOLING AIR
AIR
h
(Gin.
2 '/,-in.
Fig. 5.10
Reactor Access Nozzle Showing 10- and 2-1/2 i n . Annuli
gr
117
ORNL-DWG 63-6475
0
0
o
a
CONTROL ROD THIMBLE-
Fig. 5.11
Reactor Fuel Outlet Strainer
118
through t h e smaller of t h e two access nozzles described earlier.
A stain-
less s t e e l standpipe was l e f t attached t o t h e small graphite-sample access nozzle via a bellows.
The upper end of t h e standpipe w a s bolted t o a l i n e r
s e t i n t o t h e lower concrete roof plug.
During normal operation t h e l i n e r
opening w a s closed with a concrete plug.
When samples were t o be removed,
t h i s plug w a s replaced with a work s h i e l d which contained openings f o r tools, l i g h t s , etc.
All j o i n t s were moderately leak-tight and t h e stand-
pipe was provided with a nitrogen purge and offgas connections.
A Roots
blower located i n t h e service tunnel provided a l i g h t l y negative standpipe pressure t o assure inward leakage. Core samples were taken by removing t h e graphite sample access flange, withdrawing t h e sample i n t o t h e standpipe, placing t h e sample i n a s p e c i a l c a r r i e r , i n s e r t i n g a replacement sample specimen i n t o t h e core, and replacing t h e access flange. 5.10.2
H e a t Treatment of Reactor Vessel
After t h e reactor vessel w a s i n s t a l l e d , t e s t s of t h e p a r t i c u l a r heat of Hastelloy-N used i n the vessel showed that t h e closure weld beeween t h e top head and t h e flow d i s t r i b u t o r r i n g could have poor mechanical proper-
t i e s i n t h e as-welded condition.
Therefore i n t h e f a l l of 1965, t h e reac-
t o r v e s s e l w a s heat-treated i n place, using i n s t a l l e d heaters, f o r 90 hours at 1400'F.
5.10.3
Reactor Access Nozzle Freeze T e s t s and Effect of Circulating Bubbles
!be main purpose f o r a frozen salt s e a l i n t h e annuli of t h e 10-in. and 2-1/2-in.
dim.
RAN plugs w a s t o prevent contact of t h e s e a l i n g surfaces
with salt.
The freeze flanges used as piping disconnects have a similar
fhnction.
The main difference being t h a t t h e freeze flanges incorporated
a radial freeze j o i n t while the RAN employed a longitudinal freeze j o i n t . During development on t h e Engineering T e s t Loop freeze j o i n t , it w a s noted t h a t a frozen salt seal could not be maintained r e l i a b l y i f i n cont a c t with molten salt.
It was intended t o operate t h e MSRE RAN freeze
j o i n t with a r i n g of frozen salt above t h e normal molten-salt l e v e l .
This
would be accomplished by pressurizing t h e fuel system above normal operati n g pressure soon after f i l l i n g t h e r e a c t o r with salt.
Cooling a i r on t h e
W
119
6,
RAN woula freeze a r i n g of salt i n t h e annuli before t h e pressure w a s low-
ered t o i t s normal value, leaving a gas void between t h e frozen and l i q u i d
salt.
The gas trapped i n t h e RAN annuli w a s thus t h e p r i n c i p a l barrier
between t h e molten s a l t and t h e containment s e a l i n g flange.
The r i n g of
frozen s a l t would then serve only as a backup protection against sudden pressure increases forcing salt up i n t o t h e annuli. The first attempts t o form t h i s backup s e a l i n t h i s manner on t h e MSRE RAN w e r e unsuccessful due
n s u f f i c i e n t cooling a i r although t h e availa-
b l e flow met t h e requirements of t h e design
calculation^.^^
The flow w a s
permanently increased from about 3 cf'm t o t h e range of 1 5 t o 20 cf'm by modifying t h e control valve t r i m .
Even with t h e increased a i r flow, t h e r e
w a s s u f f i c i e n t movement of salt On t h e annuli due t o t h e turbuient flow below t o prevent forming a good frozen-salt seal.
However, gas trapped
above t h e molten salt kept t h e l i q u i d l e v e l ,well below t h e flange.
The
l i q u i d gas i n t e r f a c e changed i n height due t o any change i n volume of t h e gas pockets.
During e a r l y operation, the i n t e r f a c e w a s at least a foot
below t h e flange and t h e flange temperature w a s about 200째F.
bi-l
During l a t e r
operations, t h e r e were periods when t h e salt l e v e l w a s higher and t h e temperature of t h e flange approached 300'F. There were two mechanisms which caused gas t o be t r a n s f e r r e d t o and from t h i s pocket.
Gas was t r a n s f e r r e d from t h e pocket as a s o l u t e and
adde'd by entrapment of
l e s from t h e salt stream below.
The equilibrium d i f f e r e
e two t r a n s f e r rates determined t h e
salt l e v e l i n t h e annuli
emperature readings of t h e w a l l .
Any change i n reactor between these t w o t r a n s i n t h e annuli.
It was
be increased by any of t
lowering t h e system pres culating salt temperat
conditions which disturbed t h e balance ould thus change t h e height of salt he salt level i n t h e RAN annuli would g changes i n operating conditions:
e r gas i n t h e f u e l pump), increasing c i r g t h e amount of c i r c u l a t i n g bubbles
asing pump spee es of a l l t h r e e
s e causes a r e shown i n Figs. 5-12 and 5-13.
Please r e f e r t o Fig. 5-10 for location of t h e thermocouples used i n these
120
U
a ORNL-DWG 73-599
EACTOR OUTLET 2-
lz00
---
----------y
,
3
TE R-7B
I 400-
,
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 DATE BEGINNING FEBRUARY 3,1968
Fig. 5.12
Effect of Reactor Outlet Temperature and Fuel Pump CoverGas Pressure on Reactor-Access Nozzle Salt Level (as indicated by wall temperatures)
121
ORNL-DWG 73-600
1400
1200
lo00
800
b,
600 .
400 0
5
10
15
20
TIME, MARCH 11,1969 (htd
Fig
Effect of Fuel Pump Speed (i.e. circulating bubbles) on Reactor-Access Nozzle Salt Level (as indicated by wall temperatures)
I
\
122 i--
The e f f e c t of temperature and pressure can be seen i n Fig. 5-12.
At
W
t h e beginning of t h e observed time period, t h e salt l e v e l i n t h e RAN was r e l a t i v e l y low with t h e pump pressure higher than i t s normal 5 psig.
After
a pressure drop on February 6, t h e salt l e v e l overcame t h e immediate e f f e c t s of pressure and began t o r i s e as indicated by t h e R-43 thermocouple.
Sev-
eral days l a t e r t h e reactor o u t l e t temperature was changed i n two small steps of 30'
and 15' y e t note t h e very l a r g e e f f e c t of almost 6 0 0 ' ~ on t h e
R-7 thermocouple due t o t h e salt l e v e l increase. Later by r a i s i n g t h e pump pressure From 3 t o 5 p s i g t h e salt l e v e l i s again lowered i n t h e RAN annulus. The e f f e c t of a d i r e c t change i n t h e amount of c i r c u l a t i n g bubbles can be seen i n Fig. 5-13.
A t t h e beginning of t h i s period t h e fuel-pump speed
w a s lower than normal and it can be seen t h a t t h e RAN salt l e v e l was very high.
Other indications such as t h e r e a c t i v i t y balance l e d t o t h e conclu-
sion there-we-re no bubbles c i r c u l a t i n g a t t h i s condition.
When t h e fuel-
pump speed was increased, with no other change i n operating temperature o r pressure, t h e RAN salt l e v e l dropped very rapidiy.
(Note t h e abscissa i n
Fig. 5-13 i s now hours instead of days.) Circulating bubbles were involved i n more dramatic e f f e c t s than temperature changes i n t h e RAN, of course.
Some more important variables were
involved such as t h e e f f e c t s of c i r c u l a t i n g voids and xenon poisoning on t h e r e a c t i v i t y balance.
The e f f e c t s seen i n t h e RAI? temperatures only
helped explain the cause of some other events. One phenomenom noted i n which it is thought t h e RAN trapped gas had a d i r e c t r e l a t i o n s h i p i s t h a t of parer perturbations i n t h e MSRE i n January, 1969. There was an unusually l a r g e amount of c i r c u l a t i n g bubbles at t h i s time as v e r i f i e d by several indications including low RAN temperatures.
It i s thought some gas, clinging t o t h e core, w a s suddently swept from t h e core causing a momentary increase i n r e a c t i v i t y .
The t r i g g e r i n g mechanism
f o r r e l e a s e of gas clinging i n t h e core was presumably a very small perturbation i n f l o w o r pressure.
An occasional r e l e a s e of a b u r s t of gas from
t h e RAN which was suddenly compressed i n t h e pump could have been t h e cause of such perturbations.
There were other indications t o connect t h e RAN
trapped gas with t h e power perturbations. 47
w
123
6,
5.10.4
Core Specimen I n s t a l l a t i o n and Removal
Throughout t h e operation of t h e MSRE with s a l t i n t h e primary loop t h e r e w a s a sample a r r a y of one kind o r another i n t h e core.
The a r r a y s
t h a t were exposed between September, 1965 and June, 1969 w e r e of t h e des i g n shown i n Fig. 5-14.
The a r r a y t h a t w a s i n t h e core from t h e t i m e of
construction u n t i l August, 1965 contained similar amounts of graphite and
INOR-8 ( t o have t h e same nuclear r e a c t i v i t y e f f e c t ) b u t d i f f e r e d i n i n t e r n a l configuration.
I n 1969, during t h e l a s t f i v e months of operation, a
d i f f e r e n t array, designed t o study t h e e f f e c t s of s a l t v e l o c i t y on f i s s i o n product deposition, w a s exposed i n t h e core.
A core specimen assembly of t h e type shown i n Fig. 5-14 consisted of t h r e e separable s t r i n g e r s (designated R, L, and S).
Whenever an assembly
w a s removed from t h e core, i t w a s taken t o a hot c e l l , t h e s t r i n g e r s were
taken out of t h e basket, and a new assembly w a s prepared, usually including one o r two of t h e previously exposed s t r i n g e r s .
was reused, sometimes not.
Sometimes t h e o l d basket
The h i s t o r y of exposure of INOR-8 specimens i n
t h i s kind of a r r a y is outlined i n Fig. 5-15.
The numbers i n d i c a t e t h e
h e a t s of INOR-8 from which t h e rods i n each s t r i n g e r were made. The following s e c t i o n describes t h e experience with i n s t a l l a t i o n and removal of core specimen assemblies, with emphasis on t h e procedures, t o o l s , and systems involved.
Descriptions of t h e materials t h a t were exposed and
t h e r e s u l t s of t h e i r post-operation examinations are given i n d e t a i l i n 1
references 32-36. Pre-Power Array
ecimen a r r a y t h a t w a s i
t h e core during the
prenuclear t e s t i n g
nuclear s t a r t u p experiments d i f f e r e d i n t e r n a l l y from
t h e later s u r v e i l 1
assemblies, but e x t e r n a l l y i t was similar and i t s
i n s t a l l a t i o n and removal emp
ed t h e equipment and procedures proposed f o r
l a t e r use.
l a t i o n w a s i n 1964, before s a l t had been cir-
The o r i g i n a l i n s
culated, and, although done remotely, w a s c l o s e l y observe$ t o determine needs f o r modifications i n t h e t o o l s and procedures. moved i n August, 1965 by the remote procedure. s l i g h t l y radioactive from the'''%
The assembly w a s re-
(At that t i m e the salt was
s t a r t u p experiments.)
Only minor d i f f i -
c u l t i e s were encountered w i t h some of t h e t o o l s and f i x t u r e s .
Fig. 5.14
C:
MSRE Surveillance F a c i l i t y Inside Reactor Vessel
C
I
STilNGbR
5085 5065
L
STRINGER 6’
I.‘
I%,,
STRMER
S’
n I’
CORE TEMPERATURE’
1 I II
1lll
I
I
I
SALT tN CORE RUN NO.
Fig.
5.15
Outline
of MSRE Core Surveillaixe
Program
II,
I I
Ill
t3 WI __’
126 While t h e core access w a s open f o r t h e sample removal, a v i s u a l inspection w a s made of t h e core with a 7/8-in.-diam.
W
The inspection
scope.
revealed t h a t pieces were broken from t h e h o r i z o n t a l graphite b a r at t h e bottom of t h e core t h a t w a s supposed t o support t h e sample assembly. broken pieces were recovered, using a vacuum cleaner.
The
To circumvent t h i s
damage, t h e new sample assembly t o be i n s t a l l e d w a s modified t o be suspended from above, using a s p e c i a l f i t t i n g i n s t a l l e d i n t h e s t r a i n e r bask e t above t h e core.
This f i t t i n g (the "basket lock" i n Fig. 5-14) had
spring f i n g e r s t h a t locked i n t o the s t r a i n e r screen when i t was pushed i n t o place.
The lower end of t h e sample basket w a s extended t o reach t h e
hole i n t h e lower horizontal graphite bar t o provide lateral support.
Array 1. I n s t a l l a t i o n of t h e basket lock and t h e f i r s t standard arary i n the core i n September, 1965 was uneventful.
The array was removed
i n July, 1966 a f t e r about 1087 equivalent full-power hours (EFPH) of operation.
When t h e a r r a y w a s examined i n t h e hot c e l l , portions of t h e s t r i n g -
ers were found t o have been damaged. Some obstruction'(thought t o be s a l t i n t h e annulus) had been encountered i n l i f t i n g t h e assembly through t h e access nozzle.
The damage w a s not caused by handling, however, but by con-
t r a c t i o n during cooldown.
u
When t h e core was drained some s a l t w a s trapped
between t h e ends of graphite specimens where i t f r o z e and i n t e r f e r e d with t h e d i f f e r e n t i a l contraction of t h e p a r a l l e l graphite and metal columns during cooldown.
Array 2.
As a r e s u l t t h e graphite columns buckled.
Because of t h e damage, none of t h e t h r e e s t r i n g e r s from t h e
f i r s t array were included i n t h e second. t o prevent trapping of
The new s t r i n g e r s were modified
s a l t , but otherwise t h e a r r a y w a s l i k e t h e f i r s t .
The second a r r a y w a s i n s t a l l e d on Sept. 16, 1966, near t h e end of t h e 2month shutdown t o replace t h e main blowers. The second a r r a y w a s removed i n May, 1967.
By t h i s t i m e t h e r e a c t o r
had operated 4510 EFPH and t h e samples were removed sooner a f t e r t h e end of power operation, so t h e s a l t w a s more highly radioactive than before, r e s u l t i n g i n troublesome contamination on t h e t o o l s and i n t h e standpipe. This slowed t h e operation somewhat and i t w a s necessary t o i n s t a l l a charcoal f i l t e r i n t h e standpipe vent l i n e t o l i m i t iodine release t o t h e stack. D i f f i c u l t y was encountered i n obtaining a s a t i s f a c t o r y seal on t h e r e a c t o r access flange, which required replacement of four b o l t s .
Hot-cell examina-
t i o n of t h e a r r a y showed t h a t the basket lock had pulled out of t h e s t r a i n e r
LJ
127 screen and was s t u c k on t h e basket.
A new basket lock was t h e r e f o r e fabri-
cated and w a s i n s t a l l e d i n t h e screen. Array 3.
The t h i r d a r r a y consisted of one new s t r i n g e r and two
s t r i n g e r s previously exposed i n Array 2, so i t w a s highly radioactive a t t h e t i m e of i n s t a l l a t i o n .
No p a r t i c u l a r d i f f i c u l t y was encountered i n handling t h e a r r a y , however, s i n c e t h e equipment and procedure were designed f o r t h i s condition..
A s a t i s f a c t o r y seal on t h e core access flange
was n o t obtained on t h e f i r s t t r y , and inspection showed t h e O-ring gasket w a s dented and scratched. The b o l t holes w e r e cleaned and retapped and a new gold-plated gasket w a s i n s t a l l e d . An acceptably t i g h t j o i n t w a s then obtained. Array 3 w a s removed on A p r i l 2, 1968 a f t e r t h e conclusion of U'" operation (9005 EFPH).
Some inconvenience r e s u l t e d when t h e closure on
t h e t r a n s p o r t containment l i n e r refused t o operate properly, b u t t h e r& moval and t r a n s f e r were s a f e l y accomplished without t h e l i n e r .
While t h e
array was out of the core t
lange w a s sealed with a rubber-gasketed
blind flange t o avoid poss
damage t o t h e permanent s e a l i n g surfaces..
Array 4.
One of the o l d s t r i n g e r s and two new s t r i n g e r s made up
Array 4, which was i n s t a l l e d uneventfully on A p r i l 1 8 , 1968. was removed i n June 1969, a t 11,555
EFPH.
This a r r a y
The removal was delayed f o r
several days due t o problems with t h e removable h e a t e r t h a t w a s required
the salt from the tapered annulus between t h e removable plug and
128
5.10.5
Radiation Heating
Radiation-produced h e a t i n t h e r e a c t o r vessel walls caused t h e outer surfaces of t h e v e s s e l t o be h o t t e r than t h e adjacent s a l t by an amount t h a t w a s proportional t o t h e r e a c t o r power. would cause even higher surface temperzitures.
Any deposition of s o l i d s There were two locations
i n t h e r e a c t o r v e s s e l where s o l i d s could tend t o accumulate.
These loca-
t i o n s w e r e t h e lower head and t h e lugs located j u s t above t h e i n l e t volute and which supported t h e core matrix.
The differences between t h e r e a c t o r
v e s s e l temperatures and t h e s a l t i n l e t temperature w e r e c a r e f u l l y monitored during power operation so that this condition would be detected had i t occurred.
There w a s no i n d i c a t i o n of sedimentation during the operating l i f e of the reactor.
The temperature d i f f e r e n c e d i d increase w i t h use of zssU
f u e l ; t h i s w a s expected due t o t h e higher neutron leakage associated with t h i s fuel. 5.10.6
Discussion and Recommendations
The r e a c t o r vessel and core s a t i s f a c t o r i l y performed a l l functions f o r which it was designed.
Originally t h e operating l i f e of t h e f u e l sys-
t e m w a s t o be determined by t h e number of thermal cycles on t h e f r e e z e
flanges.
However, t h e design l i f e of t h e f r e e z e flanges w a s extended on
t h e b a s i s of development tests.
The stress-rupture l i f e of t h e r e a c t o r
vessel w a s then reviewed t o determine i f t h e operation of t h e MSRE would be limited by t h e p o s s i b i l i t y of stress-rupture cracking.
The reevalua-
t i o n indicated t h e r e a c t o r could be operated a t least another year longer than i t s o r i g i n a l l y predicted 20,000 hours stress-rupture l i f e .
Sl3,SS
The r e a c t o r access nozzle a l s o adequately performed a l l of i t s func-
tions.
It w a s obvious, however, t h a t some amount of bubbles c i r c u l a t i n g
with t h e s a l t w a s necessary t o replenish t h e gas inventory of t h e RAN ann u l i and keep the s a l t l e v e l down t o the d e s i r a b l e l e v e l .
The RAN design
w a s thus inadequate from t h e standpoint of providing a frozen s a l t seal such as was t h e case with the freeze flanges.
The i n a b i l i t y t o f r e e z e a
129
b,
s a l t seal was t h e r e s u l t of a l a r g e r than a n t i c i p a t e d s a l t flow i n t h e
RAN annuli which increased t h e h e a t load beyond t h e c a p a b i l i t y of t h e cooling air-provided.
Additional thermocouples i n t h e area of t h e RAN
would have allowed b e t t e r d e f i n i t i o n of t h e s a l t l e v e l .
If an access flange such as t h e RAN i s t o b e used again i n a molten s a l t system, t h e r e are t h r e e p o s s i b l e methods of obtaining i t s main function
- t h a t of
preventing contact of s a l t with s e a l i n g surfaces:
(1) Provide an e x t e r n a l source of gas t o t h e annulus. determination would a l s o be necessary.
A salt level
Provide an i n t e r n a l source of gas i n t h e form of c i r c u l a t i n g
(2)
bubbles which would constantly replenish t h e gas inventory i n t h e annulus.
A trapped gas pocket a t a higher pressure than t h e pump cover pressure w i l l not remain i n d e f i n i t e l y .
(3) Reduce t h e access of flowing s a l t t o t h e annulus by b a f f l e s , l a b y r i n t h o r such, so t h a t a frozen s a l t seal can be maintained. If bubbles are t o be purposely i n j e c t e d i n t o t h e f u e l s a l t f o r t h e
removal of xenon and krypton, then i t e m (2) appears t o be t h e b e s t solution.
The j o i n t could be designed t o remain e s s e n t i a l l y f u l l of gas a t
all times.
5.11
5.11.1
Fuel and Flush S a l t Drain Tanks
Description
Two f u e l s a l t d r tank c e l l f o r t h e s a f
(FD-1 and FD-2) w e r e i n s t a l l e d i n t h e d r a i n of t h e f u e l s a l t during shutdown periods.
A t h i r d tank (FFT), a l s o 1
s t o r i n g t h e f l u s h s a l t which was used f o r cleaning up t h e f u e l c i r c u l a t i n g system before and a f t e r m a
130
A f u e l drain tank i s shown i n Figure 5-16.
The INOR-8tank was 50 i n .
U
i n diameter by 86 in. high, and had a volume of 80 f t 3 , s u f f i c i e n t t o hold i n non-critical geometry a l l t h e salt from t h e f u e l c i r c u l a t i n g system. The tank was provided with a cooling system designed t o remove 100 kW of f i s s i o n product decay heat.
The cooling w a s accomplished by b o i l i n g water
i n 32 double contained bayonet ttibes and thinibles i n each of t h e tanks. The f u e l f l u s h tank w a s similar t o t h e fuel drain tanks, except t h a t
it had no cooling system.
Since t h e f l u s h salt d i d not contain f i s s i l e
material, t h e only decay heating was from t h e s m a l l quantity of f i s s i o n products t h a t it removed f r o m t h e f u e l system during t h e flushing operations. The weight of salt i n each of t h e t h r e e tanks was indicated by forced
balance pneumatic weigh c e l l s .
The weights were recorded on s t r i p charts
but could be read more accurately from i n s t a l l e d mercury manometers.
Two
conductivity type l e v e l probes (one near t h e bottom of t h e tank and t h e other near t h e t o p elevation of t h e s a l t ) were provided f o r use as reference points -5.11.2
Calibration of t h e Drain Tank Weigh C e l l s Using Lead Weights
During e a r l y t e s t i n g i n t h e f a l l of 1964 t h e weigh c e l l s were c a l i brated by loading t h e tank supporting rings with l e a d b i l l e t s , tanks were a t ambient temperature during t h i s calibrakion.
The drain
D r i f t s i n indi-
cated weight of up t o 39 l b s were noted when no changes were being made. Repairs of a i r leaks i n t h e instruments caused s h i f t s of up t o 65 l b s .
The
weigh c e l l s were f u r t h e r c a l i b r a t e d during t h e addition of f l u s h and 23%J
'fie1 salt,
5.11.3
(See below.)
Addition of Flush S a l t and..Further,Calibration .of Weigh Cells
Later i n 1964, t h i r t y - s i x batches of f l u s h s a l t . ( ' ~ 2 5 0l b s per batch)
were added from cans i n a portable furnace d i r e c t l y i n t o drain tank FD-2 through a heated l i n e and flanged d i p tube attached a t t h e inspection flange on t h e tank.
Each batch o f salt added w a s weighed on accurately
c a l i b r a t e d s c a l e s and t h e t a r e weight of t h e container was checked t o obt a i n t h e net weight of salt added. near 1200'F as possible.
A l l temperatures were maintained as
During additions some ' d i f f i c u l t y was encountered
with salt freezing i n sections of t h e addition l i n e .
On one occasion t h e
u
131
ORNL-LR-DWC 61719
r
INSPECTION, SAMPLER, AND
LEVEL PROBE ACCESS
FUEL SALT SYSTEM FILL AND DRAIN LtNE
Fuel S a l t Drain Tank
132
drain-tank vent l i n e plugged.
To l o c a t e and remove t h e plug, t h e vent l i n e
tpJ
w a s heated with a torch, s t a r t i n g a t t h e tank and working down t h e pipe u n t i l gas could be blown through t h e l i n e i n t o t h e drain tank,
The plug
blew loose while heating t h e l i n e several f e e t f r o m t h e tank.
The plug location, t h e r e l a t i v e l o w temperature required t o remove t h e plug and some o i l found i n t h e nozzle t o t h e inspection flange on t h e tank l e d t o t h e conclusion t h a t t h e r e s t r i c t i o n w a s caused by an o i l residue. During t h e addition of t h e first two batches of s a l t , weigh c e l l read-
ings were taken every f i v e minutes t o determine when t h e probe l i g h t came on.
Then t h i s salt w a s t r a n s f e r r e d through t h e t r a n s f e r l i n e s t o FFT and
FD-1 and back t o FD-2 through t h e r e a c t o r f i l l l i n e s .
This operation l e f t
salt i n all freeze valves which were then frozen t o i s o l a t e t h e t a n k s ' f o r t h e first time. While t h e salt was i n t h e FFT, two more cans of f l u s h s a l t were added t o FD-2 t o recheck t h e location of t h e lower probe.
I n t h e two checks t h e
probe l i g h t came on when 469.6 and 492 l b s had been added t o t h e tank. These figures corresponded t o weigh c e l l readings of 419 and 316 l b s .
Af-
t e r a l l of t h e f l u s h salt (9230 l b s ) had been added t o FD-2, it was transf e r r e d between t h e t h r e e drain tanks.
By using t h e weigh c e l l s on both
tanks involved during each t r a n s f e r , t h e weigh cells on FD-1 and FFT were calibrated.
The indicated weight at t h e upper and lower probes on FD-1
and FD-2 w e r e a l s o noted.
The lower probe of FTT was not functioning (See
24.15). "he weight between t h e probes w a s 7353 f 176 l b s f o r FD-1 and 7545 f 191 l b s f o r FD-2. From t h e above data and subsequent operation, it w a s concluded t h a t t h e weigh c e l l s are inadequate f o r precise inventory work.
I n general,
t h e longer t h e time required f o r a given operation, t h e l a r g e r t h e e r r o r introduced.
Transfers of small q u a n t i t i e s of salt over s h o r t time i n t e r -
v a l s appeared t o give f a i r l y good data.
Using t h e tank weight a t t h e lower
probe l i g h t as a fixed reference w a s u s e f u l i n preventing t h e t r a n s f e r of too much salt t o t h e f u e l system and as an indication of t h e amount of salt remaining i n t h e tank after a f i l l . c e l l s were at fault.
The implication i s not t h a t t h e weigh
Pipe loading on t h e tanks caused by temperature
changes of t h e tanks and adjoining pipes probably caused t h e v a r i a t i o n s . Errors as high as 200 t o 300 l b s have b e e n p o t e d during t r a n s f e r s .
ds
133
a,
5.11.4
U-235 Fuel S a l t Addition
The c a r r i e r s a l t and l a r g e r additions of 235U enriching s a l t were added by t h e same general method used f o r adding t h e f l u s h s a l t . Details of these additions and t h e 235U c r i t i c a l i t y experiment a r e given i n Ref. 40.
5.11.5
S a l t Transfers
S a l t i s t r a n s f e r r e d between drain tanks by pressurizing t h e supply tank and venting t h e receiver tank. through t h e 1/2-in.
Originally t r a n s f e r s were made only
transfer lines (lO7-llO).
I n t h e event of a drain t o
both drain tanks followed immediately by a r e f i l l of t h e c i r c u l a t i n g loop, considerable time w a s spent freezing t h e f i l l freeze valves (105 and 1061, thawing t h e t r a n s f e r freeze valves (108 and log), refreezing them a f t e r t h e t r a n s f e r and then rethawing t h e f i l l freeze valves (105 and 106).
After
c a r e f u l consideration, procedures were modified t o permit jumpering i n t e r locks and t r a n s f e r r i n g through t h e f i l l l i n e s .
No d i f f i c u l t i e s were en-
countered, however, a l l of salt could not be t r a n s f e r r e d t h i s way since t h e f i l l l i n e s did not go completely t o t h e bottom of t h e tanks.
5.11.6
A f t e r Heat Removal
On 2/17/64 and 2/24/64, t e s t s were made of t h e heat-removal capacity of t h e steam domes.
Flush salt w a s put i n t o FD-2 and t h e s a l t w a s heated
t o %l20O0F. With 40 gallons of water i n t h e feedwater tank (FWT-2), a supply valve (LCV-807) w a s opened t o admit t h i s water t o a l l 32 of t h e bayonet cooling tubes.
The water w a s refluxed f o r approximately two hours
with cooling tower water f l 40 gal/min,
o t h e drain-tank condenser kept constant a t
The i n l e t and e x i t temperatures of t h e cooling tower water
were monitored, as were t h e drain tank temperatures and feedwater tank levels.
A t equilibrium conditions, FD-2 temperatures dropped l i n e a r l y a t
a r a t e of 86'F/hr.
The heat-removal r a t e calculated from condenser cooling
water flows and temperatures
s 139 kW/hr.
The heat-removal rate calcu-
l a t e d from t h e temperature-decay r a t e of FD-2 and t h e heat capacity of FD-2 w a s 132 kW/hr.
A second sitnilar.tset was wade by opening t h e other supply valve (ESV-807) and keeping LCV-807 closed,
The flow through t h i s valve w a s
T
134 marginal, the l e v e l i n t h e FWT decreased slowly and t h e heat-removal r a t e
was only 110 kW/hr based on a water heat balance and 98 kW/hr from t h e salt temperature decay rate. These data showed t h a t the bayonet cooling tubes were more than adequate t o remove the design decay heat rate of 100 kW/hr ( f o r 10-MW operation).
Since t h e r e a c t o r was run a t only approximately 8 MW, t h e margin
of s a f e t y was enhanced. Because of leaks through t h e seats of t h e water supply.valves, t h e r e a c t o r w a s operated some of t h e t i m e w i t h t h e feedwater tanks empty and procedures were w r i t t e n t o manually valve water t o t h e s e tanks i f necessary t o remove af'terheat,
When t h e salt was divided about equally between
t h e two drain tanks, as would occur on an emergency drain, no heat removal
w a s necessary.
Turning o f f t h e drain tank heaters w a s s u f f i c i e n t t o prevent For instance, on November 2, 1970, t h e
excessive temperature increases.
r e a c t o r was drained while operating a t full power. t o FD-1 and 5130 lbs drained t o -2.
4380 l b s of salt drained
With t h e drain tank h e a t e r s turned
o f f , FD-1 temperature increased from 1060 t o 1155'F i n about 6-1/2 hours and then s t e e d decreasing.
FD-2 temperatures increased from 1075 t o
ll95OF i n about'9-1/2 hours and then s t a r t e d decreasing.
5 11.7
Tanks Plugging of HeLiLmLJ;ines.-at..the.~op..of,the,~e;l..~~ain a
The helium supply, Tent and equalizer l i n e s e n t e r t h e t o p of t h e f u e l
drain tanks via a common 1/2-in.
Schedule-40 pipe.
on FD-2 became almost t o t a l l y plugged.
I n June 1969, t h i s l i n e
The plug first appeared when t h e
main 4.2 l i t e r / m i n H e offgas flow w a s routed p a s t t h e drain tanks due t o plugging i n t h e regular offgas routes.
The plug w a s a t least p a r t i a l l y
cleared by first heating t h e tank t o Ql275'F overnight and then backblowing with helium a t 42 psig.
This technique was not successful when FD-1 l i n e
plugged i n A u g u s t 1969.
Due t o t h i s plug, f u e l drained p r e f e r e n t i a l l y t o
FD-2 during a planned drain on November 2, 1969,
During t h e l a s t drain on
December 12, 1969, t h e f u e l w a s made t o d r a i n more equally by.,deliberately closing t h e vent from FD-2 f o r p a r t of t h e time during t h e draining period. The result w a s t h a t t h e total sal$ weights i n FD-1 and FD-2 were within 5% of each other af'ter t h e drain had been completed.
135
5.11.8
Loading 233U i n t o FD-2
After t h e f u e l salt had been processed t o remove t h e 235U, t h e remaining c a r r i e r salt was t r a n s f e r r e d from t h e f u e l storage tank t o t h e f u e l drain tanks.
The method of adding 233U enrichment salt w a s d i f f e r e n t than
t h a t used f o r adding t h e f l u s h salt ana 235U f u e l salt.
The loading equip-
ment as shown i n Fig. 5-17 was attached t o t h e access flange of FD-2. containing
n ,
Cans
7 kg of uranium were lowered i n t o FD-2 which contained h a l f of
the molten c a r r i e r salt.
When t h e salt had melted out of t h e can, it w a s
withdrawn i n t o t h e shield.
Mixing w a s accomplished by t r a n s f e r r i n g t h e
s a l t between t h e two drain tanks.
The 233U addition and c r i t i c a l i t y ex-
periment i s described i n d e t a i l i n Ref. 41.
5.11.9
Heatup and Cooldown Rates
These a r e described under heaters i n Section 17. Discussion and Recommendations
5.11.10
The f u e l and f l u s h s a l t drain tanks have functioned very s a t i s f a c torily.
They w e r e very sluggish thermally and therefore easy t o control
during operation.
The a f t e r h e a t removal system functioned s a t i s f a c t o r i l y
when needed. The weigh c e l l s were not completely s a t i s f a c t o r y f o r inventory purposes.
Piping s t r e s s e s should be eliminated o r other type instruments
provided.
As with other offgas l i n e s , t h e p o s s i b i l i t y of plugging should be considered.
Perhaps two l i n e s could be used with heaters attached t o melt out
any plug which might develo
136
ORNL-DWG 68-967
-TURF
CARRIER
XHAUST BLOWER
FDT ACCESS FLANGE
.
Fig. 5.17
.
Arrangement for Adding U'"
Enriching S a l t t o Fuel Drain Tank
137
6.
COOLANT SYSTEM
C. H. Gabbard
J. K. Franzreb
M. Richardson
6.1
Description
The coolant c i r c u l a t i n g system consisted o f a centrifugal-type coolant pump and an air-cooled r a d i a t o r which were connected t o t h e f u e l heat exchanger by 5-inch INOR-8 piping. t o llOO'F
at full power.
The normal operating temperature w a s 1000
The coolant s a l t w a s c i r c u l a t e d at 850 gpm.
i n g shutdowns t h e salt was drained t o t h e coolant drain tank. had low points on each s i d e of t h e radiator. and freeze valves were required.
Dur-
The piping
Therefore, two drain l i n e s
The c i r c u l a t i n g systemwas f i l l e d by
pressurizing t h e coolant drain tank with helium and venting t h e coolant
6.2
Purging Moisture and Oxygen from t h e System
-
Before salt was charged i n t o t h e coolant drain tank, oxygen and mois-
LJ
ture were purged from t h e coolant c i r c u l a t i n g system and t h e drain tank. This was accomplished by first pressurizing t h e system t o assure t h a t it
was leak-tight , then evacuating,. and purging with helium during t h e heatup. Details of t h e heatup a r e given i n Section 17. The helium w a s introduced at t h e coolant pump and coolant drain tank and evacuated o r vented through t h e offgas system. i n t h e loop.
The coolant pump was operated t o provide c i r c u l a t i o n
Purging continued u n t i l t h e system w a s at 1200'F and essenti-
a l l y no more moisture w a s bein
released.
Later analysis of coolant salt
samples indicated t h a t t h e purging had been very effective.
. A f t e r t h e coolant salt 11 days.
been added t o t h e coolant drain tank (see
A t e s t of t h e drain time indicated t h a t complete draining could
be accomplished i n approximately 12 minutes a f t e r t h e freeze valves thawed. During t h e following f i l l t h
i r c u l a t i n g system w a s calibrated.
The drain
tank was pressurized i n increments and t h e drain tank weight and t h e d i f -
Lid
f e r e n t i a l pressure between t h e drain tank and t h e coolant pump were re-' corded.
The system c a l i b r a t i o n curve i s shown i n Fig. 6-1.
138
ORNL-DWG 73-601
30
-x
1
25 0
z
U Y
z a z
I-
azo
K
0
2
w
E
w m w
5
15
8 K
0.
2
f c
5 W
10
(L
II: 0
5
0
800
1600
2400
3200
4000
4800
POUNDS OF SALT IN COOLANT LOOP
Fig. 6 . 1
Coolant Circulating System Calibration
5600
139
6.4
Operation
Other than t h e d i f f i c u l t i e s with t h e r a d i a t o r described l a t e r , t h e coolant system operated s a t i s f a c t o r i l y . No salt leaks were detected a t any time.
6.5 6.5.1
Coolant S a l t Circulating Pump
Description The coolant pump used t o c i r c u l a t e t h e LiF-BeF2 coolant s a l t was s i m i -
l a r t o t h e f u e l pump.
It did not have a spray r i n g since f i s s i o n product
gas removal was not required. stalled.
No overflow tank D r cooling shroud was in-
A t t h e operating speed of 1750 rpm, it had an output of 850 gpm
a t a 78-ft salt head,
A small helium purge w a s introduced j u s t below t h e
shaft s e a l i n t h e bearing housing.
u
Most of t h e gas flowed downward through
t h e labyrinth between t h e s h a f t and t h e s h i e l d block t o prevent s a l t mist from reaching t h e seal. The remainder flowed upward t o prevent any o i l which leaked through t h e s e a l from g e t t i n g i n t o t h e pump bowl. O i l f o r l u b r i c a t i n g t h e bearings and cooling t h e s h i e l d plug w a s r e c i r c u l a t e d by an e x t e r n a l pumping system. Helium bubblers and a float-type l e v e l i n s t r u ment were used t o measure t h e l i q u i d l e v e l . not i n s t a l l e d on t h e f u e l pump.)
( A float-type instrument w a s
The coolant pump w a s located outside t h e
r e a c t o r c e l l and could be d i r e c t l y maintained s h o r t l y a f t e r shutdown. 6.5.2
Operation
The coolant pump ran very w e l l throughout t h e operation of t h e reactor. ow range of operating l e v e l s caused operaA s with t h e f u e l pump, t h e tional difficulties.
Prio
loop temperature could not pump l e v e l changed about ? l e c t e d f i l l point w a s not
s t a r t i n g t h e pump a f t e r a fill, t h e average ccurately determined.
Since t h e coolant-
l O O * F change i n loop temperature, t h e se-
s s a t i s f a c t o r y f o r operation.
I n which case
t h e freeze valves had t o be thawed and t h e system l e v e l adjusted. m a l operating l e v e l w a s about
The nor-
55% at full power ( t h e l e v e l instrument had
a span of 10 inches f o r 100% and c e n t e r l i n e of t h e volute was a t 35%).
140
When t h e pump w a s first s t a r t e d after a f i l l , t h e l e v e l would drop about
U
9% due t o a gas pocket i n t h e loop and about 4% due t o t h e pump fountain flow
.
During load and rod scrams, t h e system cooled rapidly which decreased the l e v e l below t h e interlocks required f o r s t a r t i n g t h e pump. c i r c u l a t i o n t h e r e w a s a p o s s i b i l i t y of freezing t h e radiator.
Without
An emergency
start switch w a s i n s t a l l e d which enabled t h e operator t o bypass all i n t e r locks and s t a r t t h e pump i f freezing o f t h e r a d i a t o r appeared eminent. Based on t h e o i l found i n t h e coolant pump offgas l i n e s , t h e r e was a small o i l leak i n t o t h e pump bowl.
The offgas l i n e a t t h e pump bowl be-
came plugged occasionally and i n Noveniber 1968, a clamshell heater was installed.
This w a s used once and c l e a r e d t h e l i n e s u f f i c i e n t l y t o permit
operation u n t i l shutdown December 12, 1969. The f l o a t l e v e l indicator was not used f o r operation. 6.5.3
Conclusions and Recommendations The coolant pump w a s used t o c i r c u l a t e salt f o r 26,026 hours at tem-
peratures between 1000째 and 1200째F. Some plugging of t h e offgas l i n e w a s encountered. sidered i n future design.
u
This should be con-
Leakage of o i l i n t o t h e pump bowl could be e l i m i -
nated by seal-welding a gasketed j o i n t as described f o r t h e fuel pump (Section 6.1).
6.6 6.6.1
Radiator
Description The nuclear power generated by t h e r e a c t o r was f i n a l l y dissipated by
a coolant s a l t - t o - a i r radiator.
S a l t flowed through t h e 120 r a d i a t o r tubes
at 850 gpm where it w a s cooled from about 1075 t o 1015'F a t f u l l power. Approximately 200,000 cfm of air from t h e two stack fans flowed perpendicular t o t h e tubes and out t h e coolant stack. In passing through t h e r a d i a t o r , t h e a i r was heated about 100째F above t h e ambient i n l e t tempera-
ture.
The heat removal was adjusted t o match t h e nuclear power produced
by r a i s i n g o r lowering t h e two r a d i a t o r doors, a d j u s t i n g t h e bypass damper, o r s t a r t i n g and stopping t h e main blowers.
LJ
141
The r a d i a t o r tubes were 0.75 i n . OD x 0.072 i n . w a l l x 30 f t long, zee-shaped t o f i t i n t o t h e insulated r a d i a t o r enclosure. enclosure supported t h e tubes and t h e e l e c t r i c heaters.
The body of t h e Two hundred
twenty-five kW of heat were provided t o prevent freezing of s a l t i n t h e
.
tubes i n case of a sudden l o s s of nuclear power ( s e e Fig. 6.2). "he upstream and downstream faces of t h e r a d i a t o r enclosure were equipped with doors t h a t could move up and down i n a "U" shaped track.
As
t h e doors moved downward i n t o t h e f u l l y closed position, camming devices operated by t h e weight of t h e door forced them against t h e seals located on t h e face of t h e radiator.
The doors were r a i s e d and lowered by means
of cables driven by a s i n g l e motor. independent operation of e i t h e r door. an e l e c t r i c a l power outage, t
Magnetic clutches and brakes permitted I n case of a load scram request o r
brakes and clutches were deenergized and
the doors f e l l closed at a r a t e which w a s limited by t h e i n e r t i a of flywheels attached t o t h e ends of t h e drive s h a f t s through overrunning clutches (s7 f t / s e c ) .
6.6.2
Early Tests and D i f f i c u l t i e s . Preliminary heatup and checkout of t h e r a d i a t o r began i n October 1964.
The main d i f f i c u l t i e s were due t o i n s u f f i c i e n t i n s u l a t i o n and warping of
As a r e s u l t , t h e doors jammed and would not operate properly and t h e r e was t o o much heat loss t o reach operating temperatures. The tem-
the doors.
perature d i s t r i b u t i o n was poor and overheating of thermocouple and e l e c t r i c a l leads w a s encountered. After unsuccessful at
t s t o correct these d i f f i c u l t i e s , t h e doors
were temporarily sealed t o allow completion of t h e c r i t i c a l i t y t e s t s while new doors were being designed and b u i l t . During t h e l a s t h a l f of 1965, extensive changes, t e s t s , and remodifications were made on t h e r a d i a t o r assembly. provided more i n s u l a t i o n f
g t h e heat source and additional expansion
j o i n t s t o prevent warpage.
e "soft" s e a l gaskets were removed from t h e
doors and relocated on t h e
i a t o r enclosure.
3/4 i n . t o 1-1/2 in. wide f o r better sealing. i n s t a l l e d on t h e doors.
LJ
New doors were i n s t a l l e d which
They were increased from A modified "hard" seal w a s
The door guide channels were repositioned t o
142
Fig. 6.2
Radiator Coil and Enclosure
143
provide more clearance.
Four cams were i n s t a l l e d on each door t o provide
more p o s i t i v e seals when they were closed. Sheet metal door hoods were i n s t a l l e d above t h e r a d i a t o r t o reduce t h e
air leakage i n t o t h e penthouse.
The synchro i n d i c a t i n g device on t h e drive
s h a f t had not proved r e l i a b l e since it w a s possible f o r t h e door t o jam and t h e shaft t o continue t o run i n t h e downward direction.
End r o l l e r s were
positioned t o prevent s i d e motion from jamming t h e doors.
Additional upper
l i m i t switches and mechanical stops were added t o prevent l i f t i n g t h e doors
i n t h e event of a l i m i t switch f a i l u r e .
L i f t i n g motor overload switches
were i n s t a l l e d i n conjunction with t h e stops. The new doors weighed about 2700 pounds each which required t h e add i t i o n of two 20,000-pound capacity, 4-in. door.
s t r o k e shock absorbers on each
Although it w a s recognized t h a t with t h e a d d i t i o n a l weight, operation
of t h e e x i s t i n g brakes and clutches w a s marginal, no changes were made i n t h e l i f t i n g mechanism.
Adjustment of t h e door-lifting cables s o that t h e
doors hung straight d i d much t o ease t h e i r operation i n t h e s l i d e s . The above modifications improved t h e heat d i s t r i b u t i o n of t h e r a d i a t o r system.
Two major d i f f i c u l t i e s were then encountered; one w a s t h e exces-
s i v e temperature rise i n t h e pent house j u s t above t h e r a d i a t o r , t h e other
was maintaining a negative pressure i n t h e pent house with respect t o t h e high-bay pressure.
This w a s needed f o r beryllium containment.
E f f o r t s t o correct t h e s e problems included relocating t h e h e a t e r l e a d junction boxes, p u l l i n g t h e thermoplastic w i r e (194'F)
from t h e junction
boxes t o t h e h e a t e r leads coming from t h e r a d i a t o r , spreading out t h e beaded heater leads t o improve t h e cooling a i r c i r c u l a t i o n through t h e wiring,
and, modifying t h e exhaust
t i n g t o d i r e c t t h e sweep of cool a i r across
t h e w i r e troughs ( ~ 5 0 0cfm)
These e f f o r t s were successful. i n reducing
t h e temperature t o about 1
u s t above t h e i n s u l a t i o n on t o p of t h e
radiator.
he door hoods and r a d i a t o r enclosure w a s
Heat leakage aro
reduced by welding up cracks, adding i n s u l a t i o n , and i n s t a l l i n g boots around t h e door-lifting cab overduct w a s i n s t a l l e d bet
the i n l e t hood.
The pent
openings i n t h e door hoods.
A 1-f't2 cross-
t h e door hoods t o r e l i e v e
e pressure with respect t o
aided b y , t h e addition of a duct from t h e pent house t o t h e i n l e t s i d e of
144
t h e south annulus blower.
W
This reduced t h e pent house pressure t o -0.14
t o -0.28 inches of water without t h e main blowers.
Vith both blowers i n
operation, t h e pressure became s l i g h t l y positive.
6.6.3
Subsequent Operation and Modifications The systems were heated and t h e approach t o f u l l power was s t a r t e d
e a r l y i n 1966.
The r a d i a t o r functioned s a t i s f a c t o r i l y , however, due t o
heat l o s s e s through t h e door s e a l s , . i t was d i f f i c u l t t o heat t h e empty r a d i a t o r t o an acceptable temperature d i s t r i b u t i o n p r i o r t o a fill. The heat t r a n s f e r c a p a b i l i t y w a s found t o be about 20% less than calculated.
The cause f o r t h i s discrepancy has not been d e f i n i t e l y established
but may be due t o t h e l a r g e surface-to-air temperature difference which existed.
This is described i n detail i n References 20 and 36.
Coolant salt was frozen i n t h e r a d i a t o r tubes on two occasions.
The
f i r s t occurred on May 19, 1966 when a building power failure caused a load
scram and l o s s of a l l electrical. equipment.
Two tubes of t h e r a d i a t o r w e r e
p a r t i a l l y frozen as indicated by t h e temperatures reading about 75OF low. These temperatures recovered a f e w minutes a f t e r t h e coolant flow w a s reestablished i n d i c a t i n g t h a t t h e plugs had thawed. The second incident occurred on June 27, 1966 when again t h e e l e c t r i c a l system failed, t h i s t i m e causing a coolant salt drain because of low r a d i a t o r o u t l e t temperature.
The coolant salt weigh c e l l s indicated t h a t
about 200 lbs of salt were held up i n t h e coolant system.
Since it w a s
probable t h a t t h e salt w a s frozen i n the r a d i a t o r which had cooled below t h e freezing point, t h e r a d i a t o r w a s reheated and t h e system refilled.
When
t h e pump s t a r t e d , t h e temperature scanners indicated t h a t 5 tubes were frozen and not c i r c u l a t i n g .
These tubes thawed after s e v e r a l minutes of salt
c i r c u l a t i o n without damage t o t h e tubes.
To reduce the p o s s i b i l i t y of t h i s happening again, t h e control system w a s revised so t h a t a rod scram would a l s o cause a load scram.
The low-
temperature s e t p o i n t for a load scram was increased from 900째F, which i s only 6 0 O ~above t h e liquidus temperature, t o 990째F.
Other revisions were
made so t h a t t h e coolant pump would not be turned o f f unless absolutely necessary.
A r a d i a t o r door scram test w a s conducted from f u l l power, and
dd
145
W
these revisions were adequate t o prevent freezing of t h e r a d i a t o r as long as t h e coolant pump remained running.
The average temperature of the f u e l
and coolant leveled out at about 1120'F. On J u l y 17, 1966, a sudden l o s s i n power indicated t h a t a main blower had been l o s t .
Main blower No. 1 had f a i l e d catastrophically and pieces
of t h e aluminum r o t o r hub and blading had gone i n t o t h e r a d i a t o r duct.
Af-
t e r t h e r a d i a t o r had cooled, examination revealed t h a t pieces of aluminum had passed through t h e r a d i a t o r tube bundle, several dented tubes were
found, and s e v e r a l pieces of aluminum were stuck t o tubes.
Metallographic
examination determined there were no punctures i n t h e tubing and t h e radia t o r w a s safe t o operate a f t e r each tube w a s thoroughly cleaned.
The exact
cause of t h e blower f a i l u r e was never s p e c i f i c a l l y determined but examinat i o n of numerous "old" cracks i n t h e blades and hub would indicate t h a t these were t h e probable point of origin.
Flying debris caused some damage
t o t h e h e a t e r l e a d wires, wire t r a y s , e t c . It w a s apparent from inspection of t h e "hard" s e a l i n g surface, which
u
w a s mounted on t h e hot s i d e of t h e r a d i a t o r door, t h a t a modification would be required.
The continuous s t r i p metal s e a l was badly buckled and frac-
t u r e d due t o exposure t o the r a d i a t o r heat.
The roughness of t h i s surface
damaged t h e "soft" s e a l (two widths of 3/4-in. Johns-Mansville Thermocore
#C-197 square asbestos packing s t r i p s mounted on t h e face of t h e r a d i a t o r ) s o as t o render t h i s s e a l ineffective. A short evaluation program of s i x types of "hardrr s e a l s r e s u l t e d i n
t h e s e l e c t i o n of a seal mad ments spaced 1/32-in.
apart
of 2-1/4-in.
long overlapping s t e e l seg-
ch w a s i n s t a l l e d on t h e new doors.
The t e s t
further indicated t h a t t h e "T" b a r , t o which t h e segments were plug-welded,
bowed as much as 3/16.in. on the 3-ft-long t e s t section. t h e new doors were provi
on t h e cold s i d e of t h e
The "T" bars i n
expansion s l o t s plus a tension bar mounted ch w a s used f o r s t r a i g h t e n i n g t h e doors
a f t e r heating. Performance of t h e r a d i a t o r enclosure and door was without incident
after .the above modificati operating period.
Heat l o
were completed f o r t h e balance of t h e E R E and mechanical operation were adequate.
Programmed maintenance on t h e l i f t i n g mechanism w a s performed when t h e opportunity presented i t s e l f .
Miscellaneous s m a l l items, such as replacement
146 of t h e "soft" i n s u l a t i o n along t h e t o p of t h e o u t l e t door, replacement of
W
broken ceramic i n s u l a t o r s on t h e h e a t e r leads within t h e enclosufe, r e p a i r ,
of grounded heaters, e t c . , w a s t h e extent of the maintenance required.
The
door on t h e inlet s i d e of the r a d i a t o r r e t a i n e d i t s shape and sealed reasonably w e l l , t h e s e a l along t h e t o p and s i d e s of t h e o u t l e t door r e t a i n e d i t s
seal except f o r replacement o f burned-out t a p e along t h e top.
The bottom
of t h e o u t l e t door continued t o d i s t o r t and, even after being straightened sevkral t i m e s , made a poor seal as a result of t h e door bowing away from t h e soft s e a l on t h e r a d i a t o r face.
As long as t h e s i d e and top seals were
i n t a c t , t h e r e w a s no chimney e f f e c t and t h e heat l o s s e s were within usable limits.
6.6.4
Automatic Load Control "he heat removal rate at t h e MSRE was dependent upon t h e positions of
the radiator doors and bypass damper and on t h e operation, of t h e main blow-
ers.
These could be manually manipulated as desired.
Instrumentation was
a l s o provided t o automatically sequence t h e operation of these components so t h a t t h e operator could use one switch f o r increasing o r decreasing heat removal rate.
dd
An intermediate r a d i a t o r door s e t t i n g w i t h one blower
i n operation and 1 MW of heat removed was t o be used as t h e s t a r t i n g point. Since it w a s found t h a t t h e heat removal with t h e doors a t t h e i r minimum s e t t i n g of 12-in. w a s about 1.9 MW, t h e automatic system did not function properly.
The r e s t of t h e sequencing functioned properly during a t e s t .
However, i n v i e w of t h e above and since manually adjusting t h e h e a t load
w a s not a disadvantage, t h e automatic load control w a s never used i n t h e operation of the reactor.
6.6.5
Comments and Recommendations Assessment of t h e r a d i a t o r containment, doors, and door-lifting mech-
anism i s complicated by t h e many changes, l a r g e and small, which were required before acceptable performance w a s achieved. The r a d i a t o r enclosure d i f f i c u l t i e s can be a t t r i b u t e d p r i n c i p a l l y t o d e t a i l s of i n s t a l l a t i o n such as gross air leakage i n t o and around t h e enclosure as a result of cracks, unsealed openings, i n s u f f i c i e n t i n s u l a t i o n , lack of a s u f f i c i e n t number of expansion j o i n t s , e t c .
The p r i n c i p a l radi-
a t o r enclosure support "H" beam members were not v e r t i c a l and t h e r e f o r e
bd
147
the adjustable "U" shapped door guides located within these members had t o be remachined t o prevent the doors from binding. The l i f t i n g mechanism was unduly complex.
A s u b s t a n t i a l increase i n
both r e l i a b i l i t y and control would have been achieved by providing each door with i t s own motor-gear reducer u n i t with a b u i l t - i n brake on t h e moA s m a l l , fast-acting brake i n t h e high-speed end of t h e
t o r r o t o r shaft.
drive would have reduced the number of electromechanical devices and reduced o r eliminated t h e problem of coordinating time constants between clutches, brakes, and motor.
The a b i l i t y t o operate t h e doors indepen-
dently w a s probably unnecessary. The u n r e a l i s t i c close tolerances between t h e doors and t h e radiator enclosure and d i s t o r t i o n s produced by thermal gradients, primarily i n t h e doors, required a massive e f f o r t t o ' c o r r e c t i n order t o provide t h e required seal.
These factors should be given careful consideration i n any
future design.
6.7
b 6.7.1
Performance of Main Blowers, MB-1 and MB-3
Description The two main blowers are a p a r t of t h e heat r e j e c t i o n equipment where
the heat generated by t h e reactor i s dissipated t o t h e atmosphere.
The
blowers were manufactured by t h e Joy Manufacturing Co. and a r e t h e "Axivane" type Model AR-600-3600-1225.
The blowers a r e direct-connected t o 250-hp,
1750-rpm motors by a short drive s h a f t with a f l e x i b l e shim coupling on each end. lthough t h e blowers r e e s s e n t i a l l y muse blowers were each o r i g i n a l l y 114,000 cfm of f r e e a i r de 100,000 cfm from each b l operating range of t
en o r i g i n a l l y purchased f o r another program, eginning of t h e MSRE program.
The
d at 82,500 cfm a t 15 inches of water o r The MSRE r a d i a t o r design requirement of
9 inches of water w a s within t h e normal , Had new blowers been proc
MSRE, t h i s same type of equipment would have been considered a s a t i s -
factory choice.
148
6.7.2
LJ
Preoperational Testing
de
preoperational t e s t i n g of t h e blowers consisted of about 25 hours
of t o t a l operation.
When t h e blowers were first delivered and i n s t a l l e d ,
they were operated a short t i m e ( l e s s than 10 hours) t o provide t h e elect r i c a l load f o r t e s t i n g t h e d i e s e l generators.
The blowers were a l s o oper-
ated f o r a s h o r t period of t i m e p r i o r t o power operation of t h e MSRE t o check t h e r a d i a t o r tubes f o r self-excited vibrations, t o permit a calibrat i o n of t h e s t a c k a i r flow instrumentation, and t o determine if control c i r c u i t r y w a s necessary t o avoid c e r t a i n abnormal operating conditions. During t h e abnormal operation t e s t i n g , the blade p i t c h was found t o be
less than t h e s p e c i f i e d 20 degrees. t e s t program was repeated.
The p i t c h w a s s e t c o r r e c t l y and t h e
The t e s t s indicated t h a t t h e blowers would
operate s a t i s f a c t o r i l y a t a l l t h e normal conditions, but t h a t a "surge" condition would occur i f t h e bypass damper were p a r t i a l l y closed when t h e r a d i a t o r doors were a l s o closed.
The manufacturer had s t a t e d during dis-
cussions of t h e t e s t program that t h e "surge" condition could be expected but t h a t no damage would occur o t h e r than overloading t h e d r i v e motors. The blowers were shut down immediately after t h e "surge" condition w a s en-
cs
countered and administrative control was u s e d t o prevent f u r t h e r operation under surge conditions.
A second p i t c h change w a s made t o 2 2 4 2 degrees
immediately after t h e first full-power operation.
This 22-1/2 degree p i t c h
gave t h e maximumblower performance within t h e power r a t i n g of t h e drive motors.
6.7.3
Operating History The aerodynamic performance of t h e blowers w a s s a t i s f a c t o r y through-
out t h e operating l i f e of t h e reactor.
However, t h e r e were s e v e r a l me-
chanical failures which are l i s t e d i n Table 6.1 and which a r e discussed below.
This table a l s o shows the e f f e c t of t n e f a i l u r e on t*dperation
of the reactor. MB-1 Coupling Failure
- After' about
550 hours of operation, t h e f l e x i -
b l e coupling on t h e motor end o f t h e f l o a t i n g drive shaft f a i l e d during operation of MB-1.
"he motor end o f t h e s h a f t w a s then supported only by
t h e coupling guard, and t h e shaft whip t h a t occurred as t h e blower r o t o r
U
le 6.1
Date
Summary of Main Blower Failures
Failure
Effect on Reactor Operation
6/14/66
Coupling Failure MB-1
Zero-Power Operation for 14 hours
7/17/66
Hub and Blade Failure MB-1
Premature Shutdown from Run No. 7 Partial Power Operation During Run No. 8 Zero-Power Operation for 3 days Partial Power Operation for 6 days Maintenance Performed During Scheduled Zero-Power Operation Completed During Scheduled Maintenance Period
9/17/69
Bearing and Mount Replacement MB-1
Zero-Power Operation for 2 days
9/19/69
Faulty Bearing Replacement MB-1
Partial Power Operation 3 days Zero-Power Operation for 3 days
* CI \Fi
150
coasted down destroyed t h e guard and a l s o t h e coupling on t h e end of t h e s h a f t near t h e blower.
-W
The s h a f t came t o r e s t embedded i n t h e sheet-metal
nose with t h e blower end of t h e coupling near, but completely severed from,
i t s mating flange on t h e rotor.
The motor end of t h e s h a f t extended radi-
a l l y across t h e blower i n l e t at about t h e 4 o'clock position. Scratches and dents on t h e blades indicated t h a t some of t h e debris from t h e f a i l e d couplings had gone through t h e blower i n t o t h e r a d i a t o r However, t h e damage, except t o t h e couplings, appeared t o be super-
duct.
f i c i a l and of l i t t l e consequence.
N o a d d i t i o n a l inspection of t h e blower
was made. The primary cause o f t h e f a i l u r e , we believe, w a s a f a t i g u e f a i l u r e of t h e f l e x i b l e shims.
The exact cause o f t h e shim f a i l u r e i s unknown.
Some
home-made washers were found i n t h e d r b r i s t h a t were not rounded as they should have been.
High s t r e s s e s at t h e corners of these washers could have
s t a r t e d t h e f a i l u r e , o r t h e coupling could have been running with excessive misalignment.
The blower r o t o r and drive s h a f t assembly had been r o t a t e d
by hand when t h e f i n a l blade p i t c h change w a s made about 216 operating hours before t h e f a i l u r e .
There may have been some cracked shims i n t h e
g4
coupling at t h a t time, but t h e shims were s t i l l supporting t h e s h a f t . The couplings were r e b u i l t with new b o l t s , washers, and shims.
The
s h a f t was realigned, and t h e blower w a s returned t o service.
The couplings
on MB-3 were also disassembled and new shims were i n s t a l l e d .
The o r i g i n a l
shims i n these couplings were not i n bad condition. MB-1 Hub and Blade Failure.-After
about 650 hours of a d d i t i o n a l op-
e r a t i o n , t h e r e w a s a second failure of -1.
ure of t h e blading and t h e hub of MB-1.
This w a s a catastrophic fail-
The o u t e r periphery of t h e r o t o r
hub disintegrated and all of t h e blading was destroyed.
Most of t h e frag-
ments were contained i n t h e blower casing, but numerous pieces of t h e c a s t aluminum-alloy hub and blades entered t h e r a d i a t o r duct and some of these a c t u a l l y passed through but did not seriously damage t h e r a d i a t o r tube bundle.
Figure 6 . 3 shows t h e f a i l e d hub and t y p i c a l pieces of t h e blading.
An inspection of t h e broken pieces of MB-1 revealed numerous "old" cracks i n t h e blades and i n t h e h& as evidenced by darkened o r d i r t y areas on t h e fractured surfaces.
One blade i n p a r t i c u l a r had f a i l e d along a
h*'
151
Fig. 6.3
Outer P e r i p h e r y of F a i l e d MB-1 Hub and Q p i c a l Blade F r a c t u r e s
152
large "old" crack.
The hub had contained s h o r t , 1-1/2 t o 2-inch, circum-
f e r e n t i a l cracks at t h e base of 8 of t h e 16 blade sockets and t h e f a i l u r e generally followed these cracks.
Figure 6.4 i s a piece of t h e hub showing
t h e darkened areas t h a t indicated previous cracking a t t h e base of t h e blade sockets.
Similar cracking w a s found i n t h e hub of t h e spare blower
t h a t had been i n storage and a continuous crack extending about 35%of t h e circumference as well as some of the s h o r t e r cracks were found i n t h e MB-3 hub.
No cracking was found i n t h e blading of MB-3 o r t h e spare blower.
The exact cause of t h e MB-1 failure i s uncertain, but a blade f a i l u r e at
one of t h e e x i s t i n g cracks seems t o be t h e most probable first event.
All three of t h e blowers were r e b u i l t by t h e manufacturer with new redesigned hubs and l i g h t e r magnesium-alloy blading.
The hubs were rein-
forced with r a d i a l r i b s t o reduce t h e bending moments i n t h e areas where t h e cracking had been observed.
There have been no f u r t h e r cracking prob-
lems i n e i t h e r of t h e r e b u i l t units i n about 11,000 eration.
- 12,000 hours of
op-
A more complete discussion of t h e f a i l u r e s and of t h e corrective
action t h a t was taken is given i n References 42 and 43. Thrust Bearing Failures -There
were a t o t a l of s i x f a i l u r e s assoc-
i a t e d with t h e main t h r u s t bearing o r i t s mounting.
Main blower MB-3 was
taken out of service on March 6, 1967 f o r a bearing replacement a f t e r t h e vibrations had increased from a normal value of about .8 t o 2.5 mils.
The
bearing contained an excessive quantity of very o l d appearing grease, and t h e bearing surfaces were severely worn and p i t t e d .
This bearing w a s ap-
parently from t h e replacement blower t h a t had been i n storage, and t h e bearing apparently had not been cleaned and relubricated when t h e r e b u i l t r o t o r w a s assembled. failure.
Improper l u b r i c a t i o n was judged t o be t h e cause of
The bearing had operated about 1800 hours.
Main blower MB-3 was shut down a second time f o r a bearing replacement on January 16, 1968, because of an increase i n bearing vibration.
Although
t h e vibration and temperature had been normal, t h e MB-1 bearing w a s a l s o found t o be defective and was replaced a t t h e same time.
The MB-1 bearing
had one p i t t e d b a l l , but t h e MB-3 bearing w a s severely damaged and one b a l l had a c t u a l l y fractured.
Figure 6.5 i s a photograph of these two b a l l s .
Since improper lubrication did not seem t o be t h e cause of these two f a i l u r e s and since three bearings had f a i l e d i n r e l a t i v e l y short operating
Fig. 6 . 4
Fragment of MB-1 Hub Showing Areas of Previous Cracking
154
Fig. 6.5
B a l l Bearing Failures Taken from MSRE Main Blowers MB-1 and MB-3 i n January 1968
bi
times, t h e expected l i f e of these bearings w a s calculated from an estimated s e t of load conditions.
The predicted l i f e w a s 5000 hours as compared t o
operating times of 6630 and 4800 hours respectively f o r MB-1 and MB-3.
As
a r e s u l t of t h i s evaluation, t h e o r i g i n a l radial-type b a l l bearings were replaced with an angular-contact type having i d e n t i c a l mounting dimensions. The angular contact bearing had a greater t h r u s t load capacity. The angular contact bearings were apparently capable of s a t i s f a c t o r y l i f e , but additional d i f f i c u l t y was experienced w i t h t h e self-aligning mount.
The MB-3 bearing was replaced during a scheduled maintenance period
because of a thumping noise t h a t was heard when t h e blower was turned by hand.
The bearing was found t o be i n good condition, but t h e wear p a t t e r n
indicated t h a t t h e outer race had been misaligned w i t h t h e shaft.
The mis-
alignment plus t h e high r a d i a l clearance i n t h i s type bearing caused t h e noise.
No f u r t h e r d i f f i c u l t y was experienced with MB-3.
Blower MB-1 was shut down on September 17, 1969 because of an increase i n vibration.
The vibration increase had been caused by excessive wear i n
the spherical surfaces of t h e self-aligning mount. mount a l s o required replacement of t h e bearing. I
The replacement of t h e
The new bearing failed by
overheating during i t s test run because of i n s u f f i c i e n t clearnaces i n t h e phenolic b a l l r e t a i n e r .
This bearing was then replaced w i t h a bearing of
t h e o r i g i n a l r a d i a l type t h a t was on hand. f o r t h e remaining operation of t h e reactor.
This bearing was s a t i s f a c t o r y
The defective bearing w a s re-
turned t o t h e vendor.
6.7.4
Surveillance of Blower Operation Following t h e hub and
f o r monitoring t h e o ups were mounted horizon motors and thermoco A vibration meter w i t
de f a i l u r e of MB-1, a surveillance program
t h e blowers w a s i n i t i a t e d .
Vibration pick-
each bearing of t h e blowers and drive s t a l l e d i n each of t h e blower bearings. put channels w a s provided f o r each blower,
and t h e vibration of
g w a s recorded i n peak-to-peak
once per s h i f t by t
personnel.
displacement
The thermocouples were moni-
tored by t h e computer and i n i t i a l l y an a l a r m would be given i f t h e temperat u r e exceeded a preselected l i m i t .
This was revised l a t e r t o alarm i f t h e
156 temperature rise above ambient exceeded a l i m i t .
This permitted much
W
t i g h t e r operating l i m i t s because t h e ambient fluctuated w i t h t h e outside
a i r temperature.
I n addition t o t h e displacement readings taken from t h e vibration meters, an output w a s a l s o available t o display t h e vibration t r a c e on an oscilloscope o r other high-speed recorder.
Oscilloscope photographs were
taken f o r reference on new bearings and when t h e vibration meters indicated an increase.
Figure 6.6 shows a comparison of vibration t r a c e s from a nor-
m a l bearing and f r o m t h e bearing containing t h e fractured b a l l .
A complete inspection of t h e blowers including a dye-penetrant examina-
t i o n of t h e hubs and blading was completed annually during shutdown periods.
6.7.5
Discussion and Conclusions Following t h e major rebuilding of t h e main blowers i n
1966, t h e oper-
a t i o n has been generally s a t i s f a c t o r y even though there have been s e v e r a l power reductions caused by bearing problems.
The t o t a l l o s t operating time
caused by t h e various bearing f a i l u r e s w a s 8 days of zero-power operation and 9 days of partial-power operation.
A b e t t e r t h r u s t bearing arrangement
Li
could have been required. The surveillance program t o monitor t h e mechanical performance of t h e
blowers has been very e f f e c t i v e .
All t h e bearing problems except t h e MB-1
p i t t e d b a l l i n January 1968 have been found by an increase i n vibration. When t h e b a l l fractured i n t h e MB-3 bearing, a temperature increase had preceded t h e vibration increase by several days.
However, t h i s tempera-
t u r e increase had gone unnoticed because t h e bearing temperature was well below t h e l i m i t t h a t had been set.
The computer program w a s t h e r e f o r e re-
vised t o alarm on t h e temperature rise above ambient, and much t i g h t e r l i m i t s were s e l e c t e d based on t h e previous operating experience.
This
alarm system was e f f e c t i v e i n detecting t h e defective new bearing on September 19, 1969.
The operating experience with t h e vibration monitoring system has c l e a r l y shown t h a t t h e displacement meter alone i s not adequate t o monitor t h e condition of t h e bearings.
Some system of maintaining a h i s t o r y of
t h e frequency spectrum i s a l s o required.
We used photographs of t h e o s c i l -
loscope t r a c e t o document the normal operating vibrations, and a d d i t i o n a l
u
157
Normal Trace -Taken 12/19/67
Trace w i t h
ured Ball -Taken 1/16/68 lower MB-3 Vibration Traces
158
photographs were taken f o r comparison when an increased vibration w a s indicated by t h e displacement meters.
A modification t o t h e computer t o in-
crease i t s sampling r a t e would have permitted t h e frequency spectra t o be generated by t h e computer.
This would have provided a more convenient and
more precise spectrum, but some degree of background experience would s t i l l be required f o r proper i n t e r p r e t a t i o n o f t h e s p e c t r a i n e i t h e r case because
there can be several normal frequencies.
There were several instances of
abnormal vibrations when t h e bearings were not a t fault.
The f i n a l de-
cision t o replace a bearing w a s made i n each case by t h e sound and f e e l of t h e blower during a coastdown from operating speed.
6.7.6
Recommendations The main blowers were conventional commercial equipment which would
normally be expected t o operate w i t h a minimum of a t t e n t i o n o r maintenance. The instrumentation of all r o t a t i n g equipment i s probably impractical i n regard t o both cost and manpower.
However, c e r t a i n c l a s s e s of equipment
should be instrumented and monitored on a routine basis.
The equipment
should be selected on t h e basis of: 1. Size and cost
2.
The a c c e s s i b i l i t y during operation
3.
The e f f e c t of shutdown on o v e r a l l p l a n t operation
4.
The p o t e n t i a l f o r self-destruction
I d e a l l y t h e monitoring should be computerized but t h i s i s not a c t u a l l y required. Each type and piece of equipment w i l l have i t s own operating character i s t i c s i n regard t o vibration o r temperature, and these c h a r a c t e r i s t i c s may vary with t h e i n s t a l l a t i o n o r operating conditions.
Preli&nary limits
may be obtained from the manufacturer o r from Reference 44, but t h e f i n a l operating limits should be selected from a background of normal operating experience. "he experiences w i t h t h e cracked hubs and w i t h the o r i g i n a l t h r u s t bearing design are indications t h a t t h e Q u a l i t y Assurance Program should extend i n t o t h e design phase as w e l l as t h e fabrication phase of major o r c r i t i c a l items of equipment.
t4
159
6.8 6.8.1
Coolant Drain Tank
Description The coolant drain tank was similar t o but smaller than t h e f u e l f l u s h
tank.
It w a s 40 in. i n diameter by 78-in. high and had a volume of 50 f t 3 .
A s t h e r e was e s s e n t i a l l y no a f t e r h e a t i n t h e coolant salt, cooling was not
provided.
Two s a l t l i n e s (204 and 206) and t h e i r associated freeze valves
were used f o r f i l l i n g and draining t h e coolant c i r c u l a t i n g loop. The weight of salt was indicated by forced balance w e i g h c e l l s .
I
The
w e i g h t s were recorded on s t r i p charts but could be read more accurately
from i n s t a l l e d mercury manometers.
Two conductivity type .probes (one near
the bottom and t h e other n e a r - t h e top elevation of t h e s a l t ) were provided
f o r use as reference points. 6.8.2
Calibration of t h e W e i a Cells Using Lead..Wei&ts The weigh c e l l s were c a i b r a t e d by loading t h e drain tank w i t h l e a d
billets.
The drain tank was a t ambient temperature during t h e c a l i b r a t i o n .
There was good agreement between the manometer readings and t h e a c t u a l weight added.. D i f f i c u l t y was encountered with a i r and mercury leaks i n t h e readout system.
The weigh c e l l s were f u r t h e r c a l i b r a t e d during t h e
addition of coolant salt. 6.8.3 -
Addition of Coolant S a l t Twenty-two batches of coolant salt (5,756 l b s ) w e r e added f r o m cans i n
a portable furnace d i r e c t
t h e coolant drain tank through a heated l i n e
attached t o t h e t o p of t h
After t h e addition w a s complete, t h i s
l i n e w a s cut off and welde
, A good l i n e a r p l o t of t h e manometer
readings vs a c t u a l weight w a s obtained except f o r a 150-lb s t e p i n t h e curve which occurred whe repaired.
The r e l a t i o n s h i p
n f i t t i n g i n t h e w e i g h c e l l tubing w a s een t h e weigh c e l l readings and t h e loca-
t i o n s o f t h e probe lights The weighing system o
more s t a b l e than on t h e
at about 12OO0F, t h e ext period of a week, was k 22 l b .
oolant drain tank proved t o be considerably n tanks.
With 5756 l b of salt i n t h e tank indicated weights, taken over a
160
6.8.4
Heatup and Cooldown Rates These are described under heaters i n Section 17.
6.8.5
Discussion and Recommendations
No unusual d i f f i c u l t i e s were encountered w i t h t h e coolant drain tank. The weigh c e l l s functioned s a t i s f a c t o r i l y , however, t h e trouble w i t h t h e f u e l drain tank weigh system emphasizes t h e need f o r c a r e f u l consideration of piping s t r e s s e s o r t h e use of other type instruments.
161
7.
COVER-GAS SYSTEM P. H. Harley
The MSRE cover-gas system consisted of a helium t r a i l e r f o r normal supply, two banks of t h r e e helium cylipders f o r emergency use, two p a r a l l e l t r e a t i n g s t a t i o n s for removing moisture and oxygen t o a nominal 1 ppm by volume, a t r e a t e d helium surge tank, and two headers, 35 p s i g and 250 psig, which supplied cover gas t o t h e various systems.
Each of t h e t r e a t i n g sta-
t i o n s consisted of a dryer (molecular sieve) , preheater, and an oxygenremoval u n i t (heated titanium sponge).
Moisture and oxygen analyzers could
be valved i n t o monitor the t r e a t e d o r untreated helium as desired.
7.1
I n i t i a l Testin&
-The components o f t h e t r e a t i n g s t a t i o n were developed and designed by t h e Reactor Division Development Section.
hj
The performance of t h e sys-
tem w a s t e s t e d by measuring oxygen and moisture content of t h e cover gas upstream and downstream of t h e t r e a t i n g s t a t i o n t o insure proper operation.
Other t e s t i n g a t t h e MSRE consisted of leak-testing t h e system and
checking flow control and temperature controls.
Thermal expansion of t h e
O2 removal beds during t h e i n i t i a l heatup caused t h e 3-in. r i n g j o i n t flanged heads t o leak.
After retightening t h e flanges, t h e u n i t s were kept
hot t o prevent thermal cy L a t e tests indicated
which w a s i n s t a l l e d t o red susceptible t o mois
Fisher Model 67-H pressure regulator from t r a i l e r pressure t o 250 p s i g w a s rough t h e diaphram. These tests indie Model RBX-204-15 were b e t t e r from
162
t h i s standpoint.
The Victor model was less complicated and therefore w a s
W
i n s t a l l e d i n t h e MSRE. The low-pressure cover-gas header was i n i t i a l l y a 40-psi header which
was protected by a 48-psig rupture disc.
During i n i t i a l t e s t i n g , t h i s rup-
t u r e d i s c f a i l e d on several occasions without apparent cause.
The header
pressure w a s subsequently lowered t o 35 p s i g which w a s s u f f i c i e n t f o r all normal operating requirements and did a l l e v i a t e t h e problem.
The s a f e t y
l i m i t s 2 0 (Lo H e supply pressure) had t o be lowered from 30 t o 28 p s i g t o
provide a margin from the normal operating pressure. Four helium control valves failed i n t h e first month of operation. Some g a l l i n g of t h e close f i t t i n g t r i m (17-4 PH plug and S t e l l i t e s e a t )
was apparently caused by the extremely clean, non-lubricating d r y helium. Other valves i n t h e system, including the four replacements have operated
sat i s f a c t o r i l y
.
7.2
Normal Operation
Since p u t t i n g the cover-gas system i n operation i n t h e f a l l of 1964, sixteen t r a i l e r s of helium have been used.
About 21,800 f t 3 of t h e
29,000 f t 3 i n each t r a i l e r was used before returning t h e t r a i l e r t o be re-
filled.
A t r a i l e r l a s t e d about
4 months on t h e average o r an average usage
of approximately 180 cubic feet p e r day. operating w a s 260 cubic f e e t per day.
Normal usage when t h e r e a c t o r was
Larger amounts were used during pur-
ging and f i l l i n g procedures and very l i t t l e w a s used during maintenance periods. Although helium w a s t h e normal cover gas, during maintenance periods t h e r e a c t o r system w a s purged with argon.
Argon w a s a l s o used during a
s p e c i a l study of c i r c u l a t i n g gas i n t h e fuel. &ppm
Bottles of argon containing
moisture were placed on one of t h e emergency cover-gas headers and
then fed through t h e heating s t a t i o n i n t o t h e regular cover-gas distribut i o n system. The cover gas w a s changed from one t r e a t i n g s t a t i o n t o t h e other only once i n t h e five years of operation.
This w a s done i n February 1967 when
trouble w a s encountered i n t h e power supply control c i r c u i t t o t h e operat i n g unit.
Although t h e dryer which had been i n operation p r i o r t o t h e
LJ
163 change was regenerated, there has been no indication t h a t t h e capacity of any dryer o r 02-removal u n i t has been exceeded. Spectrographic analysis of t h e helium received averaged 2 ppm oxygen and. 4.6-ppm moisture. helium contained
A f t e r passing through t h e t r e a t i n g s t a t i o n , t h e
0.2-ppm 02 and from 0.5 t o 10-ppm mo&ture.
The vari-
a t i o n i n moisture analysis appeared t o follow t h e ambient temperature of t h e helium t r a i l e r and t r e a t i n g s t a t i o n .
During hot weather t h e analyses
w e r e high and during cold weather t h e r e s u l t s were low.
Fluctuations could
be seen from day t o night temperature changes of several ppm moisture.
This f l u c t u a t i o n of moisture analysis w a s decreased by b e t t e r heating of the sample s t a t i o n location i n t h e winter and b e t t e r v e n t i l a t i o n of t h e
area during t h e summer months While i n v e s t i g a t i n g t h e
i s t u r e analysis-problem, another dryer was
i n s t a l l e d i n t h e main helium header between.the nonnal t r e a t i n g s t a t i o n and t h e l i n e t o t h e analyzers.
Whether t h i s second drying s t e p helped i s doubt-
ful but .it d i d give assurance t h a t t h e helium w a s as dry as possible.
The e l e c t r o l y t i c c e l l i n t h e moisture analyzer caused considerable d i f f i c u l t y during e a r l y operation.
The c e l l was replaced four times be-
tween June and October 1965, once i n June 1966, and i n January 1967. c e l l replacements have been necessary since then.
of t h e c e l l s was improved.
No
Apparently t h e q u a l i t y
Only one e l e c t r o l y t i c c e l l has f a i l e d i n t h e
oxygen analyzer and t h a t was i n June 1965.
To check t h e f i n a l performance of t h e t r e a t i n g s t a t i o n i n October 1969, a cylinder of He containing moisture was connected t o t h e analyzers. The analyzers read 20-ppm moi through t h e t r e a t i n g s t a t There has only been i n i t i a l testing.
This w
e and 0.6-ppm 02.
A f t e r t h e gas w a s run
analyzed 0.9-ppm moisture and 0.2-ppm 02.
failure of the 48-psig rupture d i s c since t h e sed by improper i n s t a l l a t i o n .
Each time the
s i d e of t h i s rupture d i s
s t was run, pressure w a s put on t h e back eveht it f r o m breaking. I n August 1969,
when t h e backpressure wa
ased, t h e rupture d i s c s t a r t e d leaking.
During replacement, t h e
support was found t o be on t h e wrong s i d e
60-psig f u e l system pres
of t h e disc.
Repeated flexing, when backpressure had been applied, was-
t h e apparent cause of t h e f a i l u r e .
More importantly, . t h e f a u l t y i n s t a l l a -
t i o n had n u l l i f i e d t h e s a f e t y function of t h e disc.
164
7.3
Conclusions and Recommendations
"he cover-gas system performed q u i t e s a t i s f a c t o r i l y w i t h a minimum amount of maintenance o r operator time.
Since t h e moisture and oxygen i n
t h e helium as received was very l o w , it i s doubtful t h a t t h e t r e a t i n g s t a t i o n s were r e a l l y necessary.
By obtaining analysis of t h e cover gas before
usage and having an on-line analyzer, t h e t r e a t i n g s t a t i o n could probably be eliminated.
I n l a r g e r r e a c t o r s , t h e t r e a t i n g s t a t i o n might be needed
t o dry t h e e a r l y recycle gas.
The MSFE charcoal beds remained "wet" f o r
a long period during e a r l y operation. W e have had considerable f l u c t u a t i o n i n t h e indicated moisture.
If
t h e t r e a t i n g s t a t i o n and analyzers w e r e located i n an a r e a which had a con-
s t a n t temperature, t h e analyzers would be more consistent.
If no t r e a t i n g
s t a t i o n i s used, t h e helium supply should be kept a t a constant temperature.
Rising temperature drives moisture out o f t h e container w a l l s and
decreasing temperature causes t h e w a l l s t o absorb moisture.
(14
165 8.
OFFGAS SYSTEM
A. I. Krakoviak 8.1
Description *
The offgas systems were comprised of those l i n e s , filters, charcoal beds, and valves through which cover-gas flowed from t h e s a l t systems t o t h e environment.
The off-gas systems f o r t h e coolant s a l t and f o r t h e
f u e l s a l t were similar i n many respects, but t h e coolant off-gas system
was simpler because i t did not have t o d e a l with t h e i n t e n s e t y radioactive f i s s i o n products found i n t h e ' f u e l off-gas during pdwer operation. 8.1.1
Coolant Off-Ras System Figure 8.1 i s t h e flowsheet f o r t h e coolant off-gas system (simplified
by t h e omission of elements which are not p e r t i n e n t t o t h e presentation Gas entered t h e coolant pump bowl
of operating experience i n t h i s r e p o r t ) .
through bubbler l e v e l i n d i c a t o r s and as a purge t o t h e pump s h a f t annulus. The s h a f t purge flow was s p l i t , with some flowing downward i n t o t h e pump bowl and t h e rest upwards and out through l i n e 526.
The upflow was t o
prevent o i l fumes from coming down the annulus i n t o t h e pump bowl. ever, some o i l (about 0.5
- 1.0
How-
cc/day) did g e t i n t o t h e pump bowl by a
&
leakage path around t h e s h i e l d plug.
This o i l was thermally decomposed
i n t o vapors and soot t h a t l e f t with t h e off-gas through l i n e 528.
tias
ILOWS
m c o ana
OUL
or r;ne pump
DOWL
w e r e CunLLnuous.
uaa w a s
admitted a t a fixed rate and t h e pump bowl pressure'was controlled by t h r o t t l i n g PCV-528 i n
r gas flow ( l i n e s normally vented f t h e c i r c u l a t i n g loop, All aiaine i n th
ess steel pipe.
The va
t i o n on experience.
The
ystem.
The d r i v i n g f o r c e for t h e
r e s s u r e d i f f e r e n t i a l ticross PCV-528. o l a n t d r a i n tank
e by p r e s s u r i z i n pas svstem w a s 1/2-inch Sch-40 s t a i n -
ters, and t r a p s are described i n t h e secas located i n t h e coolant d r a i n c e l l ,
which could be s a f e l y entered whenever t h e r e a c t o r power was s h u t down.
166
Fig. 8.1
Simplified Flowsheet of MSRE Coolant Off-Gas System
167
b,
8.1.2
Fuel Off-gas System As seen i n Fig. 8.2, t h e flowsheet f o r t h e f u e l s a l t off-gas was
b a s i c a l l y similar t o t h a t f o r t h e coolant ofâ&#x201A;Ź-gas, but w a s more complex. Three d r a i n tanks were'required i n t h e f u e l system (two f o r f u e l , one f o r f l u s h s a l t ) and an overflow tank w a s required f o r t h e f u e l pump bowl. major d i f f e r e n c e was t h a t a1
A
u e l system off-gas w a s s e n t through charcoal
beds f o r f i s s i o n product absorption before release t o t h e stack.
The gas
hold-up and cooler consisted of a 68-ft length of 4-inch Sch-40 s t a i n l e s s s t e e l pipe.
The remainde
off-gas piping with t h e exception of t h e
f l e x i b l e jumper l i n e near t h e p a r t i c l e t r a p and l i n
bowl and t h e l i n e i n t h e v i c i n i t y of 57 w a s 1/2-in.
Sch-40 s t a i n l e s s s t e e l pipe.
Therefore, a t the normal off-gas flow rate o
-1 liters per minute, t h e
f i s s i o n gases were a t l e a s t 45 minutes old o
rrival at the particle trap
and approximately 100 minutes old on a r r i v a l a t t h e main charcoal bed. An a u x i l i a r y charcoal bed w a s provided t o accommodate l a r g e venting
rates ( f o r s h o r t i n t e r v a l s ) ,such as were required during f i l l i n g and draining t h e r e a c t o r or during s a l t t r a n s f e r s .
'c,
A sampling system was a l s o
provided t o study t h e concentrations of f i s s i o n products and o i l decompos i t i o n produces i n t h e f u e l off-gas. The f u e l off-gas system was contained and shielded and t h e r e f o r e required remote maintenance t
iques f o r work on any portion upstream of
t h e charcoal beds. 8.2
Experience with Coolant S a l t Off-gas System
A summary of t h
encountered with the coolant off-gas e s c r i p t i o n of these i
n t r o l valve (PCV-528)
l a n t system, t h e o r i g i n a l pressure t t l e d t h e off-gas from t h e coolant pump ate pressure r e l i e f and w a s replaced
by a valve having l a r g e r
cient
(q =
0.02).
168
tJ
169
Table 8.1.
Summary of D i f f i c u l t i e s with t h e Coolant Off-gas System
Description
Date
I n s t a l l e d l a r g e r control valve (PCV-528) t o 0.02)
Jan. 1965
Added a 2511 f i l t e r ahead of t h e control valve
Jan. 1965
Replaced t h e 25p f i l t e r (1.4 sq in.)
Feb. 1965
I n s t a l l e d a l a r g e r 25p f i l t e r (50 sq in.)
June 1965 Jan. 1966
I U
(Cv 0.032
Jan. 1965
â&#x20AC;&#x2122;
Control valve plugged -cleaned a 1 v f i l t e r (50 sq in.) Replaced 1 p f i l t e r
Feb. 1966
11
11
11
Feb. 1967
11
I1
11
Feb. 1967
11
It
11
June 1967
11
11
11
Sept 1967
I1
11
11
Jan. 1968
I1
11
11
Sept 1968
I1
I1
11
Sept 1968
t h i s and i n s t a l l e d
Restriction a t coolant pump - c l e a r e d
by backblowing
Nov. 1968
Restriction a t coolant pump - c l e a r e d by heating with a torch. I n s t a l l e d heaters on t h e l i n e . Cleaned off gas l i n e s and replaced 1 p f i l t e r .
Feb. 1969
Restriction a t coolant pump with electrical heaters.
June 1969
Control valve, 1 p f i l t e r and check valve cleaned and back-diffusion preventer removed.
Sept 1969
Replaced 1p f i l t e r .
- cleared by heating
170
A s i n t e r e d metal f i l t e r (tubular c a r t r i d g e one-inch long by 1/2-inch
L)
diameter) capable of stopping p a r t i c l e s g r e a t e r than 2 5 w ~as installed i n the off-gas l i n e t o p r o t e c t t h e pressure c o n t r o l valve.
It w a s necessary
t o replace t h i s f i l t e r (because of excessive pressure drop) a f t e r only 24 days of service with s a l t i n t h e coolant loop.
Inspection showed t h a t
t h e plugged f i l t e r was covered with amorphous carbon containing traces of t h e coolant s a l t and INOR-8.
A second i d e n t i c a l c a r t r i d g e plugged a f t e r
20 days of service and a t the completion of Run 1, (see Table 3-2 f o r dates
of runs) i t was replaced with one having 35 t i m e s t h e surface area.
Cool-
a n t s a l t was not c i r c u l a t e d again f o r 3-1/2 months and then f o r only 118 hr.
During c i r c u l a t i o n , pressure control again became erratic which indi-
cated an obstruction of e i t h e r t h e f i l t e r o r t h e valve.
Both w e r e removed
f o r inspection, and although there was no deposit on the f i l t e r , the valve
was p a r t i a l l y obstructed by a black, granular material. Rinsing w i t h acetone restored t h e o r i g i n a l flow characteristics of t h e valve, and it w a s r e i n s t a l l e d f o r t h e next run.
The f i l t e r w a s replaced with a similar one
but capable of removing p a r t i c l e s g r e a t e r than
1 u i n diameter.
The smaller
pore size w a s chosen because the black granular material was determined t o
cd
be a mixture of glassy spheroids of coolant s a l t approximately,lU i n di-
ameter and carbon o r carbonaceous material.â&#x20AC;&#x2122; It i s n o t known whether these materials were c a r r i e d i n t o t h e l i n e continuously o r w e r e swept o u t of t h e pump bowl by sudden venting. The valve w a s r e l a t i v e l y trouble-free a f t e r t h e i n s t a l l a t i o n of t h e 1 u f i l t e r . The f i l t e r , however, experienced gradual plugging and w a s changed o r cleaned twice i n 1966, t h r e e times i n 1967, t h r e e t i m e s i n 1968, and twice i n 1969.
The f i l t e r s , when removed
from t h e system, showed no evidence of foreign matter o t h e r than a l i g h t
f i l m of o i l i n t h e f i l t e r housing and a l i g h t f i l m of o i l w a s t r a n s f e r r e d from t h e porous metal f i l t e r onto the p l a s t i c bag used i n removing t h e filter, Flow tests on the plugged f i l t e r s showed pressure drops of less than 0.1 p s i a t flow rates two and t h r e e t i m e s g r e a t e r than the normal MSRE rate ( e s s e n t i a l l y t h e same r e s u l t s were obtained on a test of t h e f u e l off-gas filter).
Plugging i n these f i l t e r s w a s a t t r i b u t e d t o t h e l i q u e f a c t i o n of
oil i n t h e pores of t h e f i l t e r .
The f a c t t h a t t h e pressure drop (at nor-
m a l MSRE flow rates) changed from
%lo p s i
before removal t o e s s e n t i a l l y
W
171 zero when t e s t e d a f t e r removal w a s a t t r i b u t e d t o t h e removal of some of t h e o i l from t h e pores onto t h e p l a s t i c bag and rubber gloves during removal and i n s e r t i o n of t h e f i l t e r i n t h e flow test r i g .
The f i l t e r s
were rinsed i n acetone before reuse. On s e v e r a l occasions when t h e f i l t e r became r e s t r i c t e d and could not
be replaced during power op I
the output from t h e pressure c o n t r o l l e r
w a s switched from PCV-528 t
536.
The coolant pump pressure w a s then
controlled with t h i s l a r g e by-pass valve (Cv = 3.5).
Although t h e valve
w a s operated a t e s s e n t i a l l y i t s f u l l y open o r f u l l y closed position, adequate pressure c o n t r o l w a
achieved because t h e check valve and back-
d i f f u s i o n preventer offered During t h e May 1967 shutdown
appreciable r e s i s t a n c e t o l a r g e flows. t h e off-gas piping downstream of HCV-536 w a s
changed t o j o i n l i n e 528 downstream of PCV-528 r a t h e r than l i n e 560.
This
-change routed t h e gas by t h e coolant system r a d i a t i o n monitor so t h a t t h e coolant off-gas could be monitored when i t became necessary t o use HCV-536 t o c o n t r o l t h e system pressure.
On one occasion l a t e i n r e a c t o r operations (September 1969) when t h e
h*,
f i l t e r becaue r e s t r i c t e d , t h e coolant system pressure w a s allowed t o reach
10 p s i g before remedial ac
w a s taken.
i n t h e system meandered ba
d f o r t h between 10 and <5 p s i g f o r approxi-
During t h i s test t h e pressure
mately one month before exceeding 10 p s i and thus requiring pressure cont r o l with t h e vent valve (HCV-536).
This pressure drop i s i n agreement
with t h e gas pressure required t o enlarge o r b u r s t a f i l m of o i l from a one-micron pore i f one
t the bubble radius i s t h e same as t h a t
of t h e pore.
t i o n f o r a s p h e r i c a l bubble i s
The e q u i l
where u = s u r f a c e t e n
r = bubble radius pore =
This c a l c u l a t i o n
l l e r than 111, once plug plugged a t pressure d i f f e r o s c i l l a t i o n experienc
bid
es/cm f o r o i l s i n t h e coolant 0.5 x 10-~ em. a t any pore i n t h e s i n t e r e d metal f i l t e r densed o i l vap
1s of less than 9.'3 p s ttributed t o a n
ould remain
e slow pressure rium condition
of condensation and evaporation of o i l a t the pores of t h e f i l t e r .
172 8.2.2
R e s t r i c t i o n a t t h e Pump B o w l I n September of 1968 approximately one month a f t e r t h e s t a r t of Run
15, i n addition t o a r e s t r i c t i o n a t t h e f i l t e r , a r e s t r i c t i o n developed i n the off-gas l i n e a t the e x i t of t h e coolant pump bowl.
This r e s t r i c t i o n
w a s blown clear once by closing t h e pump off-gas and equalizer valves and
allowing t h e pump pressure t o increase t o 15 p s i g before suddenly venting t h e pump t o t h e drain tank which w a s a t atmospheric pressure.
The r e a c t o r
w a s shut down i n November of 1968 when t h i s r e s t r i c t i o n reappeared and a similar r e s t r i c t i o n developed i n t h e f u e l off-gas system.
During the shutdown, l i n e 528 w a s c u t approximately 3 f e e t downstream
of the pump bowl and a clean-out t o o l which was i n s e r t e d toward t h e pump bowl came out covered with black, heavy grease.
A brown vapor (hydro-
carbons) was collected when t h i s s e c t i o n of l i n e w a s heated with a torch. i
The r e s t r i c t i o n w a s not cleared, however, u n t i l t h e junction of t h e offgas l i n e and t h e pump bowl w a s heated t o red heat.
There w a s p r a c t i c a l l y
no evolution of vapor from t h e l i n e a t t h a t t i m e , suggesting t h a t t h e res t r i c t i o n w a s primarily s a l t o r a combination of s a l t and hydrocarbons.
A h e a t e r w a s i n s t a l l e d on t h e l i n e a t t h e e x i t from t h e pump bowl; a l s o a t r a p which w a s packed with s t a i n l e s s - s t e e l mesh w a s i n s t a l l e d i n t h e l i n e approximately s i x f e e t from t h e pump bowl i n an e f f o r t t o condense o i l vapors a t t h i s l o c a t i o n and thus prevent o r minimize o i l transport downstream t o t h e f i l t e r s and valves.
Also during t h i s shutdown a l l coolant off-gas valves i n t h e coolant d r a i n c e l l were dismantled and cleaned.
The valves and interconnecting
l i n e s and f i l t e r housing were found t o have a l i g h t l a y e r of o i l on t h e inner surface, and approximately 10 t o 20 cc of o i l w a s cleaned from low pockets i n t h e interconnecting l i n e s . Approximately s i x weeks after resumption of operations, t h e l i n e a t t h e pump bowl again showed evidence of a r e s t r i c t i o n .
The h e a t e r w a s
energized and t h e r e s t r i c t i o n cleared when t h e temperature of t h e pipe under t h e heater (normally a t 485OC) increased t o 620째C. gas problems were encountered a t t h i s location.
N o f u r t h e r off-
However, about s i x months
later r e s t r i c t i o n s appeared i n t h e pressure c o n t r o l valve, check valve, and t h e back-diffusion preventer.
During t h e June 1969 shutdown,
tbe f i l -
ter, c o n t r o l valve, and nearby check valve were cleaned and r e i n s t a l l e d ;
i173
b,
t h e back-diffusion preventer was removed completely.
Although a l l four
components were coated with o i l which impeded off-gas flow, i t is believed t h a t the back-diffusion p r r e s t r i c t i o n i n t h e s i n t e r e d metal f i l t e r i n
With t h e exception of September 1969, no f u r t h e
oolant off-gas problems w e r
f i n a l shutdown. I n summary, t h e r e s t r i c t i o n s i n 'the coolant off-gas l i n e s were primarily r e l a t e d t o o i l leakage i n t o the pump bowl.
I n retrospect, a f i l t e r The
of an intermediate pore s i z e (say 2-4u may have been a b e t t e r choice.
r e s t r i c t i o n a t t h e pump exit w a s probably more of a s a l t mist carryover phenomenon r a t h e r than hydrocarbon r e l a t e d .
The plugging a t t h e coolant
pump bowl e x i t , although occurring less frequently, seems t o be similar t o t h a t experienced a t t h e f u e l pump which w i l l be discussed later.
8.3 A summary of t h e f u e l off-gas experience is given i n Fig. 8.3.
The
s a l t c i r c u l a t i n g t i m e s and power operations are shown and t h e type d i f f i c u l t i e s encountered are tabulated.
These are discussed in the following
sections.
8.3.1
Experience with Fuel System Off-gas System P r i o r t o Power Operation The problems encountere
i t h t h e f u e l off-gas system during t h e pre-
c r i t i c a l and low-power opera i n the coolant system. 8.3.1.1
F i l t e r an
ifficulties.
After about two months of
f l u s h s a l t c i r c u l a t i d n during t h e p r e c r i t i c a l t e s t i n g , t h e f u e l system ng t h e shutdown a t t h e end of Run 1
pressure c o n t r o l bec t h e f i l t e r and valve peared clean but t h e v own out with gas and t
both f i l t e r and v of t h e next run, t
Id
ed.
The f i l t e r (pore diameter of ' ~ 2 5 ~ )
as p a r t i a l l y plugged.
The obstruction w a s
w a s washed with acetofie, reassembled, and Approximately a week a f t e r t h e s t a r t came p a r t i a l l y plugged.
This t i m e i t
w a s replaced with er % (0.077 instead of 0.022). The smaller valve w a s then c u t open and a black deposit of amorphous carbon and glassy beads w a s found p a r t i a l l y covering the tapered stem. The acetone
bs U
.
w
0
W
175 I
LJ
r i n s e from t h e f i r s t cleaning (Run 1) and t h e deposit removed from t h e valve during Run 2 contained 1- t o 5-11 glassy beads which proved t o have the composition of f l u s h s a l t .
The s a l t beads i n t h e off-gas l i n e s were
probably frozen droplets of m i s t produced by t h e s t r i p p e r spray i n t h e f u e l pump.
J u s t as i n t h e c
a n t system, i t is not known whether t h e
m i s t was transported continuously o r w a s swept out by sudden venting. The performance of t h e f u e l off-gas system w a s s a t i s f a c t o r y f o r t h e remainder of Run 2 and t h e zero-power exper shutdown following 'Run 3, l a r g e r s i n t e r e d m
n 3).
f i l t e r s ( f i l t e r area
0.34 f t " ) capable of removing p a r t i c l e s gre i n s t a l l e d ahead of t h e pressure control valve
-
During t h e
111 i n diameter were
0 t h t h e f u e l and coolant
off -gas l i n e s . 8.3.1.2
Check valve malfunctions.
The check valves i n t h e off-gas
system are spring-loaded poppet-type valves and were designed f o r low pressure drops.
During p r e c r i t i c a l checkout of t h e system when a b i t of
debris caused one such valve t o remain open, a l l t h e other s i m i l a r valves
were removed f o r inspection, cleaning, and pressure drop checks before
b,
reinstallation.
Those which did not open with a forward pressure of a t
least 5-in. HaO and reclose with a forward pressure d i f f e r e n t i a l of 'b2-in. H a 0 were replaced with new
The next indication o
alves meeting these criteria.
a check valve malfunction w a s i n January 1966
when a r e s t r i c t i o n developed i n t h e check valve downstream of HCV-533.
The
blockage occurred a f t e r t h e f u e l system pressure w a s vented f o r a s h o r t period of t i m e through t h i s valve t o the auxiliary charcoal bed. Although some foreign material, shown i n Fig. 8.4, was found
itl
t h e poppet of t h e
check valve, i t w a s judge
t s u f f i c i e n t t o s t o p t h e gas flow (the plug
may have been dislodged d
g disassembly).
The s o f t O-ring (which makes
t h e s e a l ) and t h e cone of t h e poppet were covered with an amber varnishl i k e material which w a s i d e n t i f i e d as a hydrocarbon.'
The check valve w a s
not replaced and administrative c o n t r o l of HCV-533 w a s used t o prevent a backflow of gas i n t o t h e fuel-pump gas space during venting of t h e d r a i n tanks.
I n April 1966, d i f f i c u l t y i n venting through t h e a u x i l i a r y charcoal bed w a s found t o be due t o a check valve poppet which had become lodged i n
hd
the i n l e t l i n e t o the bed.
Apparently t h e poppet had vibrated loose from
176
F i g . 8.4
Check Valve CV-533
17.7
-u
a check valve a t one of t h e drain tank l i n e s and had been c a r r i e d down-
stream during venting
rations.
prevent reverse flow i n the
The inab'ility of t h e check valves t o l i n e s of a l l t h r e e d r a i n tanks implies
t h a t any and a l l of these valve poppets may have vibrated loose. 8.4.
D i f f i c u l t i e s with the Fuel Off-gas System During Power Ascension
The f u e l system off-gas problems became somewhat more complex during power ascension than those encountered e a r l i e r because of f i s s i o n fragment production i n t h e reactor and subsequent transport of t h e f i s s i o n gases and some s o l i d decay daughters i n t h e off-gas l i n e s .
Some of t h e l u b r i c a t i n g
o i l which leaked i n t o t h e pump bowl (0.5 t o 1.5 g/day)
, was
polymerized t o
a varnish-like substance by t h e high radiation f i e l d i n t h e off-gas l i n e s
and components.
This material, amorphous carbon from t h e cracking process,
and s o l i d decay daughters of the f i s s i o n gases were trapped i n t h e various components of t h e off-gas system such as f i l t e r s , check valves, c a p i l l a r i e s ,
u
valves, t r a p s , and presumably in t h e charcoal beds.
The d i f f i c u l t i e s ex-
perienced during t h i s period are reported by run numbers o r major shutdowns. 8.4.1
Run 4 The f i r s t indication o
power l e v e l was raised t o 0 t o t a l integrated because of a r e s t r i c t i o n s t a r t of t h i s pressure t r a
t h i s type of r e s t r i c t i o n occurred when t h e
MW on January 23, 1966 g f t e r only 3 MWhrs of
system pressure began t o rise slowly i c i n i t y of PCV-522.
Shortly a f t e r t h e
t h e response of t h e drain-tank pressures
a t a capilla'ry flow r e s t r i c t o r i n l i n e 521, t h e fuel-loo f u e l system was re
alizing line.
The excess pressure i n
t i n g gas through HCV-533 t o t h e auxili-
a t PCV-522 w a s cleared ts f u l l stroke.
These opera
n i n the line t o the coal bed, apparently a t H -1/2-hr
reactor w a s made sub
operatio maintained f o r about 24 h r , but t h e
Bd
f t e r t h e reactor power was raised t o
I
'
j 178
I
I
1 MW.
Again, t h e operation of t h i s valve through i t s f u l l s t r o k e w a s ef-
f e c t i v e i n r e l i e v i n g a t least p a r t of the r e s t r i c t i o n .
u
Also on t h i s oc-
casion, evidence of r e s t r i c t i o n s a t t h e charcoal-bed i n l e t s began t o appear. These r e s t r i c t i o n s were bypassed by p u t t i n g t h e two s p a r e charcoal-bed sect i o n s (2A and 2B) i n s e r v i c e and closing t h e i n l e t valves t o t h e two t h a t
were r e s t r i c t e d (1A and 1B).
Within about 6 h r t h e i n l e t s of beds 2A and
When t h e previously plugged beds were checked, a t least
2B a l s o plugged.
one w a s found t o be clear and off-gas flow w a s reestablished through bed 1B.
The combination of d i f f i c u l t i e s i n t h e off-gas system r e s u l t e d i n a r e a c t o r shutdown t o correct these conditions.
The r e s t r i c t i o n s i n t h e
equalizing l i n e (521) and t h e a u x i l i a r y vent l i n e (533) were relieved by removing t h e c a p i l l a r y (FE-521) and a check valve (CV-533, reported earlier). The t h r o t t l e valve and f i l t e r assembly, and a second valve t h a t had been t r i e d b r i e f l y , w e r e removed from l i n e 522.
A l a r g e , r e l a t i v e l y coarse ( 5 0 ~ )
f i l t e r and a l a r g e , open hand valve were i n s t a l l e d a t t h e PCV-522 location.-This arrangement eliminated t h e small, e a s i l y plugged passages associated and i t appeared that t h e f i l t e r would s t i l l remove any p a r t i -
with PCV-522,
LJ
cles l a r g e enough t o plug valve 522B, which w a s t o be used f o r system pres-
s u r e control.
) , t h e flow The o r i g i n a l valve, f i l t e r (pore diam. ~ ~ l p and
element were subseguently examined at the High Radiation Level Examination Laboratory (HRLEL)
.
Although nothing was done t o clear the charcoal beds o t h e r than realignment*of the operating mechanisms t o the valves, successive pressure drop measurements showed that the r e s t r i c t i o n s in the beds had decreased s i g n i f i c a n t l y and before the start of Run 5 , flows through t h e beds were back t o normal.
Flow measurements through t h e a u x i l i a r y bed showed t h a t
the bed had a l s o returned t o normal.
Examination of FE-521 -The flow r e s t r i c t o r consisted of a s h o r t length of 0.08-in.
I D tubing welded i n t o l i n e 521 a t one end and coiled t o f i t
i n s i d e a 3-in.
I D container.
i n t h e container.
The other end of t h e c a p i l l a r y w a s l e f t f r e e
When t h e container was cut open, t h e entrance region w a s
found t o be clean, and only a small deposit was found on t h e container w a l l near the end of t h e discharge region as shown i n Fig. 8.5.
These i s o t r o p i c
p a r t i c l e s were characterized as p a r t l y coalesced amber globules w i t h a
hd
179
PHOTO R27589
Fig. 8.5
I
De
Flow Restrictor Eâ&#x20AC;&#x2122;E-521.
180 r e f r a c t i v e index of 1.520.
Spectrochemical analyses indicated microgram
quantities of lithium, beryllium, and zirconium.
The r e s t r i c t o r w a s not
completely plugged when examined i n the hot cell. Examination of PCV-522 -The s t e m of the control valve shown i n Fig. 8.6A, w a s covered with an amber oil-like coating and there w a s an accumul a t i o n of o i l i n the recess formed by the bellows-to-stem
adapter.
The
tapered f l a t (flow area) on the s t e m had small accumulations of s o l i d s ; the body was coated with a s i m i l a r material and had a semisolid mass of
material on the surface near the seat port as shown i n Fig. 8.6B. scan' of t h i s material indicated the presence of "'Sr, Examination of the F i l t e r Preceding PCV-522 diameter by 15-in.
14'Ba,
- Filter-522
A gamma
and '"'La.
was a 1-in.
long cylindrical type-316 s t a i n l e s s - s t e e l sintered metal
element enclosed by a 1-1/2-in.
schedule-40 pipe.
The pore diameter of t h i s
p a r t i c u l a r f i l t e r was " 1 ~ and the element thickness w a s 1/16 in. area w a s 0.34 f t ' and t h e flow was from t h e outside i n .
The f i l t e r
The f i l t e r assem
bly was removed*on February 8, 1966 and examined a t HRLEL on t h e following day.
The upper t h i r d of -the element was steel gray i n appearance; t h e
lower two-thirds had a f r o s t y white appearance. evidence of buildup o r cake.
A t no place w a s there any
A day later t h e f r o s t y white areas had be-
come darker and tended toward the steel gray appearance.
The index of re-
f r a c t i o n of some of the material scraped from t h e s i d e s of t h e element w a s , 1,524 and a gamma scan' of t h e sample indicated t h e presence of '"'Ba,
'"La,
103
Ru, and '"Cs. The f i l t e r w a s then reassembled f o r flow tests.
The data i n d i c a t e
t h a t f o r a given pressure drop t h e f i l t e r w a s passing only 5% of the corresponding flow of a clean f i l t e r .
6
However, extrapolation of the data t o
the normal MSRE off-gas flow indicates that even though 95% plugged, t h e contribution of the f i l t e r (0.075 p s i measured a t t h e hot c e l l ) t o t h e t o t a l pressure drop (>5 p s i during Run 4) w a s negligible.
Exposure t o atmosphere
could have opened some of the pores i n t h e f i l t e r ; however, a more l i k e l y explanation is t h a t handling and sampling opened enough pores i n the element t o produce these r e s u l t s ,
g9
181 d
PHGTO R278l.2
.
>
Fig. 8.6A
PCV-522 Valve S t e m
Fig. 8.6B
PCV-522 Valve Body
182 8.4.2
Run 5 S a t i s f a c t o r y manual pressure control by t h r o t t l i n g V-522B w a s demon-
s t r a t e d during t h e preceding shutdown; however, a t t h e beginning of Run 5 a f t e r only 2-1/2 hours of power operation (1 MW), t h i s valve showed evidence of rapid plugging and required constant adjustment u n t i l i t w a s i n the fully-open position f i v e hours later.
During one such adjustment,
rapid plugging of t h e charcoal beds occurred.
A t t h e conclusion of s a l t
c i r c u l a t i o n (terminated by a space-cooler motor f a i l u r e ) , t h e r e s t r i c t i o n
a t the i n l e t t o charcoal bed 1 B could not be cleared; therefore, t h e i n l e t valve (HV-621) w a s removed f o r examination a t t h e HRLEL.
With t h e valve
out, t h e pressure drop across the bed w a s much lower but i t w a s s t i l l higher than f o r a normal u n r e s t r i c t e d bed.
An excess of helium (forward blow) w a s
forced through the bed and the pressure drop suddenly decreased t o normal. The cone of t h e i n l e t valve t o t h i s bed (shown i n Fig. 8.7) w a s shiny as though w e t with an o i l - l i k e material.
The s m a l l metallic chips near t h e
l a r g e end of t h e cone were a r e s u l t of t h e sawing operation.
There w a s
some white amorphous powder on t h e tapered s e c t i o n bf t h e valve t r i m and appeared t o be adhered f a i r l y strongly t o t h e m e t a l surface.
A similar
The material removed from t h e s t e m
material was found i n t h e valve body.
w a s described as an i s o t r o p i c , f a i n t l y colored material, varnishlike i n appearance with a pebbly surface.
was
132
The predominant isotope i n t h e material
The r e f r a c t i v e index of t h e varnishlike material w a s 1.526
Te.
COP
pared w i t h 1.50 f o r a d i s t i l l e d f r a c t i o n of t h e l u b r i c a t i n g o i l used i n t h e f u e l pump. 8.4.3
6
Ekp eriments and Alterations DurinR t h e March-1966 Shutdown
Off-gas samples taken while t h e reactor w a s shut down showed an increasing hydrocarbon content i n the gas as t h e r e a c t o r c e l l temperature w a s increased, lending support t o t h e hypothesis t h a t t h e r e w a s a r e s e r v o i r of hydrocarbons i n the holdup-volume portion of t h e off-gas l i n e . practicable t o clean t h e 68-ft-long,
4-in.-diam
It w a s not
pipe, so t h e off-gas l i n e
w a s disconnected a t t h e f u e l pump and i n t h e vent house; l a r g e q u a n t i t i e s of helium w e r e blown through t h e l i n e i n t h e forward and reverse d i r e c t i o n s a t v e l o c i t i e s up t o 20 t i m e s normal.*,â&#x20AC;&#x2122;
Very l i t t l e v i s i b l e material w a s
183
PHOTO R28500
F i g . 8.7
Charcoal Bed I n l e t Valve S t e m HV-621
collected on f i l t e r s a t t h e ends of the l i n e , but t h e r e w e r e f i s s i o n prod-
bi
ucts, and t h e amount doubled when the cell w a s heated from 50 t o 8OOC. V i s u a l observation showed t h a t the head end of t h e holdup volume w a s clean
except f o r a barely perceptible d u s t l i k e film.
A thermocouple w a s attached
t o the holdup pipe near t h e head end f o r monitoring temperatures during power operation.
(When the power w a s subsequently r a i s e d , t h e temperature
indeed r o s e from c e l l a i r temperature of about 55°C a t zero power t o about 113OC a t 7.5 MW.
The temperature increase occurred with a t i m e constant of
about 30 min which w a s not inconsistent with buildup of gaseous f i s s i o n products i n the l i n e . )
As a r e s u l t of extensive laboratory tests on organic vapor t r a p s , 7 t h e f i l t e r - v a l v e assembly w a s replaced with a p a r t i c l e t r a p and an organic vapor t r a p i n series and upstream of the pressure c o n t r o l valve (PCV-522) whose flow c o e f f i c i e n t was increased t o 0.865.
The Mark I p a r t i c l e trap,’
shown i n Fig. 8 . 8 w a s designed t o remove p a r t i c u l a t e s and mist.
G a s from
t h e fuel-pump bowl entered a t t h e bottom of t h e u n i t through a c e n t r a l pipe, reversed flow in t h e Yorkmesh, and passed i n succession through two conc e n t r i c cylinders of porous m e t a l ( f e l t metal) and a bed of inorganic f i b e r s . The f i r s t f e l t metal f i l t e r w a s capable of stopping 96.7% of p a r t i c l e s
greater than 0 . 8 ~and the second -99.4% p a r t i c l e s g r e a t e r than 0 . 3 ~ . The organic vapor trap’ consisted of 13 f e e t of 1-in. sch.-40 s t a i n l e s s -
steel pipe, arranged i n three hairpin sections of approximately equal lengths. The vapor t r a p w a s loaded with 1092 gm of Pittsburgh PCB charcoal and instrumented w i t h thermocouples a t 5, 12, 51, 59, 105, and 113 inches, respectively, from the bed i n l e t .
As a r e s u l t of the experience with and examination of HV-621 and o t h e r off-gas components, the i n l e t valves t o the mafn and a u x i l i a r y charcoal beds and V-522A were replaced w i t h ones having l a r g e r flow c o e f f i c i e n t s .
8.4.4
R u n s 6 and 7 A t t h e s t a r t of Run 6, the pressure drops ( a t normal flow rate) across
the p a r t i c l e t r a p , organic vapor t r a p , and charcoal beds (sections IA and 1 B i n p a r a l l e l service) w a s ~ 0 . 0 5 , 0.7, and 1.6 p s i respectively.
The off-
gas system performance w a s s a t i s f a c t o r y during t h e f i r s t t e n days of R u n 6 w i t h the r e a c t o r operating a t 1 MW o r less.
However, when t h e power was
he,
ORNL-DWG -6-11444R
FINE METALLIC FILTER FIBERFRAX (LONG FIBER) 7
COARSE METALLIC F I L T E R 1
I-0WER WELD3
.......... .. ............. ~~~~
\
SAMPLE 97
&y-.
-SAMPLES 5B AND 6
.
\
\
.
SAMPLE 3 A STAINLESS STEEL MESH
Fig. 8.8
THERMOCOUPLE (TYP)
J
NICKEL BAFFLE ( 8 )
Line 522 Particle Trap (Mark I) and Location of Sampling Points During Subsequent Examination
186 increased t o 2.5 MW, t h e pressure drop across t h e charcoal beds a l s o increased.
Pressurization and equalization experiments established t h a t t h e
r e s t r i c t i o n s were a t t h e i n l e t s t o t h e beds, probably a t t h e packing of steel wool above t h e charcoal.
It was found t h a t t h e pressure drop could
be reduced, usually t o near t h e normal 1.5 p s i , by blowing helium a t 35 p s i g e i t h e r forward o r backward through t h e bed; back-blowing seemed t o be more e f f e c t i v e .
Back-blowing of t h e beds w a s done whenever t h e pressure
drop of two s e c t i o n s i n p a r a l l e l approached 3 psi.
Later, higher pressure
Plugging occurred
drops were t o l e r a t e d before back-blowing was i n i t i a t e d .
when t h e power w a s r a i s e d during t h e approach t o f u l l power with s e c t i o n 1 B plugging more o f t e n than t h e others.
Plugging became less frequent later,
but a t t h e end of Run 6, it w a s s t i l l necessary t o back-blow t h e beds about once a week. P a r t i c l e Trap Performance.
After approximately 1 0 days of operation
a t t h e 5-MU l e v e l , the pressure drop across the p a r t i c l e t r a p s t a r t e d to
increase and reached 6 p s i approximately a week later.
When t h e power was
increased t o 7.5 MW, t h e pump pressure increased t o 9 p s i i n less than 12 hours a t f u l l power.
Since t h e pressure drop across t h e organic vapor t r a p
dd
w a s less than 0.5 p s i and t h a t across t h e charcoal bed w a s kept below 3 p s i by periodic back-blowing,
t h e v a r i a t i o n i n pump-bowl pressure w a s i n t e r -
preted as a corresponding r e s t r i c t i o n i n t h e p a r t i c l e t r a p .
About t h r e e
hours a f t e r an unscheduled power i n t e r r u p t i o n , t h e fuel-pump pressure decreased t o 6.5 psig and t h e following day i t decreased t o 4.3 psig.
How-
ever, a few hours a f t e r t h e r e a c t o r power w a s again r a i s e d t o 7.5 MW, t h e pump pressure increased t o 8 psig i n d i c a t i n g a r e s t r i c t i o n i n t h e p a r t i c l e trap.
A t t h i s time t h e helium purge t o t h e pump s h a f t was decreased from
2.4 t o 1.9 liters per minute i n an e f f o r t t o reduce t h e fuel-system pressure.
The r e a c t o r w a s then taken s u b c r i t i c a l f o r two days.
When t h e power
w a s again increased t o 5 MW, t h e pump pressure increased from 3 p s i g a t zero power t o 8 p s i g and eventually t o 12 p s i g a t f u l l power.
The pump
pressure gradually returned t o t h e 6 t o 8 psig range a f t e r approximately
a week a t f u l l power.
After Run 6 w a s terminated by a component cooling
pump f a i l u r e on May 28, 1966, t h e p a r t i c l e t r a p pressure drop aecreased t o
less than 1 p s i and it remained r e l a t i v e l y u n r e s t r i c t e d f o r a month a f t e r full-power operation was restored i n Run 7.
The pump pressure then gradu-
a l l y increased t o 8.7 p s i g before another power reduction
m-1 f a i l u r e )
c.
187
ki
occurred ending Run 7.
A few hours a f t e r t h e power reduction, t h e pump
pressure returned t o a normal value of 3 p s i g and remained a t t h i s pressure f o r approximately a
t h of zero-power operation before t h e r e s t r i c t i o n
gradually returned.
.
I n summary, pump pressure increases ( r e s t r i c t i o n s i n t h e p a r t i c l e
traps) were usually associated with periods of power increases.
Pressure
decreases were sometimes unexplained but usually were associated with power decreases o r with d e l i b e r a t e
tempts t o clear t h e t r a p such as reverse-
blowing with helium. Rerouting of Line 524.
H e l i u m , a t a rate of 2.4 2/min, w a s introduced
through l i n e 516 i n t o t h e fuel-pump s h a f t annulus j u s t below t h e lower s h a f t The l a r g e r p a r t of t h i s flow goes down t h e s h a f t i n t o t h e pump
o i l seal.
bowl thus preventing f i s s i o n gases and s a l t m i s t from entering t h i s radia t i o n s e n s i t i v e region of t h e pump.
The smaller portion of t h i s flow
( c O . 1 2/min) goes up t h e s h a f t t o prevent o i l vapors from migrating t o t h e
f u e l . s a l t ; i t a l s o a i d s i n transporting any o i l seal leakage t o t h e o i l catch tank.
t,
The d r i v i n g f o r c e f o r t h e smaller flow ( l i n e 524) is t h e pres-
Sure d i f f e r e n c e between t h e f u e l pump and l i n e 522 downstream of t h e f u e l system pressure c o n t r o l valve (PW-522).
However, when PCV-522 was opened
f u l l y (because of r e s t r i c t i o n s i n t h e charcoal beds), l i n e 524 had no press u r e d i f f e r e n t i a l and consequently no gas flow; therefore, it w a s rerouted t o e n t e r t h e off-gas
V-557B.
l i n e a t V-560A downstream of t h e main charcoal bed and
V-557B w a s then t h r o t t l e d t o c o n t r o l system pressure.
On two oc-
casions i n Run 6, sudden increases i n pump pressure ( a t t h e completion of
salt recovery from t h e
tank) caused a gas flow reversal a t t h e
pump s h a f t and gaseous
was c a r r i e d i n t o l i n e 524.
bypassed t h e charcoal bed,
as flow had very l i t t l e decay t i m e .
consequence, gaseous act
iggered t h e closure of t h e r a d i a t i o n block
valve on t h e main off-ga
up krypton and xenon f o
days,- respectively.
t e d of 9 f t of 3-in.
2-1/2 and 30
sched.-10
stainless-
On a t least four
steel pipe loaded with
i t t s b u r g h PCB charcoal.
occasions i n t h e subsequen
s s i o n product a c t i v i t y was blown o r dif-
fused up t h e pump s h a f t annulus.
61
As a
Therefore i n June of 1966, a small charcoal
bed' w a s added t o l i n e 524't The bed
Since t h i s l i n e
Although t h e a c t i v i t y l e v e l i n t h e o i l
catch tank and l i n e 524 increased, t h e r e w a s e s s e n t i a l l y no a c t i v i t y re-
lease, thus proving t h e effectiveness of the added charcoal bed.
However,
188 t h e last two releases from t h e pump bowl caused t h e flow element i n l i n e
524 t o plug p a r t i a l l y and then completely. The plug w a s found t o be i n t h e s i n t e r e d s t a i n l e s s - s t e e l d i s c at t h e i n l e t t o t h e matrix-type flow element.
The element was replaced with a c a p i l l a r y tube.
8.4.5
Experiments and A l t e r a t i o n During t h e June-September Shutdown (1966)
bj
A t t h e conclusion of Run 7, it was decided t o f l u s h o u t t h e r e s i d u a l
f u e l i n t h e overflow tank and t o check t h e indicated l e v e l a t which overflow occurred.
Because of an insufficiency of s a l t i n t h e d r a i n tank, pres-
s u r i z i n g gas from t h e d r a i n tank entered t h e r e a c t o r and flooded t h e pump bowl with s a l t , thus binding t h e pump s h a f t with frozen s a l t and pushing
s a l t i n t h e gas l i n e s a t t h e top of t h e pump bowl. The frozen s a l t w a s cleared from t h e reference bubbler l i n e with t h e
use of remotely-applied e x t e r n a l heaters; s a l t i n t h e sampler l i n e was melted by t h e same technique but i t ran down and r e f r o z e a t t h e j u n c t i o n with the pump bowl.
This obstruction and t h e frozen s a l t i n the annulus
around t h e pump s h a f t w e r e cleared when t h e pump bowl w a s r e f i l l e d and heated t o 650OC.
Although some s a l t entered t h e main off-gas l i n e as indi-
cated by a temporary rise i n temperature a t TE-522-2 (Fig. 8 . 9 ) , pressure drop measurements showed no s i g n i f i c a n t d i f f e r e n c e from t h e clean condition. This w a s a t t r i b u t e d t o t h e b l a s t of compressed helium from t h e d r a i n tank I
t h a t w a s released backward through t h e off-gas l i n e , before t h e s a l t had time t o f r e e z e completely, when t h e o v e r f i l l triggered an automatic drain.
Therefore t h e only work done on the main off-gas l i n e (522) during t h i s shutdown w a s t o replace t h e s h o r t , f l e x i b l e "jumper" s e c t i o n of t h e l i n e , where t h e convolutions would be expected t o hold some s a l t . The p a r t i c l e t r a p was a l s o removed from l i n e 522 and sent t o HRLEL f o r d e t a i l e d examination.
An i d e n t i c a l replacement (Mark I) w a s i n s t a l l e d t o
permit operation while t h e o r i g i n a l was examined and a new one designed and constructed.
Also during t h i s shutdown, a remotely replaceable h e a t e r w a s designed and i n s t a l l e d on t h e i n l e t s e c t i o n of t h e a u x i l i a r y charcoal bed (ACB).
8.4.6
Runs 8, 9, and 10 With t h e resumption af operations t h e off-gas flowed f r e e l y , with no
unusual pressure drop f o r 26 days of c i r c u l a t i n g helium, f l u s h s a l t , and
(B
189
ORNL-OWG 67
-
4764
SS HOLDUP ME, 4 in. D I N
L,
OVERFLOW TANK -c
TO PARTICLE TRAP AND CHARCOAL BEDS
Fig. 8.9
-
Off-Gas Piping Near Fuel Pump and Overflow Tank
,
190 f u e l s a l t a t low power.
Then, two days a f t e r power operation w a s resumed
W
a t 5.8 MW, a plug developed i n l i n e 522 somewhere between t h e pump bowl and the junction of the overflow tank vent with t h e 4-in.
holdup l i n e .
The
f i r s t indication was a decrease over a few hours from 107OC t o 7l0C a t TE522-2, as t h e plug caused t h e off-gas t o bypass through t h e overflow tank (OFT).
The presence of the plug was confirmed when HCV-523 w a s closed t o
build up pressure i n t h e OFT t o r e t u r n s a l t from t h e tank t o t h e pump bowl; pressure i n the pump bowl a l s o b u i l t up.
E f f o r t s t o remove t h e r e s t r i c t i o n
by applying a 10-psi d i f f e r e n t i a l e i t h e r forward o r backward were unsuc-
cess f u l
.
The bypassing of off-gas through t h e OFT d i d not hinder operations except f o r one s p e c i f i c job:
recovery of s a l t from t h e overflow tank.
S a l t slowly but continuously accumulated i n t h e tank during t h e e n t i r e operating l i f e of the MSRE, and it was therefore e s s e n t i a l t o r e t u r n s a l t t o t h e pump bowl two o r three t i m e s a week i n order t o maintain proper
levels.
With a plug i n t h e off-gas l i n e of t h e pump bowl, i t was necessary
t o g r e a t l y reduce helium flows i n t o t h e pump so t h a t t h e overflow tank pressure could be increased f a s t e r than t h a t i n t h e pump bowl t o make t h e salt transfer.
Through t h e remainder of Run 8 , s a l t w a s returned (burped)
from t h e OFT six t i m e s , and on a t least four of t h e s e occasions some f i s s i o n product a c t i v i t y w a s blown o r diffused up t h e pump s h a f t annulus i n t o t h e o i l c o l l e c t i o n space.
This w a s a consequence of t h e reduced helium purge
down the s h a f t annulus and t h e unavoidable, sudden pressurization of t h e pump bowl t h a t occurred a t t i m e s i n the procedure. After "Run 8 w a s terminated, s t e p s were taken t o clear t h e plug from the off-gas l i n e so that the normal s a l t recovery procedure could be used. Frozen s a l t w a s suspected as the cause of t h e plug, so t h e f u e l loop w a s flushed t o reduce r a d i a t i o n l e v e l s , t h e r e a c t o r c e l l was'opened, and specia l l y b u i l t electrical heaters were applied t o t h e l i n e between t h e pump bowl and the f i r s t flange.
Heating alone did not clear t h e plug, but when,
with t h e l i n e hot, 10 p s i w a s applied backward across t h e plug, i t blew through.
The pressure drop came down as more helium w a s blown through un-
t i l i t became indistinguishable from t h e normal drop i n a clean pipe.
I n Run 9, t h e power operation w a s begun 8 h r a f t e r f u e l c i r c u l a t i o n had commenced, but TE-522-2
came up t o only 66"C, i n d i c a t i n g t h a t t h e l i n e
.LJ
191
W
w a s again plugged.
While t o o l s and procedures were being devised, t h e re-
a c t o r w a s kept i n operation, but g r e a t care w a s taken t o avoid g e t t i n g f i s s i o n products o r s a l t spray lowering t h e power t o 10 kW
the-pump s h a f t annulus again.
This e n t a i l e d
4 h r before t h e overflow tank was t o b e emp-
t i e d , then stopping t h e pump 4 h r beforehand t o l e t t h e s a l t m i s t settle. During t h e s a l t t r a n s f e r , t h e f u e l pump was vented through t h e s a m p l e r and a u x i l i a r y charcoal bed.
After t h r e e cycles of t h i s , t h e r e a c t o r w a s drainel
and flushed again i n preparation f o r working on t h e off-gas l i n e . This t i m e heat w a s applied t o t h e s h o r t s e c t i o n of l i n e between t h e second flange and t h e top of t h e 4-in.
decay pipe.
When heating t o about
60OOC d i d not open t h e l i n e , t h e f l e x i b l e jumper w a s disconnected t o permit c l e a r i n g t h e obstruction mechanically. t h e 1/2-in.
I n t h e flange above t h e 4-in.
line,
bore was completely blocked, but t h e weight of a c h i s e l t o o l
broke through what appeared t o be only a t h i n c r u s t of s a l t .
Borescope
inspection showed t h a t t h e rest of t h e v e r t i c a l l i n e w a s p r a c t i c a l l y clean, and t h e r e w a s only a t h i n l a y e r of s a l t i n t h e bottom of t h e horizontal run of t h e 4-in.
(Ced
pipe.
Helium was blown thraugh t h e l i n e a t f i v e t i m e s t h e
normal flow, and t h e pressure drop indicated no r e s t r i c t i o n .
The flange
near t h e pump bowl contained a similar plug which w a s e a s i l y broken. 1/4-in.
A
f l e x i b l e t o o l w a s then i n s e r t e d a l l t h e way i n t o t h e pump bowl t o
prove t h a t a good-sized p
existed.
A new jumper l i n e was i n s t a l l e d
and operation was resumed. Because obtaining samples remotely wxthout spreading contamination would have been most d i
ses w e r e made of t h e material i n
t h e flanges.
had frozen i n t h e li
But i t ap
p l e t e l y blocking i t , d u r i n off-gas stream during s During t h e J u l y cleanout, pipe, b u t l e f t behind a t h i n
e r f i l l of J u l y 1966.
rs apparently melted t
t h e flanges w e r e mechan
d reassembled
off-gas system operated
r i n g Run 10.
8.4.7
*m-I P a r t i c l e
Material i n t h e
on then plugged t ge of s a l t i n t h e co
almost com-
s m a l l passages.
s a l t o u t of t h e
er flanges.
After
t h i s p a r t of t h e
Trap
The second p a r t i c l
ugh Runs 8, 9 , and 10.
This u n i t
behaved i n Runs 8 and 9 much as hadâ&#x20AC;&#x2122;the f i r s t t r a p ; t h e pressure drop oc-
6,
casionally b u i l t up t o 5 t o 10 p s i , beginning two days a f t e r power operations
192 s t a r t e d i n Run 8.
Back-blowing with helium-was e f f e c t i v e i n reducing t h e
W
pressure drop t o 2 t o 4 p s i ; however, i n Run 10, a f t e r t h e f i r s t week of power operation, back-blowing became i n e f f e c t i v e .
Various tacticse were
used t o g e t f i s s i o n gas t o t h e p a r t i c l e t r a p w i t h as l i t t l e delay as possib l e , t o see i f increasing t h e f i s s i o n product heating i n t h e t r a p would drive off t h e material a t t h e r e s t r i c t i o n .
After t h i s proved t o be inef-
f e c t i v e and recognizing t h a t heating caused t h e central i n l e t tube t o expand f a r t h e r i n t o t h e Yorkmesh f i l t e r i n t h e t r a p , t h e opposite approach
was used.
Eight hours of delay w a s obtained by routing t h e gas through an
equalizing l i n e t o t h e empty d r a i n tank, through t h e tank and s a l t f i l l
lines t o t h e o t h e r tank where i t bubbled through several inches of s a l t heel, then out through t h e drain-tank vent l i n e t o t h e p a r t i c l e t r a p .
The
pressure drop across t h e t r a p w a s 16 p s i when t h e gas w a s f i r s t rerouted, but within a few hours it was below 2 p s i .
When t h e o r i g i n a l route w a s
again t r i e d , the r e s t r i c t i o n began t o b u i l d up almost immediately; therefore, t h e delayed route w a s used u n t i l t h e end of Run 10 which w a s terminated f o r in-cell r e p a i r s and a l s o because a newly constructed p a r t i c l e t r a p (Mark 11) based on the Hot C e l l examination of t h e f i r s t t r a p w a s
LJ
ready f o r i n s t a l l a t i o n . Examination of MK-I P a r t i c l e Trap.'
Flow tests on t h e f i r s t t r a p ( i n
service from April through July of 1966) indicated t h a t (1) t h e pressure drop a f t e r service w a s about 20 t i m e s the "clean" pressure drop.
Assuming
o r i f i c e type flow, t h i s would represent a reduction i n flow area of about 80%.
(2) The bulk of t h e pressure drop occurred in t h e Yorkmesh and Felt-
metal s e c t i o n s of t h e f i l t e r .
The Fiberfrax s e c t i o n w a s e s s e n t i a l l y clean.
The area of t h e Yorkmesh which had been immediately below t h e i n l e t pipe w a s covered with a blue-gray t o black m a t which had completely f i l l e d t h e space between t h e wires of t h e mesh (Fig. 8.10).
The shape of t h e m a t
corresponded t o the bottom of t h e i n l e t pipe, and i t is l i k e l y t h a t t h i s
material w a s t h e major r e s t r i c t i o n t o gas flow while i n service. Since t h e i n l e t pipe temperature probably increased s e v e r a l hundred degrees during power operation of t h e r e a c t o r , i t is believed t h a t t h i s r e s t r i c t i o n behaved similar t o a thermal valve.
This could account f o r t h e
unexpected increase i n pressure drop while a t power and t h e decrease i n pressure drop when t h e power was reduced o r when t h e off-gas w a s delayed
via t h e d r a i n tank.
W
,
193
U
PHOTO R32824
Fig. 8.10
Inlet to Yorkmesh Section MSRE Particle Trap Mark I
194 ~
A r a d i a t i o n survey made around the outside s u r f a c e of t h e lower s e c t i o n
of t h e t r a p gave readings ranging from 6900 t o 8000 R/hr.
W
The probe indi-
cated 11,400 R/hr when inserted i n t o the p o s i t i o n formerly occupied by t h e
i n l e t pipe.
The r a d i a t i o n l e v e l dropped off t o 4100 R/hr a t t h e bottom of
t h e trap. The bottom of the i n l e t pipe had a deposit of blackish material which corresponded t o t h a t i n t h e Yorkmesh.
The i n s i d e of t h e i n l e t pipe
a t t h e upper end of t h e bellows appeared clean and f r e e of deposit.
The
e x t e r i o r of t h e bellows had some of the light-yellow powdery material on a background of dark brown.
When t h e Yorkmesh was removed, i t was found t h a t t h e surface of t h e outer wires of the mesh bundle was covered with a t h i n l a y e r of ambercolored organic material.
Much of t h i s material evaporated from t h e h e a t
of t h e floodlamp used t o make the photographs.
As t h e bundle w a s unrolled,
t h e color of the f i l m on the wire changed from amber t o brown t o black near the center.
or 3.
The black material was thicker than the wire by a factor of 2
This material w a s b r i t t l e , as w a s the w i r e , and much of i t came
loose as t h e w i r e w a s flexed.
Samples of t h i s material, designated Nos.
LJ
5B and 6 (Fig. 8 . 8 ) w e r e taken f o r examination and chemical analysis. M e t allographic examination of a piece of t h e w i r e covered with t h e black mat e r i a l showed t h a t t h e wire w a s heavily carburized with a continuous network of carbide i n t h e grain boundaries.
There w a s no evidence of melting
of the w i r e ; however, t h e grain growth and o t h e r changes indicated opera t i n g temperatures of a t least 650OC.
The nonmetallic deposit observed on
t h e w i r e mesh w a s apparently of a carbonaceous nature and appeared t o have been deposited i n layers.
These "growth rings" were probably t h e r e s u l t
of off-gas temperature, r e a c t o r power, and gas flow-rate changes.
The perforated p l a t e of t h e coarse f i l t e r s e c t i o n and t h e lower flange of t h e f i l t e r assembly were covered with a s t r a t i f i e d s c a l e (view A-A, Fig. 8 . 8 ) .
The colors varied from a very l i g h t yellow t o orange.
stratum i n t h e lower flange area appeared gray, almost black.
One
A sample
(No. 9) was taken of t h e light-colored material, including some of t h e black material.
The perforated p l a t e of t h e f i n e f i l t e r s e c t i o n w a s covered
with a t h i n , dark-brown coating, which seemed t o be evenly d i s t r i b u t e d over t h e surface of the plate.
The inner surface of t h e outer w a l l of t h e t r a p
w a s covered with a light-amber coating, which w a s a l s o evenly d i s t r i b u t e d .
-
b,
195
b,
It is believed t h a t these coatings were deposited by condensation and sputt e r i n g of t h e o i l from t h e adjacent f i l t e r and t h a t the dark brown color i n d i c a t e s t h a t t h e porous metal screen had operated a t a much higher temperature than had the
t e r - w a l l (during operation t h e t r a p w a s immersed
a tank of water f o r of t h e upper flange ( v i appearance of organic r
g by n a t u r a l convection).
The lower surface
g. 8.8) contained a deposit which had t h e e deposit w a s amber colored, and t h e frac-
tured edge (Fig. 8.11) gav
impression that the material was b r i t t l e .
There i s as y e t no expl
or formation of t h i s deposit o r how i t came location, The r a d i a t i o eve1 on t h e out-
t o be formed i n t h i s pa s i d e of t h i s s e c t i o n of The material a t
r w a s 200 t o 360 R/hr. of t h e Fiberfrax s e c t i o n showed an o i l -
l i k e discoloration, b
evidence of any s i g n i f i c a n t accumula-
of material.
weights of t h e d i f f e r e n t l a y e r s with
C
t h e weights of t h e materia than 0.2 g.
g i n a l l y loaded indicated changes of less
An i n t e r e s t i n g observation r e l a t e s t o t h e very low radioac-
t i v i t y l e v e l of t h e Fiberfrax a t t h e entrance s e c t i o n which is separated
LJ
only'by t h e Feltmetal f i
an area containing material with a c t i v i t y
l e v e l s of thousands of
only d e t e c t a b l e a c t i v i t y (above t h e exami-
nation c e l l background It is p r
f r a x section.
r) was,at t h e discharge end of t h e Fibert t h i s a c t i v i t y r e s u l t e d from back-blowing
t h e t r a p or from pressure t r a n s i e n t s which could have c a r r i e d gaseous decay products from t h e vapor t r a p back upstream and i n t o t h e p a r t i c l e trap. Even so, t h e a c t i v i t y 1 nly 1.8 R/hr above background. Analytical Results. cations (Fig. 8.8) were
1 of four samples from t h r e e d i f f e r e n t lo= t o a v a r i e t y of a n a l y t i c a l tests. The
samples were i d e n t i f i e d Taken from
Sample No.
3A 9
,
5B 6
oarse s e c t i o n of porous ale on lower flange of t a t i n l e t t o Yorkmesh t a t i n l e t t o Yorkmesh
For s e c t i o n s of
metal filter porous metal s e c t i o n section section
s found t h a t about t h e same weight 10s
(0.2% of sample weight) r e s u l t e d from heating t o 6OOOC i n helium as from
dipping i n a trichlorethylene bath.
The material removed by heating w a s
196
F i g . 8.11
Upper End of Porous Metal S e c t i o n
\
- MSRE
P a r t i c l e Trap Mark I
r
.._ . -"
197
.-a*:<
i
'*'1
,
cold trapped and found t o "e e f f e c t i v e l y Lxontaminated; however, t h e trichlorethylene wash w a s contaminated with f i s s i o n products. Samples 5B and 9 were compared f o r low-temperature v o l a t i l e s ; a t 15OoC, No. 9 l o s t 5% and No. 5B l o s t none.
When r a i s e d t o 6OO0C, t h e weight l o s s e s
were 35% f o r sample 5B and 32% f o r sample 9.
Analysis of t h e carbon content
gave none f o r sample 9 and 9% f o r sample 5B.
This i n d i c a t e s t h a t sample 9
had not reached as high a temperature as had sample 5B. The mass spectrographic a n a l y s i s of sample 6 indicated t h a t t h e r e w a s a very high f r a c t i o n of f i s s i o n products.
w t % Ba, 15 w t % S r , and 0.2 w t % Y.
These are estimated t o be 20
I n t h e same a n a l y s i s t h e s a l t con-
s t i t u e n t s B e and Z r were estimated t o be 0.01 and 0.05 w t % respectively. I n addition t h e material i n samples 5B and 6 contained s m a l l quantities of C r , Fe, and N i , while sample 9 d i d not.
The r e l i a b i l i t y of t h e s e
values w a s compromised by d i f f i c u l t i e s caused by t h e presence of organics and small s e c t i o n s of w i r e i n t h e sample. The gamma-ray spectrographic work indicated t h e presence of t h e following isotopes:
"'cs,
Sr,
109
Ru o r "'Ru,
'lomAg,
PI
Nb, and '"La.
A l l t h r e e samples were chemically analyzed f o r B e , and t h e l e v e l w a s
below t h e d e t e c t a b l e l i m i t of 0.1%. Attempts t o analyze f o r Z r were comp l i c a t e d by t h e presence of l a r g e q u a n t i t i e s of Sr. The f r a c t i o n of s o l u b l e hydrocarbons w a s determined using CS1, and t h e values were 5B, 60%; 6 , 73%; and 9, 80%. The e x t r a c t s o l u t i o n s from samples 5B and 6 were allowed t o evaporate, and a few milligrams of t h e residue w a s mounted between s a l t c r y s t a l s for i n f r a r e d analysis.
T h e sam-
p l e s were i d e n t i c a l and were c h a r a c t e r i s t i c of long-chain hydrocarbons.
There w a s no evidence of any f u n c t i o n a l groups o t h e r than those involving carbon and hydrogen, bonds.
Nor was t h e r e any evidence suggesting double o r t r i p l e
There w a s an i n d i c a t i o n of a possible mild cross-linkage.
It is
l i k e l y that t h e r e is more cross-linkage of t h e organic i n t h e gas stream than appeared i n t h e s e samples, and t h e low i n d i c a t i o n could be due t o t h e i n s o l u b i l i t y of t h e cross-1
d 'organic and t h e high operating temperatures
of t h e w i r e mesh, which would cause breakdown of t h e organic i n t o elemental carbon and v o l a t i l e s .
MK-I1 P a r t i c l e Trap
8.4.8
As a r e s u l t of t h e operating experience with and d e t a i l e d examination of t h e first p a r t i c l e t r a p , t h e MK-I1 trap,'
(shown i n Fig. 8.12) w a s de-
signed w i t h t h e following features:
1. The t r a p housing was increased from 4 t o 6-in. I D , r e s u l t i n g i n an increase i n cross-sectional area of 225% i n both t h e Yorkmesh and Fiberf r a x sections. 2.
The u n i t was i n e f f e c t turned upside down so t h a t t h e Yorkmesh
s e c t i o n is a t t h e top of t h e unit and t h e Fiberfrax s e c t i o n is a t t h e bottom.
This change permits heating of t h e Porkmesh s e c t i o n (using beta decay
heat and lowering t h e water l e v e l ) while s t i l l maintaining cooling on t h e other two sections.
3.
The d i s p o s i t i o n of t h e Yorkmesh w a s modified t o provide increased
f r o n t a l area i n t h e d i r e c t i o n normal t o t h e flaw.
4.
Since t h e f i r s t t r a p had shown very l i t t l e loading i n t h e f i n e
Feltmetal s e c t i o n , only t h e coarse Feltmetal w a s used. 5.
The t o t a l f i l t e r area was increased from 22 in' f o r t h e o r i g i n a l
t o 288 i n 2 f o r t h e new trap.
.)
The depth of the F i b e r f r a was reduced by 50% 7. The pressure 'drop a t 15 !t/m was less than 1 i n . HaO and t h e t r a p efficiency w a s 99.9X f o r p a r t i c l e s g r e a t e r than 0 . 8 ~ . 6.
8.4.9
Organic Vapor Trap It was expected t h a t accumulation of organic material i n t h e charcoal
t r a p immediately downstream of t h e p a r t i c l e t r a p would result i n progressive poisoning along t h e length of the trap.
Such poisoning would s h i f t t h e lo-
cation of maximum f i s s i o n product adsorption and produce a s h i f t i n t h e temperature p r o f i l e of t h e trap.
Except f o r t h e upward s h i f t due t o in-
creased power l e v e l , t h e b a s i c shape of t h e temperature p r o f i l e d i d not change, i n d i c a t i n g t h a t no s i g n i f i c a n t poisoning by organics occurred. Since the pressure c o n t r o l valve (PCV-522) w a s operated i n the f u l l y opened position and i t appeared t h a t nothing would be l o s t by eliminating t h e charcoal t r a p , both were removed t o make room f o r two new MK-I1 p a r t i -
cle t r a p s which were i n s t a l l e d i n p a r a l l e l i n January 1967.
ORNL- m
/
/YORKMESH ,.-T.E.
[FELTMETAL
p.
\FlElERFfW
NKXEL
M Fig.
8.12
Line 522 Particle
4 --
0
I INCHES
Trap Mark II,
2
3 I
Particle
Trap Subassembly
6?- 4766
/-T.’
BAFFLES
I
200 8.4.10
Charcoal Bed Performance
After t h e removal of t h e check valve poppet i n A p r i l 1966, venting through t h e a u x i l i a r y charcoal bed was uneventful through t h e approach t o power operation. July.
However, an i n t e r m i t t e n t r e s t r i c t i o n w a s noted e a r l y i n
During t h e f i l l and drain a f t e r t h e J u l y 17 shutdown, t h e restric-
t i o n became continuous and more severe and was d o t amenable t o back-blowing. A remotely placeable assembly of electric h e a t e r s w a s designed and placed
around t h e junction of t h e i n l e t pipe and steel wool s e c t i o n of t h e a u x i l i ary charcoal bed.
I n t h e b r i e f shutdown between Runs 8 and 9 t h e l e v e l i n
t h e p i t was lowered below t h e bed inlets and t h e h e a t e r w a s energized. Some improvement w a s observed when t h e h e a t e r temperature reached 385OC.
After t h e temperature was raised t o 670째C and then cooled, t h e pressure drop was downby a f a c t o r of 5, t o a s a t i s f a c t o r y l e v e l .
The a u x i l i a r y bed
operated r e l a t i v e l y trouble-free u n t i l i t became p a r t i a l l y r e s t r i c t e d i n March 1969 and again about two months later.
Back-blowing cleared the bed
i n March; a forward-blw i n May w a s adequate but not as e f f e c t i v e .
No fur-
l
t h e r operational problems with t h e a u x i l i a r y bed were reported. Sections lA and 1 B of the main charcoal bed had been used almost exclusively during earlier runs, and they w e r e on-line when Run 8 s t a r t e d . The pressure drop a t t h e i n l e t of these s e c t i o n s began t o increase a few hours a f t e r t h e power was r a i s e d and when t h e pressure drop reached 7 p s i , s e c t i o n s 2A and 2B w e r e a l s o brought on l i n e .
Since back-blowing had be-
come progressively less e f f e c t i v e and i n l i g h t of t h e success with h e a t e r s on t h e i n l e t of t h e a u x i l i a r y bed, an attempt w a s made t o clear t h e res t r i c t i o n i n s e c t i o n 1 B by heating t h e i n l e t with a torch (electric h e a t e r s t o f i t t h i s bed were not then available).
Although t h e bed temperature
reached 465OC, t h e r e was no improvement and a l l four s e c t i o n s of t h e main bed were operated i n p a r a l l e l f o r Run 9.
After t h i s run, electric h e a t e r s
were i n s t a l l e d on t h e i n l e t s of sections lA and 1B.
Heating t o 400*C f o r
8 hours brought t h e pressure drop back t o t h e normal range f o r clean beds.
201
u
8.5
Subsequent Operating Experience with Fuel Off-gas System
With t h e major p a r t of t h e MSRE operating t i m e y e t t o come, t h e off-gas problems during t h i s period while s t i l l present were somewhat ameliorated by more e f f i c i e n t p a r t i c l e t r a p s and methods of c l e a r i n g t h e s e r e s t r i c t i o n s . Some new r e s t r i c t i o n s appeared but b a s i c a l l y they were similar t o those already described.
Remote maintenance methods were used t o cope with some
of these.
8.5.1
P a r t i c l e Trap The MK-I1 t r a p (Unit No. 1) w a s i n service from January 1967 t o May
of 1969 with no noticeable r e s t r i c t i o n ; however, during t h e f i r s t week i n
A week later
May, t h e pressure drop across t h e t r a p increased t o 0.4 p s i .
when t h i s r e s t r i c t i o n increased t o 0.75 p s i , t h e reserve u n i t , which had been valved o u t up t o t h i s t i m e , w a s put i n p a r a l l e l service. t h e pressure drop t o 0.1 p s i o r less.
This lowered
Also temperatures i n t h e t r a p re-
f l e c t i n g fission-gas heating indicated t h a t nearly a l l of t h e gas was flowi n g through t h e f r e s h bed.
tifad
I n October 1969, pressure drop measurements on
t h e off-gas system indicated a r e s t r i c t i o n of 0.3 p s i across t r a p No. 2; however, cycling t h e i n l e t valve, V-522C, between i t s closed and open pos i t i o n s cleared t h e r e s t r i c t i o n ; t h i s indicated a r e s t r i c t i o n i n t h e valve r a t h e r than t h e trap. ~
Three days later t h e r e s t r i c t i o n reappeared and w a s
cleared by t h e same method.
N
f u r t h e r problems i n t h i s area were en-
countered.
8.5.2
Introduction of Oxypen i n t o Fuel Drain Tank No. 1 I n t h e event of a s a l t
.
n t o t h e r e a c t o r c e l l , provisions had
been made t o vent t h e reacto
t o t h e a u x i l i a r y charcoal bed through
a l i n e equipped with two b l o
es.
I
I n January 1967, t h e s e valves,
V-571 A and B, were inadvertently opened while t h e r e a c t o r c e l l w a s being pressure-tested p r i o r t o Run 11. Reactor c e l l a i r entered t h e a u x i l i a r y charcoal bed which a t t h e t r e a c t o r via FD-1.
The slow
being used t o vent t h e empty b u t h o t age,of c e l l a i r i n t o t h e ACB r e s u l t e d i n
an increase i n f u e l system p r
u r e and was i n c o r r e c t l y diagnosed as plug-
ging i n the a u x i l i a r y charcoa
To clear t h e bed, i t was pressurized
t o 50 p s i g with helium and released i n t o FD-1 (V-571A was discovered t o be
6,
about 1 / 2 t u r n open during t h e back-blowing operation).
202
i
It w a s estimated t h a t a maximum of 1.25 g-moles of 02 could have been
u
back-blown i n t o FD-1 which contained approximately 2 cu f t of molten f u e l It i s believed t h a t t h e oxygen w a s later purged out of t h e d r a i n
salt."
tank without appreciable mixing with t h e f u e l system piping. Subsequently, l i n e 571 which joined t h e two systems w a s c u t and both ends were cap-welded. 8.5.3
Charcoal Beds The r e s t r i c t i o n s a t t h e i n l e t sections of the,main charcoal bed con-
tinued t o exhibit chronic plugging throughout t h e remainder of MSRE power operations. , 1
Sections lA and 1 B began t o plug slowly a f t e r t h e two weeks of
power operation i n Run 11. After 4 weeks of operation t h e pressure drop increased from 2.5 t o 5 p s i and reached 7 p s i two months a f t e r t h e s t a r t of t h e run.
When sections 2A and 2B w e r e put i n s e r v i c e , t h e pressure drop in-
creased from 2.6 t o 9 p s i i n only t e n days.
Sections 2A and 2B were cleared
by forward-blowing with helium and sections lA and lB were cleared with t h e previously i n s t a l l e d heaters.
These operations d i d not r e q u i r e a reduction
i n r e a c t o r power but only a temporary lowering of the w a t e r . l e v e l i n t h e charcoal bed p i t t o allow the heaters t o function.
Approximately one month
d.'
later t h i s operation had t o be repeated. Owing t o t h e success of t h e h e a t e r s on t h e a u x i l i a r y bed and s e c t i o n s
IA and 1 B of t h e main bed, similar h e a t e r s w e r e i n s t a l l e d a t t h e i n l e t s of sections 2A and 2B during t h e shutdown between Runs 11 and 12. Approximately one month a f t e r the start of Run 12, i t w a s again neces-
sary t o clear a l l four sections. During MSRE s t a r t u p , i n l e c t i o n s of *'Kr
indicated t h a t t h e charcoal
bed hold-up time was 5-1/2 days f o r krypton with two s e c t i o n s i n p a r a l l e l service; t h e equivalent xenon hold-up time w a s calculated t o be 88 days. Thermocouples are located a t t h e entrance regions of t h e l-l/F,3-, and 6-inch diameter sections of t h e charcoa1,beds.
These s e c t i o n s contain 4.75,
27.4 and 67.85 percent, respectively, of t h e t o t a l charcoal i n t h e beds. On t h e b a s i s of a 5.5 day hold-up
t i m e f o r krypton, t h e l a g i n temperature
response t o f i s s i o n gas adsorption between t h e l-l/Fand 3-inch diameter sections should be 6.3 hours (4.755% of 5.5 days) when two beds are i n serv i c e and 3.1 hours when one bed is i n service.
During full-power operation
W
203
u
on September 15, 1967, t h e temperature response i n t e r v a l f o r beds lA and 1 B w a s 11-1/2 hours.
During a similar s t a r t u p on September 21, 1967, t h e
response i n t e r v a l w a s 9-1/2 hours f o r bed lA and 11-1/2 hours f o r bed 1 B i n d i c a t i n g t h a t bed 1 B had developed more of a r e s t r i c t i o n than bed lA. During t h e s t a r t u p of November 25, 1967 with beds 2A and 2B i n service, t h e response i n t e r v a l w a s 8-1/2 hours f o r bed 2A and no response i n bed 2B inBeds 2A and 2B were then
d i c a t i n g an appreciable r e s t r i c t i o n i n bed 2B.
valved off and bed lA was put i n service; t h e temperature response i n t e r v a l
was 6-1/2 hours.
Bed lA remained i n service f o r t h e remainder of Run 14
(130 days) with t h e exception of a 10-day period i n t h e middle of February when two s e c t i o n s w e r e put
s e r v i c e t o achieve t h e low fuel-pump pressure
required f o r a r e a c t i v i t y test.
The maximum pressure drop across t h e s i n g l e
s e c t i o n during Run 14 was 4
psi.
a c t i v i t y w a s detected by t h
a d i a t i o n monitors downstream of t h e charcoal
No d i s c e r n i b l e increase i n e f f l u e n t
t r a p during t h e s i n g l e bed test. No plugging i n t h e bed was experienced during t h e p r e c r i t i c a l and low power (<25 kW) operation using assU f u e l (Runs 15 and 16).
W
However, during
t h e next runs, gradual plugging recurred a f t e r t h e f i r s t 12 days of power operation (5 MW).
The electric h e a t e r s , previously i n s t a l l e d on t h e i n l e t
s e c t i o n , were used t o cle
t h e beds, although two of t h e t h r e e h e a t e r ele-
ments i n s t a l l e d on s e c t i o
lA and 1 B and one element on 2B f a i l e d .
This
procedure along with a back-blow a t t h e end of t h e 8-hour heating cycle w a s required t o clear t h e bed
ximately t h r e e months later.
during t h e valving operat
volved, t h e stem o r stem extension on t h e
Unfortunately,
i n l e t valve t o s e c t i o n 2A broke with t h e valve i n t h e closed position. p a i r s were not attempt and t h e s u f f i c i e n c y of During portions o gas t o study i t s e f f e c unusual off-gas problems t e r p r e t a t i o n of f
u
Re-
f t h e high r a d i a t i o n level a t t h e valve
ree sections.
(19) argon w a s s u b s t i t u t e d as a cover f r a c t i o n and r e a c t i v i t y changes.
No
with argon were manifest o t h e r than
re drops through t h e
tem during t h e
several switchovers
e other.
crease i n t h e e f f l u e n t
d i c a t o r ( a t t r i b u t e d t o neutron acti-
vation of argon i n t h e r e a c t o r ) was noticed.
A b a r e l y p e r c e p t i b l e in-
During t h i s run, i t w a s neces-
s a r y t o clear t h e charcoal beds twice, using t h e i n l e t h e a t e r and a
204 forward-blow with helium a t 25 psig.
No f u r t h e r problems were encountered
W
with t h e bed during the last run. 8.5.4
Restriction a t t h e Pump B o w l The r e s t r i c t i o n i n the off-gas l i n e a t t h e pump bowl reappeared i n
November 1967, one year s i n c e t h e l i n e w a s last cleared with a f l e x i b l e
reamer.
The r e s t r i c t i o n became noticeable approximately 2-1/2 days a f t e r
t h e reactor power was reduced t o 10 kW t o r e p a i r t h e sampler-enricher. During t h i s period t h e operational and maintenance valves w e r e open, and
a 0.6-liter/min
helium purge was maintained down t h e sampler tube t o t h e
pump bowl t o prevent contamination of t h e sampler by f i s s i o n gases.
The
r e s t r i c t i o n w a s evidenced by an increase (0.5 psig) i n t h e fuel-pump bowl pressure when t h e overflow tank vent valve w a s closed during a routine return of s a l t from t h e overflow tank t o t h e pump bowl.
After t h e sampler
w a s repaired, t h e r e s t r i c t i o n w a s relieved by pressurizing t h e pump bowl t o 6.0 psig and suddenly venting t h e gas i n t o a drain tank which w a s a t 3 psig.
The pressure drop then appeared notmal ( ~ 0 . 1p s i ) , b u t a f t e r t h r e e more days of low-power operation, an abnormal pressure rise (0.2 p s i ) w a s again observed when s a l t was being returned from t h e overflow tank.
Repetition
dd
of t h e mild blow-through t o t h e drain tank had l i t t l e o r no e f f e c t t h i s The l i n e w a s not completely blocked, however, and full-power opera-
I
time.
!
t i o n w a s resumed. A t f i r s t , temperatures on t h e overflow tank and t h e offgas l i n e (responding t o f i s s i o n product heating) c l e a r l y indicated t h a t
,
there was enough pressure drop a t t h e pump-bowl o u t l e t t o cause much of t h e Then,
I
off-gas t o bubble through t h e overflow tank and out i t s vent l i n e .
,
a f t e r 1 5 h r a t f u l l power, t h e bypass flow stopped, indicating t h a t t h e res t r i c t i o n had decreased s i g n i f i c a n t l y ; i t remained below t h e l i m i t s of det e c t i o n throughout t h e next nine weeks while t h e reactor w a s operating a t 7.2 MW o r 5 MW.
After two days a t 10 kW, t h e pressure drop again became
detectable and continued t o increase over t h e next s i x days a t very low power.
When power operation was resumed a t 5 MW, temperatures indicated
t h a t there was again bypass flow through t h e overflow tank.
The restric-
t i o n increased during two weeks of operation a t 5 MW, but t h e l i n e never became completely plugged.
Four days a f t e r t h e resumption of full-power
operation i n February 1968 while t h e overflow tank was being emptied, t h e
td
205
td
r e s t r i c t i o n p a r t i a l l y blew out, bringing t h e pressure drop again below t h e
l i m i t of detection.
The pressure drop remained j u s t a t o r below t h e l i m i t
u n t i l the March-1968 shutdown f o r f u e l processing. The behavior w a s q u i t e unlike the plugging t h a t occurred earlier i n t h e same s e c t i o n of l i n e as a r e s u l t of t h e o v e r f i l l with f l u s h s a l t , mainl y i n t h a t it never completely pluggeqand t h e r e w a s a suggestion of some e f f e c t of power level.
A f t e r t h e f u e l was drained, the f l e x i b l e s e c t i o n of off-gas l i n e w a s removed, and a f e l x i b l e t o o l was run through t h e 28-in. pump bowl.
of l i n e back t o t h e
The t o o l s , a 1/4-in. cable with a diamond d r i l l t i p , encountered
some r e s i s t a n c e a t f i r s t and came out with a considerable amount of s o l i d s on it.
Although t h e f l e x i b l e jumper and t h e l i n e downstream were not re-
s t r i c t e d , t h e jumper l i n e along with the f l e x i b l e reaming t o o l s were s e n t t o t h e High-Radiation-Level-Examination
Laboratory f o r cut-up and exami-
nation. The replacement jumper l i n e was equipped with four thermocouples t o provide more information during power operation.
Lp,
A slender basket contain-
ing devices of metal and graphite t o c o l l e c t material from t h e off-gas
stream was suspended i n t h e entrance of t h e 4-in. off-gas hold-up pipe. The r e s t r i c t i o n a t t h e pump bowl reappeared a f t e r approximately one month of operation with t h e n
f u e l i n September 1968.
The r e s t r i c t i o n
was blown,clear once during p
sure-drop measurements, but reappeared a
week later and could not be dislodged by back-blowing helium from t h e d r a i n drop across t h e r e s t r i c t i o n had intank t o t h e pump b creased t o 4 p s i b a n t systems were drained t o
also t o pix t h e f u e l befor The flange a t b l e cable assembly
ream out t h e off-gas li t h e r e s t r i c t i n g materia recovered on t h e f
was reconnected a
i n November 1968, when t h e f u e l and coolt t h e off-gas l i n e s of both systems and
of full-power operation.
was again disconnected, and a f l e x i l t e r and a vacuum pump, w a s used t o 1 pump and a l s o t o c o l l e c t a sample of ry small amount of blackish d u s t was t h e f l e x i b l e tubing.
The flange
f i l l e d ; however, after about t h r e e
weeks of operation, t h e r e s t r i c t i o n reappeared i n mid-January and gradually
b=
increased t o a AP of 5 . 5 p s i .
Because t h e rapid recurrence of t h e r e s t r i c t i o n
206
W
a t the pump-bowl e x i t threatened t o become a serious nuisance, a compact
2-kW heater u n i t was designed f o r i n s t a l l a t i o n on t h i s l i n e a t t h e f i r s t opportunity; however, i n t h e last week i n February during a routine trans-
fer of f u e l from t h e overflow tank t o t h e f u e l pump, t h e r e s t r i c t i o n was blown clear, r e s u l t i n g i n a complete t r a n s f e r of f u e l from t h e overflow tank t o t h e pump bowl.
The pressure drop i n t h i s s e c t i o n of l i n e w a s then
below t h e l i m i t of detection (about 0.1 p s i ) . On f i v e other occasions i n t h e next 9 weeks t h e pressure drop became
noticeable, but each time i t blew out when t h e overflow tank w a s burped. On May 1 it reappeared and remained detectable through t h e June-1969 shut-
down.
The r e s t r i c t i o n caused most of t h e pump-bowl purge gas flow t o bub-
b l e through t h e overflow tank except when t h e overflow tank w a s being pressurized t o push accumulated s a l t back t o t h e pump bowl.
As a consequence,
on May 25, t h e off-gas l i n e from t h e overflow tank became almost completely plugged and i t became necessary t o turn off t h e overflow tank bubbler flow most of t h e time.
Thereafter a l l t h e pump purge gas w a s forced out through
t h e off-gas l i n e from t h e pump bowl.
This s i t u a t i o n caused some inconveni-
4J
ence during t h e burping operation, but by reducing purge flow t o t h e pump bowl a t these times, s a l t could be transferred back t o t h e f u e l pump without exceeding 1 5 psig i n t h e pump bowl. During t h e June shutdown, t h e 2-kW heater t h a t had been b u i l t earlier
was i n s t a l l e d remotely on t h e pipe as near t o t h e pump as possible and connected t o s p a r e power and thermocouple leads i n t h e r e a c t o r cell.
The pump-
tank furnace heaters and t h e new heater were turned up t o bring t h e s e c t i o n of off-gas l i n e t o near 65OoC, then gas pressure w a s applied t o blow t h e r e s t r i c t i n g material back toward t h e pump bowl.
The heater was subsequent-
l y turned off but l e f t connected so t h a t i t could be used again without reopening t h e reactor cell. The set of specimens exposed i n t h e off-gas hold-up volume s i n c e J u l y 1968 w a s a l s o removed during t h e June shutdown. Although the r e s t r i c t i o n did not reappear during Run 19, t h e pressure modifier and recorder which was i n s t a l l e d on t h e fuel-pump pressure trans-
m i t t e r t o d e t e c t high frequency pressure noise i n t h e r e a c t o r core indicated t h a t a r e s t r i c t i o n a t t h e pump exit was again i n i t s formative stage.
14
Lid
207
u
The r e s t r i c t i o n did not become evident (by previous standards) u n t i l approximately two weeks after t h e f u e l f i l l of Run 20.
I n t h e t h r e e days
before f i n a l r e a c t o r shutdown, t h e r e s t r i c t i o n increased from 1.3 t o 2.2 p s i as measured with t h e procedure f o r s a l t recovery from t h e overflow tank. Because of t h e proximity of t h e r e a c t o r shutdown d a t e and t h e d e s i r a b i l i t y of not i n t e r r u p t i n g t h e experiment i n progress, t h e h e a t e r which had previously been i n s t a l l e d and successfully used t o clear a similar r e s t r i c t i o n ,
was not used. Examination of t h e 522 Jumper Line Removed March 1968.
I'
An 8-mg Sam-
p l e of what appeared t o be soot was c o l l e c t e d from t h e upstream flange of t h e f l e x i b l e jumper l i n e ; a
-mg sample of a similar material w a s collected
from t h e downstream flange.
he 8-mg sample indicated about 80 R/hr and t h e
15-mg sample indicated about 200 R/hr a t 'tcontact."
Each of t h e flanges and
t h e convolutions of t h e f l e x i b l e hose appeared t o have remaining on them a t h i n dull-black film.
A one-inch s e c t i o n c u t from t h e upstream end of t h e
jumper l i n e indicated 150 R/hr; a similar s e c t i o n from t h e downstream end indicated 350 R/hr.
6,
The reaming t o o l s used w e r e covered with blackish,
c l e a r i n g t h e r e s t r i c t i o n a t t h e pump bowl t y , granular material which was i d e n t i f i e d
by x-ray as f u e l s a l t p a r t i c l e s .
The r e l a t i v e l y high values of l i t h i u m
and beryllium found i n t h i s sample may i n d i c a t e pickup of some r e s i d u a l LiFa-BeFa f l u s h s a l t which had entered t h e off-gas l i n e as a r e s u l t of t h e pump o v e r f i l l i n 1966. Electron photomicrogr
of t h e upstream d u s t showed r e l a t i v e l y s o l i d
p a r t i c l e s of 1~surrounded
material of l i g h t e r and d i f f e r e n t s t r u c t u r e
which appeared t o be.amor p l e s showed 12 t o 16%
carbon.
Chemical analyses of t h e dust sam-
rbon, 28 t o 54% f u e l s a l t , and 4% s t r u c t u r a l metals.
Based on a c t i v i t y data, f i s
products could have amounted t o 2 t o 3%of
t h e sample weight.
t a b i l i t y of t h e remainder is a t t r i b u t a b l e
The una
t o t h e small amount of samp From t h e dust c o l l e c t e d from l i n e 522 onto a f i l t e r during t h e December1968 shutdown, t h e followin
,
a 3 3 ~ 1 4 0Ba,
'44Ce,
(2) t h e isotopes
89
and ''2 Sr,
91
onclusions"
were reached:
(1) t h e isotopes
ere transported only as s a l t constituents;
Y, and "'CS,
which have noble gas precursors, were
208
present i n s i g n i f i c a n t l y g r e a t e r proportions consistent with a mode of transport other than s a l t p a r t i c l e s ; (3) t h e isotopes '"Nb, "Mo, .10 6 Ru, and "'ye
w e r e present i n even g r e a t e r proportions, i n d i c a t i n g t h a t they
were transported more vigorously than f u e l s a l t .
A more complete discussion of these findings and t h e r e s u l t s of examination of t h e sample specimens removed from t h e 4-in. hold-up volume w i l l be covered i n a report t o be issued by Compere and Bohlmann e n t i t l e d "Fiss i o n Product Behavior i n t h e MSRE During
'Uranium Operation" (ORNL-TM-
2753). 8.5.5
Other Restrictions i n Line 522
On f i v e occasions during Runs 1 7 and 18, complete flow blockage occurred i n t h e main off-gas l i n e somewhere between t h e 4-in.-diam hold-up pipe and t h e junction of Line 522 and t h e auxiliary charcoal b e d ' l i n e , 533. Although t h e plug could have occurred a t t h e i n l e t p o r t of V-522A, i t is unlikely t h a t t h e plug formed i n t h e valve i t s e l f because manipulating' t h e valve stem t o i t s f u l l travel i n both d i r e c t i o n s d i d not break t h e plug loose.
I n each case t h e f u e l system pressure w a s vented f o r s h o r t periods
of time through t h e equalizer and d r a i n tank vent l i n e s t o t h e a u x i l i a r y charcoal bed u n t i l l i n e 522 was unplugged by back-flowing helium from l i n e
571, through l i n e s 561, 533, and 522 t o t h e hold-up volume. each plug became successively more d i f f i c u l t t o dislodge.
It seems t h a t To dislodge 'the
l a s t plug, it was necessary t o vent t h e f u e l system t o 2 p s i g , pressurize l i n e 533 t o 60 psig, and quickly open V-52%.
The l a s t e f f o r t apparently
restored t h e l i n e t o near its o r i g i n a l condition because plugging a t t h i s l o c a t i o n had not recurred a t the end of r e a c t o r operations.
Because of t h e
remoteness and i n a c c e s s i b l i t y of t h i s l i n e , no f u r t h e r e f f o r t was made t o l o c a t e t h e exact p o s i t i o n o r characterize t h e nature of t h e plug. 8.5.6
R e s t r i c t i o n i n t h e Overflow Tank Exit Line The f u e l pump was provided with an overflow pipe (shown i n Fig. 8.9)
and a catch tank (5.4 ft')
t o prevent s a l t from becoming,high enough t o al-
low it t o . e n t e r t h e gas and l u b r i c a t i n g passages of t h e pump as a r e s u l t of unusually high gas entrainment i n t h e c i r c u l a t i n g s a l t o r possibly as a r e s u l t of a high-temperature excursion while operating.
The overflow pipe
from t h e pump dips t o within less than 1/2-in. of t h e bottom and i n t o a
dj
209
bi
1/2-in.-deep
dimple pressed i n t h e lower tank head t o permit more complete
r e t r i e v a l of i t s contents.
During operations, s a l t spray and/or bubbles
slowly s p i l l e d i n t o t h e overflow tank (OFT);
t h e s a l t w a s p e r i o d i c a l l y re-
turned t o t h e pump bowl by closing t h e OFT gas discharge valve and allowing helium from t h e bubble-type level i n d i c a t o r s t o build up t h e required press u r e (%3.5 p s i > pump bowl pressure) t o f o r c e t h e s a l t back t o t h e pump bowl. When t h e OFT discharge is opened, t h e s a l t remaining i n t h e tank and t h a t remaining i n t h e pipe is s u f f i c i e n t t o provide a seal a t t h e bottom of t h e tank and thus prevent f i s s i o n product gases from entering t h e overflow tank. However, when t h e main off-gas l i n e , 522, became r e s t r i c t e d near t h e pump bowl, p r a c t i c a l l y a l l of t h e pump purge gas and gaseous f i s s i o n products
were bubbled through t h e s a l t h e e l i n t h e OFT and o u t through l i n e 523 t o j o i n t h e main off-gas l i n e downstream of TE-522-2.
From TE-522-2
d a t a and
pressure measurements during s a l t recovery operations, i t w a s estimated t h a t f i s s i o n product gases were f i r s t diverted through t h e OFT on October 15, 1966.
Since t h e fuel-pump ove
ill on J u l y 24, 1966, i t is estimated t h a t
most of t h e fuel-pump gases were forced through t h e overflow tank f o r ap-
LJ
proximately 28 days i n 1966, 6 days i n 1967, 76 days i n 1968, and 68 days i n 1969, before t h e overflow tank gas e x i t l i n e (523) became plugged on May 25.
Thereafter, a l l of t h e pump purge gas w a s forced o u t through t h e
r e s t r i c t i o n i n t h e off-gas l i n e a t t h e e x i t of t h e pump bowl.
This s i t u -
a t i o n caused some inconvenience during t h e s a l t recovery operation, but by reducing purge flow t o t h e pump a t these t i m e s , t h e operation could be done without exceeding 15 psig
During t h e shutdown i scanned with t h e remote g observed a t t h e air-operate the overflow tank.
he pump bowl.
e 1969, when t h e overflow tank vent l i n e was rometer, an unusually s t r o n g source w a s (HCV-523),
about 33 f t downstream from
Flanges were opened and pressure observations showed
that the restriction w
langed s e c t i o n containing both t h e air-
operated valve and a h
This s e c t i o n was removed t o a h o t c e l l
where polymerized hydrocarb i n g t h e hand valve i n l e t
PO
operated valve but no hand v
nd f i s s i o n products were found t o be blockreplacement s e c t i o n containing an aire was i n s t a l l e d .
Although t h e r e s t r i c t i o n
a t t h e exit of t h e pump bowl reappeared during t h e l a s t week of MSRE operat i o n , t h e OFT discharge l i n e functioned normally.
210 8.5.7
W
R e s t r i c t i o n s a t Exits of Drain Tanks During t h e r e a c t o r d r a i n i n June of 1969, pressure measurements indi-
cated t h a t a r e s t r i c t i o n had developed i n t h e gas l i n e somewhere between Fuel Drain Tank No. 2 (-2) t h e e x i t l i n e (575).
and t h e junction of t h e i n l e t l i n e (574) and
The f i r s t half of t h e d r a i n appeared normal a t a d r a i n
rate of %2 cu f t of s a l t per minute a f t e r which i t tapered o f f t o 1~0.25cu f t per minute when t h e trapped gas i n t h e d r a i n tank balanced t h e s a l t i n
t h e f u e l system.
Two and a h a l f hours were required f o r t h e d r a i n as op-
posed t o a normal d r a i n time of less than 40 minutes when drained i n t o one tank.
The r e s t r i c t i o n w a s l a t e r cleared by heating t h e tank t o 68OOC and
applying a pressure d i f f e r e n t i a l of 60 p s i g across t h e plug via l i n e 561.
A similar but lesser r e s t r i c t i o n i n t h e gas l i n e a t t h e o t h e r f u e l d r a i n tank w a s p a r t i a l l y cleared during t h e f u e l system pressure test i n August 1969 by heating t h e tank t o 68OOC and flowing helium from FD-2 ( a t 50 psig) t o FD-1 (at 2 psig). R e s t r i c t i o n s i n these l i n e s were most probably deposited (but not t o t h e point of detection) during December of 1966 and January of 1967 when t h e r e a c t o r off-gas w a s routed t o -1,
through FV-106 and FV-105,
t h e s a l t h e e l i n FD-2, and out through l i n e 575 (Sect. 5.6.1).
through
LJ
Fission gas
decay heating, during t h i s t i m e , was appreciable i n t h e d r a i n tanks and required downward adjustments t o t h e d r a i n tank heaters.
The off-gas flow
through t h e d r a i n tanks was reversed after t h e first week i n an e f f o r t t o minimize plugging.
;
The off-gas flow w a s again routed through d r a i n tank No. 2 gas lines f o r a day when l i n e 522 plugged i n May of 1969. I n October of 1969 during a pressure release test when t h e fuel-pump gas was released i n t o FD-1,
it w a s discovered t h a t t h e gas l i n e was again
r e s t r i c t e d near t h e d r a i n tank.
I
Two attempts w e r e made t o clear t h i s l i n e
by back-blowing helium, f i r s t a t 30 psig and then a t 50 p s i g , from l i n e 561 through HCV-573 and i n t o FD-1.
Although only marginal improvement w a s ob-
tained, t h e gas flow through t h i s l i n e was judged adequate f o r any emergency d r a i n s i t u a t i o n . During t h e subsequent d r a i n i n November 1969, approximately 4300 l b of
s a l t drained i n t o FD-1 and 5100 l b s drained i n t o FD-2 i n d i c a t i n g t h a t a s l i g h t r e s t r i c t i o n existed i n t h e exit gas l i n e of FD-1.
During t h e
u
b,
December drain, the exit gas valve from FD-2 was closed during part of the drain to compensate for the partial restriction in FD-1 gas line; however, overcompensation resulted and 5400 lbs drained into FD-1 and 4100 lbs drained into FD-2. The drain.times for both drains were less than 20 minutes which is normal for a drain when both tanks are used. 8.5.8
Restrictions at the Off-Ras Sampler A system to permit the analysis of the reactor off-gas stream was in-
stalled downstream of the particle trap as shown in Fig. 8.2. The sampler contained two thermal-conductivity cells, a copper oxide converter, and two molecular sieves, one operating at .liquid nitrogen temperature and the other at room temperature.8 Since the sampler was an integral part of primary containment during sampling operations, and since some components of the sampler did not meet the requirements of primary containment, solenoid block valves were installed in the inlet and outlet lines which connect the sampler to the reactor system. Two fail-closed valves in series were installed in each line and were instrumented to close on high sampler activity, high reactor cell pressure, and high fuel-pump pressure. Although located downstream of the particle trap, the inlet line to the off-gas sampler periodically developed a restriction in the vicinity of the safety block valves. The block valves have 3/32-in.-diam ports and the inlet piping is 0.083 in ID autoclave.tubing. The restriction was successfully cleared each time by back-blowing with helium. As encountered in other parts of the off-gas system, successively higher pressures were required to clear the restriction each time. During the latter part of uld not shut off tightly and were replaced Run 18, the,inlet block va ction of the faulty inlet valves did not rein July of 1969. Visual i g valves nor the nature of the restriction veal the.reason for the.le (the restriction had been blown clear before the valves were replaced). During the August startup, it was again necessary to blow out the restriction at the inlet to the sampler.
212 8.6
W
Discussion and Conclusions
The periodic plugging i n t h e coolant system off-gas f i l t e r and a l s o
i n t h e f u e l system f i l t e r during t h e p r e c r i t i c a l and low-power (<25 kW) operation can b e a t t r i b u t e d primarily t o t h e accumulation of o i l i n t h e 0.7 t o
lu
diameter pores of t h e s i n t e r e d metal f i l t e r s (2 t o 411 pores might
have been a b e t t e r choice).
However, subsequent f i l t e r s i n t h e f u e l system
were constructed of f e l t metal (Huyck Nos. FM-225 and FM-204) which w a s not only e f f i c i e n t a t stopping s o l i d f i s s i o n products but a l s o w a s apparently immune t o plugging by hydrocarbons.
V i r t u a l l y a l l of t h e s o l i d decay daugh-
ters were stopped by t h e Yorkmesh and f i l t e r before reaching t h e Fiberfrax s e c t i o n of t h e p a r t i c l e trap. O i l vapor i n t h e off-gas stream condensed on various p a r t s of t h e off-
gas system components and apparently enhanced t h e adsorption o r trapping of p a r t i c u l a t e matter, p a r t i c u l a r l y on the Yorkmesh portion of the p a r t i c l e trap The f a c t t h a t such l a r g e amounts of s o l i d decay daughters were trapped i n t h e p a r t i c l e t r a p i n d i c a t e s t h a t most of these s o l i d s tend t o remain i n t h e gas phase u n t i l trapped onto a surface wetted by hydrocarbons.
It i s
not clear whether t h e s o l i d decay daughters agglomerate i n t h e hold-up vol-
ume and i f so whether they agglomerate with o t h e r s of t h e i r own species, with other species, with hydrocarbons, or possibly combine chemically t o form both v o l a t i l e and non-volatile compounds.
A l a r g e f r a c t i o n of t h e
noble gas f i s s i o n products decay i n t h e 4-inch-diameter hold-up volume and t h e i r decay daughters apparently p l a t e out o r are adsorbed onto t h e f i r s t flow channel r e s t r i c t i o n where t h e gas flow changes d i r e c t i o n and/or is c l o s e t o a r e l a t i v e l y cold surface such as valves and flow r e s t r i c t o r s .
The high hydrocarbon content (>60X) of t h e material c o l l e c t e d on t h e Yorkmesh i s inconsistent with t h e high temperature (%65OoC) i n t h i s region during power operation; hydrocarbons are vaporized and cracked a t this tem'perature.
One must assume, then, t h a t t h e hydrocarbons were collected
sometime a f t e r power operation had ceased and a f t e r t h e f i s s i o n product decay heat became negligible.
The growth r i n g s on t h e Yorkmesh were probably
caused by a l t e r n a t e periods of power and zero-power operation.
''
213
bi
Since spectrographic an l y s i s d i d not show any of t h e a l k li m e t a l
1
cesium o r rubidium, on t h e Yorkmesh, they were apparently boiled o f f a l s o
by f i s s i o n product decay heating i n t h i s region.
I f t h i s be t h e case, ap-
preciable q u a n t i t i e s of these materials would be expected t o be c o l l e c t e d i n t h e charcoal beds.
On t h e o t h e r hand, l a r g e q u a n t i t i e s of barium and
strontium were reported on t h e Yorkmesh; t h i s would argue f o r e f f e c t i v e trapping of t h e i r precursors, cesium and ribidium.
However, t o be trapped
on Yorkmesh a t 65OoC, t h e a l k a l i metals would n e c e s s a r i l y have t o be chemic a l l y combined (as halides f o r instance) r a t h e r than i n t h e elemental form, otherwise they would b o i l off and be c a r r i e d downstream. Although none of t h e s o l i d decay daughters i n t h e 4-in.-diam
pipe got
p a s t t h e p a r t i c l e t r a p , another batch (though somewhat smaller number) of s o l i d decay daughters are born (and presumably remain i n t h e gas phase) i n t h e l a r g e r hold-up volume ahead of t h e main charcoal beds.
These f i s s i o n
products along with hydrocarbons are probably t h e source of t h e plugging experienced a t t h e entrance region of t h e beds. After examination of t h e MK-I p a r t i c l e t r a p , i t is understandable why t h e entrance region of t h e charcoal beds plugged periodically; t h e entrance pipe ( 1 / 4 4 n . sched-40) e n t e r s t h e bed normal t o t h e s e c t i o n where t h e s t a i n l e s s steel wool is packed, thus t h e cross-sectional flow area (trapping area) a t t h e entrance t o t h e s t a i n l e s s s t e e l wool i s only 0.1 sq. i n . Although a p a r t i c l e t r a p w a s not i n s t a l l e d between t h e second hold-up volumdand t h e charcoal bed, i t w a s considered and w a s d e f i n i t e l y needed.
Also the.entrance s e c t i o n of f u t u r e beds should be redesigned such t h a t gas e n t e r s t h e empty chamber above t h e s t e e l wool and t h u s o f f e r a l a r g e r cross s e c t i o n of s t a i n l e s s s t e e l wool t o gas flow as i n t h e MK-11 t r a p . Another argument t h a t f i s s i o n products, as w e l l as hydrocarbons, are involved i n t h e plugging mechanism a t . t h e charcoal bed entrance, is t h e f a c t t h a t t h e beds remained f r e e of plugging u n t i l power operations were begun. For a b e t t e r understanding of t h e plugging mechanism, t h e MK-I1 p a r t i c l e t r a p and t h e entrance s e c t i o n t o one of t h e charcoal beds would need t o be examined.
Also a t least one bed could be examined along i t s length t o de-
termine t h e f i s s i o n product adsorption c h a r a c t e r i s t i c s of t h e bed. The probable reason t h a t t h e overflow tank exit l i n e remained clear
k,
f o r such a long time was the f a c t t h a t pressure d i f f e r e n t i a l s (Q3.5 p s i )
-.
were periodically released through t h e l i n e each t i m e s a l t was returned t o t h e pump bowl.
The pressure pulses probably cleared any i n c i p i e n t plug
which may have been i n i t s formative stage. The r e s t r i c t i o n a t t h e f u e l pump bowl does not seem t o be r e l a t e d t o f i s s i o n products s i n c e t h e plug seems t o form as r e a d i l y i f not more so while s u b c r i t i c a l and a t low power as i t does a t f u l l power.
The gas e x i t
l i n e from t h e pump bowl should be modified t o eliminate m i s t o r l i q u i d carryover i n t o t h e off-gas stream.
A heated cyclone-type separator"
or a
l a r g e w e l l baffled and cooled region which could b e l a t e r heated t o m e l t down any s a l t formation are two p o s s i b i l i t i e s . I f hydrocarbons had not been present i n t h e pump bowl, f i s s i o n gas behavior i n t h e MSRE might have been somewhat d i f f e r e n t from t h a t experienced. Some speculations on probable r e s u l t s are (1) less s e r i o u s o v e r a l l plugging because t h e r e would be no semisolid varnish-like buildup due t o hydrocarbons, (2) t h e York mesh would probably be less e f f i c i e n t a t trapping t h e s o l i d decay daughters and more of t h i s material would then be trapped on t h e f e l t metal f i l t e r , (3) t h e f i s s i o n product "cake" would probably be loosely packed on t h e f i l t e r and more e a s i l y disrupted by backflow s i n c e i t would not contain a 'tbinder" o r p a s t e material (hydrocarbons).
LJ
215
FUEL AND COOLANT PUMP LUBE OIL SYSTEMS
9.
J. K. Franzreb
9.1
Description
Two i d e n t i c a l o i l systems served t o ~ l u b r i c a t eand cool t h e f u e l and
coolant pump bearings and t o cool t h e s h i e l d plugs located between t h e bearings and t h e bowls of t h e pumps. meet containment requirements.
Each was a closed loop designed t o
The o i l used was Gulf-Spin 35.
Each system consisted of two 5-hp, 60-gpm a t 160-ft head, 3500-rpm Allis Chalmers Electri-Cand pumps, (one normally i n operation with t h e other
i n standby), an in-line Cuno EFS o i l f i l t e r , and an o i l tank of 22-gal ope r a t i n g capacity having "brazed on'' water c o i l s capable of removing. 41,000 Btu/hr
.
The o i l flow t o t h e bearings of each of t h e s a l t pumps was about 4 gpm with about 8 gpm t o each s h i e l d plug.
The remaining 4 8 q g m was recycled
through t h e o i l tank t o a i d i n cooling.
A scavenging J e t was i n s t a l l e d i n
the o i l l i n e s near t h e s a l t pump s o t h a t t h e s h i e l d plug o i l flow aided t h e return of t h e bearing o i l t o t h e storage tank. Lubricating o i l seeping past t h e lower s h a f t s e a l of t h e s a l t pump was piped t o t h e o i l catch tank f o r measuring t h e r a t e o f leakage. The pumps, storage tank, f i l t e r and much of t h e instrumentation and valving f o r each system were mounted i n an angle i r o n frame t o f a c i l i t a t e moving t o t h e s i t e .
This was commonly r e f e r r e d t o as a lube-oil package.
The two packages were interconnected s o t h a t e i t h e r could be used i n an emergency t o supply both salt pumps.
!
9.2 Before i n s t a l l a t i o n i n a t e s t stand.
e MSRE, both lube-oil packages were operated
Heat load and pressure drop d a t a were obtained and t h e
performance of t h e systems was checked.
A problem o f gas entrainment w a s
found which affected t h e priming of t h e standby pumps.
Approximately
30 seconds were required t o prime a spare pump i f it had not been operated f o r several hours.
Priming time w a s reduced t o 5 sec by i n s t a l l i n g gas
216 -
vents from t h e pump volute casing and t h e pump discharge l i n e t o t h e gas space i n t h e o i l tank.
u
This experience l e d t o a modus operendi of' s t a r t i n g
t h e spare pump of each package once per s h i f t and running it f o r about
1 5 min. The packages were i n s t a l l e d at t h e MSRE i n 1964.
I n i t i a l operation
was hampered by f a i l u r e of t h e s t a t o r i n s u l a t i o n i n one of t h e pump motors and by low resistance t o ground i n s t a t o r s of others.
Since moisture w a s
suspected, a p o t t i n g compound (Dow Corning S i l a s t i c RTV-731) was used t o s e a l t h e motor housing j o i n t s and a moisture r e s i s t a n t coating of paint (Sherwin W i l l i a m s epoxy .white B69W6) w a s applied t o t h e e x t e r i o r surfaces of a l l four operating motors, plus two spares. The loss of prime of t h e standby pumps continued t o be a problem u n t i l
t h e scavenging J e t s used t o r e t u r n o i l t o t h e reservoirs from t h e salt pumps were replaced with j e t s of lower capacity t o reduce t h e entrainment of gas i n t h e r e t u r n o i l streams.
These replacements were done i n September 1965.
A f t e r modifications, it was possible t o reduce t h e frequency of priming of t h e standby pumps t o once a week. Because of t h e lowered capacity of t h e new r e t u r n j e t s , it was found necessary t o l i m i t t h e f l o w of o i l t o t h e fuel-pump motor t o
4 gpm, as a
flow of 5 gpm would result i n a buildup of o i l i n t h e s a l t pump motor cavity. ,
9.3
Addition of Syphon Tanks t o t h e O i l Catch Tanks
The o i l catch tanks were f a b r i c a t e d of a 46-1/2-ine-1ong 2-in. pipe topped by a 20-in.-long
section of 8-in. pipe.
section of
The lower sec-
t i o n allowed accurate measurement of t h e o i l accumulation rate whereas t h e upper portion provided s u f f i c i e n t volume t o handle possible gross leakage. The catch tanks were drained periodically t o keep t h e l e v e l i n t h e lower section,
Radiation would not permit doing t h i s during power operation and
since t h e volume of t h e lower section was around 2500 cc and t h e allowable
seal leak rate was 100 cc/day, it w a s possible t h a t t h e r e a c t o r would have t o be shut down t o drain t h e o i l catch tanks during extended runs. Therefore, i n October 1965, equipment was i n s t a l l e d t o automatically siphon t h e
o i l from t h e catch tanks when they became f u l l .
LJ
217
ti
After i n s t a l l a t i o n , t e s t s were run on these and they performed w e l l . However, they f a i l e d t o function properly a t t h e low o i l leakage r a t e s that a c t u a l l y occurred during operation.
The o i l flowed over t h e high point of
t h e syphon tubes, as over a weir, without bridging t h e tube t o form a syphon.
We therefore reverted t o manual draining during shutdowns.
Fortu-
nately t h e leak r a t e s did not get high enough t o i n t e r f e r e w i t h operation of t h e reactor. 9.4
O i l Leakage
Continuous records of o i l tank a n d ~ o i lcatch tank l e v e l s were maint a i n e d during all of t h e MSRE operating l i f e t o determine what l o s s e s were taking place.
Samples removed from t h e systems were c a r e f u l l y measured,
as were additions.
The o i l storage tanks were l a r g e and therefore s m a l l
changes i n l e v e l indication caused.large e r r o r s i n inventory. log readings could vary by 500 t o 600
CC.
Successive
F a i r l y accurate indication of
'-
t h e leakage through t h e r o t a t i n g o i l seals of t h e s a l t pumps t o t h e o i l
W
s e a l s of t h e s a l t pumps t o t h e o i l catch tanks were possible over one month o r longer periods.
The leak rates during operation varied from a few cc
.per day f o r e i t h e r s a l t pump t o around 25 cc/day.
The s e a l s seemed t o get
worse o r improve f o r no apparent reason. For t h e 24-month period through August 1969 , inventories showed unaccounted f o r l o s s e s from t h e f u e l pump o i l ' s y s t e m of 5.4 the coolant o i l system of 5.6
5e'+
-3.0
l i t e r s ; from
l i t e r s or average d a i l y l o s s e s of 7 . 5 cc -3.0 (5.25 gms), and 7.8 cc (5.4 gl~ls) respectively. Analyses indicated that
t h e r e were approximately 1 t o pump off-gas stream.
5 0 ' +
gms of o i l products per day i n t h e f u e l
Some o i l may have been held up i n t h e l i n e s , 6-ft3
holdup volime, and p a r t i c l e t r a p s t h a t were upstream of t h h sampling point. I n conclusion, t h e b e s t estimates showed an o i l loss of 5+ gms/day, of which only 1
-2
@~ms co
gas f r o m t h e primary s y s t e for.
e found as'hydrocarbons i n samples of off-
aving approximately 3 gms/day unaccounted
218
9.5
W
Change-Out of O i l Pumps
A f t e r t h e i n i t i a l trouble with t h e e l e c t r i c a l i n s u l a t i o n due t o mois-
ture, t h e pumps gave very s a t i s f a c t o r y service.
Two pumps were removed
from service, one because of excessive vibration and t h e other because of an e l e c t r i c e short i n t h e motor winding. month period ending August 31,
This w a s done during t h e six-
1967.
The p\m?p with excessive vibration had one of t h e two balancing d i s k s
loose on t h e shaft.
This loose disk was reattached, and t h e s h a f t assembly
was dynamically balanced.
The pump was reassembled, t e s t e d under operating
conditions and made ready f o r service.
The pump with t h e shorted winding
was rewound and put back i n t o service. Test Check of One O i l System 9.6 Supplying Both t h e Fuel and Coolant S a l t Pumps On .March 20, 1966, t h e two o i l systems were valved so t h a t t h e f u e l o i l package (FOP-2 running) supplied o i l t o both salt pumps.
This was done
t o check calculations, and t o be sure t h a t t h i s could indeed be done i n some future emergency.
The coolant pump was c i r c u l a t i n g salt a t 1200'F;
t h e f u e l pump was idle and at a temperature of 125'F.
u
Adequate flaws t o
both s a l t p p p s were maintained (3.35 gpm t o t h e bearings and 6.5 and 7 . 3 'gpm t o t h e coolant and f u e l pump s h i e l d plugs). o i l supply temperature e q u i l i b r a t e d a t l27'F.
Under t h e s e conditions t h e When t h e f u e l pump o i l flow
was stopped, t h e o i l supply temperature increased
Q
7'F,
indicating t h a t
t h e f u e l p u p was removing p a r t o f t h e heat load of t h e o i l system. Although t h i s test did not prove conclusively t h a t one o i l cooling system was adequate f o r an emergency wherein one package would be used t o supply both-operating salt pumps, subsequent heat exchanger tests and calculations indicated t h a t one o i l cooling system would be adequate.
9.7
O i l Temperature Problems
The t o t a l o i l f l o w from one hibe o i l pump w a s normally 60 gpm.
Of
t h i s , 8 gpm-w a s directed t o t h e s h i e l d plug of t h e salt pump i n t h e par'
t i c d a r system, 4 gpm t o t h e salt pump bearings, and t h e balance was bypassed back i n t o t h e o i l tanks and caused t o flow down khe outer tank w a l l s ,
W
219 This amount plus t h e warm o i l from t h e s h i e l d and bearings was cooled by means of water flow through c o i l s brazed t o t h e outside of t h e carbon s t e e l tanks. A s t h e system was subjected t o extended service, s c a l e b u i l t up on t h e
inside surfaces of these c o i l s , causing a drop i n t h e cooling capacity. This i n t u r n caused t h e temperature of t h e supply o i l - t o climb from i t s
This higher temperature w a s reached during March
normal 134'F t o 140'F. of 1966.
s c o i l s were flushed w i t h a 15-wgt % solution
Both of t h e lube o i l of a c e t i c acid.
Considerable material w a s removed, and t h e o i l tempera-
t u r e s , upon r e s t a r t i n g , were reduced 7'F
,
o i l system, and 3'F
i n t h e case of t h e coolant pump
i n t h e case of t h e fuel-pump o i l system.
During subsequent operatlon t h e temperature o f t h e f u e l pump o i l gradu a l l y rose, and i n August 1966, t h i s a c e t i c acid treatment w a s again given both coolers. This problem of gradual fouling of t h e water s i d e of t h e cooling c o i l s
u
r e curred throughout subs eque
operations, and by December 1967, t h e tem-
perature of t h e o i l supplied t o t h e pumps was up t o 150'F. ply was changed from tower
The water sup-
e r t o cooler untreated "process water" which
i n t h e case of t h e E R E was taken d i r e c t l y from t h e potable water system
via a backflow preventer.
The system w a s l e f t on process water f o r two
months and t h e heat t r a n s f e r improved enough, probably by physical flushing out of the s c a l e , so t h a t tower water could again be used and a f f o r d satisfactory cooling of t h e o i l
i s switching had t o be done at l e a s t twice
more before t h e r e a c t o r w
a l l y shut down i n December 1969.
permanent hose and piping d i t ious l y
.
9.8
A semi-
w a s provided s o t h i s could be done expe-
Increase i n Radiation Level at t h e Lube O i l Packages
A t the beginning of power.operation i n August 1969, a r a d i a t i o n moni-
t o r at t h e r e s e r v o i r of t
1pump o i l system i n d i c a t
t h e upper p a r t of t h e t
increasing and decreasing
U s i n g t h e on-site gamma
u
w a s 41Argon.
r a d i a t i o n from i t h r e a c t o r power.
t e r , it was determined t h a t t h e a c t i v i t y
This was apparently produced by a c t i v a t i o n of t h e blanket gas
i
i n t h e upper p a r t of t h e f u e l pump motor cavity.
Prior t o t h i s time only
Li
helium had been used during power operation, and no gas activation had occurred. '
Argon was being used at t h i s time t o investigate bubbles i n t h e
fuel loop.
Since radiation levels did not 'exceed 2 mR/hr and decreased
when t h e cover gas was changed back t o helium, t h i s was not a serious problem.
9.9
Analysis of O i l
Samples were taken of each new supply drum and f r o m t h e two lube o i l systems after c i r c u l a t i n g f o r a short time.
Analyses of these were used as
controls, t o compare with periodic samples taken from t h e systems during operation.
The main concern was t h a t t h e o i l , especially i n t h e f u e l pump
system, might undergo some changes due t o long-time circulation through t h e high flux f i e l d i n the pump o r t h a t there might be some thermal decompo-
sition. Table 9.1 i s a compilation of some t y p i c a l o i l analyses performed during t h e l i f e of t h e MSRE.
The only s i g n i f i c a n t o r even minor changes
CH
r a d i c a l increased somewhat and some C=O w a s formed, probably indicating some small degree of oxidation. T r i t i u m t h a t were found were t h a t the
buildup w a s insignificant. The o i l which leaked through t h e s e a l s t o t h e o i l catch tanks, although dark i n color, showed no s i g n i f i c a n t chemical o r physical changes. An addition f r a c t i o n a l d i s t i l l a t i o n t e s t analysis w a s run on a sample of new Gulfspin-35 o i l on September 2, 1964.
This was d i s t i l l e d under re-
duced pressure (3-4 mm Hg) i n an 18-in. reflux column.
As indicated below,
most of t h e d i s t i l l a t i o n occurred between 148 and 178OC. 9.10
Replacement of O i l
The first charge of Gulfspin-35 o i l was added i n September 1964 as t h e o i l tanks were calibrated.
Each tank took about 35 gallons.
Periodic ad-
ditions were made t o compensate f o r losses.
As indicated previously, no s i g n i f i c a n t physical ok -chemical changes occurred which would indicate t h a t t h e o i l s h o u l d b e replaced.
Various o i l
companies were contacted and they d i d not have any suggestions as t o other
ss
221
c
Table 9 -
MSRE Sample No.
MSRE
Run No.
LO-1 thru LO-5 and LO-9 t h m LO-13
Date Sampled
Description
Viscosity Centistokes At 25OC
8125165
5 ea 55-bal drums new Gulfspin-35
15.44 to 15.99
LO-6
5
2118166
FOP-CONTROL
Lo-7
5
21la/66
COP-CONTROL
8/8/66
New Oil for FOP
8/13/66
New Oil for COP
Lo-47
Viscosity
Viscosity
At lOO'F
At 210'F
ssu
ssu
Carbon
x
--
85.52 to 86.43
9.94
2.64
86.88
-N O T A N A L Y Z E D
7
N O T A N A L Y Z E D -
Lo-49
-
LO-69
11
2113167
From COP
9.94
2.64
86.21
-70
11
2113167
From FOP
10.26
2.67
86.6
N O T A N A L Y Z E D
LO-140
5116168
-ul-141
5/16/68
** ***
LO-154
Sf 19169
From FOP
Solids ppm 180
9.94
2.58
78.0
LO-155
5/19/69
From COP
180
10.52
2.67
76.7
-
N O T A N A L Y Z E D
-
*References: MSRE Sample Log for Water, Helium, Miscellaneous; and MSRE Data Pile - Section 6E-1 (Lube 0 **Oil service from 5120167 to 5/16/68 - Collected in the Waste Oil Receiver from the fuel pump oil cat1 ***Oil in in service from 5120167 to 5/16/68 - Collected in the Waste Oil Receiver from the coolant pump oil
I
1
MSRE
.1
Sulphur
GULFSPIN 35 LUBE O I L
Bromine Number
x
Flash Point
Sediment x v/v
Al
Be
- 0
Ca
,
TYPICAL ANALYSIS
Ce
Cr
Cu
1 -- Microp;rams/KL
SPECTROGRAPHIC ANALYSEd Fe La Mg Mo ti
I
1.56 to 1.95
316'F to 325'F
. e0.04
Ha+% 1.9
322'F
0.06 HaO-X n nn
0.05
1.55
Analyses) tank. tch tank.
I
.
_I_.
.
326'F 333'F
Sr
Ti
v
Zn
Zr
HaO-ppm 150 bo-ppm 200
t
<20
<20
0.5
<0.6
<2
96
<On6
2
250
<20
<20
2.5
<0.6
<2
136
<0.6
-- {-
108
---
-
124 <10
0.3
0.2
<2
0.1
<0.4
e1
<0.02
352 e10
0.51 ~0.4 1.9
<2
0.2
<0.4
3.9 <I
---
520
--- ---- --
%ZOO
1.0
648
~0.4
e2
3
<2
<2
Li 4.0
50
---
<6
K 10 <0.5
5.0 %2
10
K
%5
1.0 (0.5
Y
h
50
I
74
<Os02
0.2
1.29
Sn
2
<1
4
0.012
Pb
Comments and Miscellaneous
ANALYZED AS ESSENTIALLY TtI SAME AS LO-6 and LO-?
(all)
I
0.10
P
II
%l
%2
$5
.5
Tr 'a 5
'a 3
5.0
s1 1.0 s1 1 s1 10
3.4
-- --- --
---
-- --
4.6
130
---
200
--
2.0
Interfacial tension 15.4 (normal Interfacial tension 16.0 (normal
-
-
18) 18)
Dark appearance
%loo
Had C-0 band indication some oxidation
%loo
Had no C-0 band increase over new oil
222
means of checking f o r damage.
However, as protection against undetected
changes, t h e o i l was replaced i n August 1966 and again i n May 1967.
9.11
Discussion and Recommendations
Except f o r t h e priming d i f f i c u l t i e s w i t h t h e pumps and t h e moisture t r o u b l e s with t h e motors, t h e systems operated very s a t i s f a c t o r i l y . No noticeable change occurred i n t h e l u b r i c a t i n g o i l .
However, more
information i s needed as t o what chemical analysis o r physical t e s t s should be made t o d e t e c t undesirable changes.
A more r e a l i s t i c sampling schedule
needs t o be established. Improvements i n future heat exchangers f o r t h i s s e r v i c e should be made. Excess capacity should'be provided t o compensate f o r fouling of t h e heat t r a n s f e r surfaces o r better cooling water i n h i b i t o r s should be used.
223 10.
C,OIWONEN!T COOLING SYST?!NS P. H. Harley
The primary and secondary component coolant systems consisted of compressors and associated equipment required t o gas-cool various components i n t h e f u e l and coolant systems. Although they performed similar functions, , t h e two were completely independent and a r e therefore described separately. 10.1
10.1.1
Primary System
Description
Cell atmosphere gas (95% N2, 5% 02), commonly c a l l e d " c e l l air" w a s r e c i r c u l a t e d by one of two 75-hp belt-driven p o s i t i v e displacement pumps (component coolant pumps). a l l y s t a r t e d i f needed.
The o t h e r pump w a s a spare which could be manu-
Each pump was located i n a containment enclosure
o r dome which could be opened f o r maintenance.
After compression by the
component coolant pump, t h e c e l l a i r passed through a water-cooled s h e l l and-tube heat exchanger and then t o t h e reactor and drain tank c e l l s f o r d i s t r i b u t i o n t o t h e freeze valves, control rods, r e a c t o r neck, and fuel-pump cooling shroud.
A s i d e stream w a s c i r c u l a t e d past a r a d i a t i o n moni-
t o r t o i n d i c a t e c e l l - a i r a c t i v i t y and could be exhausted t o t h e containment stack f o r evacuating t h e c e l l s .
10.1.2
I n i t i a l Testing;
"he i n i t i a l t e s t i n g w a s primarily a check of t h e system design.
t o equipment were measured and interdependence of flows w a s checked. system was leak-tested and
Flows
The
heat t r a n s f e r c o e f f i c i e n t of t h e gas cooler
was measured. Permanent flowmeters were not i n s t a l l e d since most of t h e flows were s e t i n i t i a l l y and then not changed.
During i n s t a l l a t i o n of piping, various
flows were measured with temporar l y i n s t a l l e d rotameters.
t o t h e fuel-pump shroud nee s t a l l e d when t e s t s i n d i c a t e
Since a i r flow
e varied, a permanent o r i f i c e w a s in-
a i r loading on t h e control valve w a s not
an adequate indication of a i r flow. I n i t i a l flows were adequate except f o r cooling t h e control rods.
g,
f i c i e n t flow w a s obtained by removing a discharge block valve.
Suf-
This block
224
valve w a s intended t o close on high c e l l - a i r a c t i v i t y t o prevent gross contamination of t h e c e l l i n case a control rod thimble ruptured.
U
I n l i e u of
the block valve, t h e discharge a i r was routed through a 55-gallon s t a i n l e s s s t e e l drum located i n t h e r e a c t o r c e l l t o r e t a i n any salt which leaked due t o the rupture. The measured flows a r e given i n T a b l e 10.1
Tests indicated t h a t chang-
ing one flow had no e f f e c t on t h e o t h e r flows even when t h e c e l l was being evacuated at about 200 scfm.
The pressure control valve (PCV-960) which
discharges excess a i r back t o t h e c e l l adequately compensated f o r t h e varying flows and maintained a constant discharge pressure. 10.1.3
Blower Capacity
-
Design of i n - c e l l MSRE components i n i t i a l l y required a gas-cooling capacity of 885 scf'm.
However, development t e s t s indicated t h a t cooling
air would not be required on t h e freeze flanges.
Therefore, t h e blower
capacity was lowered t o 6 1 7 scfm by i n s t a l l i n g a smaller sh'eave on t h e
1
drive motor. After about two-years operation, more c i r c u l a t i o n was deemed necessary between drain tank c e l l and r e a c t o r c e l l .
g?
This was t o give a more rapid
response on t h e c e l l air a c t i v i t y monitors i n case of a leak i n t o t h e drain tank c e l l .
A l a r g e r diameter drive sheave w a s i n s t a l l e d t o give a blower
capacity of 740 scf'm and a 2-in.
l i n e w a s i n s t a l l e d from t h e freeze valve
supply header t o t h e drain tank c e l l .
This gas (%200 scfm) flowed from t h e
drain tank c e l l i n t o t h e r e a c t o r c e l l and back t o t h e component coolant pump suction.
This
740 scf'm w a s
used f o r all subsequent operation.
During
rapid evacuation of t h e c e l l , it w a s necessary t o close a hand valve i n t h i s l i n e i n order t o maintain an adequate flow t o other components. 10.1.4
Gas Cooler
The heat removal by t h e gas cooler was adequate throughout t h e operation.
The i n i t i a l heat t r a n s f e r c o e f f i c i e n t calculated t o be
f't2 OF w i t h a flow rate of 885 scfm.
After two years of operation and a t
a flow r a t e of 740 scfm, it was 28 Btu/hr f t 2 proximately t h i s same value.
36.5 Btu/hr
OF
and has remained at ap-
No inspection has been made t o determine how
much o i l and/or b e l t dust i s deposited on t h e gas s i d e ( s h e l l s i d e ) of .the tubes.
LJ
225
Table 10.1 Flaw Distribution of Primary Component Cooling System
Measured Flow scf’m
Equipment Supplied
No. 1 Control Rod Motor
1.33
No. 1,Control Rod
3.8
Reactor Neck Inside
1.45 3.9 1.4 3.9 18 18
Graphite Sampler
16.2
No. 2 Control Rod Motor No. 2 Control Rod No. 3 Control Rod Motor
No. 3 Control Rod Reactor Neck Outside
FV-104
Design
5 5
5 15
15 15
75
FV-104
26.6
15
FV-105
27.7 24.8
15 15 15
FV-106 FV-107 FV-108
15 15
FV-109 Cell A i r Radiation Monitor (~!&565)
100
W52* ** Fuel Pump
260
*** PdcV-960
30
100
*L-952
was i n s t a l l e d i n September 1966 t o increase circul a t i o n between reactor and drain tank c e l l s .
**I n i t i a l
design I n i t i a l normal use w a
***PCV-960
scfm was lowered t o 100 scf’m. cfm but l a t e r no flow w a s required.
discharged excess capacity t o maintain a cons t a n t discharge pressure.
226
10.1.5
U
CCP Drive Units
Originally t h e component coolant pumps were driven by f i v e matched b e l t s using a 15.9-in.
drive sheave an@ a 21-in.
drive sheave.
blower capacity w a s reduced by changing t o a 10.9-in. d i f f i c u l t i e s developed.
When t h e
drive sheave, b e l t
The first s e t of b e l t s f a i l e d after 1450 hours of
operation and t h e second s e t only 11 hours later.
The smaller diameter
drive sheave increased the bending s t r e s s e s t o values i n excess of t h e b e l t To correct t h i s , i n February 1966, t h e drive and drive sheave dia-
ratings.
meters were increased t o 12.5 and 24 in. respectively.
To obtain a b e t t e r
b e l t loading, t h e i n i t i a l matched b e l t sets were replaced by a poly-V-belt. Then i n August 1966, the drive sheave was increased t o 15-in. diameter t o increase blower capacity.
This f u r t h e r reduced bending s t r e s s e s on the
belt e The original b e l t s were rated a t 200째F whereas the poly-V-belts were
only rated at 130째F.
This caused d i f f i c u l t y since the ambient temperature
i n the domes was about 170째F due t o motor heat and leakage of 320째F gas from t h e discharge relief valves.
hsi
Rupture discs were i n s t a l l e d t o prevent
leakage but they would not withstand t h e shock received when t h e blowers were s t a r t e d .
Baffles were then i n s t a l l e d which deflected t h e cooler in-
coming gas across the b e l t s and thus lowered t h e ambient temperature t o
around 150'F. To magnify t h e drive b e l t d i f f i c u l t i e s , t h e motor support w a s not satisfactory.
The motor could s l i p on i t s support p l a t e and flexing of t h e
motor support legs caused a pulsating load on t h e drive b e l t s .
Small blocks
were welded t o t h e support bracket t o keep t h e motor from slipping and heavy bracing of t h e stand minimized t h e vibration.
Since t h e l a r g e s t l s t r a i n m-
-
curred on t h e drive b e l t s when t h e blower w a s s t a r t e d , one blower w a s operated continuously during a run instead of a l t e r n a t i n g t h e blowers twice a month as was done i n i t i a l l y .
This increased t h e b e l t l i f e t o more than
8000 hours and uninterrupted runs of over 3000 hours were made without b e l t
adjustments 10.1.6
O i l System
One of t h e component coolant pumps (CCP-1) had t o be stopped during
Run 10 i n January 1967 because of low o i l pressure, and CCP-2 w a s used f o r
bd
227
W
t h e remainder of t h e run.
Af'ter Run 10, a cracked copper f i t t i n g i n t h e
o i l system was repaired, and 2 gallon of o i l w a s added t o bring t h e o i l l e v e l back t o normal.
Drains were a l s o
n s t a l l e d t o r e t u r n any o i l s e a l
leakage back t o t h e o i l reservoir. More trouble was encountered i n t h e CCP-1 o i l c i r c u l a t i n g system during Run 12.
F i r s t , a loose tubing connection c a u s e d t h e loss of ~2 gal. of o i l .
This was repaired i n August $1967. When t h e blower was r e s t a r t e d , intermit-
t e n t low oil-pressure alarms again occurred,
An investigation indicated
no s i g n i f i c a n t l o s s of o i l , but a slow oil-pressure response w a s observed when t h e blower was s t a r t e d .
The suspected o i l pump and pressure-relief
valve w e r e replaced with spare units t o correct t h e trouble.
Later inspec-
t i o n indicated t h a t t h e o i l pump was not damaged but t h e pressure-relief valve was r e l i e v i n g before the normal o i l pressure developed.
I n s p i t e of
t h i s d i f f i c u l t y , adequate l u b r i c a t i o n had been maintained and t h e r e was no
damage t o t h e blower,
The o i l pressure r e l i e f valve on CCP-2 was inspected
and a l s o found t o r e l i e v e a t too l o w a pressure and was readjusted. I n September 1967, a f t e r s i x days of flush-salt c i r c u l a t i o n and two
b,
days of nuclear operation, CCP-2 was shut down by low o i l pressure.
Since
t h e discharge valve would not close leak-tight, t h e run was terminated. Repairs consisted of repairing t h e leaking discharge valve, capping a leaking drain l i n e and tightening several packing nuts and f i t t i n g s . blower, CCP-1,
w a s used f o r 873 hours i n t o November 1967.
The other
A t t h i s time a
serious o i l leak was indicated by low o i l pressure and accumulation of about
2 gal. of o i l i n t h e condensate c o l l e c t i o n tank connected t o t h e blower containment.
The standby u n i t , CCP-2,
w a s immediately s t a r t e d up and oper-
ated without further d i f f i c u l t i e s throughout t h e remainder of t h e run,
3,036 hours. I n J u l y 1968, t h e brazed tubing o i l system on both blowers was replaced by welded pipe and blems w i t h t h e o i l system.
l e hose.
This eliminated most of t h e pro-
eaky shaft o i l s e a l had t o be replaced i n
December 1969.
10.1.7
Strainer
I n August 1966, a temporary s t r a i n e r with a 1/16-in, screen was in-
LJ
s t a l l e d i n t h e component-cooling-pump discharge l i n e t o c o l l e c t rubber dust
'
228
from t h e drive b e l t s , and prevent blowing t h i s material onto t h e hot metal components being cooled.
Although t h e temporary s t r a i n e r worked s a t i s f a c -
t o r i l y a new s t r a i n e r made of 100-mesh screen, which had been on order f o r
a year, w a s i n s t a l l e d i n t h e l i n e i n May 1967.
Over an eight-month period,
t h e temporary s t r a i n e r had accumulated 30 t o 50 g of black, dry powdery ma-
t e r i a l t h a t appeared t o be dust from abrasion of t h e drive b e l t s .
After
being decontaminated, the s t r a i n e r w a s examined and was found t o be i n very good condition.
The surface was s l i g h t l y etched, but no more than would
be caused by t h e decontamination process. The new s t r a i n e r developed excessive pressure drop i n September 1967
and t h e screen was replaced.
I n 1,800 hours of operation t h e 100-mesh screen
had become p a r t i a l l y plugged with rubber dust f r o m t h e d r i v e b e l t s .
It w a s
damaged i n removal and a replacement basket w a s made using 16-mesh, 0.023in. w i r e screen.
This showed no pressure buildup over t h e subsequent 2
years of operation.
10.1.8
Condensate Collection
A f t e r a reactor c e l l space cooler began t o leak water i n t o t h e c e l l i n
March 1966, condensate accumulated i n t h e 10-in. suction l i n e t o t h e component coolant pumps at a rate of 1 t o 2 gpd.
The water w a s vaporizing i n
the reactor c e l l and condensing on cold surfaces at t h e component coolant pumps. Simple drains were i n s t a l l e d , but these could be used only when the reactor was s u b c r i t i c a l due t o r a d i a t i o n i n t h e coolant d r a i n tank c e l l . Handling of t h e drained water w a s complicated by the tritium (up t o 915 vc/ml. produced from the 6 L i i n t h e treated-water corrosion i n h i b i t o r o r from diffusion through primary piping.
During t h e shutdown i n August 1966,
piping w a s i n s t a l l e d t o permit draining t h e domes during power operation t o a tank i n t h e sump room.
It was periodically pumped from t h i s tank t o
t h e l i q u i d w a s t e tank. l0,l.g
Electrical
I n June 1966, t h e wiring t o CCP-1 drive motor shorted out i n t h e penet r a t i o n i n t o t h e containment enclosure.
The s h o r t t r i p p e d t h e main breaker
t o generator bus No. 3 as w e l l as t h e component coolant pump breaker. pump and other equipment on bus No. 3 stopped.
The
Emergency d i e s e l power was
229
bsi
s t a r t e d immediately, but because of a misinterpretation of t h e problem, the r e a c t o r drained before cooling was restored t o t h e freeze valves by switching t o CCP-2.
The short occurred i n an epoxy s e a l on t h e end o f t h e
copper-sheathed mineral insulated cable and w a s caused e i t h e r by moisture from i n - c e l l leakage o r a breakdown of t h e epoxy p o t t i n g compound. Wiring changes were made t o both component cooling drive motors t o prevent a reoccurrence.
Each phase of t h e t h r e e phase c i r c u i t s was brought
through a separate penetration and t h e epoxy w a s omitted from t h e end s e a l s .
No f u r t h e r d i f f i c u l t i e s were encountered. 10.1.10
Valve Problems
I s o l a t i o n valves were provided on each component coolant pump t o per-
m i t r e p a i r i n g one while the other w a s i n operation. (10 in. on t h e i n l e t and
These l a r g e valves
6 i n . on t h e o u t l e t ) had back-seating stems on t e f -
lon seats t o prevent leakage when t h e valves were open. be very s a t i s f a c t o r y .
A f t e r about 3-years service, one of t h e 6-in.
l a t i o n valves leaked i n excess of 100 cfd.
Li
These p r o v e d t o iso-
An fnspection revealed several
l a r g e p i t s i n t h e s e a t i n g surface of t h e valve which had been f i l l e d with epoxy during manufacture.
These were repaired by f i l l i n g w i t h weld metal
and then r e f i n i s h i n g t h e seating surface.
When reassembled, t h e leakage
through t h e valve was $0.5 cfd which was considered t o be s a t i s f a c t o r y . Check valves were i n s t a l l e d on t h e discharge of t h e blowers t o prevent These were 6-in. butterfly-type valves w i t h s i l i c o n e rubber hinges holding two flappers i n place. I n January 1966,
backflow through t h e stand-by unit.
after 1640 hours of s e r v i c
he hinge on t h e CCP-2 valve broke allowing
one flapper t o f a l l o f f ,
pares were available s o flappers and hingers
were made at ORNL.
This had not f a i l e d a f t e r 1800 hours but w a s replaced
by factory-built p a r t s .
I n November 1966, t h e b u t t e r f l y check valve from
CCP-1 was found t o be inoperative and was repaired.
The f a i l u r e a l s o oc-
curred at t h e s i l i c o n e rubber hinge which supports t h e two wings of t h e check valve. On one occasion during t h e period when b e l t problems were causing l o s s of blower capacity, t h e stem of t h e i n - c e l l pressure control valve (PdCV960) froze i n a p a r t i a l l y open position and f a i l e d t o close when t h e valve
Ld
operator a i r loading increased.
This was t h e only f a i l u r e of an i n - c e l l
230
component i n t h i s system.
The valve was removed remotely and repacked.
The stem was polished and l i g h t l y l u b r i c a t e d before r e i n s t a l l a t i o n .
W
To
prevent t h e valve from remaining i n one t h r o t t l e d p o s i t i o n f o r an extended
t i m e , t h e pressure c o n t r o l l e r was subsequently cycled periodically t o be sure t h e valve w a s functioning.
Moisture i n t h e c i r c u l a t i n g a i r probably
contributed t o t h e valve failure, and t h e periodic draining of t h e collected moisture has helped prevent a reoccurrence. 10.1.11
Conclusions and Recommendations
Although many di'fficulties have been encountered with equipment, only two r e a c t o r drains have been caused d i r e c t l y by component cooling system failures.
One w a s l o s s of one blower while the other w a s i s o l a t e d f o r main-
tenance and t h e second w a s a f a i l u r e of e l e c t r i c a l cable t o t h e motor of one compressor (CCP-1).
One run w a s r e s t a r t e d a f t e r r e p a i r i n g an o i l leak
which.developed early in the run.
Better design s p e c i f i c a t i o n s would have assured more serviceable items. For instance, t h e pressure relief valves had a s p e c i f i e d capacity a t 1 5 psig, however, they discharged a small amount of gas continuously a t norm a l operating pressure.
containment domes.
This caused excess heating of t h e equipment i n t h e
Combining overpressure protection w i t h an unloading
valve and controls would have permitted blower s t a r t u p without load and caused less w e a r on the drive b e l t s .
A more durable hinge would have pre-
vented t h e check valve f a i l u r e s . More instrumentation on gas flows and on t h e o i l system would have given earlier indication of pending d i f f i c u l t i e s .
O i l sump l e v e l and dis-
charge pressure with indicators outside t h e containment would have helped here. Although s u f f i c i e n t capacity was a v a i l a b l e , only minimum capacities
were available f o r cooling at some locations because of l i m i t i n g pressure drops of piping and control valves.
More design consideration should have
been given t o t h e p o s s i b i l i t y of needing a d d i t i o n a l cooling.
A f t e r cor-
r e c t i n g blower drive problems during t h e first two years of operation, sev-
eral r e a c t o r runs were made i n excess of 3000 hours w i t h only preventative maintenance between runs.
81
231
10.2
1
Secondary System
,1 Description
Ambient a i r f o r cooling out-of-cell
a
~ O - S Cp ~o s i t i v e
pressor (AC-3).
components w a s supplied by e i t h e r
displacement pump (CCP-3)
o r a piston-type a i r com-
"he a i r was d i s t r i b u t e d t o t h e coolant system drain freeze
valves and t o t h e chemical processing plant freeze valves. 10.2.2
Operating; Experience
Early t e s t i n g revealed t h a t t h e flow t o FV-204 and FV-206 was insufThe air-operated
f i c i e n t t o e s t a b l i s h good plugs i n t h e freeze valves.
control valve t r i m w a s enlarged t o reduce t h e pressure drop and t h e blower discharge pressure w a s increased from 8 t o 9.25 psig.
With these changes,
adequate a i r capacity, 25 scfh, was obtained t o each freeze valve. I n April 1967, t h e drive s h a f t of t h e r o t a r y blower, CCP-3,
seized i n
t h e brass sleeve bearing of t h e blower housing and d i d extensive damage t o t h e drive s h a f t .
Operation was continued using t h e service a i r compressor,
ACT-3, which w a s t h e standby supply t o keep FV-204 and FV-206 frozen.
A
second bearing f a i l u r e occurred i n CCP-3 a f t e r only 200 hours of operation following t h e first f a i l u r e .
Since t h a t time, AC-3 has been used as t h e
normal cooling a i r supply with CCP-3 as a standby unit.
CCP-3 was used
during programmed maintenance on AC-3 and while freezing FV-204 and FV-206 when more a i r w a s required than t h e a i r regulator from AC-3 would supply. A brass swinging gate check valve which prevented backflow through CCP-3 when t h e service a i r
ressor was used, failed because of wear
caused by pulsating flow wh
CP-3 w a s i n service.
replaced by a hand valve.
This check valve was
Then, when t h e a i r compressor was changed t o t h e
normally operated supply, a new check valve w a s .installed s o t h e blower could be r e s t a r t e d remotely i f a low header pressure a l a r m occurred.
The
hand valve was l e f t i n s t a l l e d so t h e blower could be i s o l a t e d i f t h e check valve f a i l e d again. 10.2.3
Conclusions
The p o s i t i v e displacement pump (CCP-3)
air mover functioned poorly.
which was intended as t h e prime
However, t h e service a i r compressor (AC-1)
was very r e l i a b l e and thus l i t t l e o r no operating time was l o s t due t o t h i s system.
232
11.
VENTILATION SYSTEM P. H. Harley
11.1
Description
The purpose of t h e v e n t i l a t i o n system w a s t o exhaust a i r from areas where there was a p o t e n t i a l danger of release of radioactivity.
The a i r
flowed through a network of ducts t o t h e f i l t e r p i t which contained 3 paral-
l e l banks of roughing f i l t e r s and absolute f i l t e r s .
It w a s pulled through
t h e f i l t e r p i t by one of t h e two stack fans and w a s then discharged up a
100-ft stack t o t h e atmosphere.
The stack was continuously monitored t o Sen-
assure t h a t t h e r a d i o a c t i v i t y released w a s within acceptable limits.
s i t i v e d i f f e r e n t i a l pressure gages w e r e i n s t a l l e d t o i n d i c a t e t h e d i r e c t i o n of.air flow between t h e various areas. 11.2
OperatinK Experience
After i n s t a l l a t i o n of t h e v e n t i l a t i o n system w a s complete, a series of t e s t s w a s run t o insure t h a t t h e c r i t e r i a would be met. Several minor changes were made i n t h e ducting, c e l l blocks were caulked, and leaks repaired.
The biggest problem encountered w a s s e a l i n g t h e hi-bay.
Gaskets
were i n s t a l l e d on t h e doors, all j o i n t s i n t h e l i n i n g were taped, and a number of l a r g e holes were plugged. ured as follows.
The leakage i n t o t h e hi-bay w a s meas-
The dampers i n t h e v e n t i l a t i o n ducts from all other areas *-
were closed.
All i n l e t vents and doors t o t h e hi-bay were closed.
The
dampers between t h e hi-bay and t h e stack fans were t h r o t t l e d u n t i l t h e hibay pressure was 4 . 3 inches of water. found t o be 4250 ft3/min.
The in-leakage ( s t a c k flow) was
Although t h i s w a s above t h e design value of
1000 ft3/min, it w a s accepted. The p i t o t t d e stack gas flow meter was c a l i b r a t e d using a hot wire anemometer.
Several t r a v e r s e s were made under various f l o w conditions.
The flow under normal operating conditions w a s found t o be around 23,000 cFm.
The stack c a l i b r a t i o n curve and d e t a i l s o f t h e c a l i b r a t i o n a r e given
i n MSRE Test Memo and T e s t Report 2.3.15. Tests indicated t h a t adequate v e n t i l a t i o n could be maintained with 2 of t h e 3 p a r a l l e l f i l t e r banks i n service. 9
Therefore, it w a s possible t o
replace one bank of f i l t e r s without i n t e r r u p t i n g operations.
I \ W
233
After i n s t a l l a t i o n , t h e absolute f i l t e r s d i d not pass t h e standard ORNL-DOP smoke test.45
(99.95% e f f i c i e n c y i s required.)
Visual inspection
indicated leakage around t h e frames which required considerable caulking and some minor revisions.
On 11/14/64 a l l t h r e e banks passed t h e DOP t e s t
(East = 99.970%; Center = 99.972%; and West = 99.967% e f f i c i e n t ) .
The next
t e s t i n May of 1965 indicated t h a t a l l t h r e e banks had dropped t o less than 99.90% e f f i c i e n t .
Therefore, they were removed and extensive recaulking
and revisions were made t o could be replaced remotely.
rove t h e i r efficiency and assure t h a t they e results of subsequent DOP tests a r e shown
i n Table 11.1. It has not been necessary t o replace t h e absolute f i l t e r s since 1965.
The pressure drop through them had changed very l i t t l e during t h e 4 years of operation which indicates t h a t t h e roughing f i l t e r s performed well and removed most of t h e p a r t i c u l a t e s . The roughing f i l t e r s were changed when t h e pressure drop across them I
LJ
reached 5 t o 6 inches of water at which time t h e s t a c k flow would be down t0-'~18;000cfm.
During e a r l i e r operation while considerable maintenance
and construction was s t i l l being done, these plugged r a t h e r rapidly.
Small
e a s i l y changed f i b e r g l a s s f i l t e r s i n s t a l l e d on s e v e r a l of t h e i n l e t ducts reduced t h e plugging r a t e . 11.2.
Subsequent f i l t e r changes a r e shown i n Table
The l a s t set of roughing f i l t e r s read %5OO mEt/hr a t 3 inches a f t e r
they were removed. Some d i f f i c u l t y was encountered i n e a r l y operation w i t h s t a c k fan No. 1 (SF-1) bearings.
The bearings were i n i t i a l l y grease-lubricated and t h e r e
w a s d i f f i c u l t y i n maintaining proper l u b r i c a t i o n ,
was about four months,
The normal bearing l i f e
I n 1966, an o i l - l u b r i c a t i n g system was i n s t a l l e d
and t h e subsequent bearing l i f e has been about 18 months. continuously except when necessary t o do maintenance on it.
SF-1 was run SF-2 served
as a standby u n i t . Sensitive d i f f e r e n t i a l pressure gages ( f u l l s c a l e = 2 inches of water) were provided t o indicate d i r e c t i o n of a i r flow.
These were recorded each
s h i f t on t h e building l o g and dampers were adjusted as necessary.
Venti-
l a t i o n flow was from t h e less hazardous t o t h e more hazardous area except
h.,
i n t h e coolant c e l l during power operation.
When t h e main r a d i a t o r blowers
were operating, considerable air leaked from t h e r a d i a t o r and pressurized
234
U Table 11.1
Results of DOP Tests of Absolute F i l t e r s
Efficiency Bank
East Center West
99 0999% 99 994 99.994 9
3/66
6/67
99 998% 99 -995 99 997
99 979% 99 -998 99 994
Table 11.2
99.994% 99 9% 99 994
.
99 993% 99 995 99 985 9
Roughing F i l t e r Changes
East -
Center
West -
October 1964
X
X
X
October 1966
X
X
X
August 1967
X
X
Septeniber 1968 August 1969
X X
X
X
235
t h e coolant c e l l .
In an e f f o r t t o correct t h i s , t h e i n l e t t o one of t h e
r a d i a t o r a u x i l i a r y blowers (MB-2) was connected t o t h e coolant c e l l .
This
reduced t h e pressure below atmospheric but it w a s s t i l l above t h e hi-bay pressure.
Since t h e only hazard was fromberyllium and since t h e coolant
stack was continuously monitored f o r beryllium, t h i s w a s considered acceptable. During maintenance, containment i s provided by t h e flow of a i r i n t o the cells.
The system has adequately supplied t h i s need f o r a l l mainte-
nance operations. There has been very l i t t l e a c t i v i t y released t o t h e atmosphere except during maintenance periods. ORNL permissible limits.
Even then, t h e releases have been below t h e
The permissible r e l e a s e r a t e f r o m t h e 5 ORNL
stacks i s %l c u r i e per week.
Table 11.3 i s a tabulation of t h e amount of
a c t i v i t y released from t h e MSRE stack.
The maximum released i n any week
at t h e MSRE was 568 m C i during t h e week ending December 21, 1969.
This
w a s mainly from t h e core specimen removal plus about 50 m C i from t h e p r i -
mary system leak which occurred during t h e f i n a l shutdown.25 Due t o a galled b o l t , t h e core specimen hold-down flange could not be r e i n s t a l l e d and was l e f t hanging i n t h e standpipe. s t a l l e d t o close t h e system.
Another flange was in-
S a l t on the hold-down assembly probably con-
tributed t o t h e l a r g e r than usual release.
The standpipe vacuum w a s a l s o
inadvertently l e f t o f f on two occasions when it should have been on.
11.3 Conclusions and Recommendations The v e n t i l a t i o n system has functioned.satisfactorilyduring operation and maintenance.
The i n s t a l l t i o n of e a s i l y removable f i l t e r s a t i n l e t s t o
t h e v e n t i l a t i o n ducts prolonged t h e l i f e of t h e roughing f i l t e r s a n d ' s h o u d
be considered i n f u t u r e designs. The stack monitoring system could be improved. were very d i f f i c u l t t o m a k e notes on.
The recorders used
This, together with t h e range chang-
ing f e a t u r e of t h e system, made l a t e r i n t e r p r e t a t i o n of t h e charts very difficult.
I n addition t o t h i s , t h e monitors were never calibrated.
This
should have been done and limits on t h e i r readings should have been est a b l i s h e d before s t a r t i n g operation.
236
U Table 11.3
/
Dates
i
I
Particulate
Gaseous
mCi
mCi
From
To
3/66 9/66
8/66
0
97
2/67
9
206
3/67
8/67
<0.3
4.0
9/67
2/68
2.3
3/68
8/68
<0.3 <0.3
9/68
2/69
<0.3
3/69
8/69 1/70
~0.3
9/69
i
Semiannual Stack Release
<0.3
343 0.4 12.5 8.2
Main Source of Activity Miscellaneous maintenance Off-gas line maintenance
Off-gas line maintenance
OFT vent maintenance Removal of core specimens and primary system leak
237 WATER SYSTEMS
12.
P. H. Harley
Water f o r various types of usage w a s supplied at t h e MSRE by t h e following systems:
(1)t h e potable water system d i s t r i b u t e d ORNL water f o r
f i r e protection, general building services, and was t h e supply f o r t h e process water system, ( 2 ) t h e process water system d i s t r i b u t e d water (potable water which had gone through a backflow preventer t o provide i s o l a t i o n ) f o r process usage, such as d i l u t i o n of t h e l i q u i d waste, f o r temporary o r emergency cooling and f o r makeup t o t h e cooling tower, ( 3 ) t h e cooling tower
water system r e c i r c u l a t e d t r e a t e d process water through various out-of-cell components and provided cooling f o r t h e t r e a t e d water, ( 4 ) t h e t r e a t e d wat e r system r e c i r c u l a t e d t r e a t e d steam condensate i n a contained loop t o provide cooling f o r a l l i n - c e l l o r contained components, ( 5 ) t h e steam condensate system condensed ORNL steam t o produce pure condensate which was used i n t h e drain tank a f t e r h e a t removal system and f o r t r e a t e d water system makeup, and (6) t h e nuclear instrument thimble water system recirculat e d t r e a t e d steam condensate t o provide cooling f o r t h e nuclear instruments
b.,
and b i o l o g i c a l shielding f r o m t h e r e a c t o r vessel. The operating experience, d i f f i c u l t i e s encountered, and recommenda-
t i o n s a r e given separately f o r each sub-system. . 12.1
Potable Water System
Two six-inch cast i r o n l i n e s supply w a t e r t o t h e MSRE area from t h e
supply main,
One l i n e suppl
water t o t h e f i r e sprinkler system and
general building services and t h e o t h e r supplies makeup t o t h e process wat e r system.
\
The underground supply l i n e t o t h e f i r e s p r i n k l e r system and building services ruptured i n t h e f a l l of 1967.
Repairs required s h u t t i n g off both
potable water supplies t o t h e a r e a so process water makeup w a s supplied by
a f i r e hose from a vent l i n e on t h e valley water main.
F i r e protection w a s
provided by c1/4 mile of f i r e hose from t h e Nuclear Safety P i l o t Plant and posting of a f i r e watch a t t h e l o c a l f i r e alarm box i n t h e building.
Iso-
l a t i o n valves where t h e l i n e s t e e d o f f t h e water main would have simplified
4d
t h e - r e -p- a i r s . ground.
I n December 1969, t h e other 6-in. supply l i n e ruptured under-
Process water was supplied from t h e other 6-in. main through a
238 temporary backflow preventer during repairs.
I n both cases, r e p a i r s were
made using a s t a i n l e s s s t e e l A d a m clamp. Although no serious malfunctions have occurred i n t h e automatic f i r e The
sprinkler system, a b e t t e r designed system could have been provided. MSRE system was i n s t a l l e d from national f i r e protection specifications
without regard t o area services.
I n some areas containing e l e c t r i c a l equip-
ment, more &anger existed from possible water damage than f r o m t h e possib i l i t y of fire.
I n these areas the s p r i n k l e r s were disconnected o r iso-
l a t e d with manual valves.
Portable extinguishers s u i t a b l e f o r e l e c t r i c a l
f i r e s have been provided i n areas containing e l e c t r i c a l equipment. 12.2
Process Water System
This system has performed very w e l l .
On a f e w occasions, process wa-
t e r was used f o r cooling c r i t i c a l equipment when r e p a i r s necessitated shutt i n g down t h e cooling tower water system. Process water which w a s %20째F cooler than t h e cooling tower w a t e r was e f f e c t i v e i n cooling t h e f u e l pump and coolant pump o i l tanks when t h e c o i l s became fouled with scale.
The
use of process water tended t o f l u s h t h e c o i l s so t h a t adequate cooling could be obtained using cooling tower water. The process water system w a s supplied by ORNL potable water through one backflow preventer.
Another backflow preventer was i n s t a l l e d i n s e r i e s
with t h i s i n t h e process water supply t o t h e l i q u i d waste tank. backflow preventers were scheduled t o be t e s t e d semiannually. i n t occurred a t i n t e r v a l s of 3 t o 14 months. never required any repairs.
These Actual t e s t -
The main backflow preventer
"he one i n t h e l i n e t o t h e waste tank required
minor r e p a i r s on two occasions.
Valving w a s provided t o i n s t a l l a second
u n i t i n p a r a l l e l i f r e p a i r s of t h e main operating backflow preventer were required.
However, t h e second u n i t w a s never i n s t a l l e d .
cess water flow was shut o f f t o make t h e periodic tests.
Instead t h e proFor a more com-
plex water system, two backflow preventers should be i n s t a l l e d . 1
W
239
12.3
Cooling Tower Water System
Standard c e n t r i f u g a l pumps were used t o c i r c u l a t e t h e cooling tower water.
Flow was adequate and no pump performance t e s t s were made.
The
cooling tower w a s a packaged commercial u n i t and provided adequate cooling. Very adequate temperature control w a s obtained using a temperatyre cont r o l l e d valve which allowed p a r t of t h e water t o bypass t h e cooling tower. One major pipe leak occurred which required shutting t h e e n t i r e system down. This leak w a s i n t h e underground six-inch r e t u r n l i n e t o t h e cooling tower.
Nalco-360 b a l l s , a mixture of sodium and potassium chromate containing some phosphates, were used as a corrosion i n h i b i t o r . I n t h e summertime, a one-pound b a l l of Nalco-2lS, an algae i n h i b i t o r , w a s added d a i l y and w a s s a t i s f a c t o r y i n preventing algae formation on t h e
cooling tower.
Hardness was controlled by bleeding o f f about 2 gal. per
minute of cooling tower water.
This water plus about 8 gallons per minute
of process water w a s used t o cool t h e charcoal beds. Over 1500 samples of cooling tower water were taken and analyzed.
bi
Most of these were analyzed at t h e s i t e t o assure adequate treatment. odically t h e samples were sent t o t h e laboratory f o r i r o n analysis. remained l e s s than 1 ppm throughout the operation.
PeriThis
Due t o t h e continual
makeup and bleedoff f r o m t h e system, good corrosion r a t e figures a r e not available, however, no .leak has occurred which could be a t t r i b u t e d t o corrosion. A considerable amount of calcium phosphate s c a l e coated t h e rotameter
tubes and cooling c o i l s and
ludge i n t h e cooling tower basin and low
velocity areas of t h e t r e a t
r cooler.
t e r a solution of p o t a s s i
mate and zinc-sulfate i n h i b i t o r i n t o t h e
cooling tower water by a prop makeup water flow.
d e n this
i n i n h i b i t o r w a s abandoned. t o f a c i l i t a t e reading and
Equipment was i n s t a l l e d t o me-
ioning pump powered by t h e cooling tower f a i l e d t o operate propertly, t h e change Lights were i n s t a l l e d behind t h e rotameters e frequency of cleaning.
240
12.4 12.4.1
U
Treated Water Systems
I n i t i a l F i l l and Operation
The t r e a t e d water system w a s f i l l e d by adding steam condensate and corrosion i n h i b i t o r batchwise.
The t o t a l volume w a s about 4000 gallons
with about 2300 gallons i n t h e thermal shield.
12.4.2
Circulating PumpS
Standard c e n t r i f u g a l pumps were used t o c i r c u l a t e t h e t r e a t e d water i n a closed loop. were made.
The capacity w a s adequate and no pump performance t e s t s
The pumps have operated s a t i s f a c t o r i l y w i t h l i t t l e maintenance
other than repairing t h e r o t a t i n g seals.
Leakage from these was collected
i n p l a s t i c b o t t l e s due t o t h e induced a c t i v i t y i n t h e water.
12.4.3
Thermal Shield
I n i t i a l t e s t i n g i n d i c a t e d t h a t eFcessive pressure would e x i s t i n t h e thermal s h i e l d and that the flow was not adequate t o t h e thermal s h i e l d s l i d e s (sections of t h e s i d e s which could be removed f o r replacing t h e rea c t o r vessel).
cd
Piping changes were made t o lower t h e back pressure and t o
provide a separate supply l i n e and pump t o assure adequate flow i n t h e slides.
During e a r l y operation, several rupture d i s c s i n t h e thermal I
s h i e l d discharge l i n e s were ruptured due t o t h e supply and discharge valve closing simultaneously.
To prevent t h i s , t h e supply valve w a s modified so
t h a t it would close before t h e discharge valve closed. The water level w a s noted t o increase i n t h e surge tank when t h e rea c t o r power w a s increased and t h e l e v e l decreased when t h e r e a c t o r w a s subcritical.
Decomposition of t h e water was producing %2 cubic feet per hour
of H2 gas a t full power and about 8 f t 3 of t h i s gas collected i n one of t h e TS s l i d e s i n which it i s postulated t h e water piping was reversed, i.e., water entering t h e t o p and leaving from a d i p l e g near t h e bottom.
A small
deaerating t a n k , i n s t a l l e d i n the water supply t o t h e TS s l i d e permitted
%2 f t 3 of gas t o redissolve when t h e system reached equilibrium.
Later a
l a r g e degasing tank was- i n s t a l l e d t o deaerate t h e water leaving t h e TS t o permit more H2 t o be dissolved.
This reduced tlie accumulate gas t o Q3 f t 3 .
I n i t i a l l y a purge of 32 and later a i r was introduced i n t o t h e surge tank
LJ
241
bi'
and degasing tank t o flush t h e gas space and d i l u t e t h e Hp t o below t h e explosive l i m i t .
No f u r t h e r d i f f i c u l t i e s were encountered, however,
%3/4-gal. of water/day w a s l o s t by evaponation i n t h e purge gas. 12.4.4
Reactor and Drain Tank Cell Space Coolers
Two space coolers were i n s t a l l e d i n t h e r e a c t o r c e l l and one i n t h e drain tank c e l l t o keep t h e c e l l a i r temperatures below 150째F.
The meas-
ured capacity was only 40 t o 60 kW compared t o t h e design value of 75 kW. However, t h e c e l l temperatures normally ran at about 140째F f o r t h e r e a c t o r c e l l and 130째F f o r t h e drain tank c e l l . The 3-hp fan motors o r i g i n a l l y i n s t a l l e d on t h e space coolers, were adequate f o r t h e normal operating pressure of
+ p s i g but
were too small
f o r higher pressures such as could be encountered i n t h e event of an acciThey were, therefore replaced with 10-hp motors.
dent.
One of these on
t h e drain tank c e l l cooler f a i l e d a f t e r 2900 hours of operation due t o t h e r o t o r s l i p p i n g on t h e shaft.
Since t h e trouble app'eared t o be caused by
f a u l t y construction r a t h e r than poor design, it was replaced by an identic a l motor. Before t h e c e l l s were sealed, a reactor c e l l cooler (RCC-2) developed a leak a t a header-to-tube brazed f i t t i n g .
Repairs proved t o be d i f f i c u l t
because,heat required f o r brazing one leak would cause an adjacent tube t o
start leaking.
After t h e c e l l was sealed, a 1-1/2 gallon per day water
leak w a s traced t o another brazed f i t t i n g ,leak on RCC-1. paired.
I n , J u l y 1967, RCC-2 again s t a r t e d leaking and w a s replaced by one
of t h r e e newly obtained un
n which t h e copper tubes were heliarc-welded header.
( i n s t e a d of brazed) t o t h 12.4.5
This was re-
In-Cell Water Leaks
In-cell water leaks have p e r s i s t e d throughout t h e r e a c t o r operation. Before t h e c e l l s were sealed, components could be checked Visually but aft e r power operation, l o c a t i n g t r e a t e d water leaks required hydrostatic t e s t i n g of i n - c e l l components
Although all known leaks such as a t t h e
space coolers were repaired,
continued t o have a leak i n t o t h e c e l l of
1 / 2 t o 1 gallons per day.
kid
Water from t h i s small leak d i d not c o l l e c t i n
t h e c e l l sump but evaporated i n t h e 130 t o 140째F c e l l gas and then condensed i n a colder portion of t h e containment, t h e suction l i n e s and gas
242
cooler of t h e component cooling system. t a i n considerable amounts of tritium.
This condensate w a s found t o con-
W
Drains were i n s t a l l e d t o permit-
draining t h e condensate t o a measuring tank i n t h e sump room from which it could be pumped t o t h e l i q u i d waste storage. The nuclear instrument penetration was one of t h e suspected sources
of t h e leak.
Since it could not be hydrostatically t e s t e d , 12 kg of D 2 0
was added as a t r a c e r .
This increased t h e deuterium concentration t o about
seven times the normal water concentration.
Since t h e deuterium concentra-
t i o n d i d not increase i n samples of t h e c e l l condensate, a leak f r o m t h e nuclear instrument penetration was discounted. 12.4.6
Treated Water Heat Exchanger
The heat exchanger i n s t a l l e d t o t r a n s f e r t h e heat from t h e treated wa-
t e r t o t h e cooling tower water w a s a surplus tube-and-shell
unit.
During
preoperational tests t h e tubes had t o be r e r o l l e d and tube sheet gasket re-
placed t o eliminate intermixing of t r e a t e d water and cooling tuwer water.
A f t e r a year of operation, 17 tubes which had developed cracks where t h e tubes were r o l l e d i n t o t h e tube sheet were plugged. caused by t h e ' e a r l i e r tube repairs.
This w a s probably
When leaks again developed i n t h e Tw
cooler, more tubes were plugged but when attempts t o s e a l t h e flanged tube sheet were unsuccessful, t h e heat exchanger w a s replaced w i t h another sur-
plus u n i t which w a s smaller but s t i l l had adequate cooling capacity.
Af-
t e r replacing t h e heat exchanger, sodium from t h e cooling tower system had t o be removed f r o m t h e t r e a t e d water by d i l u t i o n with steam condensate and new i n h i b i t o r was added.
The a c t i v a t i o n of t h e 1.6 ppm of sodium l e f t i n
t h e treated water a f t e r t h e d i l u t i o n raised t h e r a d i a t i o n l e v e l of t h e sys-
t e m t o 80 mR/hr at surge tank and 260 mR/hr a t t h e f i l t e r but d i d .not s e r i ously a f f e c t r e a c t o r operation. 12.4.7
Valves
Containment of the t r e a t e d water system depended upon air-operated block valves and check valves. annual containment checks, leaked excessively.
None of t h e block valves leaked during t h e
However, several of t h e piston-type 'check valves
Foreign material and s c a l e caused some of t h e valves
t o bind i n t h e open position due t o t h e close clearance between t h e valve
hj
24 3
body and t h e plug.
The valves were mounted i n horizontal pipes which
probably magnified t h e problem. Pressure r e l i e f valves were i n s t a l l e d t o prevent damage t o in-cell components i n case t h e block valves closed.
waste tank.
These discharged t o t h e l i q u i d
Leakage from these occurred several times.
The first indi-
cation was usually an increase i n r a t e of condensate makeup.
Pinpointing
the l e a k involved disconnecting piping from each r e l i e f valve.
Power re-
duction w a s necessary t o accomplish t h i s .
12.4.8
Corrosion I n h i b i t o r
The t r e a t e d water system w a s first f i l l e d w i t h condensate containing
2000 ppm of a 75+51
mixture of potassium n i t r i t e - p o t a s s i m tetra-borate. Induced a c t i v i t y 42K a t low reactor power w a s extrapolated t o radiat i o n l e v e l s of 400 mR/hr at t h e surge tank i n t h e water room and f i l t e r i n t h e d i e s e l house when t h e r e a c t o r was at f u l l power. To prevent th3s radia t i o n , t h e potassium i n h i b i t o r w a s changed t o lithium.
Lithium-'7 w a s used
t o minimize t h e production of tritium which would be formed i n t h e thermal
b,
shield.. The potassium was removed by d i l u t i o n with demineralized water from ORmL but had t o be repeated using steam condensate because t h e demine r a l i z e d water which contained 1 ppm of sodium became radioactive during t h e next run.
The new i n h i b i t o r w a s composed of lithium n i t r i t e , b o r i c
acid, and.lithium hydroxide. During e a r l y power operation, t h e t r e a t e d water l o s t a c t i v e i n h i b i t o r . Samples indicated t h a t 300500 ppm H202 was being produced by r a d i a t i o n damage which oxidized L i N O
w
equilibrium of -NO2--NO3 Approximately 365 s these were analyzed at t
iNO3.
Additional LiN02 w a s added u n t i l
ched with Q700 ppm of -NO3 i n t h e water. f t h e t r e a t e d water were taken.
Most of
o assure t h a t adequate treatment was be-
i n g maintained. 12.4-9
Corrosion
Corrosion - i n ,t h e treat i n g most of t h e operation, h i b i t o r , pH and.iron.
I
s,
r system has not been s i g n i f i c a n t . Dures were analyzed monthly f o r loss of in-
I n 1969, chromium and nickel analyses were added
t o f u r t h e r check f o r corrosion on t h e in-cell s t a i n l e s s s t e e l .
The i r o n
244
analysis has remained near t h e lower detectable l i m i t of t h e analysis indi-
L)
cating no corrosion and t h e C r on N i analysis a l s o indicates no s i g n i f i c a n t corrosion.
Although t h e chromium and nickel analyses haven't been followed
continuously, any corrosion would have caused an increase i n t h e amount present,
The present chromium analysis of .4 ppm i s equivalent t o 0.05
pounds of s t a i n l e s s s t e e l . One f a c t o r which-clouds t h e i n t e r p r e t a t i o n of t h e corrosion results has been t h e continual leakage from t h e system which has required periodic makeup.
The decrease i n concentration of i h i b i t o r of 25% i n two years
agrees reasonably with t h e known leakage.
An increase of t h e i r o n , chrom-
i u m , and nickel by t h e same amount would s t i l l i n d i c a t e a negligible amount of corrosion. If corrosion products p r e c i p i t a t e d as oxides, they would be caught on t h e t r e a t e d water f i l t e r .
This f i l t e r , located i n t h e d i e s e l
house, b u i l t up excessive pressure and w a s cleaned twice during p r e c r i t i c a l t e s t i n g but has not required f u r t h e r cleaning i n t h e past four years. 12 4.10
Recommendations
The biggest problem i n t h e t r e a t e d water system w a s t h e r e s u l t of t h e
l o w design pressure of t h e thermal s h i e l d (20-30 p s i ) .
Care should be ta-
ken i n f u t u r e designs t h a t all components withstand t h e dead-head pressure
of t h e c i r c u l a t i n g pump.
The piping changes required f o r protection of t h e
thermal s h i e l d r e s u l t e d i n other p a r a l l e l flow paths having marginal capacity. Consideration should be given t o t h e design of a system which i s comp l e t e l y contained.
This would eliminate t h e need f o r so many block valves.
Check valves, of t h e type i n s t a l l e d a t t h e MSRE, should not be r e l i e d on as containment block valves.
If similar valves are used, they should be i n
v e r t i c a l piping and have s u f f i c i e n t clearances t o insure t h a t they w i l l close when needed. A corrosion i n h i b i t o r should be s e l e c t e d t o minimize induced a c t i v i t y .
Provisions f o r handling gaseous a c t i v i t y such as tritium and r a d i o l y t i c hydrogen should be provided.
Large components should be shielded o r lo-
cated so as t o minimize personnel exposure t o t h e induced a c t i v i t y .
u
245
6,
12.5
mdensate has been a very s a t i s f a c t o r y supply of de-
Building steam mineralized water.
QWdensate System
The steam i s condensed i n a s t a i n l e s s s t e e l tube-and-
s h e l l heat exchanger and stored i n two s t a i n l e s s s t e e l tanks. I n i t i a l l y condensate makeup was added by gravity feed through an auto-
matic l e v e l control valve t o maintain a constant l e v e l i n t h e t r e a t e d water To obtain b e t t e r t r e a t e d water inventory data, t h i s was
surge tank.
changed t o periodic manual additions when t h e surge tank l e v e l decreased below 50 gal. Condensate i s used t o f i l l feedwater tanks which supply cooling water f o r r e c i r c u l a t i n g evaporative cooling of t h e f u e l drain tanks. i n h i b i t o r i n t h i s service would p l a t e out during boiling.
A corrosion
The system has
worked well except f o r 2 occasions when leaking valves between t h e feedwater
tanks and steam dome cause
i d e n t a l addition of water t o a steam dome.
A t one time t h e high drain
temperature ( 1 3 O O O F ) interlocks actuated
t h e supply valves t o FD-2 steam dome following a normal drain.
6.i
This could
have been prevented by turning off e l e c t r i c heaters t o t h e drain tank s h o r t l y after t h e drain.
Du
o refluxing, all of t h e water could not be
removed from t h e steam dome when cooling was no longer required.
To cor-
r e c t t h i s , a pump and storage drum were i n s t a l l e d i n t h e north e l e c t r i c service area t o permit complete drainage of t h e steam domes when required. 12.6
Nuclear Instrument Penetration
Instruments f o r l o c a t e d i n a 36-in.-di t o t h e inner edge of t h e about 1700 gallons
tassium n i t r i t e and potass 2000 ppm.
A s with
ithium n i t r i t e - l i t h i u m t e w a s i n i t i a l l y by natur
uclear c h a r a c t e r i s t i c s of t h e system were
a f t which extended f r o m t h e high-bay f l o o r shield.
This was o r i g i n a l l y f i l l e d with
water t o which a
75+5% mixture of
e t r a b o r a t e w a s added t o a concentration of
r system, t h e i n h i b i t o r was changed t o r a t e t o lower t h e induced
ivity.
a t i o n but gamma heating wh
the reactor was
taken t o full power increased t h e water temperature %7O0F.
kd
PO-
Cooling
This decreased
t h e neutron attenuation causing a divergence between t h e indicated nuclear
I
246 J
power and t h e heat balance power.
A cooling system using a 5-gpm pump and
a small available heat exchanger were i n s t a l l e d which reduced t h e increase
i n water temperature t o
and kept t h e nuclear power measurements and
% 1 8 O ~
heat balance calculations within %5%. I n i t i a l l y t r e a t e d water w a s used f o r makeup t o t h e nuclear instrument penetration.
Since t h e main losses were by evaporation, t h i s caused t h e
i n h i b i t o r concentration t o increase.
Steam condensate w a s then used f o r
water makeup. Approximately 215 samples were t a k e n of t h e nuclear instrument penet r a t i o n water.
Most of these were analyzed at t h e s i t e t o assure t h a t ade-
quate treatment w a s maintained.
As i n t h e treated water system, t h e i r o n
analysis of samples sent t o t h e laboratory remained near t h e lower detectable l i m i t and t h e chromium content never exceeded 0.4 ppm indicating very
l i t t l e corrosion. 12.7
General Water Systems Conclusions
The water systems were q u i t e adequate and as a whole presented very f e w problems.
The biggest problems resulted from t h e marginal pressure
coâ&#x20AC;&#x2122;
r a t i n g of t h e thermal s h i e l d and piping e r r o r t o one of t h e thermal s h j e l d s l i d e s which resulted i n t h e formation of a l a r g e gas pocket.
The piping
changes required t o prevent over-pressurizing t h e TS and additional equipment needed f o r gas s t r i p p i n g weakened t h e performance of t h e t r e a t e d water system. Corrosion i n h i b i t o r s i n both cooling tower water and t r e a t e d water systems proved t o be excellent i n h i b i t o r s but they had undesirable s i d e effects -the
s c a l e and sludge i n t h e CTW system and induced a c t i v i t y and
tritium formation i n t h e TW system. Equipment and instrumentation performance was very good,
The only
equipment troubles were the t r e a t e d w a t e r heat exchanger and brazed tubing i n t h e c e l l space coolers.
The in-cell water leakage (presumably from t h e
space cooler brazed f i t t i n g s ) , w a s annoying, but afâ&#x20AC;&#x2122;ter i n s t a l l a t i o n of t h e condensate drainage system, did not i n t e r f e r e w i t h performance of the reactor.
u
247
13.
LIQUID WASTE SYSTEM P. H. Harley
13.1 Description The l i q u i d waste system consisted of equipment which handled t h e aqueous wastes from t h e area. Non-hazardous wasts were pumped t o small catch bas$ns and then t o a drainage f i e l d west of t h e area.
This water was from
building foundation drains and various f l o o r drains i n t h e building, over-
SLOW from t h e charcoal bed c e l l , etc.
Hazardous wastes, primarily ,the
leakage i n t o t h e reactor c e l l s , and auxiliary c e l l s were pumped o r j e t t e d t o a waste storage tank f o r sampling and treatment and then transferred t o t h e Melton Valley Waste Disposal System. The waste system design included a component decontamination f a c i l i t y
and a water-shielded c e l l i n which maintenance of radioactive components could be performed.
Some of t h e equipment w a s i n s t a l l e d but as an economy
measure, t h i s section of t h e system was not completed.
13.2
(bi
Preliminary Testinq
Piping and tanks were leak-tested while running performance t e s t s on t h e pumps and j e t e j e c t o r s used t o t r a n s f e r liquids.
The l i q u i d waste
storage tank and c e l l sumps which have bubbler-type l e v e l indicators were calibrated and t h e e j e c t o r s were then t e s t e d t o remove t h e water.
No d f f f i c u l t i e s w e r e encountered w i t h t h e steam-operated ejectors i n t h e auxiliary c e l l s but t h e capa of t h e air-power j e t s i n t h e reactor and drain-tank c e l l was excee formance of t h e e j e c t o r
low. Attempts were made t o improve the ging t h e venturi diameter but no sig-
cant improvement was
pressure of t h e long discharge l i n e
limited t h e j e t capacity. but t o obtain more capacit
-
l i n e s a t t h e Denetration use e i t h e r a i r o r steam j e t s using a i r w a s 25+3
e out o f t h e c e l l , a i
W
if'ying piping i n t h e waste c e l l helped some
steam supply w a s piped t o the j e t supply -tank c e l l .
Provisions were made t o
) f o r t h e j e t supply.
W
and using steam 8.5 d t h e preferred supply.
Although t h e decontamination f a c i l i t y was not completed, t h e waste f i l t e r f o r clarif'ying shielding water w a s i n s t a l l e d and tested.
~
~~
Filtering
248
required pumping 30 g p m through t h e f i l t e r using t h e waste pump but backflushing t h e f i l t e r required %140 gpm.
Process water was used f o r t h i s
operation and t h e flow was c a l i b r a t e d versus t h e process water pressure required (15 p s i = 140 gpm)
.
I n i t i a l t e s t i n g of t h e waste pump f o r c i r c u l a t i n g (mixing) waste tank contents and t r a n s f e r r i n g w a s t e s t e d s a t i s f a c t o r i l y but t h e pump had a tendency t o gas bind a t low tank l e v e l s and process water was required f o r priming These i n i t i a l t e s t s w e r e used t o check and improve operating procedures as well as t e s t t h e equipment.
13.3
Operating Experience
The primary aqueous a c t i v i t y found i n t h e MSRE waste has been 3H. A * continuing water leak i n t o t h e c e l l since e a r l y power operation has been drained i n t o a condensate c o l l e c t i o n tank and t r a n s f e r r e d t o t h e waste tank a t l e a s t once a week. t h e c e l l sump.
The water from t h e i n - c e l l leak did not c o l l e c t i n
The water evaporated i n t h e 135'F
gas i n t h e c e l l and then
c4
condensed out a f t e r being compressed t o 6 p s i g from +2 p s i g and cooled t o
%11O0F i n t h e component cooling system gas cooler. up t o 6 curies/gal.
This water contained
Most of t h e water i n t h e waste tank was from r a i n wa-
t e r which drained from t h e v e n t i l a t i o n stack and f i l t e r e d i n t o a c o l l e c t i o n tank and w a s pumped t o t h e waste tank.
Although t h i s water could have been
contaminated from material on t h e f i l t e r s , very l i t t l e a c t i v i t y was detected.
Other sources of contaminated water i n t h e l i q u i d waste were from
t h e c a u s t i c scrubber during f u e l processing and wash water from cleaning t h e hi-bay f l o o r and a u x i l i a r y c e l l s following r e a c t o r maintenance periods. Table 13.1 shows t h e calculated a c t i v i t y of material t r a n s f e r r e d t o t h e Melton Valley Waste Disposal System.
The 3H a c t i v i t y c o r r e l a t e s t o
reactor pawer f a i r l y w e l l as being one c u r i e of 3H f o r every 45 MWh of power operation. The equipment i n t h e l i q u i d waste system worked q u i t e well.
Some
d i f f i c u l t i e s were encountered i n keeping t h e building sump l e v e l switches
*
See Section 12 on Water System.
W
249
Table 13.1 Liquid Waste Tank Transfers And Total Contaminants
Date
C Tritium
Curies
---
C F P
Vol
Curies
Gal.
0.092
5030
141
1800
10
8850
457 868
5060
_--
7000
0.031
5020
0.005
7440
0.0002
6500 6400
aligned.
These were operated by f l o a t s w i t h %5 f e e t v e r t i c a l tube which
actuates an ON-OFF switch.
Misalignment usually caused t h e f l o a t t o "hang"
and t h e pump would f a i l t o shut o f f on low water level. On several occasions, t h e water l e v e l i n t h e charcoal bed c e l l w a s lowered so t h e i n l e t s t o t h e beds could be heated.
This was accomplished
by draining % l 5 O O gal. of water i n t o t h e sump and allowing t h e two sump pumps t o pump t h e water t o t h e catch basin. During t h e r e a c t o r construction, t h e r e w a s only one building sump pump and it failed causing t h e coolant drain c e l l t o flood through a f l o o r drain. (The CD c e l l and sump room f l o o r are at t h e same elevation, 820 ft.) A s m a l l sump was cut i n t o t h e coolant drain c e l l f l o o r and a float-operated
alarm and steam j e t were i n s t a l l e d . failure of both sump pumps
The system was maintained t o warn of
d provide the steam j e t t o keep t h e sump l e v e l
down during a possible e l e c t r i c a l power f a i l u r e . A second sump pump was ordered t o replace t h e u n i t which was l e f t from
b,
earlier building operations.
Before receiving t h e new pump, t h e o r i g i n a l
250
one was overhauled and t h e new pump was i n s t a l l e d t o provide an operating spare.
The tritium t r a n s f e r r e d through t h i s system w a s %lo% of t h e dis-
charge from ORNL and no effo& w a s made t o l i m i t t h e r a t e of discharge. However, f o r a l a r g e breeder r e a c t o r where l a r g e r q u a n t i t i e s of 3H would be involved, some control of t h e tritium r e l e a s e would be required. If air-operated sump j e t s axe required i n other r e a c t o r s , care should
be used t o insure t h a t t h e e j e c t o r s have s u f f i c i e n t capacity.
251
14.
HELIUM LEAK DETECTOR SYSTEM
J. K. Franzreb
14.1
Description and Method of Operation
A leak detector system was used .to monitor a l l flanges i n t h e MSRE , system which could.permit.the-escape of radioactive materials.
I n addition,
a l l flanges which had t o be maintained by remotely-operated t o o l i n g were provided with leak-detected type j o i n t s t o serve as an indication of s a t i s factory reassembly.
The 119 leak-detected flanges were pressurized with
helium by 65 l i n e s from 8 headers.
As shown on Fig. 14-1, some leak-
detector l i n e s were used f o r more than one flange.
The principles of con-
s t r u c t i o n of a t y p i c a l leak-detected flange closure i s shown on Fig. 14-2. I n normal operation, t h e e n t i r e system w a s interconnected and t h e helium supply was valved off.
Periodic repressurization w a s necessary t o
keep t h e pressure within the l i m i t s of 90 t o 100 psig. leak, helium flowed i n t o t h
b,
I n t h e event of a
affected systems and t h e r e s u l t a n t l o s s i n
pressure i n t h e leak detector headers actuated an alarm i n t h e control room. If t h e records indicated t h a t t h e r a t e of pressure decrease was greater
than t h e 0.66-psi/hr
a r b i t r a r y l i m i t , t h e headers were i s o l a t e d t o narrow
down t h e location of t h e leaking flange.
The leak-detector l i n e s were then
valved o f f individually or i n groups u n t i l t h e location of t h e leak was de-
A reference volume and a d i f f e r e n t i a l pressure t r a n s m i t t e r was provided to aid i n i s o l a t i n g leaks and for determining leak rates. The altermined.
lowable leak r a t e from each flange o r group of flanges was set a t l x
The volumes i n
were determined by pressuriz-
ious p a r t s
unknown q u a n t i t i e s of gas with
an accurately
calculated by using t h e calculated using t sections of t h e system.
td
unknown system volumes were
Cali
given i n Table 14-1.
1 ~ + 1 ~
2
=~P3(v1 2
+
~ 2 ) . Leak
rates were
of pressure decay from these The volumes of various p a r t s of t h e system a r e
252
I
ANY LINE REQUIRING LEAK DETECTOR SERVICE COMPONENT REMOTE
OISCONNECT FLANGE PERMANENTLY INSTALLED PIPE
LEAK DETECTOR LINE
--
I ,
Fig. 14.1
L.
FROM LEAK -C-~OETECTOR
STAT ION
Method of U t i l i z i n g One Leak Detector Line t o Serve Two Flanges i n Series
ORNL DWG 64-8833
N
cn w
\ t Fig. 14.2
Schematic Diagram of Leak-Detected Flange Closure
254 Table 1 4 , l
Volumes i n Leak Detector System
Volume of Line 400 = 184 cc Volume of High Pressure s i d e of DP c e l l = 730 cc Volume of Header Line Line Line Line Line Line Line Line Line Line
401 = 156 cc 410 = 244 411 = 216 412 = 224 413 = 144
414 415 416 417 418 419
= OPEN = OPEN
88 OPEN OPEN
1 4 1 cc
Header 403 = Line 430 = Line 431 = Line 432 = Line 433 = Line 434 = Line 435 = Line 436 = Line 437 = Line 438 = Line 439 = Header 404 Line-440 = Line 441'= Line 442 = Line 4 4 3 = Line 444. = Line 445 = Line 446 = Line 447 = Line 448 = Line 449 =
Line
.
.
NOTE:
Line Line . Line
.
149 cc
I
'
'
117 117
'
108 106
66 63
- 143
-
Header Line Line Line Line Line Line
130
421 = OPEN 422 = 111 4 2 3 ' ~125 424 = 120 425 = 115 426 = 115 427 = 115 428 = 138 429 = 135
123 122 122
195 193 189 178 122 108
405 = 144 450 = OPEN
451 = 113 452 = 108 453 = 106
454 = 455 = 456 = 457 = Line 458-= Line 459 =
= 65
= = = Header 402 = Line 420 = Line Line Line Line Line Line Line Line Line
Volume of Header Line Line Line Line Line Line Line Line
I'
Header Line Line Line Line Line Line Line Line Line Line
406 = 460 =
461 = 462 = 463 =
464 465 466 467 468 469
109 104 109 102
99 98 148 154 99 148 99
= 107 = 108 = OPEN = OPEN
= OPEN = OPEN -
LJ
407'= 150 470 = OPEN 471 = 291 472 = 235
473 = 474 = 475 = 476 = 477 = 478 = 479 =
518 143 135 138 109 305
466 Header 408 = 147 Line 480 = 144 Line 481 = 159 Line 482 = . Line 483 = Line 484 = Line 485 = Line 486 =
123 123 310 210 SPARE L\ne 487 = SPARE Llne 488 = SPARE Line 489 = SPARE
The "B" valves i n t h e leak detector l i n e s remained open during all tests.
U
255
bi
14.3
Operating Experience
During construction and pre-operating leak t e s t i n g , some d i f f i c u l t y was encountered i n a t t a i n i n g acceptable leak rates on some flanges, part i c u l a r l y t h e very l a r g e s a l t f'reeze flanges.
During t h i s and subsequent
maintenance periods, t h e leak detector system proved very valuable i n checking t h e progress and degree of t i g h t n e s s of a l l flanges. Operating experience showed t h a t once a flange was made t i g h t , it stayed t i g h t during service.
During t h e e n t i r e l i f e of t h e MSRE, no leak
ever developed t h a t was serious enough t o cause a shutdown of t h e reactor. Some problem was posed by a f'reeze flange (FT-201) located near t h e point where t h e coolant s a l t l e f t t h e heat exchanger i n t h e r e a c t o r c e l l .
It leaked more than t h e maximum allowable r a t e of 1 x
lom3 cc/sec
was not i n t h e piping and t h e coolant system had been cooled down.
when salt This
leak w a s measured a t about 0.2 cc/sec on Augugt 2 , 1966 while t h e system
was drained and cold.
The flange always became acceptably sealed when t h e
system w a s heated p r i o r t o
u
t i n g molten s a l t i n t o t h e piping and remained
so when t h e r e a c t o r was at power.
The s e a l i n g action was probably due t o
the thermal expansion of t h
t e r n a l s of t h e flange being g r e a t e r than
t h a t of t h e e x t e r n a l clamp.
This would cause an increase i n bearing pres-
sure on t h e s e a t i n g surfaces of t h e metal "0" ring. During normal operation w i t h t h e e n t i r e system interconnected, t h e pressure decreased at about rvals of one t o t
s i per day, thus requiring repressurizing n a 52-day p e r i ystem operation
e r a t u r e i n Octobe
e r 1968, t h e p r
decay rate w a s
I n a 60-d
power operat i o
ptember and Oc-
i/day.
tober 1969, t h e decay r
0.97 psi/day.
November 30, 1968, t h e 1
e did increase t o 2.1
thermal expansion combined
During
system cooldown on sifhr
mal pressure decay p
.
Differential
l y caused t h e
high indicated leak r When an abnormal t o perform
&n
entire s
then each l i n e of t h e
4 t o 8 hours
red, it would onsisted of i s o l a t i n
h header and
t o be leaking.
ng t h e leak
te
r a t e of an individual l i n e took about an hour o r two, hcyever, a rough check t o guide maintenance personnel could be obtained i n about 10 minutes.
1I
256
The individual l i n e s , which monitored more than one flange, gave an i n t e r p r e t a t i o n problem.
It w a s not possible t o t e l l at t h e outset i f all
t h e leak w a s through one flange o r apportioned between multiple flanges on t h e same l i n e .
When t h i s s i t u a t i o n arose, t h e p r a c t i c e became t h a t of al-
t e r n a t e l y tightening one flange, then t h e other, incrementally, and running a combined leak rate on t h e l i n e between these operations.
By observing
when improvement w a s made, the apportionment of t h e leak between flanges could be estimated. It w a s necessary t o exceed t h e maximum recommended torque i n some in-
stances.
The technique was used t o overtorque where necessary t o seat t h e
sealing faces, then t o loosen- t h e b o l t s and retorque them t o t h e nominal o r , i f necessary, t o t h e maximum value l i s t e d i n Table 14.2. ceptional cases, it was necesiary t o exceed t h i s maximum.
I n a few ex-
Some of t h e
m e t a l O-rings had t o be replaced w i t h gold-plated O-rings i n order t o ob-
t a i n acceptable leak rates.
Some o f t h e 1/2-in. b o l t s i n small flanges
were torqued as high as 150 ft-lbs and operated at about 80 ft-lbs.
The
maximum recommended torque was 70 ft-lbs.
LfiJ
The record of measurements of system and flange leak rates w a s kept
i n a "Leak Detector Log Book" at t h e LKD control cabinet.
Flange leak
rates, plus a record of all work done involving any of these flanges, was logged i n the "MSRE Flange Log" notebook. each flange on a separate page. t e c t o r Valve No.,
This l a t t e r l o g gave d e t a i l s on
Information given w a s Flange No.
, leak
de-
s i z e of b o l t s ( o r i n t h e case of freeze flanges, t h e
force exerted by t h e removable c l i p i n t o n s ) , r i n g s i z e , t h e maximum permissible and recommended torque i n ft-lbs.,
t h e location of t h e leak de-
t e c t e d j o i n t , plus t h e date of action, t h e reason f o r t h e a c t i o n , a c t u a l torque applied, and any general comments.
14.4 Discussion and Recommendations The leak detector system functioned s a t i s f a c t o r i l y as designed.
Due
t o t h e two s e a l i n g surfaces on t h e "0" r i n g s , primary containment w a s maintained as long as t h e leak detector system overpressure w a s g r e a t e r than any possible pressure that could develop i n t h e primary system.
Therefore,
leak r a t e l i m i t a t i o n s were s e t based on convenience i n maintaining an adequate overpressure.
257
Table 14.2
Bolt Diameter
1-112
Threads Per inch
MSRE Flange Bolt Torque Chart
Nominal Torque ( f t - l b .
Maximum Torque ( ft-lb. )
13
47
70
11
90
135
10
150 240
225
533
800 1200
9 7 6
900
360
Wrench Size
314" 718" 1-1/ 8" 1--5/16*~ 1-112" 2- 318"
Tightening Procedure:
1. Torque b o l t s t o nominal torque value following a c r i s s c r o s s sequence. Approach f i n a l value i n successive increments of 10 t o 20 f t - l b . 2. If leak e x i s t , torque b o l t s t o maximum torque value, back o f f and retorque t o nominal value per s t e p 1.
3. Repea't s t e p 2 as required, increasing nominal value by 10% each successive s t e p , NEVER EXCEED MAXIMUM VALUES without permission of t h e MSRE Operations o r Maintenance Chief.
258
U Remote welding o r brazing techniques should be used where maintenance i s infrequent.
This should be more economical and would reduce t h e number
of d e t a i l s t o be handled by t h e operating personnel. When opening primary system flanges, it i s desirable t o have t h e leak detector vented t o atmospheric pressure t o minimize t h e spread of a c t i v i t y . Each header should have a l i n e which i s connected t o t h e c e l l t o f a c i l i t a t e doing t h i s since t h e gas i n t h e l i n e could be contaminated. ,
259
15.
INSTRUMENT AIR SYSTEM T. L. Hudson
15.1 Description The instrument air system supplied clean, dry, compressed air f o r pneumatic instruments and other s p e c i a l uses.
Two Joy compressors, AC-1
and AC-2, were used t o compress t h e a i r , which then passed through an a f t e r cooler and entrainment separator t o a common l i n e supplying two p a r a l l e l receiving tanks.
From t h e receiving tanks, t h e a i r passed through one of
two p a r a l l e l drying s t a t i o n s containing T r i n i t y h e a t i e s s dryers.
The dry a i r was d i s t r i b u t e d through headers t o l o c a l l y mounted f i l t e r and reducing
stations.
The service a i r compressor, AC-3,
could be valved i n t o t h e in-
strument a i r system upstream of t h e receiver tanks i f required. 15.2
Operating Experience
The capacity of t h e instrument a i r system w i t h one a i r compressor i n operation w a s adequate at all times. O f t h i s amount,
Normal a i r usage was about 70 scfm.
18 scfm was used f o r purging t h e instrument a i r dryers,
12 scfm f o r supplying n o n c r i t i c a l instruments, and 40 scfm f o r supplying t h e c r i t i c a l instruments.
The c r i t i c a l instruments were supplied by t h e
emergency instrument a i r supply, which consisted of two banks of s i x nitrogen cylinders, when t h e normal instrument a i r supply w a s l o s t . The capacity of t h e emergency instrument a i r supply w a s t e s t e d during
normal operation with f l u s h s a l t t o assure an orderly shutdown of t h e rea c t o r i n case of loss of both instrument a i r compressors (see MSRE Test Report 2.3.11.3).
With an emergency a i r usage of 40 scfm and one bank of
cylinders at t h e i r lowest operating, 1500 p s i g ( i n i t i a l low-pressure alarm point) and t h e other bank fW.1 (52000 p s i g ) a t t h e time both a i r compressors stop would.provi.de t h e following times: (1) A i r i n t h e compress
surge tank would l a s t 5 4 min.
( 2 ) F i r s t bank of nitrogen cylinder ( i f used t o TOO p s i g - a t
this
pressure t h e control valve stopped controlling) would l a s t 513 min,
u
( 3 ) Subsequent f u l l banks of nitrogen cylinder ( i f used t o 700 p s i g ) would l a s t 520 min.
260
From t h e test it w a s concluded t h a t t h e emergency instrument a i r system
was adequate.
t)
It would have been necessary t o i n s t a l l a l a r g e r control
valve t o more f u l l y u t i l i z e t h e nitrogen cylinders. Leaks i n t h e emergency nitrogen headers were repaired several times and t h e control valve w a s repaired once.
On t h e average one bank of nitro-
gen w a s changed out every two weeks t o keep t h e cylinders above t h e i n i t i a l
alarm pressure of 1500 psig.
15.3, Conclusions The operation of t h e normal instrument a i r system was very satisfactory.
A t no time w a s it necessary t o use t h e emergency nitrogen system
during operation of t h e reactor.
One instrument air compressor was oper-
ated with t h e other i n standby except f o r during r e p a i r s o r maintenance. Even w i t h a complete power outage, it was possible t o s t & t one air corn-
,
pressor on diesel power before t h e receiver tank pressure dropped enough t o cause a t r a n s f e r t o the emergency system.
Although some extensive main-
tenance was done on t h e a i r compressors, only one was ever inoperative at a time.
LJ
261
6,
16.
ELECTRICAL SYSTEM-*
.-
T. L. Hudson
16.1 Description .
The MSRE e l e c t r i c a l power was supplied from t h e ORNL substation by e i t h e r of two 13.8-kV TVA l i n e s , a preferred l i n e o r an a l t e r n a t e .
These
were separated by two interlocked. pole-mounted motbr-operated l i n e switches. .-Automatic t r a n s f e r w a s porvided from t h e preferred l i n e t o t h e a l t e r n a t e Both t r a n s f e r switches could be manually operated remotely from t h e
line.
a u x i l i a r y control room.
.
The ac power entered the MSRE building through two transformer subA 1500-kVa, 480-V, %phase substation served t h e process equip-
stations.
ment, and a 750-kVa, 480-V, >phase a u x i l i a r y s t a t i o n w a s i n i t i a l l y f o r building services, such as l i g h t i n g , v e n t i l a t i o n , e t c .
After t h e process
substation became overloaded, bus 5 which supplies e l e c t r i c a l heaters w a s connected t o t h e a u x i l i a r y system.
&,
Three diesel-generators were i n s t a l l e d
f o r emergency ac use. There were two separate area dc systems, a 48-v system and a 250-V system.
These were normally operated by ac-dc motor-generator sets and
had b a t t e r y supplies f o r emergency use. The 48-v dc system provided power f o r e l e c t r i c a l controls and one channel of t h e r e a c t o r s a f e t y systems.
The system c o n s i s t s of two 3-kW motor-generator sets (each r a t e d at 53 ampere at 56 V ) and a bank of 48-V b a t t e r i e s (600 amp-hr).
The two generators normally supplied t h e dc power
and charged t h e b a t t e r i e s .
I n an emergency, when no ac power was available
t o run t h e motor-generators
8-V dc power was supplied d i r e c t l y from t h e
batteries.
The 250-V dc system provided power t o t h e r e l i a b l e power system, process breakers t r i p power, 13.8-kV t r a n s f e r control power, and emergency lighthg.
The system consisted
f one 125-kW motor-generator s e t and a
bank of 240-V b a t t e r i e s (364 amp-hr). The r e l i a b l e power sys
b
provided power t o c e r t a i n important process
equipment and one channel of t h e s a f e t y systems.
I n i t i a l l y , t h e power was
supplied from a 25-kW dc-ac motor-generator s e t .
The dc motor w a s supplied
262
from t h e 250-V dc system and t h e generator supplied t h e 120/240-V ac rel i a b l e power system.
The operation of t h e MG s e t was unreliable because
of failures i n e l e c t r i c a l control components and was replaced w i t h a s t a t i c i n v e r t e r (see 16.4). Thk MSRE e l e c t r i c a l i n s t a l l a t i o n made extensive use of t h e e l e c t r i c a l
f a c i l i t i e s t h a t were i n s t a l l e d f o r t h e ARE and ART operations i n t h e 7503
area.
Many modifications-were made t o t h e e x i s t i n g equipment, but only a
l i m i t e d ambunt of supplemental equipment was required. 16.2 .
~
Alternating Current System
The operation of t h e ac e l e c t r i c a l system w a s very s a t i s f a c t o r y .
There
were eleven important failures, t e n of which caused unscheduled interrupt i o n s - i n t h e power operation of t h e reactor:. f i v e were-caused by e l e c t r i c a l storms, one by a cable f a i l u r e at a component-cooling pump, one by an overload of t h e main process-power breaker, one by an arc between a 13.8-kV l i n e and an a c t i v a t o r rod t o t h e l i n e fuse, one by a f a i l u r e of t h e main transformer primary fuse, one by a f a i l u r e of t h e a u x i l i a r y transformer primary fuse, and one by a failure o f t h e drain-tank space cooler motor.
These
interruptions varied i n - l e n g t h from a f e w minutes t o s e v e r a l days.
L.i
In a l l
cases t h e necessary r e p a i r s o r modifications were made, and no damage t o reactor equipment was incurred.
These a r e discussed i n more d e t a i l below.
On two occasions while t h e r e a c t o r w a s operating at power, momentary e l e c t r i c a l outages during storms caused control-rod scrams by rangeswitching t h e nuclear power s a f e t y channels because of dips i n t h e fuelpump-motor current.
I n both these cases t h e nuclear power w a s quickly re-
stored t o t h e value t h a t e x i s t e d j u s t p r i o r t o t h e outage.
Since t h e
s a f e t y analysis indicated t h a t l o w range of t h e s a f e t y channels w a s needed only while f i l l i n g t h e r e a c t o r (when t h e f u e l pump was o f f ) , r a p i d response of t h e range-switching was not required.
Therefore, time-delay r e l a y s were
subsequently incorporated i n these c i r c u i t s t o prevent t h e i r a c t i v a t i o n on momentary power dips. Another storm-induced f a i l u r e occurred when l i g h t n i n g s t r u c k t h e main
ORNL power substation and parted one wire of t h e normal MSRE feeder. I n t h i s case t h e e l e c t r i c a l load was automatically t r a n s f e r r e d t o an a l t e r n a t e
LJ
263 feeder, and all e s s e n t i a l equipment was r e s t a r t e d i n time- t o prevent draining either t h e f u e l o r coolant system. However, operation at high nuclear power could not be resumed u n t i l service was restored on t h e main feeder. The e n t i r e 13.8-kV supply system was subsequently reviewed and improved t o make it l e s s susceptible t o damage by storms. During a thunderstorm i n June 1967, t h e reactor operation was i n t e r rupted by t h e loss of both feeders.
The r e a c t o r period s a f e t y system func-
tioned properly and scrammed t h e r e a c t o r but two amplifiers were damaged. Approximately 32 hours l a t e r , t h e reactor was returned t o c r i t i c a l operation a f t e r t h e period s a f e t y amplifiers had been repaired.
.
I n J u l y 1967, an interruption w a s caused by t h e loss of t h e preferred feeder during another thunderstorm.
Emergency power from t h e d i e s e l gener-
a t o r s was i n service within 2 min, and low-power nuclear operation was resumed i n about 13 min.
After r e p a i r s had been completed on t h e preferred
feeder, full-power operation was resumed i n approximately 6 hr. The short i n the component-cooling pump cable s e a l i s described i n Sect. 10 e n t i t l e d "Component Cooling System."
bi
When t h i s short occurred,
t h e r e was a massive flow of current and t h e breaker supplying t h e e n t i r e bus t r i p p e d before t h e individual breaker f o r t h e motor. During t h e i n i t i a l full-power operation of t h e r e a c t o r , t h e main process-power breaker (breaker R) was loaded t o near i t s s e t p o i n t value. When the load on t h e coolant-radiator main-blower motors was increased by increasing the p i t c h on t h e blower fan blades, the.increased load on breaker R caused it t o t r i p
overcurrent.
This condition was relieved
by t r a n s f e r r i n g bus 5 (%200
) from t h e main process transformer t o t h e
a u x i l i a r y power transformer.
The load transfer decreased t h e current
through breaker R *om
about 1700 -/phase
A building power failur when an a r c developed betwee
t o about 1550.
ccurred on a foggy morning i n October 1967, 13.8-kV l i n e t o t h e main transformer f o r
t h e building and p a r a l l e l metal a c t i v a t o r rod t o t h e 13.8-kV l i n e fuse. ined.
was not unusually wet.
ult caused the, operation of t h e preferred
Th
feeder overcurrent ground
u
Although t h e weather was foggy, it
The cause o f t h e a r c i s un
located a t t h e ORNL substation, which
t r i p p e d t h e feeder breaker before t h e l i n e overcurrent r e l a y located at t h e MSRE could operate t o prevent an automatic t r a n s f e r t o t h e a l t e r n a t e feeder.
264
When t h e MSRE was t r a n s f e r r e d , t h e a l t e r n a t e feeder overcurrent .ground re. l a y operated and t h e a l t e r n a t e feeder breaker a l s o tripped, (To prevent
W
an automatic t r a n s f e r t o t h e a l t e r n a t e feeder on a similar f a u l t , an overcurrent ground r e l a y was later i n s t a l l e d at t h e MSRE.)
A f t e r an interrup-
t i o n of 37 min, low-power nuclear operation was resumed on emergency elect r i c a l power from t h e d i e s e l generators.
The-damage w a s repaired, t h e
spacing between t h e l i n e and t h e a c t i v a t o r rod wa$ increased, and normal service w a s restored after an i n t e r r u p t i o n of 2 h r . I n April 1969, t h e r e was a failure of a fuse on t h e primary side of t h e MSFtE main power transformer.
This r e s u l t e d i n t h e loss of voltage on
one of t h e transformer windings.
The power supply was changed from a three-
phase t o a single-phase supply.
This caused a drop i n secondary voltage
across t w o windings and i n c r e a s e - i n voltage across t h e other winding.
.
The
motor-operated equipment continued .to operate u n t i l t h e individual breakers were t r i p p e d by t h e high current i n two of the l i n e s ,
A l l breakers did not
t r i p at t h e same time which made it d i f f i c u l t t o l o c a t e t h e trouble immediately.
The reactor w a s scrammed from f u l l power and t h e d i e s e l gener-
a t o r s were s t a r t e d , but both the fuel and coolant systems were drained .*
before cooling air could be restored t o t h e freeze valves.
No apparent
cause f o r t h e fuse failure w a s found other than possible overheating from poor e l e c t r i c a l contacts. About t h e middle of February 1966, a ground developed i n t h e drain-tank c e l l space cooler 10-hp motor af'ter abaut 2900 h r of operation,
An iso-
l a t i o n transformer w a s i n s t a l l e d t o i s o l a t e t h e ground from t h e e l e c t r i c a l A f t e r a f e w hours of operation, t h e motor stopped and couldn't be
system.
restarted. drained.
The r e a c t o r was taken s u b c r i t i c a l and t h e f u e l system w a s The drain-tank c e l l was opened t o provide c e l l cooling.
The
cooler and motor were removed, and examination showed that t h e r o t o r had slipped along t h e shaf't u n t i l t h e r o t o r fan blades rubbed, causing t h e \
s t a t o r windings t o be destroyed.
The f a i l u r e appeared t o be due t o im-
proper manufacture and not a design weakness.
Therefore, t h e damaged motor
was replaced w i t h one p r a c t i c a l l y i d e n t i c a l . On one occasion t h e a u x i l i a r y power systemwas out of s e r v i c e f o r ab9ut 2 h r s due t o a failure of t h e a u x i l i a r y power transformer primary fuse.
During t h e fuse replacement t h e operation o f , t h e r e a c t o r continued
bd
265
at power with bus 5 supplied from diesel-generator 5.
The fuse f a i l u r e
apparently w a s caused by overheating from poor e l e c t r i c a l contact.
Due t o
low r e s i s t a n c e readings , t h e t h r e e a u x i l i a r y power t r a n s f e r s were replaced i n 1968.
16.3
250-V dc System
The operation of t h i s system has been very s a t i s f a c t o r y .
There have
been some tube f a i l u r e s i n t h e automatic voltage control system f o r t h e generator, but each time t h e u n i t was operated on manual control u n t i l t h e tubes i n t h e automatic controls could be replaced. Once a month t h e charging voltage w a s increased t o 275 V f o r 24 h r s t o give t h e b a t t e r i e s an equalizing charge. The b a t t e r i e s were given a 3-hr l i f e t e s t under full load.
tage dropped t o 215 v o l t s over t h e three-hour period.
The vol-
Extrapolating t h e
voltage curve indicated t h a t t h e lower l i m i t of 210-V would have been reached i n 'L3.5 hours which i s well above t h e 2-hr l i f e upon which t h e design was based.
16.4 Reliable Power System The 120-V ac r e l i a b l e power system was o r i g i n a l l y supplied by an MG set.
I n e a r l y operation, t h i s MG s e t ( o r i g i n a l l y i n s t a l l e d f o r t h e ARE)
proved very unreliable due t o failures i n t h e e l e c t r i c a l c o n t r o l components. To improve t h e r e l i a b i l i t y and
w a s replaced with a new 6 on t h e r e l i a b l e power sup some three-phase load, was
ncrease t h e capacity, t h e motor generator three-phase s t a t i c i n v e r t e r .
The t o t a l load
including t h e on-line computer, which has Va.
The s t a t i c i n v e r t e r offered a number
of advantages over t h e old r
i n v e r t e r , including higher e f f i c i e n c y
(increased 35$), smaller s i z
moving p a r t s , q u i e t operation, and ex-
c e l l e n t voltage and frequenc t h i s equipment, both t h e c
ation.
P r i o r t o t h e i n s t a l l a t i o n of
.
d voltage regulation of t h e r e l i a b l e
power supply were inadequate f o r t h e operation of t h e on-line computer. During t h e checkout and t e s t i n g o f t h e s t a t i c i n v e r t e r i n April 1966 a f t e r t h e i n s t a l l a t i o n had been completed, trouble occasionally developed
266
t h a t blew t h e load fuses.
Two times t h e manufacturer's f i e l d engineer made
u
exhaustive t e s t s and could not f i n d any trouble, but a l l symptoms indicated t h a t t h e trouble was i n t h e low-voltage l o g i c power supply, which supplies
24-V dc control power,
On April 25, t h e i n v e r t e r failed again while the
reactor w a s operating at l o w power.
Apparently t h e t r a n s f e r caused a tran-
- s i e n t on t h e ac system, thus a l s o dropping out t h e rod scram r e l a y supplied from t h e ac system, causing a control rod scram.
The i n v e r t e r load was
automatically t r a n s f e r r e d t o t h e normal ac power supply.
After t h i s f a i l u r e
a new power-supply module w a s i n s t a l l e d . The i n v e r t e r output voltage regulation w a s 208 f 1 / 2 V, and t h e output frequency b e t t e r than 60 cps f 0.01%.
The i n v e r t e r was tested f o r 2-1/4 h r
operating from t h e 250-V b a t t e r y system without t h e dc generator operating. Although t h e input t o t h e i n v e r t e r varied 20 V, t h e output varied only
0.5 V. There have been t h r e e other failures associated with t h e s t a t i c inv e r t e r t h a t have automatically t r a n s f e r r e d t h e normal i n v e r t e r load t o TVA The first was caused by failure of load fuses which r e s u l t e d i n an
supply.
unscheduled control-rod scram.
The second w a s caused by a f a i l u r e of a
t h y r i s t e r i n the inverter logic circuit.
d3
The t h i r d w a s caused by a momen-
t a r y ground during maintenance on t h e computer. When a t r a n s f e r occurred, many alarms were received i n t h e control room.
Some were caused by l o s s of operating equipment, and others were These
due t o t h e momentary l o s s of control voltage t o c e r t a i n monitors. had t o be r e s e t t o stop t h e alarm and control action.
After t h e second
i n v e r t e r f a i l u r e , all operating equipment was r e s t a r t e d , but an emergency f u e l drain occurred before t h e reactor-cell a i r a c t i v i t y monitors were reset.
No drain o r other serious condition r e s u l t e d from t h e other i n v e r t e r
failures
16.5 48-V dc System The
48-V dc system w4s very r e l i a b l e .
The only maintenance on t h e
system w a s caused by excessive arcing of t h e generator brushes.
Initially
one generator was operated near r a t e d capacity t o supply t h e load,
Filters
and viewing windows were i n s t a l l e d on t h e generators and both generators
LJ
267
were operated i n p a r a l l e l t o reduce t h e individual generator current a f t e r t h e load s l i g h t l y exceeded t h e capacity of one generator.
These changes
g r e a t l y reduced t h e generator brush problems Once a month, t h e charging voltage was increased t o 54 v o l t s f o r ~8 h r t o give t h e b a t t e r i e s an equalizing charge. The b a t t e r i e s were given a 6-hr l i f e t e s t .
47.5 amps and t h e l a s t 2.5
h r at 19.5 amps.
The first 3.5 h r was at
The b a t t e r i e s expended 215
amp-hour of t h e 600 amp-hour l i f e with e s s e n t i a l l y no voltage drop.
16.6 Diesel Generators The t h r e e diesel-power generators, each with a capacity of 300 kW, supplied emergency ac power t o motors and heaters i n t h e MSRE. t o be q u i t e r e l i a b l e .
They were s t a r t e d and operated unloaded f o r about an
Once a month they were t e s t e d under load.
hour each week.
They proved
f a i l e d t o start when required during a power outage.
They never
Under emergency con-
d i t i o n s , they were usually running within 2 min a f t e r a f a i l u r e of t h e norm a l power supply.
I n February 1966, a crack w a s found i n t h e block of D E 3 a t one of Attempts a t r e p a i r s were not successful, but t h e u n i t was
t h e b o l t holes. operable.
Later D E 3 w a s replaced with a surplus d i e s e l generator from
another f a c i l i t y .
The replacement unit was similar t o t h e o r i g i n a l except
t h a t it w a s s t a r t e d by compressed a i r instead of by an e l e c t r i c starter.
16.7
100-kVa Variable Frequency Motor-Gen r a t o r Set
A temporary i n s t a l l a t i o n of a used variable frequency MG s e t w a s made
at t h e MSRE t o supply t h e f'uel-pump motor during s p e c i a l t e s t s .
The fuel-
pump motor was operated from 610 rpm t o 1165 rpm (maximum speed with MG
s e t ) over a period of several weeks. The automatic controls f o r t h e variable speed and generator voltage did not work properly durin load.
e i n i t i a l checkout using a dummy resistance
After replacement of some'defective components i n t h e control c i r -
c u i t s , t h e f u e l pump was operated from t h e MG set.
bi*,
The operation of t h e
'automatic controls were s t i l l not completely s a t i s f a c t o r y and some manual
268
adjustments were required.
W
Frequently t h e fuel-pump motor was stopped be-
cause of over adjustment of t h e manual controls.
.,
Also during t h e lower-
speed operation, t h e speed of t h e f u e l pump w a s not constant enough t o make t h e desired t e s t s .
Therefore, t h e automatic controls were bypassed and
manual controls of speed and voltage were i n s t a l l e d ,
-
Satisfactory ope=?
t i o n w i t h manual controls was not obtained u n t i l t h e f u e l pump was operated
at about 35 V above rated voltage, assuming t h e fuel-pump motor voltage i s d i r e c t l y proportional t o speed.
16.8 Conclusion Y
Considering t h e age and the number of modifications made t o t h e existing e l e c t r i c a l system within t h e E R E building, t h e performance has been very satisfactory.
The main weakness of t h e e n t i r e e l e c t r i c a l system was
with t h e 13.8-kV feeders and transformers supplying t h e MSRE.
equipment w a s over 15 years old.
I
Most of t h e
During 1966, t h e r e were t h r e e momentary
interruptions of t h e power t o t h e MSRE caused by thunderstorms.
The e n t i r e
13-8-kV supply system was reviewed and improved t o m a k e it less susceptible t o damage by storms.
Since t h e improvements were made, there have been
only two interruptions due tâ&#x20AC;&#x2122;o storms i n t h e l a s t t h r e e years of operation. ! I
ie*s
269
17.
HEATERS
T. L. Hudson
17.1
Description
E l e c t r i c a l heaters were provided f o r a l l p a r t s of t h e salt system t o permit preheating of t h e c i r c u l a t i n g loops and t o maintain t h e temperature (In practice the of t h e salt at 1200'F during zero-power operation. heaters were kept energized during power operation as well.)
The t o t a l in-
s t a l l e d capacity was %go0 kW; t h e a c t u a l power requirement w a s less than h a l f t h i s amount.
It w a s possible t o supply a l l heaters e s s e n t i a l t o an
orderly shutdown, including enough t o keep thy s a l t molten i n t h e drain tanks, from a 300-kW diesel-driven generator.
A summary of t h e heaters
and i n s u l a t i o n on t h e salt systems i s given i n Table 17.1. The type of i n s t a l l a t i o n was determined by t h e location.
I n t h e re-
actor and drain-tank c e l l s , radiation from long-lived f i s s i o n products pre-
cd
cluded d i r e c t maintenance a f t e r start of power operation.
There, t h e heat-
ers were designed for removal and replacement using long-handled t o o l s inserted through t h e c e l l roof.
I n t h e coolant c e l l , t h e piping could be
approached within a few minutes a f t e r a shutdown, and t h e heaters and insul a t i o n were t h e conventional type used i n component t e s t f a c i l i t i e s . Removable heater u n i t s i n the reactor and drain-tank c e l l s were of two types.
On t h e piping and heat exchanger, t h e heaters were nichrome ele-
me5ts i n ceramic p l a t e s , located on t h e inner surfaces of t h e removable u n i t s t h a t a l s o insulated t h e s i d e s and t o p of t h e equipment.
On t h e re-
a c t o r vessel, f u e l pump, and d r a i n tanks, t h e removable heaters were ins e r t e d i n t o penetrations i n permanently mounted insulation. All e l e c t r i c a l i n s u l a t i o n w a s ceramic type,
ecause of t h e high gamma-radiation f i e l d s (up
t o lo5 R/h i n t h e reactor c e l l during operation). Thermal i n s u l a t i o n i n t h e removable heater-insulation u n i t s was of t h e metallic, multiple-layer, without dusting.
r e f l e c t i v e type, which could withstand handling
The permanently mounted i n s u l a t i o n w a s low-conductivity,
high temperature low sullfur content,
ceramic fiber of expanded s i l i c a ,
with a t h i n metal sheet between t h e i n s u l a t i o n and t h e pipe surfaces (see
ORNL Drawing E-MM-â&#x201A;Ź3-51676).
r
.
_.
~
.
_ . _" _
- - ."
..
.
....
. . . . . . . . . . . . . I
"
.. .
.. .~ -.. .. .
..
.
,
.
...
.-
.- .
.. .
__
-. . . .. ..
.
- .
~
Table 17.1
Insulation
Number Heater Units
Elements
Removable
Permanent'
9
9
1135 ft of 3/8-in. by 0.03Fin. w a l l Inconel tubing. Each removable unit consists of 7 U-tubes.
3
I
Remewsble
Permanent'
5
14
3/b-in.diam, stainless s t e e l sheath 5-it long with ceramic insulation and nichrome elements.
2
R
Removable
Removable
31
177
Ceramic plate with nichrome elements.
23
19
Permanent
Permanent'
3
8
Tubular 0 . 3 1 5 4 ~OD Inconel sheath With nichrome elements.
3
I
Heat Exchanger
Removable
Removableb
3
24
Ceramic plate with nichrome elements.
3
R
Freeze Valve
Removable
Removable'
1
1
Tubular 0.3l5-in. OD Inconel sheath With nichrame elements.
1
I
Reactor Beck
Permanent
Permanentu
2
2
Tubular 0.3l5-in. OD Inconel sheath With nichrome elements.
2
11
Drain Tanks
Remvable
Permanent'
23
156
Ceramic plates with nlchromc elements
6
nm
l-l/P-in.-diam Piping ( F i l l Lines )
Removable
Removable
13
75
Ceramic plates with nichrome elements
15
R&E
m e
Location Reactor C e l l
Heaters and Thermal Insulation
System Reactor
-1
pump
fi-in.-dIsm Piping
Typc
Thermal
Heaters
Number Heater
Number controls
Type Heater Elementsd
Powere Supply
h)
Draln Cell
'Ceramic fiber of expanded silica. bMetallic, multiple-layer. d&!eramlc plates maximum i n s t d l w a t t a g e 30 watts/in2 of surface. H' normal parer supply; E emergency power supply.
-
c
-
-
-
c
TubuSar 0.315-in.
OD
-J
0
- normal rating 500 watts/ft of heater. ,
c.
.
,
-
._.
.,
...
d
8'
Table 17.1
Heaters
Type Thermal Insulation
Freeze Valves (Fill Lines)
Removable
Removableb
3
F i l l Line
See N o t e c
Permanent'
1
Type
Location
Drain C e l l
Coolant C e l l
System
Number Heater Units
(Continued)
Number Heater Elements 27 1
d
Type Beater Elements
9
N&E
Direct resistance heating of lc1/2-in. by 0.195-in. wd.1 INOR-8 pipe 65 rt long.
1
H&E
12
E&E
Ceramic p l a t e s with nichrome elements
6
H&E
14
Ceramic plates with nichrome elements
2
H
10
87
Ceramic plates and tubular 0.315-in. OD Inconel sheath with nichrome elements.
10
N
Permanent'
8
162
Ceramic plates and tubular 0 . 3 1 F i n . OD Inconel sheath with nichrome element 8 .
8
N&E
a Permanent
3
24
Ceramic plates and tubular 0.315-in. OD Inconel sheath with nichrome elements
3
N&E
Permanent'
12
32
Tubular 0.315-in. OD Inconel sheath with nichrone elements
Freeze V a l v e s (Transfer l i n e s )
Permanent
Permanent'
3
32
Coolant Pump
Permanent
a Permanent
2
Permanent
Permanent'
Radiator
Permanent
Drain Tank
Permanent
Piping
Powere Supply
Ceramic p l a t e s with nichrame elements
Permanent 1/2-in.-diam iping (Transfer ines )
>in.-diam
Hmber Controls
.
'Ceramic
f i b e r of expanded s i l i c a .
bMetallic, multiple-layer. 'The fill l i n e was heated by passing current through t h e wall of t h e pipe. The pipe was removable. normal r a t i n g 500 w a t t s / r t of heater. dCeranric plates maximum i n s t a l l wattage 30 watts/in2 of surface. Tubular 0.315-in. OD
-
N' -normal power supply; E
-
-
emergency pmer supply.
-
-
.......
..........
..
. . . . . . . . . . . . . . . . . . .
Table 17.1
Type
Location ~-
System ~~
Coolant C e l l (continued)
Heaters
Type Thermal Insulation
Number Heater Units
..j
......................
(Continued)
Number Heater Elements
~
Number Controls
Powere Supply
Ceramic p l a t e s with nichrome elements.
4
NW
Ceramic plates ana tubular 0.35-in.
6
A m
Type Heater Elementsd ~
Freeze Valves
Permanent
Permanent'
2
8
F i l l Lines
Permanent
Permanent'
6
30
~
~ _ _ _
~~
~~
OD Inconel sheath With nichromc elements.
Flow Transmitters
Permanent
Permanent'
4
8
Tubular 0,315411. OD Inconel sheath with nichrome elements.
8
N
Level Element
Permanent
Permanenta
1
4
Tubular 0.315-in. OD Inconel sheath with nichrome elements
2
n h)
4
'Ceramic 'Ceramic e A
fiber of expanded. s i l i c a . plates d m u m i n s t a l l wattage
-
h)
-
30 watts/inz of surface. Tubular 0.315-in.
- normal power supply; E - emergency power supply.
OD
-
normal r a t i n g 500 watts/f't
of heater.
273
Maintenance requirements were minimized by designing t h e heaters w i t h excess capacity and operating them a t reduced voltage.
If some of t h e
h e a t e r elements f a i l e d , t h e other elements could be operated a t a higher voltage o r t h e power could be increased t o t h e adjacent heaters.
Heater
u n i t s t h a t were less accessible had spare heater elements i n s t a l l e d and connected, ready f o r use w i t h only minor, out-of-cell wiring changes. Control of t h e e l e c t r i c a l input t o t h e heaters w a s e n t i r e l y manual, i n response t o system temperatures displayed -on e i t h e r a multipoint tempera-
ture recorder o r a cathode-ray tube by a multipoint scanner.
A t l e a s t one
thermocouple was attached t o the piping o r equipment under each h e a t e r u n i t , with additional couples at p o t e n t i a l cold spots.
17.2 Preoperational Checkout To assure t h a t t h e heaters would perform properly and t o provide reference data f o r future trouble-shooting, and after i n s t acl l a t i o n .
h,
a number of t e s t s were done before
A v i s u a l examination of t h e heaters and t h e wiring
i n t h e junction boxes was made and all obvious e r r o r s were corrected.
Con-
t i n u i t y checks were completed and measurements of t h e h e a t e r c i r c u i t res i s t a n c e s and resistances t o ground were recorded. *
Each h e a t e r c i r c u i t was
then energized individually t o assure t h a t t h e proper thermocouple o r thermocouples responded. The e f f e c t i v e resistance of each h e a t e r c i r c u i t w a s calculated and
these were compared with measured r e s i s t a n c e of t h e heaters before and aft e r being i n s t a l l e d i n t h e c e l l .
Several of t h e r e s i s t a n c e measurements
were made using a volt-ohm-meter but t h e accuracy of these readings was usua l l y not good enough.
The f i n a l r e s i s t a n c e measurements were made with a
Wheatstone bridge and were recorded t o t h e nearest one-hundredth ohm. Table 17.2 i s a summa
f r e p a i r s made t o t h e heaters during t h e pre-
operational checkout.
17.3 System Heatup The heatup of t h e coolant, f u e l , and drain tank systems was s t a r t e d as
b
soon as construction and checkout of t h e systems were completed.
During
t h e first heatup, a l l heaters operated s a t i s f a c t o r i l y and t h e r e were no
274
Table l7,2
Summary of Preoperational Heater Checkout
Repair o r Replacement O f Heaters
Reactor Cell
Corrected wiring Replaced open heater element
'
Drain Cell Coolant C e l l
4
3
--
3
Replaced broken ceramic beads on removable unit I
9
7
yo&â&#x201A;Źf i e d removable heater box to eliminate possible short c i r c u i t i n ceramic bead insulated leads
3
16
Modified heat box t o permit remote maintenance
1
4
Replaced broke M I cable end seal
2
2
Replaced 1 MI cable
--
1
Does not apply t o coolant c e l l ,
c
275
b,
heater failures, but t h e r e were a few areas where t h e temperature w a s low because of a i r leakage and lack of thermal insulation.
These areas were
t h e i n l e t t o t h e fuel-pump furnace, north end of heat exchanger, t r a n s f e r l i n e entrance t o t h e drain-tank furnace, and t h e coolant r a d i a t o r .
After
t h e a i r leakage was reduced and thermal i n s u l a t i o n i n s t a l l e d , s a t i s f a c t o r y temperatures were obtained with t h e heaters as i n s t a l l e d .
The voltage s e t -
t i n g required f o r each heater w a s determined during t h e s t a r t u p t e s t i n g and stops were s e t t o l i m i t t h e controls t o hold t h e system temperatures t o 1300°F o r below.
The general heatup o f t h e systems and t h e d i f f i c u l t i e s
a r e discussed i n t h e following t h r e e sections,
17.3.1
Coolant and Coolant Drain.-Tank Systems Heatup..
The i n i t i a l heatup of t h e coolant system i n . t h e coolant c e l l and.coo1-
ant , d r a i n c e l l was s t a r t e d on October 13, 1964.
The coolant pump was run
during t h e heatup with normal helium purgeaan'd t h e heaters were adjusted t o l i m i t t h e heatup r a t e t o about 100°F/hr.
Heaters located i n t h e c e l l
penetration sleeves on l i n e s 200 and 201were operated at a reduced s e t -
u
t i n g t o provide a temperature gradient u n t i l t h e heatup of t h e r e a c t o r c e l l was s t a r t e d so as t o reduce t h e thermal s t r e s s e s at t h e anchor point, Heater and thermocouple leads on top of t h e r a d i a t o r enclosure were observed t o be overheating as t h e system was being leveled at a reference The r a d i a t o r was cooled t o ambient temperature and
temperature of 350°F,
a section of t h e t o p of the r a d i a t o r w a s found t h a t had not been insulated. The i n s u l a t i o n w a s i n s t a l l e d and a second heating of t h e r a d i a t o r was started.
Some of t h e r a d i a t o r heaters were disconnected i n an unsuccessful
e f f o r t t o achieve a s a t i s f a c t o r y temperature d i s t r i b u t i o n . w a s 1073 t o 1245OF and t h e o u t l e t tubes were 620 t o 1000°F.
The r a d i a t o r This condition
was reached nine days a f t e r
heatup w a s s t a r t e d ,
remainder of t h e coolant sy
i n t h e coolant c e l l was heated t o 1047 t o
During t h i s time t h e
59OF except on l i n e 202 between t h e r a d i a t o r and t h e coolant pump.
There
s 400°F difference among temperatures along t h i s section of l i n e under a
s i n g l e h e a t e r control.
The s p
p a r t o f t h e permanent i n age.
The coolant drain
culty with a t
cj
was reduced t o about 175OF by replacing e a r t h e r a d i a t o r o u t l e t t o reduce inleak-
was heated . t o about 1200'F without any d i f f i -
e r a t u r e differenlce of 30°F between t h e highest and lowest
thermocouple reading.
'
27 6
1p The r a d i a t o r w a s cooled t o room temperature f o r inspection and t o re-
A f t e r r e p a i r s were made, t h e r a d i a t o r w a s again
p a i r t h e doors and seals. heated,
a 500'F
With t h e best temperature d i s t r i b u t i o n obtained, t h e r e was about spread on t h e r a d i a t o r o u t l e t tubes.
t i v e l y l a r g e heat leak existed,
This indicated t h a t a rela-
An inspection was made of t h e r a d i a t o r
enclosure near t h e o u t l e t h e a t e r and s i x compartments i n t h e enclosure wall were found t o be without insulation.
After t h e proper i n s u l a t i o n w a s in-
stalled and gaps i n t h e v i c i n i t y of t h e o u t l e t piping were caulked, satis-
factory temperatures were obtained with a l l h e a t e r s i n service as o r i g i n a l l y designed.
Localized overheating i n t h e areas above t h e r a d i a t o r enThe heat leaks were primarily a t openings
closure w a s again observed.
through which t h e ceramic-beaded insulated e l e c t r i c a l leads entered t h e enclosure,
Some minor i n s u l a t i o n damage w a s sustained on t h e wiring i n s i d e
t h e junction boxes mounted about 12 inches above t h e t o p of t h e radiator enclosure During t h e following shutdown, l a t e i n 1965, r a d i a t o r doors of improved design were i n s t a l l e d ,
Heat l o s s e s due t o air leakage around t h e
doors continued t o present a problem,
To obtain an acceptable temperature
u
d i s t r i b u t i o n , it w a s necessary t o shim t h e gaskets and disconnect some of t h e heaters, so as t o give a non-uniform heat input over'the face of t h e radiator,
When t h e main blowers were operated t o t e s t t h e doors, other
d i f f i c u l t i e s became apparent.
One of t h e s e w a s leakage of hot a i r i n t o t h e
region j u s t above t h e r a d i a t o r enclosure which caused overheating of elect r i c a l i n s u l a t i o n on t h e numerous thermocouples and h e a t e r leads i n Junction boxes located i n t h i s v i c i n i t y ,
It w a s necessary t o r e l o c a t e Qunction boxes
i n cooler locations along t h e c e l l w a l l and i n s t a l l higher temperature elect r i c a l i n s u l a t i o n on the thermocouple and h e a t e r leads over t h e enclosure, During-the approach t o f u l l r e a c t o r power, heating t h e empty r a d i a t o r t o an acceptable temperature d i s t r i b u t i o n p r i o r t o a f i l l had become increasingly d i f f i c u l t , and on t h e last f i l l , some of t h e tubes could not be heated above t h e melting point of salt u n t i l t h e r a d i a t o r was f i l l e d and circulatfon was started. door seals were modified.
After t h e shutdown i n J u l y 1966, t h e r a d i a t o r The e l e c t r i c a l heat required t o heat t h e empty
r a d i a t o r was reduced from 108 kW t o 62 kW by t h e door seal modification. I
6,
I .
277
u
The four flow transmitters located i n t h e enclosure above t h e radiator required a higher heater s e t t i n g when e i t h e r o r both blowers were on because of a i r leakage past t h e transmitter. 17.3.2
Fuel System Heatup
The i n i t i a l heatup of the f u e l system and t h e coolant piping i n t h e reactor c e l l w a s s t a r t e d November 1, 1964.
A t t h i s time t h e coolant system
i n t h e coolant c e l l had already been heated. About nine days were spent on t h e first heatup and leveling of t h e temperatures.
The heatup r a t e w a s limited by t h e heatup rate of the re-
actor because there i s about 8000 pounds of INOR and 8000 pounds of graphi t e t o be heated.
To a i d i n heating up t h e graphite and t o help purge out
any remaining oxygen, t h e f u e l pump was run during t h e heatup with normal I
helium purge.
Also the f u e l system piping and heat exchanger were operated
about 150'F higher t o help heat up the reactor. The system temperatures were brought up t o an acceptable range (1000 t o l25O'F)
with a few exceptions.
Temperature a t t h e freeze flanges and
on t h e coolant l i n e s at the c e l l penetrations were, as expected, below t h e liquidus temperature of t h e salt, but only over short sections.
Thermo-
couples under the heater located between freeze flange 100 and t h e fuelpump f'urnace could be heated only t o 700 t o 900' reduced by closing gaps i n the heater base.
u n t i l heat losses were
There was about 300'F
spread
between t h e coolant i n l e t nozzle and the f u e l o u t l e t nozzle on t h e heat exchanger.
The spread was reduced by closing gaps between t h e heat-exchanger
base and adjacent heater.
The time required t o heat up t h e reactor and f u e l
system a f t e r t h e i n i t i a l heatup w a s about t h r e e days. Fuel Drain Tank System Heatup
17.3.3
The i n i t i a l heatup of t h e drain-tank c e l l was s t a r t e d November 1 2 ,
1964. This heatup included drain tanks, FD-1, FD-2, FFT, and l i n e s 103, 104, 105, 106, 107, 108, 109, and 110. heat ed
.
The helium purge was s
This w a s t h e l a s t of t h e systems
from t h e f u e l system through l i n e 103 t o
the drain-tank system and then exhausted a t l i n e 110. The heaters were adjusted t o limit t h e heatup r a t e t o less than 100'F/hr.
A l l of the system w a s heated t o an acceptable range (1000 t o
278
1250째F) except where l i n e s ( l i n e s 107, 108,-and 109) penetrated t h e tank On l i n e 107, t h e temperature was brought up t o 95OoF, but only
furnaces.
by operating t h e heaters i n t h e penetration at 220% of t h e i r r a t i n g ,
The
other two l i n e penetrations were somewhat b e t t e r , but s t i l l unsatisfact"0z-y. The insulation on these l i n e s near t h e penetrations was removed,
Then t h e
heaters were t i e d closer t o t h e pipes and additional i n s u l a t i o n was i n s e r t ed between t h e heaters and t h e furnace w a l l s ,
Temperatures a t these pene-
t r a t i o n s then reached 1200'F with t h e heaters a t l e s s than t h e i r r a t e d capacity.
17.4 Heater Performance The system was operated at high temperature f o r about 10 months before t h e prepower shutdown. heater elements f a i l e d .
During t h e f i r s t 1500 h r of operation, only two Both o f these were in removable heater units on
t h e main f u e l c i r c u l a t i n g l i n e s and had spare elements i n t h e ceramic heat-
er p l a t e s . changes.
"he spare elements were put i n t o s e r v i c e by out-of-cell wiring A summary of t h e h e a t e r failures and repairs, a r e included i n
Tables 17.3, 17.4, and 17.5.
Six a d d i t i o n a l heater-element f a i l u r e $ and
m e l e c t r i c a l ground at a disconnect were discovered during t h e chebkout
f o r Run 2.
I n four of t h e heaters, t h e f a i l u r e s occurred at t h e Junction
of t h e extension l e a d and t h e h e a t e r element i n s i d e t h e ceramic p l a t e . These elements and t h e e l e c t r i c a l ground were repaired before s t a r t u p , and t h e other two elements were l e f t out of s e r v i c e f o r Runs 2 and 3* During subsequent operation, one i n - c e l l h e a t e r f a i l u r e was noted, and one f a i l u r e occurred at a h e a t e r power supply,
(The l a t t e r was repaired
It may be noted t h a t a number of h e a t e r f a i l u r e s could occur
immediately.)
without s i g n i f i c a n t l y a f f e c t i n g t h e operation of t h e r e a c t o r system, Three more heater-element f a i l u r e s were found during t h e prepower shutdown,
I n addition, a number of minor defects (grounds, damaged connectors)
were found.
All t h e defects, including those l e f t from e a r l i e r operations,
were corrected. An important cause of ceramic heater-element f a i l u r e has been separa-
t i o n of t h e extension lead from t h e h e a t e r element.
This i s apparently
due t o a combination of a design weakness and excessive flexing during
c
c Table 17.3
Date
System
Summary of Heater Failures i n the Reactor Cell
Heater
Type Failure
Repairs Made
CRITICAL EXPERIMENT PERIOD
11/9/64 12/ 31/64
Reactor
R-1-2
5-in. piping
H-100-1
One U-tube grounded Failure of top element
1/5/65 3/22/65
5-in. piping
H-102-5
Failure of top element
5-in. piping
H-100-1
Connected up spares t o continue Run 1. Replaced ceramic elements before s t a r t of Run 2.
3/22/65
%in. piping
H-102-5
Replaced ceramic elements before start of Run 2.
H-201-4B
Lead grounded Lead grounded
4/30/65 7/20/65
Heat exchanger
Hx-2
U-tube shortened 3 i n . Connected up spares t o continue Run 1.
Realigned disconnect. Removed sharp edge i n lead. R>
PREPOWER SHUTDOWN
i2
Two fuel-pump heater units resistance increase from 12 t o 17 ohms since the preoperational checkout. These w e r e removed from t h e reactor c e l l and t h e mechanical j o i n t between the s o l i d and flexible nickel lead w a s highly oxidized. A l l five fuel-pump heater units were removed from the reactor c e l l and a welded connection was made at each lead extension.
7/30/65
Fuel pump
FP1&2
7/30/65
5-in. piping
H-102-1
Broken disconnect replaced.
7/30/65 7/30/65
5-in. piping 5-in. piping
H-200-10
Broken disconnect replaced.
H-200-11
Broken disconnect replaced.
7/30/65
5-in. piping
H-201-6
Broken disconnect replaced. POWER OPERATION
12/29/65
Heat Exchanger
HX-1
Heater leads t o 2 out of 8 elements f a i l e d
Operation continued. 4/9/68 (see below)
Repaired
9/15/66
Heat Exchanger
HX-1
Heater leads of 2 more elements failed.
Operation continued. Repaired 4/9/68 (see below).
Table 17.3 (continued)
Date
System
Heater
Type Failure
Repairs Made
POWER OPERATION
(continued) 10/13/66
Heat Exchanger
HX-1
Heater leads of 2 more elements failed.
Operation continued. Repaired 4/9/68 (see below).
Heat Exchanger
HX-1
A l l elements out of' service due t o heater leads.
Operation continued. Repaired 4/9/68 (see below).
5-in. piping
H-102-1s
1 of 3 i n s t a l l e d spares
No repairs required. Adequate capacity i n others.
failed
9/13/67
Freeze valve
FV-103
Complete f a i l u r e
12/12/67
Heat Exchanger
HX-2
Open lead good
12/19167
He a t Exchanger
Hx-2
Heater leads of 6 out of 8 elements f a i l e d
4/9/68
Heat Exchanger
HX-1 HX-2
Heaters were removed from reactor c e l l and were repaired by welding the heater leads t o the terminal s t r i p s . A l l heater elements were good.
10/1/68
5-in. piping
H-100-1
Failure of 1 side element
Connected up spares.
10/21/6 8
Heat Exchanger
HX-1
Heater grounded
Taken out of service.
8/2/69
5-in. piping
H-10 2- 5
Failure of 1 side element
Connected up spares.
No repairs required. Adequate temperature without heater.
- a l l elements
Operation continued. Repaired 4/9/68 (see below). .
Operation continued. Repaired 4/9/68 (see below).
,
c
c
c Table 17.4
Date
Summary of Heater Failures i n the Fuel Drain Tank C e l l
Heater
System
Type Failure
Repairs Made
CRITICAL EXPERIMENT PERIOD
W65 5/65
F i l l Line
H-106-1
Failure 1 side element
Replaced element before Run 2.
F i l l Line
H-104-1
Failure 1 side element
Replaced a l l elements -new design during prepower shutdown.
Bad M I cable
Replaced, M I cable during prepower shut down.
- loose
Repaired 8/6/65 (see below).
Top element connectidn
5/65
Freeze Valves
FV-105-3
Top and 1 side element loose connections
5/65
Freeze Valves
FV-106-1
Top elements -loose connection
-
Hepaired 8/6/65 (see below). Repaired 8/6/65 (see below).
PREPOWER SHUTDOWN
8/5/65
Drain Tank
8/6/65
Freeze Valves
Replaced a l l heater elements (new design) i n freeze valve heaters FV-104-1, FV-104-3, FV-105-1, FV-105-3, FV-106-1, FV-106-3, and H-104-5.
10/10/65
F i l l Line
H-106-1
â&#x20AC;&#x2122;
FD2-1
Female disconnect grounded
Failure 2 elements due t o bad thermocouple
Added 3 in. of ceramic beads t o t h e 3 leads Q,
Replaced a l l elements -new design.
POWER OPERATION 1/21/67
F i l l Line
H-106-4
Failure 1 side element
Continued operation using two good elements and additional heat from adjacent heaters.
Table 17.5
Date
System
Heater
Summary
of Heater Failures i n the Coolant Cell
Type Failure
Repairs Made
PREPOWER SHUTDOWN CHECKOUT
8/12/65
5-in. penetration H-200-14s
5-in. penetration H-201-11 5-in. penetration H-201-11s Radiator
CR-2
Radiator
CR- 3
5-in. piping 5-in. piping
H-201-12 H-202-2
out of 8 elements failed loose connection
-
out of 8 elements failed loose connection out of 8 elements failed loose connection element failed loose connection element failed loose connection Element grounded Failure 1 element
-
Replaced a l l heater elements (new design) i n heaters H-200-14, H-200-15, H-201-10, H-201-11, and spares.
-
Replaced element.
-
Replaced element. Replaced element, Replaced element.
POWER OPERATION
5-in. piping Checkout after Run 7 8/17/66 Radiator 8/17/66 5-in. piping 8/17/66 Radiator inlet header 8/17/66 Coolant Pump
H-202-2
Broken lead
Reconnected.
CR-3 H-202-2 CR-7
Broken extension wire Failure 1 element Failure 1 element
CP-2
Failure 1 element
8/17/66
H-206-1s
Failure 1 spare element
Replaced element. Replaced element. Disconnected element capacity in others. Disconnected element capacity in others. No repairs necessary. capacity i n others.
1/17/66
c
F i l l line
c
- adequate - adequate Adequate
Table 17.5 (continued)
Date
System
Heater
Type Failure
Repairs Made
Repairs af’ter R u n 11
3/30/67
Radiator
CR-6
Failure 1 element
Replaced element.
Broken ceramic 2 broken lead wires. Nickel lead wire had crystallized.
Replaced element. Removed defective sections and welded leads and dye-checked.
1 element grounded
Disconnected element. capacity i n others.
Repairs af’ter Run 13
512
-
”
.
-
I
.
-.
Adequate
284
i n s t a l l a t i o n of t h e elements.
The j o i n t w a s redesigned and new elements,
Ld
which incorporate t h e change, were i n s t a l l e d when such failures occurred. Two fuel-pump h e a t e r u n i t s resistance increased from 12 t o 17 ohms. These were removed from t h e r e a c t o r c e l l and t h e mechanical j o i n t between t h e h e a t e r s o l i d and flexible n i c k e l extension lead was found t o be highly oxidized,
A l l f i v e fuel-pump h e a t e r u n i t s were removed from t h e reactor
c e l l and a welded connection was made a t each l e a d extension.
No other
trouble w a s encountered with these heaters. Before and during t h e prepower shutdown, d i r e c t maintenance w a s performed on t h e heaters i n the r e a c t o r and drain-tank c e l l . 17.4.2
During Power Operation
A l l heaters performed properly during t h e heatup of t h e f u e l and
coolant loops i n December 1965 f o r t h e s t a r t of power operation, with one minor exception.
There was a broken lead on a h e a t e r between t h e r a d i a t o r
and t h e coolant pump.
It w a s simply reconnected.
A f t e r t h e s t a r t u p , one
of t h r e e groups of elements i n one of t h e heat-exchanger h e a t e r u n i t s failed.
No action was taken a t t h i s t i m e because t h e other elements were
enough t o produce t h e desired temperature.
A summary of t h e h e a t e r f a i l u r e s
LlJ
and r e p a i r s made during power operation i s included i n Tables 17.3, 17.4, and 17.5. During t h e shutdown i n September 1966, a r e s i s t a n c e check w a s made on a l l i n - c e l l h e a t e r s at junction boxes located outside t h e r e a c t o r and
drain-tank c e l l and compared with i n i t i a l resistances.
Other than t h e
heat-exchanger h e a t e r , only one h e a t e r w a s found that had f a i l e d .
This was
one of t h r e e i n s t a l l e d spare heaters on t h e v e r t i c a l section of pipe under t h e heat exchanger.
No r e p a i r s were required, since t h e r e i s adequate ca-
pacity i n t h e other heaters on t h i s section of pipe. The resistance of t h e coolant-system h e a t e r s located outside of t h e r e a c t o r c e l l was a l s o checked.
Five heaters were found t h a t had f a i l e d .
,
One w a s on t h e r a d i a t o r , one on t h e coolant system 5-in. piping, one on t h e cdolant-pump furnace, one i n t h e r a d i a t o r o u t l e t header, and t h e other one, a spare heater, was on t h e f i l l - l i n e piping.
The h e a t e r s on t h e 5-in.
piping and on t h e r a d i a t o r were replaced, t h e r e w a s adequate h e a t e r capacity at t h e other locations.
W
285
6.i
Aside from heater-element f a i l u r e s , a remote disconnect f o r one of t h e fuel-pump heaters was damaged during remote operation associated w i t h t h e thawing of a salt plug i n t h e fuel-pump offgas l i n e .
The h e a t e r was
plugged i n t o a space disconnect t o r e s t o r e it t o service. . By t h e last of October
1966, all of t h e h e a t e r elements i n one of t h e
heat-exchanger h e a t e r u n i t s had f a i l e d .
Satisfactory heat-exchanger tem-
peratures were maintained without t h i s h e a t e r , even with t h e f u e l loop empty.
However, continuous c i r c u l a t i o n of t h e coolant salt was maintained
u n t i l after Run 11, so t h e full e f f e c t of t h e h e a t e r failure could not be determined.
Tests were performed w i t h both t h e f u e l and coolant loops
empty after Run 11 t o determine t h e need f o r t h i s h e a t e r i n preheating t h e system from a cold condition.
When helium c i r c u l a t i o n was stopped i n t h e
f u e l system, t h e temperature d i s t r i b u t i o n was s a t i s f a c t o r y f o r a f i l l w i t h out t h e h e a t e r operating.
With helium c i r c u l a t i o n i n t h e f u e l system, one
s a t i s f a c t o r y temperatures temperature decreased t o below 8 0 0 ~ ~ Since . could be achieved without it, the f a i l e d h e a t e r was not repaired at t h i s time.
b,
While t h e r e a c t o r system was being heated up f o r Run 13, Heater FV-103 was l o s t by an open c i r c u i t .
This 1.5-kW bent-tubular-type
h e a t e r normally
supplies about 200 W of heat t o a b i n . section of t h e f u e l drain l i n e within t h e r e a c t o r vessel furnace, between t h e freeze valve and t h e resistance-heated section of t h e l i n e .
The purpose of t h e h e a t e r w a s t o
help control the temperature p r o f i l e through t h e freeze valve.
After t h e
heater f a i l e d , tests with f l u s h s a l t shoued t h a t by proper adjustment of t h e cooling a i r controls, t w i t h thaw times i n an acce
eeze valve could be maintained r e l i a b l y range.
(Thaw time w a s around 1 2 min when
t h e r e a c t o r vessel was at 1 1 8 0 ~ and ~ t h e center of t h e freeze valve was at
495OF.)
Therefore, t h i s h e a t e r was not replaced.
I n December 1967, a l e a d t o t h e adjacent f a i l e d heat-exchanger h e a t e r opened, thereby reducing t h e
utput of t h i s h e a t e r by 50%. A week l a t e r a
second p a r t i a l f a i l u r e f u r t h e r reduced.the output t o only 33%of normal. Although these f a i l u r e s d i d not a f f e c t operattons as long as s a l t w a s kept c i r c u l a t i n g , t h e lack of heat i n t h e two adjacent heaters made it necessary
b
t o r e p a i r them during t h e shutdown af'te/ Run 14.
These two u n i t s were
286
L)
removed from t h e reactor c e l l e a r l y i n t h e shutdown (on t h e fourth and f i f t h day a f t e r t h e end of full-power operation) t o see i f r e p a i r s were possible o r i f replacement u n i t s would be required, The trouble was found i n t h e Junction boxes mounted on t o p of t h e h e a t e r s , where t h e l e a d wires
from several of t h e h e a t e r elements had burned i n two at t h e i r screwed connections t o t h e terminal s t r i p s ,
Repair w a s complicated by t h e induced ac-
t i v i t y i n t h e assemblies, which produced a gamma r a d i a t i o n f i e l d of about
3 R/hr at 1 ft, A temporary work s h i e l d was s e t having concrete block w a l l s and a top of s t e e l p l a t e and lead block.
Direct maintenance on t h e Junc-
t i o n was done through a small opening i n t h e top.
The terminal s t r i p s and
connections were severely oxidized, apparently due t o high temperatures This s i t u a t i o n w a s corrected by i n s t a l l i n g nickel term-
during operation,
i n a l s t r i p s and welding t h e h e a t e r leads d i r e c t l y t o them.
The copper wire
from t h e terminal s t r i p t o t h e disconnect w a s a l s o replaced with No, 12 nickel wire welded t o t h e terminal s t r i p .
Aside from t h e damage i n t h e
Junction boxes, both heat-exchanger heaters were i n good condition, and a f t e r t h e r e p a i r s , both were r e i n s t a l l e d through t h e maintenance s h i e l d
-
without unusual d i f f i c u l t y . Af'ter t h e u n i t s were r e i n s t a l l e d i n t h e c e l l , it was found t h a t t h e
current on one phase of south end u n i t was zeroo This proved t o be due t o a fault i n t h e permanently mounted l e a d wire i n t h e c e l l .
The fault was
circumvented by i n s t a l l i n g a Jumper cable between t h e h e a t e r disconnect and
a spare disconnect.
The heaters functioned s a t i s f a c t o r i l y during t h e sub-
sequent heatup of t h e system. One of t h e elements i n t h e h e a t e r u n i t on l i n e 106 adgacent t o t h e Junction of l i n e 103 with l i n e s 104, 105, and 106 f a i l e d during t h e middle of January 1967.
This was not replaced since it was possible t o maintain
aaequate temperature using t h e other elements and adjacent heaters. There have been t h r e e failures of r a d i a t o r headers due t o broken nickel lead wires.
The leads look l i k e t h e nickel had crystallized.,
Several bro-
ken ceramic bushings have been replaced. The heater located on t h e south end of t h e heat exchanger developed a p a r t i a l ground i n t h e l a t t e r p a r t of October 1968.
I n order t o c l e a r t h e
ground detector l i g h t s on Bus 5, t h e h e a t e r w a s taken out of service.
This LivJ
287
was possible because t e s t s had been performed a f t e r Run 11which determined t h a t t h i s h e a t e r was not e s s e n t i a l f o r preheating t h e system from a cold condition.
The cause of t h e grad was suspected t o be a broken ceramic
bead on the l e a d wire between a h e a t e r element and t h e junction box mounted on t o p of t h e unit. About a 30% drop i n current on one heater on t h e f u e l l i n e between t h e reactor and t h e f u e l pump (H-100-1) occurred l a t e i n September 1968.
After
a r e s i s t a n c e check indicated a s i d e h e a t e r element had f a i l e d , t h e spare
heater elements were placed i n service. About a 30% drop i n current on t h e h e a t e r on t h e f u e l l i n e between t h e heat exchanger and t h e reactor (H-102-5) occurred i n t h e e a r l y p a r t of Augu s t 1969.
A f t e r a resistance check indicated t h a t one of t h e t h r e e h e a t e r
elements had f a i l e d , t h e spare h e a t e r elements were placed i n service. The individual ammeters f o r each h e a t e r c i r c u i t were read and recorded daily.
Since operation s t a r t e d , most h e a t e r f a i l u r e s caused a s u f f i c i e n t
change i n ammeter reading s o as t o be noticed.
Variation i n voltage sup-
p l i e d t o t h e MSRE w a s usually l e s s than 2% over a 24-hour period.
2% change caused l e s s than a 2'F/hr isothermal conditions. tings.
change of t h e system temperature under
This w a s e a s i l y adjusted by changes i n h e a t e r s e t -
After t h e system had been brought t o equilibrium at temperature it
was not usually necessary t o m a k e f u r t h e r adjustments. including power requirements f o r 1200'F Table
A total
17.6.
I
Typical h e a t e r data
isothermal conditions a r e given i n
'
The four types of f a i l u r e s experienced w i t h the h e a t e r leads a r e given i n Table 17.7 along with methods used t o make repairs.
17.5 Before nuclear power operation was started, system cooldown t e s t s were run t o determine t h e e f f e c t s on t h e system i f e l e c t r i c a l power w a s l o s t . The cooldown of t h e f u e l and coolant systems w a s done simultaneously since
these are connected at t h e heat exchanger.
These t e s t s which were done
with t h e r a d i a t o r doors closed a i d not show any d i f f i c u l t i e s .
A much more
serious condition occurred on two occasions a f t e r nuclear power operation was s t a r t e d when t h e cold r a d i a t o r doors were scrammed and t h e coolant pump
,
<
Table 17.6
Heater Numbers R1, R2, & R3 FT-1 & FT-2 FFT-1 & m - 2 FD1-1 & FD1-2 FD2-1 & FD2-2
Equipment
Typical Heater Data
1200'F Operating Data Average Heat Loss Heat Loss Per f t 2 of Surfacb % of Max k W Installed Watts
Maximum Installed kW
Insulation' Thickuess in.
Reactor Vessel Fuel Pump
69.6 18.2
m
45 47 45
FD-1
FD-2
29.9 12 12.8
16.1 14.9
81
$6
112
82
28.5
59
34.3
74.2
33.1
68.7
47 ' 57 57
43 66
,
30' of 5" piping 23.5' of 5" piping
H202-2
29' of 5" piping
RCH-1
7' of 5" piping
RCH-2
5' of 5" piping
7.5' of 5" piping
RCH-3
4 4 4 Metallic MultipleWYer Metallic MultipleLwer Metallic MultipleLayer ~
'Ceramic
'
Per f t of
Per f t of Pipe H200-13 H201-22
Calculated Heat Loss Per ft2 of Surface Watts
15.3
150 140
Pipe 126 126
4.7
19.8
162
126
13
2.5
19.2
357
b
10
2.0
20
400
b
23.75 21.6
5.4
22.8
3.3
23.75
R)
03 03
t
15 ~~
~~~
3.4
22.7 ~
b
453 ~
~
~
fiber of expanded s i l i c a .
'Design based on heat loss of 600 w a t t / f t of pipe on t e s t unit.
e
c
289
Table 17.7
Heater
I
.
Type of Heater Lead Failure and Repairs
Failure
Correction Made
(1)Ceramic
Separation of extension lead from heater element
Designed leads with a cross bar
( 2 ) Fuel pump tubular
Excess oxidation at mechani c a l j o i n t between s o l i d and f l e x i b l e head
Made welded j o i n t s and dye-checked
( 3 ) Heat exchanger
Excess oxidation at mechani c a l j o i n t s a t terminal s t r i p i n junction box
Made welded j o i n t s and dye-checked
(4) Radiator
Nickel extension lead c r y st a l l i zed
Removed defective sect i o n , welded and dyechecked
290
stopped.
S a l t w a s frozen i n some of t h e r a d i a t o r tubes but was succesfully
melted out with no apparent damage t o t h e radiator.
These coolant tests
and radiator d i f f i c u l t i e s are discussed i n t h e following four sections,
17.5.1
Fuel and Coolant Systems Cooldown Rate, Radiator Door Closed
Three cooldown t e s t s were run.
In the f i r s t t e s t , t h e cooldown rate
was determined f o r t h e f u e l and coolant systems with s a l t circulat'ing i n both systems and with all t h e reactor and coolant c e l l heaters o f f except \
the heaters on t h e f i l l l i n e s ,
The f u e l and coolant system reference tem-
perature dropped from 1160 and ll69'F one hour and 52 min.
t o 1105 and 1115'F respectively i n
This i s a cooldown rate of 0,5'F/min.
In t h e second t e s t , t h e same heaters were o f f as i n t h e f i r s t t e s t and the f u e l and coolant pumps were turned off.
A f t e r t e n minutes, t h e radi-
a t o r heaters and t h e f u e l and coolant pumps were turned back on.
This s i m -
ulated a power f a i l u r e followed by t h e s t a r t u p of t h e d i e s e l generators, The f i e 1 and coolant system reference temperatures dropped from 1156 and
1185'~ t o 1089 and 1112'F respectively i n 2 hours and 50 minutes.
This i s
a cooldown r a t e of 0.4'F/min, In t h e t h i r d t e s t , all t h e heaters and t h e f u e l and coolant pumps were turned off and remained o f f f o r t h e duration of t h e test,,
The reference
temperatures dropped from 1150 and 1170'F t o 1120 and 1141'F respectively f o r t h e f u e l and coolant system i n 38 minutes.
This i s a cooldown r a t e of
0.,8~~/min. The reference temperature on t h e r a d i a t o r dropped from 1170 t o 1090'F i n the same period,
This is a cooldown r a t e of 2,1째F/min,
In all three tests t h e temperatures of t h e coolant f l o a t l e v e l indicators dropped rapidly and t h e i r heaters were turned back on. ~
17,5,2
Fuel Drain Tank Cooldown Rate
With flush salt i n FD-2, a test w a s made of t h e cooldown r a t e by turning off all FD-2 heaters.
w a s 1085'~.
The average temperature a t t h e s t a r t of t h e test
Ten and one-half hours later, t h e average temperature was 1020'F.
From t h i s data, it is estimated t h a t it wouid take at l e a s t 56 hours f o r FD-2 t o cool down from 1200'F t o 850'F upon l o s s of e l e c t r i c a l power t o the tank heaters.
Decay heating would'extend t h i s time.
.
291
u
17.5.3
Coolant Drain Tank Cooldown Rate
With coolant s a l t i n CDT, a t e s t was made of t h e cooldown r a t e by turning o f f a l l CDT heaters.
The average temperature of f i v e thermocouples
t o~ 1 0 8 5 ' ~ i n sixteen hours. dropped from 1 1 6 0 ~
From t h i s data it i s e s t i -
mated t h a t it would take at l e a s t 70 hours f o r CDT t o cool down from 1200'F t o 850'3'
upon l o s s of e l e c t r i c a l power t o t h e tank heaters.
17.6 Discussion The f u e l loop operated above 9OO'F loop above 9OO'F
f o r 27,438 hours.
f o r 30,848 hours and t h e coolant
There were 1 3 heating and cooling cy-
c l e s of f u e l loop and 12 f o r t h e coolant loop.
The four drain tanks, in-
cluding t h e drain and fill l i n e s , were operated near 1200'F continuously since t h e s a l t was added t o t h e tanks except f o r two times when FD-2 was cooled down f o r i n s t a l l i n g and removing equipment f o r t h e 233U addition. There w a s not an interruption of t h e power operation of t h e reactor due t o a heater failure.
k-,
The main weakness of t h e ceramic heaters was t h e separation of t h e extension leads from t h e heater elements.
20 correct t h i s , a t o t a l of 70
(30 i n t h e r e a c t o r and f u e l drain tank c e l l and 40 i n t h e coolant c e l l ) ceramic elements of an approved design were i n s t a l l e d during t h e prepower shutdown.
Performance since then has been excellent.
There were only
5 failures which could be a t t r i b u t e d t o t h e ceramic elements out of 221 installed units.
292
18.
SAMPLERS , ,
*
><
? .
- "
-
A. I. Krakoviak
I
I
6s
.-
I n general, . . . . of both t h e f u e l and coolan ystem w a s accom. . sampling plished by a power-driven c i b l e which'lowered a s m a l l capsule i n t o t h e c i r culating l i q u i d of t h e p G p bowl.
The capsule w a s then .raised and t h e s a l t
was permitted t o s o l i d i f y before f i n a l r e t r i e v a l and shipment t o t h e Anal y t i c a l Laboratory f o r .analysis.
Lead shielding (%I-0 inches) w a s i n s t a l l e d
around t h e major components of t h e - f u e l sampler and a stainless-steel-clad Double containment
lead shipping cask was used t o transport t h e s q l e .
was accomplished by moving t h e sample ( a f t e r i s o l a t i o n from t h e pump bowl) through a s e r i e s of containment areas while maintaining at l e a s t two valves o r membranes between t h e f u e l system o r sample and .the environment, Since t h e induced a c t i v i t y i n t h e coolant s a l t i s very short-lived, no shielding w a s required during sampling and only s i n g l e containment w a s necessary during sample transport. A t o t a l of 81 coolant salt samples were taken and a t a t a l . o f 745 f u e l
sampling cycles were completed during t h e f i v e years of MSRE operation. Included i n t h e f u e l sampling t o t a l were 152 f u e l enrichments of either uranium o r plutonium,
I n addition t o sampling and enriching, t h e f u e l sam-
p l e r was a l s o used t o sample t h e gas space above t h e l i q u i d i n t h e pump bowl and t o make chemical additions such as beryllium, niobium, and FeF2 t o t h e fuel,
18,i Fuel Sampler-Enricher Since t h e fuel-sampler usage was more frequent and t h e complexity of i t s components were g r e a t e r than those of t h e coolant sampler, t h e major p a r t of t h i s report describes t h e operating experience with t h e samplerenricher,
A b r i e f description of t h e coolant sampler and a discussion of
i t s operation are included at t h e end of t h i s r e p o r t ,
18.1.~Description of t h e Fuel Sampler-Enricher The component arrangement of t h e f u e l system sampler-enricherS0 i s
shown i n Fig. 1 8 , ~The vacuum pumps and helium purge gas system required f o r pressure and atmosphere control a r e not shown.
The drive motor f o r
u
293
Fig. 18.1
Compo
for Sampler-Enricher
294
the cable, the l a t c h onto which capsules were hung were located i n primary
u
containment area 1 C which w a s approximately 17-1/2 f e e t from t h e pump bowl. C r i t i c a l closures such as t h e operational and maintenance valves (motordriven gate valves)
s
access port t o area 1 C (doubly-sealed door), and t h e
removal valve ( b a l l valve) were buffered w i t h helium t o assure t h a t i f any sealing surfaces leaked, t h e leakage would be i n e r t gas i n t o t h e system. Metered quantities of helium were supplied t o these seals and t h e leakage
rates across t h e seals were measured by t h e f i n a l equilibrium buffer s e a l pressure.
Excess s e a l leakage i n t o t h e system o r t o t h e containment area
could be detected by a decrease i n buffer seal pressure and a pressure increase i n t h e adjacent volume.
To prevent t h e accidental opening o r closure
of compartment doors or valves i n improper sequence during sample t r a n s f e r , a system of e l e c t r i c a l interlocks based on buffer seal pressures and-me-
chanical signals along with a detailed check l i s t (6A-3) ( R e f . 51) w a s used. I n general, a sample capsule, similar t o t h a t shown i n Fig. 18.2, w a s loaded i n t o a doubly sealed transport tube which w a s lowered through t h e removal valve i n t o area 3A.
The lower p a r t of t h e transport tube containing
the capsule w a s unscrewed and l e f t i n area
3A whereas t h e upper p a r t was
withdrawn above t h e removal valve and t h e valve closed.
The access port
t o area 1 C was then opened and the capsule w a s hung on t h e l a t c h w i t h t h e manipulators (the l a t c h and l a t c h pin were so designed t h a t t h e pin could not be disengaged during sampling except i n t h i s area).
The access port
w a s closed and area 1 C w a s first evacuated then pressurized t o equal t h a t i n t h e pump bowl before the maintenance ( M V ) and operational (OV) valves were opened.
The cable drive motor w a s actuated u n t i l t h e capsule w a s im-
mersed i n the salt i n the guide cage of t h e pump bowl (lower p a r t of Fig.
18.1).
The sample w a s withdrawn 18 i n . where t h e salt w a s allowed t o freeze
i n the v e r t i c a l section of unheated 1-1/2 in. Sched-40 pipe before f i n a l r e t r i e v a l i n t o area 1 C .
As t h e capsule o r sample was moved.from one com-
partment t o another there were evacuation and pressurization s t e p s which removed adsorbed oxygen during capsule i n s e r t i o n and removed gaseous fission products during sample removal.
Capsule t r a n s f e r s inside t h e sampler
were viewed with a periscope. Enriching w a s accomplished by lowering an enriching capsule (Fig. 18.2) i n t o t h e pump bowl and allowing t h e frozen s a l t i n the capsule t o m e l t and
h.d
c
Fig. 1 8 . 2
Sample Capsule and Enriching Capsule
be washed out before r e t r i e v a l of t h e empty capsule ( t h e w a l l s of t h e enriching capsules w e r e d r i l l e d through t o t h e frozen salt i n several places
j u s t p r i o r t o loading t h e capsule i n t o t h e sampler). The sampler-enricher w a s d e s i w d t o make 1000 sampling or-enriching cycles f o r a period of one year;52 however, t h e MSRE and sampler were operated f o r 20 runs over a period of 5 years. 18.1.2
Experience with t h e Fuel Sampler-Enricher During Pre-Critical and Low-Power Operation
The sampler-enricher w a s i n s t a l l e d and put i n operation i n May 1965.
Sampler t e s t i n g , shakedown, and operator t r a i n i n g sessions were conducted concurrently with t h e sampling and enriching operations of Run 2.
During
t h i s period, 53 f u e l samples were taken, 87 enrichments made, and 20 opera t o r s t r a i n e d i n t h e use of t h e sampler. â&#x20AC;&#x2122;
Some of t h e problems encountered
during t h i s period were: 1.
B u f f e r gas leaks at t h e Operational (OV), maintenance (MV), and removal (RV) valves.
2.
Failure of a solenoid valve on t h e removal valve actuator.
3.
Failure of t h e removal valve t o close completely requiring manual closureo.
4.
Failure of t h e access port (AP) t o close completely.
5. 6.
Momentary failure of cable drive motor on capsule r e t r i e v a l . Rupture of t h e flexible containment membrane (boot) a t t h e manipulator
7. 8.
.
Deformation of t h e manipulator a r m and fingers. Accidental drop of an empty capsule onto t h e operational valve.
These and additional problems encountered with t h e subsequent 86 sampling cycles made during t h e approach t o power runs a r e discussed i n more d e t a i l
i n t h e following paragraphs. B u f f e r G a s Leaks at t h e Operational and Maintenance Valves -During
t h e sampling operations of Run 2, leaks developed through t h e upper m e t a l to-metal
seats of both t h e operational and maintenance valves.
The oper-
a t i o n a l valve w a s subsequently removed; examination showed t h a t a t h i n black ring, which was e a s i l y removed, had formed at t h e upper s e a l i n g surface of t h e valve gate and a small quantity o f salt spheres ( < 1 gm) had
LJ
297
collected between t h e s e a t s of t h e gate.
Apparently s a l t p a r t i c l e s were
dislodged from t h e capsules during t h e operation where t h e capsules ( a f t e r having been dipped i n t h e pump bowl) were disengaged from t h e l a t c h above t h e valve.
When t h e stem and gate were lubricated, t h e valve sealed almost
completely; however on removal of t h e l u b r i c a n t , t h e buffer gas leak r a t e increased t o 2 cc/min through t h e upper gate s e a l and remained a t zero through t h e lower s e a l . A few sampling cycles a f t e r the valve w a s r e i n s t a l l e d , t h e leak r a t e
again increased t o 20 cc/min through t h e upper s e a l .
Apparently more s a l t
p a r t i c l e s had lodged between t h e buffer-gas s e a l i n g surfaces.
Repeated ef-
f o r t s t o blow t h e p a r t i c l e s from the sealing surfaces f a i l e d .
Since t h e r e
were t h r e e other s e a l i n g surfaces between t h e pump bowl and t h e sample access area and since t h e leak did not increase with continued sampling, t h e valve was not replaced. After approximately 86 additional sampling cycles, an opportunity t o clean t h e operational valve arose when t h e r e a c t o r was drained t o r e p a i r t h e cable d r i v e motor i n April 1966 ( t o be discussed l a t e r ) .
Cleaning t h e
s e a t i n g surfaces decreased t h e leak r a t e f o r a while u n t i l an empty cap-
sule w a s accidentally dropped on t h e gate.
Subsequent valve operation
again r e s u l t e d i n a high buffer-gas leak r a t e .
The b u f f e r gas s e a l a t t h i s
surface continued t o d e t e r i o r a t e slowly throughout t h e remainder of MSRE operations; t h e only consequence was a slow pressure buildup i n t h e containment volume above t h e valve which required periodic venting.
From
February 1967 u n t i l f i n a l shutdown, t h e b u f f e r supply pressure t o t h i s s e a l
w a s used only when t h e sampler was i n operation thereby obviating t h e periodic venting of area 1 C . During t h e l a t t e r par leakage of buffer gas t h r o
966 t h e r e had been a gradual increase i n e upper s e a t of t h e maintenance valve also.
Although t h e procedure s p e c i f i e d t h a t t h e maintenance valve be opened before t h e operational valve is opened, and that t h e operational valve be closed before t h e maintenance valve is closed, on a t l e a s t one occasion t h e reverse sequence w a s followed during closure and thus foreign matter could have been dislodged f r o m t h e upper valve onto t h e lower valve.
While t h e
leak r a t e remained r e l a t i v e l y small (Q25 cc/min), it was s u f f i c i e n t t o
-
298 cause d i f f i c u l t y with proper operation of t h e interlocks (an e l e c t r i c a l s i g n a l which indicated t h a t t h e valve was indeed closed was based on t h e a b i l i t y of seating surfaces t o maintain a specified b u f f e r pressure),
Since
cleanup would have been d i f f i c u l t and replacement of t h e valve w a s not warranted by t h e leak rate alone, a mechanical method of assuring t h a t t h e valve w a s closed was s u b s t i t u t e d f o r t h e pneumatic method.
No additional
problems with these valves were reported, Area 1 C Access Port -The
access port door was operated by s i x clamps
which r e l i e d on pneumatic operators and a spring f o r opening and a scheduled time-delay of these pneumatic operators f o r closure ;. however, once closed, t h e mechanical linkage was such t h a t t h e clamps no longer require gas pressure t o keep t h e door t i g h t l y closed.
The a b i l i t y of two concen-
t r i c neoprene gaskets (between t h e access port door and housing) t o contain helium at 55 p s i a s a t i s f i e d an interlock and indicated t h a t a gas-tight s e a l did indeed e x i s t between areas 1 C and 3A.
On s e v e r a l occasions during
t h e p r e - c r i t i c a l run, t h e access port f a i l e d t o close properly.
Also, one
o r two of t h e s i x access port operators f a i l e d t o open and t h e manipulator
was used t o release t h e s t i c k i n g operators.
During t h e shutdown following
t h e low-power experiment, t h e clamps were readjusted and an extension was added t o t h e pin of each clamp; t h i s extension made it e a s i e r t o open any clamp manually with t h e manipulator i n t h e event of a pneumatic malfunction. Flexible Containment Membranes
- There
are two 0.020-in.
thick flexi-
b l e urethane membranes (boots) which p a r t i a l l y enclosed and provided mob i l i t y t o t h e manipulator arm used i n t r a n s f e r r i n g capsules t o and from t h e l a t c h i n area1C and t h e bottom portion of t h e t r a n s p o r t c a r r i e r i n a r e a
3. A vacuum w a s maintained between these boots t o monitor t h e i n t e g r i t y of t h e boots and assure t h a t double containment did indeed exist between t h e sampler operator and t h e sample.
A loss of vacuum and a pressure re-
duction i n area 3 A would indicate a leak i n t h e outer boot whereas a loss of vacuum with no pressure change i n a r e a 3A would implicate t h e inner boot (atmosphere s i d e ) o r possibly both boots. There were t h r e e boot f a i l u r e s during t h e first 140 sampling cycles
of Run 2.
One w a s caused by an inadvertent evacuation of area 3A while
the manipulator cover was off; t h e r e s u l t a n t pressure gradient ruptured the boot.
On another occasion, t h e boot was snagged on t h e bottom piece
LiJ
299
t-,
of t h e transport tube i n area 3A when t h e manipulator was used t o release a stuck access port clamp.
The t h i r d f a i l u r e r e s u l t e d from pinching t h e
boot between t h e manipulator arm and t h e housing. A t t h e end of t h e next run ( R u n 3 ) , a pressure switch ( w i t h an alarm
and i n t e r l o c k ) was i n s t a l l e d t o detect and prevent a negative pressure gradient g r e a t e r than 1/2 p s i between area 3A and t h e manipulator arm. S t e e l rings were i n s t a l l e d i n t h e convolutions of t h e inner boot t o hold
it f r e e of t h e manipulator a r m and possibly prevent t h e pinching-type failure.
Also t o ensure t h a t at l e a s t two b a r r i e r s e x i s t e d between primary con-
tainment and sampling personnel, a pressure switch was i n s t a l l e d on t h e manipulator cover which required a negative pressure of 4-in.
Hg i n the
cover before e i t h e r the operational o r maintenance valve could be opened. I n April 1966, t h e manipulator cover pressure was inadvertently evacuated t o -25-in.
Hg without lowering area 3A pressure simultaneously; t h e
12-psi d i f f e r e n t i a l caused a s m a l l puncture of t h e inner boot.
This s e t
of boots had been used f o r 48 sampling cycles since power operation was begun and t h e maximum power reached a t t h a t time was 2.5 MW.
The r a d i a t i o n
l e v e l of t h e manipulator assembly and boots w a s 10 R/hr at 3 inches from t h e f i n g e r s , but was reduced by a f a c t o r of 10 by scrubbing w i t h soap and water.
About 2 hours were required f o r replacement of both boots.
Deformation of t h e Manipulator A r m and Fingers
- During Run 2 ,
the
manipulator arm and fingers were bent which caused d i f f i c u l t y i n gripping the l a t c h cable and moving t h e manipulator arm.
a.rm was replaced and a
A t t h e end of Run 3, t h e
l / 4 x 114 inch projection was welded t o t h e bottom
of each finger t o aid i n grasping t h e capsule cable from thâ&#x20AC;&#x2122;e f l o o r of area
3A.
The manipulator a r m w a s replaced and t h e clearances between t h e arm
and c a s t l e j o i n t were increased t o reduce t h e force required t o operate the manipulator.
The manipulator and fingers operated s a t i s f a c t o r i l y f o r 421 additional sampling cycles until A u g u s t 1968 at which time t h e manipulator assembly
was replaced because t h e t i p
t h e fingers no longer closed t i g h t l y .
old assembly w a s decontaminated and repaired.
The
The replacement fingers op-
e r a t e d s a t i s f a c t o r i l y u n t i l September 1969 when t h e double metal bellows
300
w
developed a leak, The sampler on t h e f u e l processing ystem t 5 cannibal i z e d t o make t h e r e p a i r s ; t h i s u n i t operated s a t i s f a c t o r i l y f o r t h e subsequent three months of reactor operation before shutdown. Capsule Recovery from t h e Operational Valve
-On
one occasion during
t h e p r e c r i t i c a l run, an empty sampling capsule w a s accidentally knocked i n t o area 1 C before t h e l a t c h pin was completely engaged i n t h e l a t c h , and the capsule dropped onto t h e gate of t h e operational valve,
The capsule
was r e t r i e v e d by removing t h e manipulator assembly from 3, opening t h e access p o r t , and snaring t h e capsule cable w i t h a wire hook., Since t h i s type of accident could reoccur, t h e brass l a t c h pins used t o a t t a c h t h e capsule t o t h e l a t c h w e r e replaced with nickel-plated mild s t e e l pins so t h a t capsule recovery could then be effected with a magnet. After approximately a month of full-power operation i n July 1966, an empty capsule again was dropped onto t h e gate of t h e operational valve when t h e manipulator slipped during the l a t c h i n g operation.
A magnet w a s l o w -
ered i n t o t h e t r a n s f e r l i n e and the capsule assembly was recovered as t h e magnet w a s withdrawn.
The radiation l e v e l of t h e magnet a f t e r t h e recovery
operation was 10 R/hr at 2 ft.
,
(pi
A l l sampling capsules of t h e bucket-type assembled since t h e l a t t e r part of 1966 contained a nickel-plated mild s t e e l top so t h a t capsule re-
covery could be made with a magnet. Removal Area, Valve, and Seals
- During t h e p r e - c r i t i c a l
run, t h e
removal s e a l and valve required realignment and increased tolerances bef o r e t h e transport container and removal t o o l assembly would s l i d e through these u n i t s f r e e l y without binding.
I n addition, t h e removal valve f a i l e d
t o s e a l properly even though t h e b a l l and seals were replaced.
Therefore,
at t h e end of t h e next run, t h e valve assembly was replaced with a modified version t o improve t h e s e a l i n g c h a r a c t e r i s t i c s and improve f u t u r e access t o t h e valve.
This u n i t served s a t i s f a c t o r i l y except t h a t a b u f f e r gas
leak developed i n t h e top t e f l o n s e a l of t h e valve during t h e l a t t e r p a r t of 1967.
A s a consequence of t h e leak, t h e equilibrium b u f f e r s e a l pres-
sure reached only 30-35 p s i a instead of t h e normal 5 0 5 5 p s i a ; t h e associated pressure switch which permits t h e opening of t h e operational and maintenance valves and t h e access port was lowered from 50 t o 30 and f i n a l l y i n December 1967 t o 25 p s i a where it remained f o r t h e remainder of r e a c t o r operations
hd
301
u
Latch, Capsule, and Cable Malfunctions
- Early i n Run 2, t h e cable
drive motor s t a l l e d during t h e withdrawal of an empty capsule.
The cap-
sule was r e i n s e r t e d about 12 i n . and then withdrawn successfully. reason f o r t h e malfunction,was unknown.
The
Approximately 100 sampling cycles
l a t e r i n December 1965, t h e motor s t a l l e d again w h i l e withdrawing a 10-gm sample.
Af'ter repeated attempts, t h e capsule was withdrawn completely and
t h e cable was examined.
The capsule was empty and t h e cable w a s found t o It
be backed up i n t o t h e drive u n i t box and caught i n t h e motor gears.
was assumed t h a t t h e l a t c h had hung on t h e gate of t h e operational o r maintenance valve causing t h e cable t o c o i l up i n s i d e a r e a 1 C . The cable was straightened, t h e l i m i t switches on t h e operational and maintenance valves were reset t o open t h e valves wider and t h e diameter o f t h e l a t c h was reduced.
I n April 1966, after approximately a week a t a power l e v e l of 5 MW, t h e capsule drive-unit motor s t a l l e d as t h e l a t c h w a s being r e t r i e v e d through t h e pipe bend near t h e pump bowl.
LJ
d r a w a l s , t h e l a t c h was retrieved.
After several i n s e r t s and with-
The l a t c h and p a r t of t h e cable were
then pulled through t h e access p o r t , t h e removal valve, and i n t o t h e sample transport cask which provided shielding while t h e l a t c h was replaced with one designed t o provide an additional 1/8-inch diametrical clearance, Approximately 3 months l a t e r t h e capsule stuck again temporarily f o r some unexplained reason on withdrawal and no sample w a s obtained.
Subse-
quent hangups with more serious consequences a r e discussed i n Section 18.1.3. -While
t e s t i n g t h e new l a t c h ,
t h e cable position indicator stopped when t h e l a t c h was approximately two
feet from t h e pump bowl; t h e upper limit switch a l s o actuated at t h i s time. E l e c t r i c a l continuity chec d r a w windings of t h e cab1
motor and a l s o t o t h e
All these leads penetrate
ontainment w a l l through
ptacle.
Since r e p a i r c
without v i o l a t i n g contai l a t c h were then pulled i transport container and
W
owed open c i r c u i t s t o t h e i n s e r t and with-
er l i m i t switch. ommon 8-pin re-
t be made while f u e l was i n t h e r e a c t o r he r e a c t o r w a s drained.
The cable and
s i n g t h e manipulator t o p u l l and t h e
rt operators t o hofd t h e cable so t h a t
t h e cable could be regrasped with t h e manipulators f o r another p u l l .
The
302
operational and maintenance valves were then closed.
The assembly includ-
u
ing t h e drive motor, cable, and l a t c h were removed, p a r t i a l l y decontaminated, and repaired with t h e use of shadow shielding techniques,
(Three
connector pins i n one of t h e 8-pin receptacles were burned o f f , )
A l l six
similar receptacles at t h i s location were f i l l e d with epoxy r e s i n t o provide additional i n s u l a t i o n and strength,
The cable which w a s bent i n sev-
eral places during r e t r i e v a l w a s decontaminated with soap and water t o reduce t h e radiation l e v e l t o 5 R/hr a t 3 i n . before it was straightened. A f t e r reassembly and checkout, normal sampling was resumed. Contamination of t h e Removal Seal - A n
a r e a a t t h e t o p of t h e sampler
was provided s o t h a t t h e lower p a r t of t h e t r a n s p o r t c a r r i e r and capsule could be evacuated and pressurized several times t o remove oxygen and moist u r e from t h e capsule and c a r r i e r before i n s e r t i o n i n t o t h e dry box (Area 3A) of t h e sampler.
The bottom s e a l f o r t h e removal a r e a w a s provided by
t h e removal valve and a s e a l at t h e t o p was made when t h e transport tube was i n s e r t e d through a s e t of O-rings at t h e t o p of t h e removal volume. During sampling the shipping cask w a s aligned and placed over t h e removal a r e a e A s m a l l amount of vacuum grease was smeared onto t h e lower p a r t of t h e transport tube before it w a s lowered through t h e shipping cask and i n t o the O-rings of t h e removal seal,
When t h e t r a n s p o r t tube was f u r t h e r lo-
wered i n t o Area 3A (a contaminated a r e a ) t o i n s e r t an empty capsule and again t o pick up t h e sample, t h e outer surfaces of t h e transport tube and removal t o o l became contaminated with s o l i d f i s s i o n products from t h e cap-
sule, f l o o r of Area 3, and/or manipulator,
Some of these p a r t i c l e s were
wiped o f f by t h e O-rings of t h e removal s e a l as t h e removal t o o l and transport tube were withdrawn from Area 3A; t h a t remaining on t h e removal t o o l
was wiped o f f with a damp cloth.
The shipping cask w a s gradually c o n t a d -
nated as both t h e removal t o o l and t r a n s p o r t tube were drawn i n t o it,
By t h i s process t h e removal seal becomes progressively more contambated with each sample,
During p r e - c r i t i c a l operations t h i s t r a n s f e r mechanism f o r
p a r t i c u l a t e matter was not obvious; however, after power operations had begun and especially after t h e l a t c h maintenance work, where t h e l a t c h was manually r e t r i e v e d from t h e v i c i n i t y of t h e pump and out through t h e removal area f o r r e p a i r , t h e removal area and t h e top of t h e sampler became contaminated.
gi
303
u
The adoption of t h e following procedure proved e f f e c t i v e i n preventing t h e spread of contamination during sample removal and transport throughout t h e remainder of reactor operations: (1) The top of t h e sampler w a s established as a Contamination Zone. (2) The removal t o o l was wiped w i t h a damp c l o t h during w i t h d r a w a l
from t h e removal area.
( 3 ) The shipping cask was wiped with a damp cloth p r i o r t o enclosure i n a p l a s t i c bag f o r shipment t o t h e a n a l y t i c a l laboratory.
(4)
The t o p of the sampler was wiped w i t h a damp cloth a f t e r each
sample.
( 5 ) An exhaust hood was erected near t h e removal area and p a r t i a l l y enclosed t h e shipping cask t o maintain a s l i g h t l y reduced pressure i n t h i s area s o t h a t any airborne p a r t i c l e s would be drawn i n t o t h e f i l t e r and exhaust system of t h e building.
(6) The removal seal area was cleaned periodically Control C i r c u i t r y Changes -During
LJ
w i t h damp wipes.
t h i s period f i v e changes were made
t o t h e sampler control c i r c u i t s : (1) A pressure switch was added t o prevent evacuation of Area 3A t o
more than 10 i n . of water g r e a t e r than t h e manipulator cover area.
(2) An i n t e r l o c k was added t o require t h a t both t h e operational and maintenance valves be closed before the access port can be opened.
( 3 ) A permissive switch and l i g h t were i n s t a l l e d t o indicate t h a t 1 C pressure i s equal t o o r
less than Area 3A pressur (4) A fuse w a s add
sule drive motor c i r c u i t t o protect
les from excessive currents. oss t h e two motor windings
t o l i m i t any high voltage pe
i n g s t a r t i n g and stopping. -During
t h e low-power experi-
an opening near t h e t o p and capable of nto t h e pump bowl but f o t long enough t o permit Subsequent longer assembl i e
u
obtained s a l t except one which was believed
t o have hung on t h e l a t c h stop at t h e pump bowl and did not e n t e r t h e pump. The capsules were subsequently modified t o allow them t o hang s t r a i g h t .
,
304 â&#x20AC;&#x2DC;Most of t h e samples taken a f t e r the reactor exceeded 1 MW (nuclear power) were highly radioactive and exceeded 1000 R/hr at 3 i n .
Fission
products adhering t o t h e outside of t h e capsules contaminated t h e bottom cups of t h e transport containers.
For a l i m i t e d time disposable p l a s t i c
l i n e r s i n s e r t e d i n the cups were e f f e c t i v e i n reducing t h e radiation l e v e l
i n t h e bottom portion of t h e transport tubes.
However, as more f i s s i o n
products b u i l t i n t h e f u e l and t h e i n t e r i o r of t h e sampler became more contaminated, it w a s more economical t o f a b r i c a t e disposable bottom cups of mild s t e e l r a t h e r than decontaminate and reuse t h e s t a i n l e s s s t e e l cups, The new cups were s h o r t e r than t h e previous ones and t h e p l a s t i c l i n e r w a s
used t o confine t h e capsule, wire, and l a t c h pin so t h a t t h e top p a r t of t h e transport c a r r i e r could be more e a s i l y slipped over t h i s combination t o engage the threads and double O-ring s e a l s at t h e bottom of t h e mild s t e e l cups. On one occasion while the t o p of t h e transport container was being mated with t h e bottom which contained a 50-g capsule, t h e wire on t h e capsule caught i n t h e threads of t h e mating pieces and galled before t h e two
pieces were sealed completely.
Thereafter t h e l i n e r s were lengthened t o
accommodate t h e t h e 50-g capsules. I n J u l y 1966, t h e f u e l pump w a s accidentally o v e r f i l l e d w i t h flush
salt and salt w a s pushed up i n t o t h e sampler tube approximately two f e e t above t h e f u e l pump where it froze and thus prevented t h e capsule and l a t c h from being lowered i n t o the pump bowl.
The salt w a s melted and allowed t o
drain back i n t o t h e pump by energizing a set of remotely-placed heaters i
around t h e sampler tube and t h e normal heaters surrounding t h e pump bowl. Normal sampling w a s subsequently resumed. Contamination of t h e area sGrounding t h e sampler was minimized during maintenance by designating t h e area surrounding t h e sampler a Contaminat i o n Control Zone and at times-by erecting a p l a s t i c t e n t around t h e sampler.
Repair work was done inside t h e zone with appropriate protective
clothing, gloves, and shoe scuffs.
Contamination w a s found outside t h e
zone twice and on both occasions it was a t t r i b u t e d t o contaminated shoes which indicates that t h i s type of contamination i s not r e a d i l y airborne.
305
u
18.1.3
Subsequent Operating Experience with t h e Sampler-Enricher
The major sampler problems encountered during t h e remainder of operat i o n s were caused by t h e l a t c h and/or capsule lodging at t h e entrance t o t h e sampler tube o r at one of t h e i s o l a t i o n valves immediately below Area 1 C with subsequent unreeling and tangling of' t h e cable i n Area 1 C and i n
t h e gears of t h e drive unit. Ruptures o r pin-hole leaks i n t h e boots enclosing t h e manipulator arm were a chronic source of trouble requiring periodic replacement.
Life of
t h e boots varied widely but averaged 44 sample cycles before failure.
The
cause of most of these leaks was undetermined because t h e boots could not be examined e a s i l y due t o t h e high radiation l e v e l of t h e manipulator assembly. Gradual d e t e r i o r a t i o n of t h e buffer gas s e a l s at t h e operational, maintenance and removal valves and a t t h e access port continued and on occasions t h e time required t o take a sample was lengthened because a longer time i n t e r v a l was required f o r t h e buffer pressure t o b u i l d up suff
:
W
f i c i e n t l y t o actuate t h e proper r e l a y f o r t h e next s t e p i n t h e procedure. Loss of Capsules i n t h e Pump Bowl - I n
August 1967 during what ap-
peared t o be a normal sampling operation, t h e capsule and/or l a t c h had apparently lodged at t h e maintenance valve while being lowered, causing t h e cable t o c o i l up i n Area 1 C and i n t o t h e motor drive area where it had become snarled and kinked.
The f i r s t indication of a malfunction was during
t h e capsule r e t r i e v a l cycle when t h e motor s t a l l e d w i t h
off the
8 f t 3 i n . of cable
Repeated i n s e r t s and withdrawals recovered only 5 more in.
of cable before t h e motor s t a l l e d w i t h only an inch of t r a v e l i n e i t h e r direction ( t h i s had happene
efore i n 1965 but t h e cable and capsule
were.retrieved completely If t h e capsule and c
a$ t h e sampler tube entrance be-
low Area lC, then t h e operational and maintenance valves could be closed and sampler r e p a i r s could
roceed without a r e a c t o r d r a i n ; otherwise a
drain would be necessary.
After consideration of several p o s s i b i l i t i e s , it
was decided t o disengage t h e motor drive from t h e operational valve and close t h e valve slowly with t h e hand wheel with t h e idea t h a t i f t h e cap-
kd
sule were below t h e valve, t h e cable would o f f e r enough r e s i s t a n c e t o clo-
sure t o be detected i n which case t h e valve would be reopened and t h e
306 reactor drained before repair. was p r a c t i c a l l y closed,
No resistance w a s detected u n t i l t h e valve
W
Therefore, both valves were closed and when Area
1C was opened, as expected, t h e cable was coiled up within; however t h e l a t c h , capsule, and about 6 i n . of cable were missing.
The r e a c t o r was
then drained and grappling t o o l s were fabricated of f l e x i b l e tubing and cables which could be i n s e r t e d through t h e sampler i n t o t h e pump bowl, Four of t h e f i v e t o o l s which were most successful at grasping a dummy l a t c h a r e shown i n Fig. 18.3. When t h e noose t o o l was lowered, it snared t h e l a t c h but t h e 1/32-inO s t e e l cable broke because t h e l a t c h was apparently glued i n place at t h e l a t c h stop,
with salt
The dislodging t o o l w a s then used but apparently t o no
a v a i l because a second noose grasped t h e l a t c h but could not budge it with a p u l l of 25 l b .
The pump w a s then heated u n t i l t h e l a t c h was at a tem-
perature of 700째F.
It was then l i f t e d several f e e t without d i f f i c u l t y be-
fore it lodged i n t h e tube, apparently at t h e expansion j o i n t between t h e f u e l pump and t h e sampler.
A corkscrew t o o l w a s next i n s e r t e d but t h e
l a t c h again hung on r e t r i e v a l , t h i s time a t ' t h e first bend near t h e f u e l pump. The l a t c h r e t r i e v a l t o o l which w a s designed t o f i t over t h e severed cable and s t e m and thus prevent t h e cable from jamming against t h e w a l l during r e t r i e v a l w a s successful i n p u l l i n g up t h e cable, l a t c h , and l a t c h pin,
The capsule apparently snapped i t s cable and dropped t o t h e bottom
of t h e sampler cage when t h e l a t c h came t o a sudden s t o p at t h e l a t c h stop. The dotted o u t l i n e of a capsule. i n Fig. 18.4 shows how a capsule, once detached from i t s cable, could s l i p out of t h e sampler cage and then be trapped by t h e mist s h i e l d i n t h e pump bowl. Since t h e capsule t o p i s made of n i c k e l p l a t e d mild S t e e l , a magnet i n combination with a go-gage ( F i g . 18.3) was lowered i n t o t h e pump bowl t o veri* sule.
t h a t t h e sampling tube w a s unobstructed and t o pick up t h e cap-
Capsule recovery w a s unsuccessful and since no adverse chemical o r
mechanical e f f e c t s were envisioned should t h e capsule be allowed t o remain i n t h e mist s h i e l d , f u r t h e r r e t r i e v a l attempts Were abandoned. While r e t r i e v a l e f f o r t s were i n progress, a replacement 1 - C assembly which was on hand, w a s equipped with a l a t c h made of 430 s t a i n l e s s s t e e l and with a sleeve t o prevent f'uture tangling of t h e cable i n t h e gears of the drive unit.
However, subsequent examination o f t h e o l d drive u n i t i n
a;l
307
ORNL-OWG
6?-44?08
1 3 2 -in. CABLE
CABLE SHEATH
CABLE BEING PUSHED OUT
NOOSE READY FOR FISHING DISLODGING TOOL
NOOSE TOOL
#'/46-in.
CABLE
412 in. SCHED 40 PIPE
1.375 in. DlAM
LATCH RETRIEVAL
Fig. 18.3
Tools De
GO- GAGE AND CAPSULE RETRIEVAL TOOL
Retrieval of Latch and Capsule from
308
ORNL- DWG 67-10766
OFF
,
' i,
LATCH ASSEMBLY NORMAL POSITION
LATCH STOP,
TOP OF FUEL PUMP
i
I
AMPLE CAPSULE SAMPLE CAPSULE
SAMPLE CAPSULE
SAMPLE CAPSULE LOST SAMPLE CAPSULE
INCHES
Fig. 18.4
Location of Sampler Latch and Capsules i n Fuel Pump Bowl
W
309
hii
a hot c e l l revealed t h a t t h e cable was not tangled i n t h e gears as had been assumed; t h e f a i l u r e t o withdraw o r i n s e r t was due t o a sharp k i n k i n t h e cable which became lodged i n t h e 17/32-in.
diameter channel of t h e
l a t c h positioner. When normal operations and sampling were resumed, a commercially available device was t e s t e d which was capable of sensing t h e l a t c h (now made of magnetic material) inside t h e sampler tube.
A design was prepared t o mount
t h i s proximity switch about 10 i n . below t h e maintenance valve t o detect
the l a t c h as it i s lowered during normal sampling; however, before it could be i n s t a l l e d , t h e l a t c h again lodged i n t h e sampler tube without detection
i n March 1968.
The usual 17 f t 5 in. of cable was reeled o f f and w a s tan-
gled i n Area 1C. the reel.
A s it was rewound, t h e motor s t a l l e d w i t h 13 f t 5 i n . o f f
When repeated withdrawal attempts f a i l e d , t h e r e a c t o r was drained
and the capsule access port was opened.
The capsule was v i s i b l e through t h e
lower corner of t h e opening and t h e l a t c h could be seen i n t h e back of t h e chamber w i t h many loops and c o i l s , some of which appeared t o extend down i n t o t h e sampler tube.
Rather than r i s k c u t t i n g t h e cable, t h e i s o l a t i o n
valves were l e f t open; however, when an attempt was made t o lift t h e caps u l e out of t h e chamber, t h e manipulator brushed some of t h e c o i l s causing them t o spring out through t h e port and p u l l t h e capsule wire from t h e manipulator fingers.
The capsule dropped i n t o t h e sampler tube and t h e l a t c h
tipped over, thus releasing t h e key; t h e capsule and key then disappeared down t h e sampler tube i n what appeared t o be a bottom-up position. A f t e r several unsuccessful attempts were made t o r e t r i e v e t h e capsule
by lowering magnets of various s i z e s i n t o t h e pump bowl, t h e reactor was f i l l e d w i t h f l u s h salt.
A 50-g capsule ( 5 in. long w i t h an opening 4-1/8
i n . from t h e bottom of t h e
apsule) w a s lowered i n t o the. pump but came up
empty; salt droplets clinging t o t h e outside indicated t h a t t h e capsule had been only h a l f submerged.
Later a 10-g capsule d i d c o l l e c t a sample.
I n t h e meantime a f u l tube, mist s h i e l d , and s
e p l a s t i c and metal mock-up of t h e sampler age was constructed and used t o s e l e c t t h e
b e s t t o o l s and magnets which could be lowered v i a t h e sampler i n another attempt t o r e t r i e v e t h e capsule.54 proved t o be 1 / 2 and 3/4-in.
bi
The simplest and most e f f e c t i v e t o o l
diameter Alnico-5 magnets.
When t h e f l u s h
salt was drained and r e t r i e v a l e f f o r t s resumed, sounds from a contact
- I
310 microbhone on t h e f i e 1 pump indicated t h a t t h e 1/2-in.
an object a few inches and then dropped it.
d i a , magnet l i f t e d
U
A f t e r mhny attempts, t h e mag-
net came up with an object which l a t e r proved t o be t h e corroded top of t h e old capsule.
Further e f f o r t s with e i t h e r magnet were unsuccessful and t h e
second capsule w a s abandoned. During t h i s shutdown, t h e proximity switch was i n s t a l l e d about 4 i n . below t h e maintenance valve and t h e sampling procedure w a s modified t o s t i p u l a t e t h a t if t h e switch is not actuated when t h e capsule position indicator shows t h a t s u f f i c i e n t cable has been r e e l e d o f f t o reach t h e proximity switch, t h e drive w i l l be stopped.
Also a vakiac was added t o t h e
drive motor c i r c u i t so that t h e operating voltage could be changed t o 80 v o l t s t o lessen t h e p o s s i b i l i t y of seriously kinking t h e cable, During t h e next s t a r t u p , with f l u s h salt i n t h e r e a c t o r , salt w a s trapped i n a 10-g capsule but none trapped i n a 50-g capsule which requires
an immersion depth of 4-1/8 i n . t o t r a p a sample.
However, a 30-g capsule
requiring only 2-1/2 i n . of immersion did t r a p a sample.
These results
agree with t h e measurements taken during capsule r e t r i e v a l which indicated that t h e abandoned capsule w a s 1-3/4 in. above t h e bottom of t h e cage.
bd
However, with f u e l i n the r e a c t o r , t h e a c t u a l l e v e l i n t h e pump bowl was somewhat higher than t h a t indicated because of t h e high void f r a c t i o n of t h e l i q u i d i n t h e pump bowl.
Consequently no problems were encountered
taking 50-g f u e l samples throughout t h e remainder of operations except during the l a t t e r p a r t of 1969 when t h e f u e l pump speed was lowered t o
980 rpm.
A t t h i s speed the f u e l salt void f r a c t i o n i s approximately equiv-
a l e n t t o that f o r f l u s h salt at normal pump speed.
The failure t o t r a p a
50-g sample under these conditions indicated t h a t t h e position of t h e abandoned capsules had not changed after approximately one year of operation, I n October 1969, a serious cable tangling problem w a s averted by t h e use of t h e proximity switcho When t h e switch f a i l e d t o actuate a f t e r 3 f t 10 i n , of cable was paid out during an otherwise normal capsule i n s e r t , t h e cable w a s r e t r i e v e d u n t i l t h e position i n d i c a t o r indicated t h a t a l l of t h e cable was back on t h e takeup reel and t h e capsule was f u l l y withdrawn i n t o i s o l a t i o n chamber 1 C .
The operational and maintenance valves were then
closed and t h e access port was opened.
The cable was indeed fully re-
t r i e v e d as indicated and t h e capsule was f u l l y v i s i b l e but was lodged
W
311
diagonally between t h e ledges of t h e doorway t o t h e chamber.
Apparently
t h e capsule had lodged at t h e sampler tube entrance o r a t one of t h e isolat i o n valves and as t h e cable w a s paid out, it somehow encircled t h e capsule because t h e capsule w a s l i f t e d and relodged a t a higher position (during cable r e t r i e v a l ) than it w a s at t h e s t a r t of t h e sampling procedure.
The
capsule was then lowered t o i t s normal position and a l l subsequent cable operations proceeded without interruptions. Wiring Fault
- I n November of 1967 as a sample w a s being withdrawn,
the 0.3-amp fuse i n t h e drive motor f a i l e d .
Subsequent t e s t s showed t h a t
t h e i n s e r t mode was normal at 0.2 amp but during w i t h d r a w a l , t h e current increased t o 1.0 o r 1.5 amp, indicating leakage'to ground.
During f u r t h e r
t e s t s , t h r e e c i r c u i t s opened, disabling t h e i n s e r t and w i t h d r a w c i r c u i t s and t h e upper l i m i t switch.
After t h e erection of a t e n t and other meas-
ures t o prevent t h e spread of contamination, a 3-in. dim hole w a s sawed
i n t h e cover p l a t e of Area 3A d i r e c t l y above t h e cable penetration i n t o the inner box (Area 1C).
b,
The f a i l u r e was found where t h e wires were bent
back against t h e s i d e of t h e plug on t h e lower end of t h e cable between t h e inner (1C) and outer (3)containment boxes.
When a temporary con-
nection indicated t h a t everything within Area 1C was operable, t h e sample was r e t r i e v e d and t h e i s o l a t i o n valves were closed.
The damaged section
was abandoned and a new cable was i n s t a l l e d having a penetration through a 4-in. -
pipe cap welded over t h e sawed hole i n t h e t o p cover.
While t h e pipe cap w a s being welded i n place, a heavy current evi-
dently went t o ground through an adjacent penetration, destroying t h e plug and receptacle. tration.
Repairs we
As a result o f t h
was added t o each of t h e t h
made by c u t t i n g out and replacing t h i s peneexperience, an i s o l a t i o n transformer and fuse drive u n i t cables which allows one ground
without interference with operation.
After t h e r e p a i r s , a l l c i r c u i t s were
operable except t h e upper l i m i t switch which stops t h e drive motor when t h e l a t c h reaches t h e l a t c h stop.
Since t h e motor and c i r c u i t can e a s i l y w i t h -
stand blocked-rotor current, t h e upper l i m i t switch w a s bypassed and therea f t e r t h e motor w a s turned o f f when t h e position indicator indicated t h a t the l a t c h was f'ully withdrawn. ..
LJ
'
..
'RepalP 'df Cable Drive Gear
- I n December 1968 during an
attempt t o
remove a capsule from t h e l a t c h , it w a s found t h a t t h e cable could be
312 pulled o f f t h e r e e l while t h e drive motor w a s stationary.
The t e n t a t i v e
diagnosis was t h a t one of t h e p a i r of drive gears was s l i p p i n g on i t s shaft.
I n order t o gain access t o t h e drive u n i t , it w a s necessary t o re-
move t h e s h i e l d blocks over t h e reactor c e l l which, i n t u r n , required t h a t t h e f u e l be drained.
A f t e r a temporary containment enclosure and contami-
nation zone had been s e t up around t h e sampler, t h e containment box, lC, containing t h e drive u n i t was disconnected, l i f t e d , and a 3-in. hole sawed through t h e s i d e adjacent t o t h e gears.
One gear w a s found t o be loose be-
cause i t s two setscrews had come loose.
The gear was repositioned and tack-
welded on i t s shaft and t h e other gear w a s fastened with jam screws on t o p of i t s setscrews. stalled,
A patch w a s then welded on t h e box and t h e u n i t rein-
The work was done without excessive exposure of personnel despite
t h e high radiation l e v e l s (10 R/hr at 12 i n . from t h e box) by use of shielding and extended t o o l s designed e s p e c i a l l y f o r t h e job. Since other parts of t h e sampler were e a s i l y accessible during t h i s shutdown, t h e manipulator hand was repaired, t h e illuminator port and viewing window l e n s were replaced because of discoloration by r a d i a t i o n , and an imperfect s e a l i n the removal valve was replaced. Vacuum F+mp Problems
- The sampler was
LJ
equipped with two vacuum pumps.
Vacuum pump No. l w a s used t o evacuate contaminated areas such as containment areas 1C.and 3A and discharged i n t o t h e a u x i l i a r y charcoal bed; vacuum pump No. 2 was used t o evacuate non-contaminated areas such as t h e manipul a t o r cover and t h e plenum between t h e manipulator boots and discharged int o t h e containment air system f i l t e r s and t o t h e stack, During t h e July 1966 shutdown, t h e o i l l e v e l s i n t h e pumps were checked and a small amount of o i l w a s added. I n January 1969 t h e r e w a s an a c t i v i t y r e l e a s e t o t h e stack (~0.08mc) which was a t t r i b u t e d t o a gas leak around t h e s h a f t of vacuum pump No. 1. The pump w a s replaced because t h e o i l s e a l at t h e s h a f t could not be repaired without extensive decontamination procedures,
Four months l a t e r ,
repeated vacuum pump motor outages (due t o overload) were apparently caused by low o i l level.
On one inspection, t h e pump w a s p r a c t i c a l l y empty of o i l
with no evidence of external o i l leaks and was consequently r e f i l l e d t o t h e proper level.
On t h e next inspection, t h e pump contained approximately
crii
313
2 q t s more than t h e normal inventory.
drained back i n t o t h e pump.
Apparently t h e previously l o s t o i l
High pressure (>13p s i a ) i n Area 1C w a s nor-
mally vented i n t o t h e auxiliary charcoal bed through a l i n e bypassing t h e pump.
Apparently high pressure w a s vented through t h e vacuum pump instead,
thus carrying over a l a r g e portion of t h e o i l i n t o t h e holdup tank and offgas l i n e above t h e pump discharge.
After t h e o i l l e v e l w a s readjusted,
pump operation returned t o normal and t h e l a s t o i l check w a s normal. Heated Shipping Cask
-A t
temperatures l e s s than 400째F, radiation
from f i s s i o n products can produce f r e e f l u o r i n e i n frozen f u e l s a l t .
To
minimize f l u o r i n e production and thus avoid t h e e f f e c t s of fluorine on t h e r e s u l t s of oxide and t r i v a l e n t uranium analyses, a shipping cask w a s fabricated which maintained t h e f u e l sample at approximately 500'F
from t h e time
it was removed from t h e sampler u n t i l it was unloaded a t t h e a n a l y t i c a l laboratory.
The cask used molten b a b b i t t both as shielding and as a heat
reservoir.
Built-in e l e c t r i c heaters melted t h e babbitt before t h e sample
was loaded and t h e heat of fusion maintained t h e sample a t a r e l a t i v e l y constant temperature f o r about 9 hours during shipment and storage.
W
For
t h e elevated temperature, O-rings made of Viton A were s u b s t i t u t e d f o r t h e standard neoprene O-rings i n t h e transport tube.
I n addition t o t h e double
O-ring seal provided by t h e transport tube, t h e transport tube w a s capped inside t h e cask during shipment. A l l samples were unavoidably cooled t o room temperature i n t h e sampler
where they were sealed i n t h e transport tube a t atmospheric pressure.
When
t h e transport tube was i n s e r t e d i n t o t h e heated cask, t h e pressure i n t h e tube increased t o approximately 10 p s i g because of t h e temperature increase. On one occasion when t h e cap w a s removed from t h e cask, airborne a c t i v i t y was released from t h e cask cavity i n t o t h e v e n t i l a t i o n system of t h e unloading s t a t i o n at t h e a n a l y t i c a l laboratory.
Subsequently t h e s e a l i n g
surfaces of t h e t o p p a r t of t h e transport tube were found t o have been eroded by repeated decontamination procedures t o such an extent t h a t t h e double O-ring s e a l s were i n e f f e c t i v e .
After examination of t h e sealing
surfaces of t h e remaining f i v e transport tube tops and reevaluation of t h e decontamination c o s t s , it w a s decided t o discard t h e reusable s t a i n l e s s s t e e l transport tubes and use "throw-away" transport tubes made of mild
tyj
314
s t e e l f o r all subsequent samples.
A valve was l a t e r added t o t h e cap of
U
t h e heated cask so t h a t any future pressure buildup i n t h e cask cavity could %e released t o a charcoal f i l t e r before cap removal. Other Capsules and Samples -The
s i z e of t h e capsules t h a t could be
accommodated by t h e sampler was l i m i t e d t o 0.75 inches i n diameter and
6-1/4 inches i n length.
Approximately 1 / 4 inch g r e a t e r diameter could be
accommodated by t h e sampler but not by t h e transport tube.
Four of t h e
various types of capsules which were lowered i n t o t h e pump bawl a r e shown i n Figures 1 8 . 5 ~through ~ 18.5d, e i t h e r salt 'or gas,
The capsule i n Fig. 18,5a was used t o t r a p
The capsule was first evacuated and sealed with a mea-
sured amount of salt similar t o t h e fuel.
On entry i n t o t h e f u e l pump, t h e
s e a l i n g salt melted and salt o r gas (depending on t h e capsule elevation i n t h e pump bowl J u s t p r i o r t o salt l i q u e f a c t i o n ) was drawn i n t o t h e capsule, When t h e beryllium addition capsules were withdrawn, they were encrusted with s o l i d s as sham i n Figure 18.5b.
A s a r e s u l t , t h e capsules
were d i f f i c u l t t o load i n t o t h e bottom p a r t of t h e t r a n s p o r t tubes even though t h e p l a s t i c l i n e r was removed.
Also t h e enriching capsules were
sometimes d i f f i c u l t t o load because of t h e i r length (6-1/4 i n * ) * The ca-
b
b l e s of these long capsules were occasionally caught i n t h e threads of t h e mating parts of t h e transport container and were d i f f i c u l t t o unload at t h e hot c e l l o The r a d i a t i o n levels of t h e beryllium capsules and t h e empty enriching capsules were more than four times higher than t h e standard 10-g sample <
when delivered t o t h e a n a l y t i c a l laboratory, .-
Miscellaneous Sampler Problems
- During t h e l a t t e r p a r t
of 1966, a
c o t t e r key had come out of t h e t o p of t h e lower hinge p i n on t h e access port door.
The hinge pin had then worked out of t h e t o p of t h e hinge,
Using only t h e manipulator, t h e pin w a s repositioned i n t h e hinge, a new c o t t e r key was i n s e r t e d i n t h e pin and t h e key w a s spread t o lock it i n place,'
During t h i s r e p a i r , t h e lower key w a s dislodged and was a l s o re-
paired. Operation o f t h e pheumatic clamps on t h e access port door occasiona l l y r e s u l t e d i n gaseous f i s s i o n products being vented through t h e operator discharge l i n e s .
Therefore a small charcoal f i l t e r was added t o t h e op-
e r a t o r discharge l i n e s t o prevent t h i s a c t i v i t y from being discharged t o
W
315
c
J
ONE qe-in. HOLI THREE '4in. HOi THREE (/e-in. HO THREE '/3*-in. H(
'4in.
ORNL-DWG 69-!3!05
BETWEEN WINDOWS
SOLID NICKEL SPACER
34-in. OD x '/=-in. WALL NICKEL TUBING
ORNL-OW0 61-4184
-WINDOWS
STAINLESS STEEL CABLE
ORNL-rn 67-4785
5/a-in. OD I3-in. TALL GRAPHITE CYLINDER
c
-5/4-111.
1
TOPOF
OD NICKEL
1
VOLUME x)CC---,
h 4
WELU MIST SHlELDs, SHOULDER FOR GRA CYLiNDER TO SIT
CuO IN STAINLESS STEEL CLOTH BASKET
34-in.LOD Ni TUBE
'/a-in.-DIAM PERFORATIONS ABOUT EVERY '4in, 3 Be RODS '&-in. OD x
1 INLET I
/
L'2seF4\
NICKEL CAPILLARY
P Fig. 18.5a Freeze Valve Capsule
SECTION 6 - B
Fig. 8.5b Nickel Cage Beryllium Rod Assembly a f t e r Exposure t o MSRE Fuel S a l t for 10.5 hr
Fig. 1 8 . 5 ~Hydrocarbon Analyzer ,
c ,
c .g. 18.5d Surface Tension Capsule
316
t h e atmosphere via the stack.
Also t h e vent valves from t h e gas operators
were kept closed except during openings and closures of t h e access door. The neoprene seal at t h e access door continued t o d e t e r i o r a t e .
In
April of 1967, t h e increased b u f f e r gas leakage across t h e inner seal was countered by lowering t h e b u f f e r pressure requirement i n t h e s a f e t y i n t e r lock system from 50 t o 40 psia.
I n J u l y 1967, t h e s a f e t y switch s e t t i n g
w a s lowered t o 35 p s i a and t h e helium flow rate t o t h e b u f f e r seal was a l s o
increased so t h a t a s a t i s f a c t o r y b u f f e r pressure could be maintained. During t h e intensive sampling of t h e l a s t run, t h e access p o r t door frequently required repositioning w i t h t h e manipulators before t h e second set of clamps were actuated t o lock t h e door and obtain an adequate seal. The cause of t h e d e t e r i o r a t i o n of t h i s seal has not been determined.
The r a d i a t i o n l e v e l of some of t h e samples taken w a s i n excess of 1000 R/hr
at 3 inches and during boot replacements when t h e sampler was empty, t h e radiation l e v e l i n s i d e Area 3A was approximately 100 R/hr. Although some long-term d e t e r i o r a t i o n may be a t t r i b u t e d t o r a d i a t i o n damage, more l i k e l y t h e reduction i n buffer pressure during t h e l a s t two runs can be a t t r i b u t e d t o t h e f a c t t h a t t h e l e f t center h u - v i s e clamp w a s loose and thus ineffec-
3
t i v e i n i t s closed position. The f l e x i b l e containment membranes (boots) on t h e manipulator continued t o be a chronic source of trouble.
Generally, when a leak developed,
t h e sampling cycle would be completed before r e p a i r s were made; however, during t h e remainder of t h e sampling cycle, containment was abetted by t h r o t t l i n g t h e boot evacuation valve t o assure t h a t any leakage would be i n t o t h e plenum between t h e boots and through t h e vacuum pump t o t h e s t a c k r a t h e r than t o t h e work area.
After f u l l power w a s achieved, replacement
of t h e boots became somewhat more involved because t h e manipulator assembly was quite contaminated.
During a boot replacement i n e a r l y 1967, t h e radi-
a t i o n l e v e l of t h e manipulator arm was 300 R/hr a t 3 i n .
It was l a t e r de-
contaminated and saved f o r possible f u t u r e use. Also while sampling during e a r l y full-power operations, t h e r a d i a t i o n l e v e l at t h e operating area increased t o approximately 30 mR/hr. The radia t i o n l e v e l was reduced by an order of magnitude by changing t h e sampling procedure so t h a t a purge of helium from Area 1C t o t h e pump bowl w a s
9
317 maintained when t h e operational and maintenance valves were open.
Also on
r e t r i e v a l i n t o Area lC, t h e samples were purged f o r 1-1/2 h r t o t h e off-gas system before f u r t h e r removal operations were continued.
On two occasions during t h e routine t r a n s f e r of capsules between Areas 1C and 3,a capsule was inadvertently dropped and r o l l e d out of sight and out of manipulator range i n Area 3A.
These were l a t e r found and
recovered by t h e use of a mirror and a grappling fork which were lowered through t h e removal valve. During t h e l a t t e r p a r t of Run 19, routine sampling was suspended f o r approximately four days and t h e sampler w a s adapted t o accommodate a collimator and a germanium c r y s t a l detector atop t h e sampler where t h e c a r r i e r cask i s normally positioned.
A p l a s t i c plug i n t h e removal area permitted
t h e unobstructed view i n t o Area 3A where samples r e t r i e v e d f r o m t h e f u e l pump were positioned.
Gamma-ray spectrometry data were then collected on
short-lived fâ&#x20AC;&#x2122;uel f i s s i o n products within 50 minutes of t h e i r removal from t h e c i r c u l a t i n g salt stream.
bi
18.1.4
Discussion and Conclusions
The main operational problems encountered w i t h t h e sampler-enricher r e s u l t e d from 1) manipulator boot f a i l u r e , 2 ) lodging of t h e capsule o r l a t c h at t h e sampler tube entrance, operational valve, o r maintenance valve with subsequent cable tangling, and 3) b u f f e r s e a l leaks at t h e operation and removal valves and a t t h e capsule access door. . The cause of some of t h e boot f a i l u r e s were not determined because t h e leaks were small and t h e actions which produced t h e ruptures were not always coincident with t h e discovery of t h e leak.
Also t h e boots were not
examined i n d e t a i l because of t h e high r a d i a t i o n l e v e l of t h e boots and manipulator assembly.
It i s
elieved t h a t t h e boot failures which could
not be a t t r i b u t e d t o any s p e c i f i c action occurred during t h e capsule loading operation where t h e manipulator i s r e t r a c t e d and pushed downward and
/
t o t h e l e f t i n order t o load t h e sample i n t o t h e l i n e r of t h e transport tube bottom.
A t times ( a
sampler operator had t o r e s
invariably w i t h t h e long capsules), t h e
o p u l l i n g t h e manipulator w i t h both hands
i n order t o r e t r a c t it far enough t o i n s e r t t h e capsule and l a t c h pin i n t o
u
the liner.
Had t h e capsule loading s t a t i o n been mounted t o t h e l e f t of
-
318 -
t h e manipulator c e n t e r l i n e , it could have been located c l o s e r t o Area 1 C
LJ
Another possible a l t e r n a t i v e
and thus require l e s s manipulator t r a v e l .
would be t o f a b r i c a t e t h e boots w i t h fewer convolutions; however, t h e convolution width would have t o be increased t o maintain t h e same extended length,
The r e t r a c t e d thickness of t h e boots would then be decreased per-
mitting t h e manipulator t o be pulled back f a r t h e r than previously. Why and where t h e capsule o r l a t c h occasionally lodged i n t h e sampler tube during capsule i n s e r t i o n has not been s a t i s f a c t o r i l y answered.
How-
ever, t h e use of t h e proximity switch minimized t h e p o s s i b i l i t y of serious cable tangling problems.
Perhaps two a d d i t i o n a l detection devices mounted
above and below t h e operational valve would have determined t h e exact trouble spot e The buffer seal leak at the access port door started t o increase i n
January 1967 and continued t o increase slowly f o r t h e remainder of E R E operations,
Perhaps some of t h i s d e t e r i o r a t i o n can be a t t r i b u t e d t o radiThe r a d i a t i o n l e v e l i n Area 3A was
ation damage t o t h e neoprene gaskets. 10HOO R/hr during intensive sampling.
Also, t h e access door w a s always
i n t h e closed position except during capsule t r a n s f e r ; t h e r e f o r e t h e neoi
prene w a s almost always compressed which oveP t h e years has probably caused
it t o y i e l d t o a thinner protrusion.
I n addition, t h e s t e l l i t e p l a t e s on
t h e access door appear t o be gouged s l i g h t l y by the s t e l l i t e - t i p p e d Knu-vise clamps,
ATSO the l e f t center clamp w a s discovered t o be loose i n i t s closed
position during t h e last run.
Therefore, it i s believed t h a t ordinary wear-
and-tear of these components was t h e primary cause of seal leakage.
The
beating t h a t t h e door took could possibly have been lessened by a flow res t r i c t o r t o t h e clamp actuating cylinders, The top s e a l s of t h e buffered valves developed leaks whereas t h e bottom seals apparently suffered only slight d e t e r i o r a t i o n which indicates t h a t p a r t i c u l a t e matter which occasionally drop on t h e valves w a s t h e cause
of seal deterioration.
Consequently t h e volumes above these valves showed
a gradual increase i n pressure and required constant venting.
desirable of these leaks was t h a t at t h e removal valve.
The l e a s t
On removal of a
sample, t h e transport tube and removal t o o l which a r e withdrawn from Area 3A a r e locked i n position above t h e removal valve and t h e removal valve i s then closed.
If an appreciable delay i s encountered between t h e
u
319
time t h a t t h e valve i s closed and t h e time t h a t t h e transport tube i s w i t h drawn f r o m t h e removal s e a l i n t o t h e shipping cask, appreciable pressure (due t o s e a l leakage) can b u i l d up i n t h e removal a r e a and blow contamination out of t h e removal area t o t h e top of t h e sampler and shipping cask when t h e transport tube c l e a r s t h e two O-rings of t h e removal s e a l . Swabbing the removal area w i t h wet wipes on a monthly basis g r e a t l y reduced t h e contamination problem encountered i n t h i s area.
During intensive sam-
pling, t h i s area should be cleaned more frequently. The s p e c i f i c a t i o n s f o r t h e e l e c t r i c a l i n s u l a t i o n on t h e lead dires near t h e penetrations a t Areas 1C and 3A should be reviewed.
A failure
a t t h e connector pins occurred i n May 1966 r e s u l t i n g i n open c i r c u i t s i n t h e i n s e r t and withdraw modes and i n t h e upper l i m i t l i g h t . During in, spection of t h e 1C assembly which was removed i n August 1967, t h e i n s u l a t i o n on .the lead wire was found t o be very b r i t t l e and most of it w a s unavoidably: stripped during removal of t h e drive unit.
I n November 1967 an
e l e c t r i c a l fault occurred i n t h e same c i r c u i t s as those i n May 1966 except t h e f a i l u r e occurred i n t h e wires between t h e two containment boxes.
It
i s not c l e a r whether these f a i l u r e s can be a t t r i b u t e d t o radiation damage
t o the i n s u l a t i o n o r possibly t o excess temperature caused by resistance heating i n t h e lead wire due t o blocked r o t o r current (0.3 amps) i n combination w i t h t h e heat supplied by t h e illuminator lamp. To increase t h e precision of uranium i s o t o p i c analyses, provisions
should be made t o clean Area 3A periodically, especially a f t e r enrichments a r e made.
During t h e enriching procedure, small q u a n t i t i e s of enriching
salt could e a s i l y adhere t o t h e manipulator and/or be 3arred loose from t h e capsule and f a l l t o t
oor of Area 3A where a l l capsules a r e l a i d
temporarily while swappin
s u l e s t o and from t h e l a t c h and t h e transport
tube.
A very small amount of enriched salt adhering t o a sample submitted
f o r i s o t o p i c analysis could d i s t o r t t h e i n t e r p r e t a t i o n of t h e r e s u l t i n g analyses.
320
18.2
Coolant Sampler
In principle, sampling of t h e coolant system was t h e same as t h a t f o r the f u e l system.
However, t h e induced a c t i v i t y i n t h e coolant salt was
very short-lived and there w a s no residual radioactivity i n t h e sample; therefore, no shielding was required and only single containment was necess a r y during transport.
Direct manipulation of t h e sample by one hand
through a glove p a r t simplified sampler construction and operation. 18.2.1
Description o f t h e Coolant Sampler
The coolant salt sampler, shown i n Fig. 18,6 consisted e s s e n t i a l l y
Two b a l l
of a dry box connected t o t h e coolant pump by a transfer tube.
valves i s o l a t e d t h e sampler from t h e pump bowl which has a mist s h i e l d and capsule guide similar t o t h a t i n t h e f u e l pump.
The 1-1/2 i n . Sched-40
t r a n s f e r tube was similar t o t h a t used f o r t h e f u e l system except t h a t no expansion j o i n t was required because t h e coolant pump position d i d not change with system temperature as did t h e fuel pump.
The sampler was pro-
vided with l i g h t i n g and viewing ports and t h e necessary capsule t r a n s f e r s inside t h e sampler were done with one hand through a glove port.
Sliding
gas seals at t h e top of t h e sample c a r r i e r along with a vacuum pump and a helium supply were used t o reduce oxygen contamination t o a minimum during sampling and during sample transport t o t h e a n a l y t i c a l laboratory. A check-list type procedure along w i t h a system of locks and keys f o r
valve manipulation w a s used t o assure proper sequence of operations, key was used t o unlock one valve, which could then be opened.
A
When t h e
valve w a s opened, it also locked the first key i n position and released a second key,
Removal of t h e second key (which must be removed t o unlock
t h e ne& valve i n sequence) assured t h a t t h e first valve was locked i n the proper position, 18.2.2
t
Operating Experience with t h e Coolant System Sampler
The coolant salt sampler operated r e l a t i v e l y trouble-free throughout MSRE operations.
Only 81 coolant system samples were taken as opposed t o
745 cycles f o r t h e f u e l system.
321
U
F i g . 18.6
Coolant Salt System Sampler
1
322 Early i n 1966 during one sampling operation, t h e l a t c h f a i l e d t o s l i d e
No reason f o r t h e d i f f i c u l t y could be However, the outside diameter of t h e l a t c h w a s reduced and no simi-
f r e e l y through t h e t r a n s f e r tube. found,
lar t r o u b l e w a s reported. Also i n the first p a r t of
ig66, one pin i n an i n d i c a t o r
shorted out at t h e penetration during a sampling cycle.
light circuit
The receptacle w a s
removed and a new one welded i n place. During t h e same period, t h e leak rate f r o m t h e lower seal of the removal valve increased.
A f t e r t h e valve w a s disassembled, cleaned, and re-
assembled, the s e a l w a s s a t i s f a c t o r y ; however, after approximately eight additional samples, t h e seal leakage again increased and was corrected by replacing the valve seals.
Also an extra washer w a s added t o the valve t o
permit more mechanical pressure t o be applied t o t h e seals.
Thereafter the
coolant sampler operated trouble-free f o r t h e f i n a l t h r e e years of MSRE operation.
18.2.3
Conclusions and Recommendations
From an operational viewpoint, two improvements may be worth considering: (1) The glove port cover should be hinged t o swing out horizontally.
On several occasions when t h e clamps w e r e removed, t h e glove port cover slipped f r o m t h e operator's hand and dropped f r e e l y t o i t s stopped position.
I n doing so, it could have pinched a bundle of nearby insulated
electrical wires.
(2) The sample c a r r i e r should be lengthened approximately two inches t o accommodate t h e longer capsules which were used during t h e l a s t run.
323
u
19.
CONTROL RODS
M. Richardson ~
19.1
Description
Three control rods were used i n the MSRE core vessel.
The rods were
not exposed d i r e c t l y t o t h e f u e l salt but were contained i n t h r e e "blind well" thimbles which were located i n three of t h e four channels surrounding t h e v e r t i c a l centerline of t h e graphite core.
The thimbles were straight
within the core vessel, but above t h e reactor access flange they were offset
- using two 16-in.
containment standpipe. tion.
radius bends t o provide space f o r t h e core sample A r o l l e r was located at each bend t o reduce f r i c -
The control rod drive mechanisms were v e r t i c a l l y positioned around
t h e standpipe.
The control rods were fabricated of flexible metal hose
w i t h hollow center c y l i n d r i c a l poison elements (38 per rod) beaded over
t h e lower 60 i n .
Rod motion w a s supplied by motor driven 1/4-in.
chain and sprocket drive units.
ki
roller
Included i n t h e reverse locking power
t r a i n i s an overrunning clutch which permits power transmission i n the inA magnetic clutch w a s used t o release each rod from
s e r t d i r e c t i o n only. '
t h e power t r a i n gearing and permitted f r e e f a l l of t h e rod i n t o t h e core at k0.4 g. Continuous indication of each rod position w a s provided by two synchro torque transmitter-receiver p a i r s , one "fine" and one "coarse."
A potenti-
ometer provided position indication f o r s a f e t y interlocks i n t h e reactor f i l l circuits. The upper and lower limits of rod t r a v e l were controlled by four (2 .
upper and 2 lower) mechanically actuated switches. '
The poison elements were cooled by c e l l a i r (95%N2) from t h e compon-
ent coolant system.
This entered t h e upper end of t h e hollow control rod
and exhausted r a d i a l l y i n t o t h e thimble from a nozzle at t h e lower end.of the rod. Changes i .,n pressure drop as t h i s nozzle movedthrough a flow r e s t r i c ; t o r b u i l t i n t o the bottom of t h e thimble provided a rod position reference point ( f i d u c i a
ero).
Comparison of t h i s with t h e rod position as indi-
cated by t h e synchro rod position instrumentation w a s used t o monitor s t r e t c h i n g of t h e rods and other malfunctions.
324
c-)
The rod drive units were i d e n t i c a l as were t h e rods except f o r a slight variation i n lengths of the individual rods due t o t h e differences
of the thimble offsets. or trouble-shooting.
This allowed interchange of components f o r repairs
For i d e n t i f i c a t i o n of a rod assembly, it was neces-
s a r y t o specify the thimble number (T-1,
o r R-4)
R-2, R-3,
11-2, o r T-31,
and the drive rimer (V-1, V-2, V-3,
the rod number (R-19
o r V-4).
!he rod
assembly connected t o t h e servo controller was considered t o be the regul a t i n g rod and the other two rods were t h e shim rods.
The rod assemblies
could be used interchangeably as shims o r regulating rods by shi-fting t h e out-of-cell
rod drive disconnects.
Any one rod was capable of taking the
reactor subcritical. 24 The rods and drive assemblies were not designed f o r in-cell maintenance,
Repairs on t h e drive units were done a f t e r disassembling the rod
from the drive and removing the drive unit from t h e reactor c e l l by remote maintenance methods,, The control rod shock absorbers used t h e general principles of a ty-pic a l hydraulic shock absorber but differed i n t h a t t h e working "fluid" con-
sisted of 3/32-in.-diameter
steelballs.
A t t h e end of a scram, t h e bottom
face of the shock absorber cylinder struck permanently-mounted s t e e l blocks which were bolted t o t h e housing.
The shock absorber plunger, t o which the
control rod was attached, continued t o move downward and was decelerated by the forces developed by t h e buffer springs and by the flow of .the s t e e l b a l l s around a knob on the plunger.
bers were adjusted t o ~
The stroke length of t h e shock absor-
3 inches ~ 5 f o r each rod, 19.2
I n i t i a l Testinq
The rods were i n s t a l l e d i n the reactor vessel i n January of 1965.
The
l i m i t switches were adjusted t o coincide with t h e rod position indicators on the main console and e l e c t r i c a l continuity t e s t s were performed on all drives p r i o r t o assembly t o t h e control rods.
Satisfactory anibient tem-
perature performance was checked with t h e r e s u l t s given i n Table l g o l 0 /
325
Table 19.1
I n i t i a l Tests of Control Rod Assemblies Spare Components
Thimble No.
T-1
T-2
T-3
Rod No.
R-1
R-2
R- 3
Drive No.
v-3 0.6
v-2
v-1
Shock stroke, inches
0.59 2.85
0.6 0.58 3.6
0.6 0.58 3.7
Scram time, secs
0.835
0 752
0.775
3.8
3.9
3.9
---
0.53
0.52 50.9
Motor current w i t h d r a w Motor current i n s e r t
A i r flow, rod
- amps
- amps
- scfln
Rod t r a v e l speed, in./sec
0.53
N l rod t r a v e l , inches
50.91
50.95
50.95
Fiducial zero, inches
1.4
1.4
1-35 .
19.3
1
0.53
~
R-4 v-4 0.6 0.59
3.0
---
Periodic Testing
During reactor operation and especially a f t e r power operations commenced, it w a s not possible t o run extensive t e s t s on t h e rods.
However, it w a s necessary t o do s u f f i c i e n t t e s t i n g t o assure t h a t t h e rods would scram i f needed.
It was a l s o important t o know t h a t t h e rod position indicators were functioning properly as these were used i n making r e a c t i v i t y balances. To accomplish t h i d t o aid i n a n t i c i p a t i n g necessary maintenance , t h e following periodi A.
Rod Scrams -Scram
s t s were made,
I
t e s t s were routinely conducted t o determine
rod scram times before a f i e 1
11 and before c r i t i c a l operation.
Each
rod w a s withdrawn t o a height of 50 i n .
The magnetic clutch current was broken by t r i p p i n g t h e manual scram switch. The rod would f r e e - f a l l t o t h e fully-inserted position. The 50-in. drop w a s e l e c t r o n i c a l l y timed f r o m t h e moment of release t o t h e lower l i m i t switch.
The maximum permissible scram
time w a s 1.3 sec/50 i n . rod scram t e s t s could not be made during B. Rod Exercise -Since nuclear operation, each rod w a s exercised d a i l y t o demonstrate t h a t they
would move on command and a l s o t o f l e x t h e metallic hose.
W
The f l e x i b l e
hoses tended t o s t i f f e n with time i f allowed t o remain i n one position f o r Observation by t h e reactor operator of t h e rod-position
prolonged periods.
indicators during t h e exercise ascertained t h a t t h e synchros were working properly and t h a t t h e rods operated f r e e l y i n t h e thimbles.
C.
Fiducial Zero - A s
described above, t h e f i d u c i a l zero was a fixed
reference position i n each of t h e control rod thimbles which r e l a t e d t h e actual rod position t o t h e rod position indicated by t h e position-in&cating potentiometers.
Fiducial zero t e s t s were routinely made before a f u e l f i l l
and before c r i t i c a l operation.
19 4 Operating Experience Except f o r d i f f i c u l t i e s with t h e rod assemblies i n thimble T-1,
opera-
t i o n w a s s a t i s f a c t o r y from both an o p e r a t i o n d and maintenance standpoint There w e r e no unscheduled shutdowns because of a c o n t r o l rod f a i l u r e .
The
t e s t i n g program was adequate t o allow advance maintenance planning and work
was done i n conjunction with maintenance i n other areas a t t h e time of a scheduled shutdown. During over 3,000 rod scrams from 50 inches (see Table 19.2) and numerous others from d i f f e r e n t elevations, t h e r e was only one time when a rod f a i l e d t o scram.
Except f o r t h i s and one other period, t h e scram times
were all l e s s than t h e 1.3 seconds s p e c i f i e d i n t h e s a f e t y limits and were usually less than 1 second.
known.
The number o r amount of rod movement is not
Once positioned, there w a s l i t t l e movement of t h e shim rods.
The
regulating rod, normally t h e assembly i n thimble T-1, moved l i t t l e at steady power u n t i l bubbles appeared i n t h e reactor.
From t h a t point on t h e regu-
l a t i n g rod moved frequently under servo control t o compensate f o r t h e action of bubbles on t h e r e a c t i v i t y .
Several rod jog tests were made and it is
estimated t h a t t h e regulating rod moved about 720 jogs per hour at 0.4 in. per jog f o r a t o t a l of 32,400 jogs. inches
The t o t a l movement was about 14,000
W
327
Table 19.2
Estimated Number of 50-inch Rod Scrams
Thimble No.
Rod No.
Drive No.
No. of Scrams
T-1
R-1
v-3
750
T-1
R-1
v-4
150
T-2
R-2
v-2
930
T-3
R-3
800
T-3
R-4
v-1 v-1
. >
430 -
Total
3060
The estimated l i f e of t h e control rod motors was 10,000 hours at a
u
reactor power of 10 MW (Ref.
49).
The condition of a l l drive motors, lu-
b r i c a t i o n , wiring, and gears remained s a t i s f a c t o r y throughout t h e e n t i r e operation.
These were l a s t inspected i n J u l y 1969 at which time t h e reac-
t o r had accumulated 92,805 MWhrs
.
The h i s t o r y of t h e MSRE control rods and drives a r e b e s t summarized i n Figs. 19.1, 19.2, and 19.3.
Control rods R-1 and R-2 remained i n ser-
vice t h e e n t i r e operating period i n thimbles 1 and 2. and R-4 were interchanged f a i l u r e s o f t h e drives a r e Heat-up and p r e - c r i t i
Control rods R-3
l e 3 as shown i n Fig. 19.3.
The component
a t e d i n Table 19.3.
t c i r c u l a t i o n commenced i n January 1965
and during t h e e a r l y perio
s operation, d i f f i c u l t y was encountered
with t h e l i m i t switches.
cations were t h a t t h e switch operators
were s t i c k i n g o r g a l l i n g randomly on all t h r e e rods. were made during t h e M vealed t h a t t h e s l i d i n g
Temporary r e p a i r s
April 1965 shutdown when investigation re-
rs of t h e switch operators had galled causing
t h e f a u l t y operation. The drives were removed from t h e r e a c t o r c e l l i n July 1965 f o r i n s t a l l a t i o n of modified l i m i t switch operators on a l l t h e u n i t s including t h e
FIDUCIAL ZERO (in.) MaintemHistory
,
Installedtemperatunswitch Replaced limit switch operator
AUGUST Replaced potentiomaer
I
Replaced ternparatun switch Replaced potentiometer
SEPTEMBER
UI
APRIL MAY JUNE
n I D
JULY
d
I B r <
AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY
3
tJANUARY
FEBRUARY
MARCH
.
.
.
Finc synchro failed Potentiometer noisy Rod No. 1 used aa shim rod
Replaced V-3 driw with V - 4 Rod No. 1 repulsing rod
8ZC
i
z
FIDUCIAL ZERO (in.) Maintenma Historv
i
F,
s
8
SCRAM TIME
y
I
Installed temperetun switch Replaced limit switch operaor
g f
MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER. DECEMBER JANUARY FEBRUARY
kJ
P-
MARCH APRIL
op
P
MAY
'p
JUNE
h3
JULY AUGUST
I MARCH APRIL MAY
Replaced fine synchro
JUNE JULY AUGUST
w 0 a
SEPTEMBER OCTOBER NOVEMBER
0
Replaced finr synchro
E SEPTEMBER OCTOBER
t Replaced potentiometer
0
6Z€ r
r9
FIDUCIAL ZERO (in.)
SCRAM TIME (sed
I
â&#x20AC;&#x153;s ffi
Maintenam History
I
1
Installed temperature switch Replacedlimit switch operator
.
MAY
II
Repairedroller in T-3 Repaired R-3 upper hose
Bent limit switch omator
JUNE
DECEMBER JANUARY FEBRUARY
P= APRIL
-
R-3 sticking in thimble ReplacedR-3 with spare rod R-4
d
(D 0) CD
Bent limit switch operator
lunRccl I APRIL
<
I P
I
JULY AUGUST
D
SEPTEMBER OCTOBER
z
NOVEMBER
P
D rn -<
-
MAY JUNE
r m
DECEMBER
JANUARY FEBRUARY
50 inch scram time increase
Added 1.5 Ib weight to V-1 0.9 scram time Repairedrod R-3 No. 3 assembly used as reg rod
No. 3 failed to scram on demand ReplacedR-4 with R-3 Replaced potentiometer
331
Table 19.3
Drive Unit No.
Drive U n i t Component Failures
v-2
Service Time as Reg. Rod Service Time as Shim Rod
4 mo. 56 mo.
v-2
--60 mo.
v-4
v-3
5 1 mo. 4 mo.
(Spare Drive) 5 mo.
Temperature switches L i m i t switch operator
1
L i m i t switch operator push rod
2
Total
---
1
1
1
1
4
-
-
2
4
c
5
1000-R potentiometer
-
-
Fine position synchro transmitter
0
2
1
-
3
Coarse p o s i t i o n synchro t r a n s m i t t e r
1
-
-
-
1
spare
(V-4).
The new operators contained case-hardened bearing surfaces,
t h e r e were no more d i f f i c u l t i e s of t h i s type with t h e switch operators. Examination of t h e control rods during t h i s period revealed t h a t t h e upper w i r e sheathed upper hose of t h e No. 3 rod was badly worn at a point
28 i n , below t h e flange.
Examination of t h e No. 3 thimble revealed t h a t
t h e upper r o l l e r , on which t h e control rod operates, d i d not r o t a t e which caused t h e severe wear t o t h e rod.
The r o l l e r was replaced and a new hose
i n s t a l l e d o n â&#x20AC;&#x2122; t h e No. 3 control rod. During t h i s period temperature switches were a d d e d t o t h e lower drive mechanism.
These small bi-metallic switches were t o alarm should t h e tem-
perature within t h e drive u n i t cases exceed 200°F,
Without t h e downward
sweep of gas through t h e housing, convection of t h e heat r i s i n g from t h e thimble could cause a rapid temperature increase and r e s u l t a n t damage t o t h e drive assembly.
W
After t h e rod drives had been r e i n s t a l l e d and t h e re-
actor heated and f i l l e d with s a l t , t h e switches were t e s t e d by turning off
332
t h e gas flow.
Since no ala,rms occurred, another method'of measuring t h e
drive assenibly was devised which, although not precise, w a s adequate f o r operation.
Thls'was done by s h u t t i n g o f f t h e i n t e g r a l drive motor cooling
fan and measuring t h e resistance of t h e windings of t h e fan motor. " A s l i g h t l y greater than 10% increase i n r e s i s t i v i t y equaled %50째F change. winding temperature.
From t h e measured r e s i s t a n c e and* using t h e temperature . , coefficient of resistance of t h e wire, a curve w a s p l o t t e d r e l a t i n g t h e .
I
.
measured resistance t o t h e approximate motor temperature.
This value w a s
rdughly r e l a t e d t o t h e drive housing temperature. I n e a r l y operation, t h e control-rod servo system had not functioned properly due t o coasting of t h e r e m a t i n g rod motor and s h i m locating motor,
Brakes were i n s t a l l e d i n both of t h e s e (see 24.5).
19.4.1
5
%
Rod Assembly i n Thinible T-1
Figure 19.1 gives t h e rod drop t i m e s and f i d u c i a l zero values as w e l l
as t h e maintenance h i s t o r y f o r t h e rod assemblies located i n thimble T-1. This was first made up of Rod R-1 and drive u n i t V-3 and was used as t h e
regulating rod, except f o r a four-month period, f o r t h e e n t i r e r e a c t o r ope r a t i o n and consequently performed t h e bulk of t h e control rod service.
hd
Other than replacement of f a u l t y drive components, t h e r e was l i t t l e d i f f i c u l t y with thi-s assembly..
The.JOOO-R,potentiometer position i n d i c a t i n g potentiometer f a i l e d 3 times due t o an o p e n - r e s i s t a n c e ' c o i l and 1 t i m e due t o a worn r e s i s t a n c e _.
coil.
This d i f f i c u l t y had been a n t i c i p a t e d as it i s a common failure f o r
t h i s type,component i n continuous service, One temperature switch f a i l e d due t o an e l e c t r i c a l ground a t t h e wire connections at t h e switch.
The ground w a s created by bumping t h e switch ,. .. during assembly of t h e drive i n t o t h e case. .
There w a s an a p p a r e n t . s h i f t . o f 1/2-fn. i n November of 1965.
I
i n t h e f i d u c i a l zero position
The change was a t t r i b u t e d t o slTppage of t h e chain
on t h e sprocket but examination. and t e s t i n g f a i l e d to' reproduce t h e s l i p o The 1/2-in0,, d e d a t i o n w a s ,recovered involuntarily
July of 1967 which
&in
would-indicate t h a t t h e exact cause of t h e s h i f t 5 s not known.
The fidu-
c i a l zero position p l o t i? Fig, lg.1 shows the gradual increase, .%1/2 inch,
-.
.. " . I
,
. . .
c)
333
rod from t h e e f f e c t s of usage.
i n t h e length of t h e contro
The change
i n length which occurred i n September 1969, i s t h e result of changing t h e drive u n i t . The rod scram times remained at 0.8 k .05 seconds from i n s t a l l a t i o n u n t i l t h e spring of 1969. 1.03 sec for a 50-in.
I n March of 1969, t h e scram time increased t o
drop and a l s o d u r i n g t h i s period, t h e r e w a s a failure
of t h e f i n e position synchro t r a n s m i t t e r and t h e 1000-R pot.
Analysis (see
19.5) of t h e rod acceleration and velocity during a scram revealed that t h e r e was an area of drag i n t h e lower 20 i n . of t h e scram. evidence, t h e V-3 drive was replaced w i t h t h e spare drive V-4.
From t h e above With drive
V-4 and rod R-1 i n thimble T-1 t h e scram time returned t o ~ 0 . 8 sec u n t i l shutdown.
A d e t a i l e d inspection and maintenance program i s planned f o r
drive V-3 but i s as yet incomplete. 19.4.2
Rod Assembly i n Thimble T-2
As shown i n Fig. 19. drive, R-2 rod.
t h e assembly i n thimble T-2 w a s made up of V-2
This assembly w a s used as a shim rod, except f o r short
t e s t periods-, f o r t h e entire.MSRE operation.
,
Other than t h e mechanical
f a i l u r e s shown i n Fig. 19.2, t h e r e were no operating d i f f i c u l t i e s . The mechanical f a i l u r e s included two f i n e position synchro f a i l u r e s and one potentiometer.
Examination of one of t h e synchros ( t h e radiation
level of t h e svnchro was Q l O mR/hr) revealed t h e failure t o be t h e w i o e r arm contact between t h e r o t a t i n g shaft and t h e windings. The f i d u c i a l zero h i s t o r y shown i n Fig. 19.2 c l e a r l y shows t h e progressive s t r e t c h i n g o f t h e rod t o be *1/2 i n . f r o m t h e time of i n s t a l l a t i o n t o t h e end of operations.
19.4.3
Rod Assembly i n Thimble T-3
This assembly was made s t a l l a t i o n as shown i n Fig. i n t h e figure, t h i s assemb control rods.
-
of V-1 drive and R-3 rod a t t h e time of inAs indicated by t h e scram h i s t o r y shown
t h e most troublesome of t h e t h r e e MSRE
The drive u n i t s were removed during t h e p r e c r i t i c a l mainte-
nance period i n t h e summer of' 1965 t o i n s t a l l t h e modified l i m i t switches described elsewhere.
The No. 3 c o n t r o l rod w a s inspected a t t h i s time and
found t o have a t o r n section i n t h e upper hose, 28 i n , below t h e flange. The damage t o t h e rod was t h e result of t h e f a i l u r e of t h e upper r o l l e r
1
334 i n t h e No. 3 thimble which had jammed and would not r o t a t e .
The r o l l e r
and rod were repaired and after reassembly, t h e rod scram time w a s C0.8 sec
. During t h e low power period of operation, it w a s found t h a t after a f u l l scram t h e lower l i m i t switch would not c l e a r when t h e rod w a s withdrawn.
It was necessary t o push t h e switch actuator by withdrawing t h e
It would function The bottom end of the. l i m i t
rod t o t h e upper l i m i t t o c l e a r t h e , l o w e r l i m i t switch. normally u n t i l t h e rod was again scrammed.
switch push rod w a s found t o be bent below t h e lower r e t a i n i n g bushing. The rod had been bent by s t r i k i n g t h e lower housing flange as t h e r e s u l t of a weak push rod spring.
Repairs consisted of straightening t h e push rod
and i n s t a l l i n g a stronger spring. Later t h e control rod commenced t o s t i c k i n t h e thimble a t a point %1-1/2 in. above t h e f i d u c i a l zero position.
The rod could be freed by
jogging and would hang i n no other position.
Direct examination of t h e
thimble was not possible, but t h e e x t e r i o r of t h e rod w a s examined and appeared t o be i n good condition.
The d i f f i c u l t y appeared t o be t h a t a sharp
projection such as a broken weld at t h e t h r o a t section of t h e f i d u c i a l zero position e x i s t e d on which t h e rod could snag. was replaced with t h e spare rod R-4,
However, when t h e R-3 rod
it d i d not snag.
This rod operated within limits but t h e scram time increased slowly from 0.85 sec i n May 1967, up t o 1,26 s e c i n December 1968. of both t h e drive and rod revealed no apparent defects.
Examination
Replacing t h e V-1
drive with t h e V-2 drive u n i t i n t h e T-3 p o s i t i o n f o r t e s t purposes made l i t t l e change i n the scram t i m e .
Therefore a 1.5-lb weight w a s added t o
t h e shock absorber assembly cf t h e o r i g i n a l V-1 drive u n i t and it was reinstalled.
This r e d u c e d t h e scram t i m e t o l e s s than 0.9 see.
The reactor was shut down i n June 1969 by manually scramming all t h e control rods.
The assembly i n thimble T-3 (R-4 and V-1) was a t 35.0 in.
and did not scram.59
The rod w a s i n s e r t e d 0.2 fn. and scrammed again, t h i s
time t h e rod dropped freely.
Repeated attempts t o make t h e rod s t i c k fol-
lowed t h i s incident but were unsuccessful. scram tests at 2-in. I
Fig. 19.4 shoys t h e r e s u l t s of
increments taken on June 2 before t h e r o d was exam-
ined and c l e a r l y shows t h e area .of high drag,
hd
335
1.8
I .6
1.4
.2 c
:a:
Y
W
-
.O
I I-
LJ
0.6
0.4
0.2
ORNL-DWG 69-8976A
336
To correct t h i s d i f f i c u l t y , Rod R-4 w a s removed and replaced by Rod
R-3.
This w a s the rod which had earlier snagged i n t h e thimble.
u
Before
r e i n s t a l l i n g it, t h e air exhaust tube (a possible cause of t h e snagging) The defective potenti-
w a s f i l e d smooth t o eliminate all t h e sharp edges. ometer w a s also replaced on the drive (V-1).
The greatly improved perform-
ance of t h i s assembly (R-4 and V-1 i n T-3) suggests t h a t t h e trouble was i n t h e rod (R-4) rather than i n t h e drive.
The extreme radioactivity of
rod 3 (R-3) made it impractical t o examine. Figure 19.4 a l s o shows the results of drop t e s t s conducted on August 4, a f t e r a l l repairs had been accomplished f o r all t h r e e control rods.
This
was the condition of t h e assemblies during t h e f i n a l period of operation.
19.5
Improved Rod Scram T e s t i n g
It was obvious from t h e No. 3 rod failure t o scram that t h e method of
t e s t i n g each rod by a single 50-in.
scram was inadequate.
f a i l e d t o expose any l o c a l areas of high drag.
The method
The incremental method of
t e s t i n g , scram time vs s t a r t i n g - p o s i t i o n , as shown i n Fig. 19.4, showed no region of excessive drag f o r any of the rods.
cp,
In an e f f o r t t o provide
a quick t e s t i n g procedure t h a t would not require numerous rod drops, yet would reveal regions of.abnormal drag, a procedure developed f o r t h e EGCR
w a s renovated.
The output from t h e 1000-ohm position potentiometer was
amplified and transmitted by wire t o t h e main ORNL area, where it w a s passed through a f i l t e r w i t h a 5-Hz time constant t o remove transmission noise, d i g i t i z e d a t 2000 samples/sec, and stored on magnetic tape.
The
data was then sent t o t h e IH-360 computer f o r smoothing and analysis by a program especially developed f o r t h i s purpose.
This procedure when a c t i -
vated during a rod drop from 50 in. gave about 1800 data points during t h e drop.
From t h i s velocity and acceleration curves were generated.
Data
taken before r e p a i r of t h e s t i c k i n g rod assenibly (R-4 and V-1 i n T-3) showed a region of near zero acceleration between 26 and 40 i n . , i n good agreement
with t h e incremental r e s u l t s shown i n Fig. 19.4.
After r e i n s t a l l a t i o n , all
three assemblies were t e s t e d by both incremental drops and t h e single drop computer analyzed method.
Both methods showed reasonably constant acceler-
ations of 10ft/sec2 o r more. i s shown i n Fig. 19.5,59
A t y p i c a l p l o t of acceleration vs position
u
i
!
337
P
z
..
u
338 The signal f i l t e r i n g and data smoothing that a r e required because of the noise that is inherent i n t h e position s i g n a l generator and transmis-
sion system made it impossible t o detect w i t h c e r t a i n t y a zero acceleration zone less than about s i x inches long i f it i s more than about s i x inches below the scram s t a r t i n g point.
Revised c r i t e r i a were established f o r t h e monthly t e s t i n g of t h e rods. This consisted of scramming t h e rods from 40 and 50 inches. These two points w e r e just above the normal operating positions of t h e r e g u l a t i n g The c r i t e r i a were t h a t t h e 50-inch scram t i m e must be less
and shim rods.
than 1.00 sec, the 40-in.
scram time must be less than 0.90 see, and t h e
acceleration curves must show no area which indicates abnormal drag.
If
an area of high drag was noted, t h e rod would be dropped from a point within t h i s area.
This scram t i m e must be no more than 20% longer than normal f o r
that point.
Prior t o a f u e l salt f i l l , t h e rods would be scrammed from 24
inches, which i s t h e i r normal position f o r a f i l l .
The scram t i m e s *om
t h i s point must be less than 0.7 seconds and t h e acceleration curve show no abnormality.
G.
19 6 Maintenance Experience Repairs performed on t h e drive u n i t s usually consisted of a simple replacement of worn components which were readily accessible a f t e r t h e u n i t had been removed from t h e reactor c e l l .
I n t h e event of major d i f f i c u l t i e s ,
both a spare drive assembly and a spare control rodwere available as replacement items.
It w a s possible t o perform d i r e c t out-of-cell maintenance t o t h e drives without excess exposure t o personnel.
The radiation l e v e l was about 600
mR/hr a t t h e drive case ( t h e drive u n i t w a s removed from i t s case during
maintenance).
The overall l e v e l of radiation at t h e drive mechanism w a s
about 200 mR/hr and, as an example, t h e l e v e l a t t h e f i n e synchro transmitter which was mounted on the gear case w a s about 30 mR/hr.
Maintenance
on t h e control rod i t s e l f after c r i t i c a l i t y w a s very limited.
Activity on
t h e upper hose w a s about 400 mR/hr at contact and the section containing t h e poison elements was 20 R / h r a t t
18 in. a f t e r a two-year storage period.
339
Since t h e four drive assemblies were i d e n t i c a l and t h e spare p a r t s w e r e common t o all t h e drives, t h e maintenance problems regarding replacement of f a u l t y components was not complicated.
However, investigation as t o t h e
reasons f o r slow scram t i m e s , even when t h e areas of high drag were located (see 19.5)
,w a s
by trial and e r r o r method.
c e r t a i n areas such as t h e 7/16-in.-diam
Past experience had shown t h a t
a i r tube, on which t h e control rod
s l i d e s , could become s l i g h t l y @led o r bent.
These areas were examined,
adjusted, and t h e u n i t operated on a t e s t stand f r o m t h e main console.
The
f i n a l t e s t was t o reassemble t h e complete assembly i n i t s normal operating position i n t h e reactor c e l l and proceed w i t h t h e scram tests as described elsewhere.
19.7
Discussion and Recommendations
The o v e r a l l operation of t h e rods and drives w a s q u i t e good.
There
was only one time when a rod failed t o scram i n t h e four years of use.
At
no time w e r e reactor operations terminated s p e c i f i c a l l y f o r control rod maintenance.
Construction of t h e assemblies w a s r e l a t i v e l y simple since
very rapid response w a s not a requirement.
The flexible controls rods were
adequate but t h e p o s s i b i l i t y of binding, breakage, etc., always existed. It was possible t o control t h e rod position closely, within 0.1 inch, from
t h e "fine" position indicator on t h e console but t h e exact physical locat i o n of t h e poison elements was subject t o weaknesses inherent t o a semiconfined flexible m e t a l hose of t h i s type.
The only known position was at
t h e f i d u c i a l zero position i n t h e thimble t o which all other positions were related. It required about 12 hours t o open t h e reactor c e l l access t o approach
t h e control rod assemblies , Location o f t h e drives i n a
a c t u a l removal required about one hour. r e a d i l y accessible location i s recommended.
The manner of attachin
control rods t o t h e drives and t h e drive
u n i t s t o t h e thimbles w a s h
ed by ( a ) poor l i g h t i n g a t t h e b o l t loca-
t i o n , (b) lack of space t o i n s e r t more r i g i d t o o l s t o perform t h e b o l t i n g which w a s done from a distance of about 18 f e e t , and ( c ) poor v i s i b i l i t y f r o m t h e t o o l operator position at t h e maintenance shield.
A modified
340 external, r a t h e r than i n t e r n a l , b o l t e d flange attachment between t h e rod housing and thimble which is c l e a r l y v i s i b l e i s recommended. The "fine" position cynchro t r a n s m i t t e r s and potentiometers, such as were used i n these u n i t s should be of a more durable type f o r extended service.
'
h3
341
FREEZE VALVES
20.
M. Richardson 20.1
Introduction
The flow of salt i n t h e MSRE drain, f i l l , and process systems was cont r o l l e d by freezing and thawing short plugs of s a l t i n f l a t t e n e d sections of l-l/2-inch
pipe, c a l l e d "freeze valves."
This method of control was Although
adopted because of a lack of a proven r e l i a b l e mechanical valve.
mechanical type valves would have had t h e advantage of f a s t e r action and t h e a b i l i t y t o modulate flow, t h e "freeze valve" concept had a good record of s a t i s f a c t o r y performance, t h e freezing and thawing times were s a t i s f a c t o r i l y s h o r t , and t h e "off-on" type control did not impose any p a r t i c u l a r hardship.
~~ Design and development of t h e MSRE freeze valves began i n 1 9 6 ~ 1 .The basic design was established and had been t e s t e d by t h e time t h e MSRE cons t r u c t i o n had begun.
However, considerable e f f o r t was expended at t h e re-
actor s i t e before t h e valves were ready f o r routine operation.
6,
Successful
freeze valve operation i s t o t a l l y dependent on t h e manner i n which t h e heat and/or coolant gas i s applied and controlled,
The e f f e c t of t h e heaters,
type of s e r v i c e , l o c a l environment, e t c . , made it necessary t o ' h n e " each of t h e valves f o r i t s s p e c i f i c location.
This e f f o r t continued u n t i l
criticality.
20.2
Description of t h e Design of t h e Freeze Valves
The valves were located i n t h e system as shown i n Fig. 20.1. twelve valves were similar arid were made of 1-1/2-in.
All
Sched-40 INOR-8 pipe.
The valve body, o r f l a t t e n e d section, w a s cold-formed t o m a k e a 2-in.-long f l a t , i2-in.
wide, with a 1/2-in.
i n t e r n a l thickness o r flow area.
A cool-
ing gas shroud w a s welded around t h e valve f l a t t o d i r e c t t h e gas f o r freezing t h e salt. was 1/2-in. shroud.
The annulus formed between t h e valve body and shroud
The 3/4-in.
gas i n l e t and exhaust l i n e s were welded t o t h e
See Fig. 20.2.
A p a i r of thermocouples located a t t h e center of t h e valve inside t h e
a i r shroud, and a p a i r upstream and downstream j u s t outside t h e a i r shroud
34 2
ORNL-DWG 73-605 TO COOLANT SALT DRAIN TANK
REACTOR
TO AND FROM COOLANT SYSTEM FV-204
FV-206
COOLANTSYSTEM FILL AND DRAIN LINES
'FV-104
FFT
FD-2
FD-1 V4IFFERENTIA.L CONTROLLER
FUEL SYSTEM FILL DRAIN AND TRANSFER LINES
Fig. 20.1
Location of Freeze Valves
omL4m3.
ALL NOT
THERMOCOUPLES SHOWN.
THERMOCOUPLES
Fig.
20.2
Typical
Freeze Valve
84-68$g
344 were connected t o the control modules and t o t h e cooling a i r modulating controller.
A s i n g l e thermocouple spaced
5 i n . from t h e valve c e n t e r l i n e ,
upstream and downstream on t h e top side of t h e 1-1/2-in. information f o r s e t t i n g t h e manual heater controls.
pipe, supplied
The heaters on e i t h e r
side of t h e valve were individually controlled. The thermocouples at t h e valve body were connected t o an E l e c t r a Sys-
tems modular control network ( s i x modules per valve) which operated automatically at preset temperatures t o maintain each valve within i t s specif i e d condition.
The control network of t h e c r i t i c a l valves, described
below, included an automatic cooling a i r flow c o n t r o l l e r which maintained the valve center a t a constant temperature.
The ultimate control, however,
remained with t h e modules. 20.3
Description of t h e Operation of t h e Freeze Valves
Valves were usually operated i n one of t h r e e steady-state conditions o r modes. (1) Deep-frozen
- t h e heaters were adjusted t o maintain t h e valve a t 40W0O0F without cooling a i r .
(2) Thawed
-The
heaters were adjusted t o maintain t h e valve at
'U200째F without cooling air.
(3) Frozen
-The
heaters remained as i n t h e thawed condition but
the cooling gas flow w a s automatically adjusted t o
hold t h e frozen valve i n condition f o r a rapid thaw. Operation of t h e valve required only a hand switch ( o r control a c t i o n ) t o freeze o r thaw t h e salt. 11
When t h e valve w a s switched t o "thaw" from t h e
frozen" condition, t h e air was cut o f f and t h e heat supplied by t h e shoul-
der headers would m e l t t h e salt i n t h e valve body.
Similarly, when t h e
valve was switched t o t h e "frozen" from t h e "thawed" condition, t h e a i r was turned on and t h e salt i n the valve body would freeze. Each valve had t h r e e coolant gas flow conditions: o r reduced flow :(hold). t u r e control modules.
o f f , full on ( b l a s t ) ,
The flow condition w a s determined by t h e temperaAlthough the exact module s e t p o i n t s varied, f o r t h e
purpose of explanation, t h e following conditions a r e used:
FV-103 control. )
(See 20.4.4 f o r
345
Settinq
Module No.
FV-lAl FV-lA2 FV-2A1 FV-2A2
850°Ft, 8 0 0 O ~ ~
FV-3A1
850°F+, 8 0 0 O ~ ~
Shoulder temperature range
700°F+
Low shoulder temperature
500°F
Center temperature
-- low alarm
1300°F+
Center temperature
-- high
alarm
Shoulder temperature range Low shoulder temperature
700°F+
Fv-3A2
Function
When it was switched t o freeze, Assume t h a t t h e valve was thawed at 1200'F. the b l a s t air ( ~ 2 4 scfh) came on. When both valve shoulder temperatures decreased t o 800°F, t h e "blast" a i r was cut o f f and t h e "hold" a i r automatically came on and held t h e valve shoulder temperatures between 8 0 0 ~ ~
. t h e shoulder temperatures continued downward f o r any reason, and 8 5 0 ~ ~If all a i r w a s cut o f f at 7OO0F by modules lA2 o r 3A2.
If hold a i r flow was
i n s u f f i c i e n t and e i t h e r of t h e shoulder temperatures exceeded 85OoF, t h e b l a s t a i r automatically switched on by t h e action of t h e lAl o r 3A1 module and reduced t h e temperature t o 8 0 0 ~ ~ . On those valves used most frequently, FV-103, 105, 106, 204, and 206, a temperature controlled d i f f e r e n t i a l a i r flow adjustment was provided f o r t h e hold a i r i n addition t o t h e overriding on-off module control.
These
c o n t r o l l e r s maintained the valve center a t a constant temperature. The normal valve condition during r e a c t o r operation w i t h f u e l s a l t was as follows:
(See Fig. 20.1.)
u t i n t h e event of an emergency drain, it was required t o thaw i n less than 1 5 minutes. (2) FV-105 and FV-106 were thawed and were required t o remain thawed (1) FV-103 was frozen
even i n t h e event of a power outage.
I n t h e event of an emergency f u e l
drain, t h e s a l t flowed t o both f u e l drain tanks through FV-105 and FV-106. A normal drain required freezing one of these valves p r i o r t o t h e d r a i n t o
direct the s a l t t o the s (3)
cted receiver tank.
FV-104, and FV's 107 t o 112 were deep-frozen.
(4) FV-204 and FV-206 were frozen during coolant s a l t c i r c u l a t i o n . These valves were required t o thaw rapidly t o prevent t h e coolant salt from freezing i n t h e r a d i a t o r tubes. -
... 346
During flush salt operation, FV-103 was frozen and FV-104 was thawed,
All other valves were deep-frozen.
FV's lo7 t o 112 were used only during shutdown periods f o r salt transfers and additions. The thaw and freeze time f o r these valves w a s not important.
When not i n use, they were frozen o r deep-frozen depending on t h e
operation i n progress.
20.4 ..
Operating Experience
After t h e i n i t i a l d i f f i c u l t i e s described below, operation of t h e .
freeze valves was quite satisfactory. . .
Table 20.1 i s a tabulation of t h e
fYeeze and thaw cycles f o r each of t h e valves.
Table $20.1 Freeze Valve Freeze-Thaw Cycles
Freeze Valve No.
No. of Cycles
103 104
91
105
46 97
106 107
78
108
45
109 110
53 14
111
10
112
3
204
57
206
57
36
The valves which were required t o melt rapidly, less than 1 5 minutes,
were timed whenever t h e system w a s drained.
These valves were FV-103, FV-
204, and FV-206. Periodically they were t e s t e d under complete power failure conditions t o assure t h a t t h e valves would melt within t h e maximum tlme allowed.
The normal thaw time without power f o r fi-103 was 9-41 min, f o r
FV-204 and FV-206 t h e normal thaw t i m e was 12-14 minutes.
347 During e a r l y operation, FV-104, FV-105, and FV-106, which were required t o remain thawed (>850째F) f o r %30 minutes, were a l s o t e s t e d under A t an i n i t i a l temperature of 1150째F,
t h e same power f a i l u r e conditions.
t h e s h o r t e s t period of time required t o reach t h e s a l t freezing point was
36 minutes, with no salt flowing through t h e pipe. It required approximately 8 hours t o bring a valve out of t h e deepfrozen condition (40wOO0F) i n t o t h e frozen condition.
Heat was applied
s e l e c t i v e l y during t h e heatup t o insure t h a t t h e salt w a s melted progress i v e l y from e i t h e r end towards t h e center of t h e freeze valve.
This was
t o avoid having molten salt trapped between two frozen plugs and possible The siphon pots were interlocked s o t h a t t h e pot.
danger of pipe rupture.
temperature was required t o be g r e a t e r than 95OoF before t h e associated valve could be thawed. 20.4.1
Testing of a Frozen Valve
After a valve had been frozen, it was pressure t e s t e d t o assure t h a t a s o l i d , l e a k t i g h t frozen plug had been obtained.
Early t e s t i n g revealed
t h a t due t o t h e f u e l drain l i n e geometry, it w a s possible t o have insuf-
bi
f i c i e n t salt remaining i n t h e freeze valve a f t e r a drain t o make up a s o l i d Investigation, using a glass pipe model, demonstrated t h a t after t h e drain l i n e had been blown down through t h e head of s a l t i n t h e drain tank, t h e salt remained i n the siphon pot and would not flow back i n t o t h e valve body t o f i l l t h e void.
After blowing down l i n e 103, a 4 o r 5 p s i AP e x i s t -
ed between t h e drain tank and r e a c t o r , when attempts were made t o equalize
t h i s AP, t h e salt w a s force
f t h e pot and past t h e freeze valve back
i n t o t h e 103 l i n e . The method which was
t successfully w a s t o reduce t h e r e a c t o r
system pressure by %1/2 p s i g which usually allowed s u f f i c i e n t flQw-back t o f i l l t h e valve and make a good s e a l .
A similar procedure woz-ked s a t i s f a c -
t o r i l y for the transfer free 20.4.2
Control Modules
During t h e i n i t i a l chec sequent operations and t e s t s
of freeze valve' control c i r c u i t s and ~uboblems were encountered i n t h e operation
of t h e modules which make up t h e E l e c t r a Systems Corporation alarm
(ICJ
348 monitor system,
The setpoints d r i f t e d o f f t h e pre-se,
Id
tempera-ure t r i p
value, double t r i p points occurrea, and failures t o respond t o alarm*signals were encountered. Failures and malfunctions of t h e modules were t r a c e d t o i n f e r i o r quali t y components and corrclsion o r oxidation a t t h c p r i n t e d c i r c u i t board conThese faults were mostly eliminated by replacement of f a u l t y com-
tacts.
ponents, gold-plating a l l module contacts, and minor c i r c u i t changes.
The
setpoints f o r the 72 modules were checked a t l e a s t once p e r year thereafter.
I n general, t h e modules remained within 15% of the p r e s e t values
after t h e i n i t i a l d i f f i c u l t i e s were.resolved. 20,4.3' Heat Control t o Freeze Valves
As i n s t a l l e d , t h e heaters on e i t h e r s i d e of t h e freeze sections were on a common control. Also t h e heat supplied t o t h e lower section of t h e siphon pots adjacent t o t h e freeze valve heaters w a s marginal due t o t h e manner of i n s t a l l a t i o n .
Therefore, a balanced temperature gradient on ei-
t h e r s i d e of t h e freeze plug was not obtained.
Since t h e s e t p o i n t s f o r
.the shoulder temperature control modules were adjusted f o r a balanced heat d i s t r i b u t i o n , it w a s d i f f i c u l t t o maintain the valves within t h e module
limits.
Separate heater controls were added which permitted separate con-
t r o l of t h e heat t o each s i d e of t h e freeze plug,
The h e a t e r wattage of
< t h e shoulder heaters was increased from 1 5 W/in2 t o 30 W/in2.
These added
features relieved most of t h e above d i f f i c u l t i e s . ?
The proximity of FV-105 t o FV-106 (35 inches a p a r t ) closely related t h e temperature e f f e c t s of one valve t o t h e other beyond the limits of good control.
This could only be corrected by a change i n t h e piping.
Since
t h i s was not p r a c t i c a l , t h e problem e x i s t e d throughout t h e r e a c t o r operation.
This was especially t r u e during a salt f i l l from FD-I o r FD-2 since 7
t h e hot salt i n l i n e 103 w a s 19-3/4 in. from FV-106 o r 15-1/4 i n . from Fy-105. 20.4.4
Comments on Operation of Individual Freeze Valves
"FV-103
- Since t h i s was t h e main f u e l salt drain freeze valve,
it
was required t o thaw i n less than 1 5 minutes. A coolant gas f l o w r a t e of 75 scf'm was .available f o r freezing. Heat w a s supplied t o t h e valve from t h e ambient temperature within t h e reactor furnace and therefore no.valve heaters were required.
Operation of t h i s valve w a s complicated by a dual
ds
349
bi'
(1)Immediately following a r e a c t o r f i l l , t h e
set of operating conditions:
line-103 on t h e downstream s i d e of t h e valve remained full of s a l t f o r a period of several hours; (2) Line-103 w a s then emptied of salt leaving no s a l t on t h e drain tank side.
Figure 20.3 shows t h e temperature distribu-
t i o n f o r both conditions. I n order t o meet t h e fast thaw requirement and s t i l l meet both above conditions, t h e module control points were set t o f'unction as l i s t e d below: FV-103-lAl
-- Blast
FV-lO3-1A2
-- Low &temperaturea i r cut
off
temperature). FV-103-2A2
- (shoulder
a i r on at 10IO°Ff, no h y s t e r e s i s temperature. )
-- High alarm 6 0 0 ~ ~ "2 1 . of
- 465OF+ - (shoulder
3" alarm matrix (center
temperature).
FV-lO3-3A1
-- Blast
FV-103-3A2
-- Low temperature
a i r on at 765OFf (shoulder temperature )
.- 71g°F+, 50°F h y s t e r i s i s ,
a i r cut o f f
- 515'F+,
(shoulder
temperature).
Bd
The "hold air" d i f f e r e n t i a l a i r flow c o n t r o l l e r w a s normally set t o hold t h e center temperature (TE-FV-103-2Al)
a t 375 t o 40O0F.
The temperature adjustments required f o r t h e proper valve operation r e s u l t e d i n only one shoulder control module being e f f e c t i v e i n each cond i t i o n of t h e valve shown i n t h e figure. t h e TS-FV-103-3Al
During t h e i n i t i a l freeze cycle,
w a s t h e p r i n c i p a l c o n t r o l and after t h e line-103 w a s
drained of salt, t h e TS-FV-103 t h e r e w a s no s a l t below TS-
perature responded very sl on t h a t ' s i d e of t h e valve.
w a s t h e key temperature c o n t r o l since
3-31. During a thaw cycle, t h e IAl tence t h e mass of heat ( r e a c t o r v e s s e l ) w a s '
The 3Altemperature respo
since t h e source of heat w a s through t h e frozen plug perature.
t h e a&ient ten-
Selection of these temperature a l a r m points w a
after many tests. Once t h e valve c o n d i t i
were established, FV-103 o
cept for one unscheduled drain which occurred on April 15, 1969.
"his w a s
caused by an upward 5O0F drift of t h e s e t p o i n t of t h e FV-103-3A1 module from 75OoF t o 8 1 0 plus ~ ~ an administrative change of t h e FV-103-lAl
hj
point from 10IO°F t o 1050°F,
set-
.ORNL-DWG
73-806
w
cn
0
w
5l-
800
2
600
o NORMAL FREEZE, NO SALT IN LINE i o 3 ON DRAIN TANK SIDE
a K
- 9-11
min MELT -
2
0 VALVE FROZEN, SALT BOTH SIDES OF FREEZE PLUG
400 200
I
I 8- 1
I
I
I
I
A1 B
38
28
1B
I R27
- 30
TE NUMBERS
-.
c
Fig. 20.3
Temperature Distribution at FV-103
c
C
351
Gi
The upward d r i f t had occurred p r i o r t o t h e previous freezing operation and a short plug (<2 i n . ) of salt w a s formed i n t h e valve. and freeze t e s t indicated a good freeze.
The controls
A t t h e time of t h e incident, t h e
salt temperature was not at 1210'F f o r which t h e controls were o r i g i n a l l y adjusted, but at lllO'F with t h e cooling gas flow controlling t h e valve center temperature at 420'F. Figure 20.4 i s a graphic h i s t o r y of t h e incident and it can be seen t h a t , with t h e FV-103-lA1 adjusted t o 1050째F, t h e r e w a s l i t t l e i f any warning t o t h e operator p r i o r t o t h e melt. "E-FV-103-lAl
was about 1030'F1040'F
the reactor outlet.
"he valve normally melted when
when t h e f u e l salt w a s a t 1210'F a t
The lA1 module administrative change t o 1050'F w a s
based on an untested theory t h a t the valve should be adjusted on t h e rea c t o r i n l e t salt temperature r a t h e r than t h e o u t l e t . valve had a normal 2-in.
As long as t h e freeze
freeze plug, t h e cooling a i r could be adjusted t o
maintain t h e valve within t h e normal operating range.
I n t h i s instance,
due t o short plug, t h e valve melted as a r e s u l t of changing t h e cooling
a i r flow which i n i t i a t e d a heating cycxe' beyond the'rahge of t h e cooling a i r t o control.
The a i r flow was routinely adjusted 50'F upward when t h e
fuel salt was at a reduced temperature but t h e combination of r a i s i n g t h e 1Al alarm point, t h e short plug r e s u l t i n g f r o m t h e 3 A l upward d r i f t before
t h e previous freeze, and r a i s i n g t h e cooling setpoint r e s u l t e d i n t h e f u e l drain. FV-104
- This valve operated without d i f f i c u l t y since normally it
was deep-frozen o r thawed. It was s u f f i c i e n t l y spaced'away from t h e l i n e 103 so as t o - b e unaffected by a fuel drain. See Fig. 20.1. FV's 105 and 106 -FV-lOS and FV-106 were d i f f i c u l t t o control during a f i l l o r drain operation because of t h e proximity of t h e valves t o each
other and t o t h e 103 l i n e which was common t o both freeze valves. Fig. 20.1.
See
Heat conduction from t h e hot s a l t passing through line-103
strongly affected t h a t valve which was t o remain frozen.
I n an emergency
drain s i t u a t i o n , both valves were thawed and t h e f u e l would drain t o both tanks.
A scheduled drain required t h a t one of these valves be frozen t o
d i r e c t t h e salt t o t h e selected-drain tank only.
LJ
I n severaJ, instances
during a scheduled reactor drain t h e frozen valve thawed which r e s u l t e d
352
I
r
ORNL-DWG 73-607 SALT IN REACTORVESSELAT 111PF
SIMULTANEOUS ALARM AND THAW-SALT FLOWING THROUGH 1/2 in. EMERGENCY DRAIN TUBE ALVE CENTER TEMPERATURE RISING, COOLANT GAS LOW INCREASE NOT FAST ENOUGH TO CONTAIN HEAT-UP
TIME
Fig. 20.4
Temperatures During Unplanned Thaw of FV-103
r
353
ci
i n an added s a l t t r a n s f e r operation t o put a l l t h e salt i n t o t h e proper The condition was most troublesome during a r e a c t o r f i l l operation,
tank.
a time-consuming function a t b e s t , when t h e frozen valve would go i n t o a l a r m and control interlocks would s t o p t h e f i l l o r t h e freeze-valve thaw with t h e system p a r t i a l l y f i l l e d and d r a i n t h e salt t o both storage tanks. By proceeding very c a r e n i l l y during t h e i n i t i a l stages of t h e f i l l t h i s
condition could, and usually was, overcome by stopping t h e f i l l w i t h t h e 103 l i n e f i l l e d t o above t h e t e e and allowing t h e frozen valve t o approach equilibrium.
However, the last f a i l u r e of t h i s s o r t occurred as l a t e as
April 11, 1969. A leak i n t h e primary system became evident a f t e r t h e f i n a l system
drain of December 12, 1969 which appears t o be i n t h e v i c i n i t y of FV-105. This i s discussed i n Section
5. were no d i f f i c u l t i e s o r unscheduled
FV's lo7 through 112 - m e r e thaws with these t r a n s f e r valves.
FV's 204 and 206 -The
CiJ
coolant valves have t h e same inherent fault
as FV-lo5 and FV-106 of being too close t o each other.
Since t h e valves
were always operated as a p a i r , e i t h e r frozen o r thawed, t h i s did not cause any d i f f i c u l t y .
There were no operating d i f f i c u l t i e s with these valves
after criticality.
I
1. Freeze valve
so located by distance o r
separate -lines so as 2.
interaction.
They should be s
ioned t h a t t h e adjacent piping i s posi-
t i v e l y full of salt.
l e d i n a v a r i e t y of ways,
v e r t i c a l l y , inclined,
e no void spaces i n the piping.
3.
Those valves
r i t i c a l should be
r o l feature.
nating t h e automatic expensiv9, less time-cons
l i f i e d by elimi-
These valves would be l e s s
t o a u u s t , and e a s i e r
control by simple
manual air control.
4
.
The . aut omat i
sidered f o r a l l c r i t i c a l valves.
s c o n t r o l l e r should be con-
This f e a t u r e g r e a t l y reduced t h e amount
354
W
of time required t o maintain the valves.within'limits and reduced t h e post-freeze thermal cycling.
5 . The shakedown period f o r adjustment of t h e valve temperature cont r o l s at the reactor was time-consuming f o r a variety of reasonsp I n i t i a l l y the control modules were found t o d r i f t o f f setpoint and had double t r i p points.
These d i f f i c u l t i e s were e s s e n t i a l l y , but not completely, elimi-
nated by l o c a l redesign,
There were s i x control modules per valve f o r a
t o t a l of seventy-two individual modules t o be maintained.
Each module con-
t a i n s two indicating lamps, "Alarm" and "normal" which a r e wired i n s e r i e s with the control c i r c u i t .
There were frequent f a i l u r e s of these bulbs
which i n turn affected the module function,
Many of t h e operators had dif-
f i c u l t y i n understanding t h e functions of t h e modules and were at a l o s s as t o what corrective action was t o be taken when an alarm sounded.,
Al-
though the Electra modular system succeeded i n controlling t h e valves, a simpler system i s recommended,,
6. Individual control should be providedfor each of t h e heaters which a f f e c t t h e valve temperatures.
An excess of heat i s needed t o overcome heat
losses i n t h e valve areao Placement of t h e heaters should be given caref'ul consideration t o eliminate cold spots i n ad3acent piping.
70 valve.
An adequate supply of cooling gas should be available t o each A supply of 4050 scfm f o r 1-1/2-in.
pipe "blast air" cooling
would be s u f f i c i e n t f o r good control.
8. The modular control setpoints were established f o r normal 1210째F reactor operating temperature at steady s t a t e ,
The c r i t i c a l "fast thaw"
valve control setpoints were adjusted f o r t h i s temperature as was t h e cont r o l point f o r t h e automatic cooling gas control.
Deviation f r o m t h e nor-
m a l temperature f o r which the control points were adjusted (within 75OF) required ad3ustment of t h e cooling air flow only.
This raised o r lowered
the valve center temperature t o maintain t h e shoulder temperature control modules within limits,
During t r a n s i e n t conditions, such as heatup of t h e
system, t h e valves were severely cycled because of a l a r g e temperature gradient between two valve shoulders on t h e same valve.
One shoulder would
be above the module alarm point and t h a t control would t u r n on t h e b l a s t
air.
The other shoulder would be below t h e low temperature alarm point
and cut a l l the cooling a i r o f f .
The module controls would allow the
355
valve t o cycle back and forth across t h i s range u n t i l t h e lâ&#x20AC;&#x2122;ow temperature shoulder would heat enough (*70O0F) t o permit t h e d i f f e r e n t i a l controller t o control the a i r flow.
In practice all a i r w a s manually turned o f f
during heatup.unti1 t h e valve shoulder temperatures were above 70O0F.
The
valve then would remain i n t h e frozen condition but heat much more rapidly and t h e thermal cycling would be greatly reduced.
356
21.
L)
FREEZE FLANGES
R. H, Guymon I
21,l
Description
Mechanical-type j o i n t s were provided i n t h e 5-in.
salt piping inside
the reactor c e l l t o permit major components t o be removed f o r maintenance. Figure 21.1 i s a sectional view of t h e so-called "freeze flange" type of The design was such t h a t t h e "0" r i n g was i n a r e l a t i v e l y cool
j o i n t used. location.
A frozen salt s e a l protected t h e r i n g j o i n t seating surfaces
from contact with sal-t which could cause corrosion when t h e s a l t w a s exposed t o moisture.
Each freeze flange had s i x thermocouples; four were
located on a 4-1/2-in. on a 10.2-in.
radius, 9,0째 apart, one on a 7-in.
radius and one
radius. 21.2
Operat ion
It w a s never necessary to.remove a major component and therefore one
important function of the freeze flanges was never t e s t e d at t h e E R E . E a r l i e r tests on a prototype flange indicated t h a t making and breaking
these could be done remotely without appreciable d i f f i c u l t y , The leak rates of t h e five freeze flanges during e a r l y operation are
given i n Table 21.1.
Similar leak r a t e s were observed,throughout operation
except t h a t one of t h e freeze flanges (FF-201) located near t h e point where t h e coolant salt l e f t the heat exchanger leaked more than the allowable leak r a t e o f 1 x
cm3/sec when salt was not i n t h e piping and t h e cool-
ant system hpd been cooled down., cm3/sec on August 2, 1966.
This leak was measured at about 0.2
The flange always became acceptaply sealed when
the system w a s heated. Although there w a s some variation between individual flanges and some variation w i t h time, the thermocouples located 4-1/2 ine' from t h e centerline of t h e pipe were usually between 750 and 95OoF0 Those on 7-in. r a d i i were about 550 t o 65OOF and those at 10.2-in.
were about 450 t o 55OoF.
The thermocycle history i s shown i n Table 21.2.
When it was decided
t o extend t h e operation of the MSRE longer than t h e o r i g i n a l plans, t h e
LJ
357
ORNL-LR-DWG 63248R2
1
b
)
BUFFER CONNECTION (SHOWN ROTATED) MODIFIED R-68 RINGGASKET
FROZEN SALT SEAL-
>
-4r, in.+
5-in. SCHED-40 PIPE
ze Flange and Clamp
i
358
Table 21.1
Observed Leak Rates of Buffer Gas from Freeze Flanges
Leakage .Rahe (E&
cm3/gec) ?
Freeze Flange
-
Initial Heatup
Circulating
salt
10-3
System H o t DraYned After Run 1 1
x 10-3
io-3
System Cold After . Run 1 x 3 ' 0 1
System Cold After
Run 3 x 10-3
0.57
0.7
1.5
2,o
10 3
0.4
0.25
2.26
1.2
102
0*5
0.3
0.30.
2.33
1.0
200
1.0
0.21
0.41
1.48
201
0.6
0.22 .
0.21
1.23
100
200
101
.
f
003
1.8
freeze flange t e s t loop whibh had cycled a prototype flange f o r 103 cycles
was restarted.
Af%er cycle 268, examination using dye-penetrant revealed
a crack i n t h e bore i n t h e ' v i c i n i t y of t h e weld attaching t h e alignment
stub t o t h e face of t h e flange.
The crack appeared about t h e same a f t e r
t h e f i n a l thermal cycle, number 540. 21.3
Conclusions I
The freeze flanges performed s a t i s f a c t ' o r i l y throughout t h e r e a c t o r operation.
The higher leak r a t e on FF-201when cold caused some concern
but did not require repairs.
.
U
359
Table 21.2.
Summary of Unscheduled Scrams a t MSRE with Reactor Criticala
Operatinp Hours
Year
Quarter
Core in
Critical
1966
1 2 ' 3 4
672 1293 554 1266
62 1070 1221
1 2 3
4
1861 1254 1318 2159
1 2 3 4 1 2 3 4
1967
1968
u
1969
Tot a1
amere
Number of Unscheduled Rod Scrams Human Power Total Error Failures L & C Otherb 0 3 2 1
1 6 0
3.
2 2 0 1
1
1 2 0 0
1852 1186 1292 2144
2 2 1 2
1 1 0 0
0 1 1 1
1 0 0 1
0 0 0 0
2048 0 88 1000
2045 0
0 0 0 1
0 0 0 1
0 0 0 0
0 0 0 0
0 0 0 0
1850 1385 1076 1203
1800 1375 1054 1176
2 2 0
0 0 0 0
0 1 0 0
0 0 0 0
0 2 2
d
19027
17425
37
8
10
10
9
413
0 735
4 13 2
3
is no record of any unscheduled s c r a m s during 1965, w h e n f u e l e reactor was c r i t i c a l ( a t 1 kW o r less)
was i n t h e core f o r 1062 h f o r 230 hr.
% o s t l y equipment fau Fgr example, f i v e of t h e . l a s t s i x scrams due t o "other" causes occurred when t h e speed of t h e variable-frequency generator being used t t o d r i v e t h e f u e l pump sagged below a prescribed l i m i t .
360 22.
CONTAINMENT
P. H. Harley 22.1
Description and Criteria
Containment a t t h e MSRE w a s required t o be adequate t o prevent t h e escape of multicurie amounts of r a d i o a c t i v i t y o r dangerous amounts of other materials t o t h e atmosphere.
A minimum of two b a r r i e r s w a s provided,
During operation, t h e primary b a r r i e r was t h e piping and equipment which contained o r w a s connected t o t h e f u e l s a l t system and off-gas system.
Block valves were i n s t a l l e d i n a l l t h e helium supply (cover gas) Primary containment extended out t h e .. l i n e s and through t h e charcoal beds. The e n t i r e primary system
l i n e s t o prevent backup of a c t i v i t y . off-gas
was of all-welded construction with leak-detected flanged J o i n t s except at a f e w l e s s vulnerable locations where autoclave f i t t i n g s w e r e used. Ess e n t i a l l y zero leakage w a s permissible from t h e primary system. The secondary containment currounded t h e primary containment,,
It con-
s i s t e d mainly of t h e reactor and drain-tank c e l l s and appendages t o them, There are ~ 7 0 0penetrations f o r service l i n e s i n t o t h e c e l l s .
Figure 2 2 , l
is a simplified diagram showing t h e various type penetrations and methods of s e a l i n g o r blocking these. Under conditions of t h e MCA, t h e secondary containment would be limited t o a pressure of 39 p s i g by t h e vapor condensing ( s e e 22.4).
The c e l l
temperature would r i s e t o 2 6 0 ~Under ~ ~ t h e s e conditions t h e maximum a l lowable leakage rate as specified i n t h e MSRE Safety Analysis ReportSs was 1% per day of t h e secondary containment volume.
During maintenance, t h e reactor was shut down and t h e amount of radioac'tivity available f o r t h e chances of i t s r e l e a s e were considerably reduced, When t h e primary system w a s opened, t h e c e l l s became t h e primary containment with t h e high-bay area acting as secondary containment. 22.2
Methods Used t o Assure Adequate Containment and Results
Most of t h e time t h e primary and secondary containment w a s operated well below i t s design c a p a b i l i t i e s .
Various means were used t o assure t h a t
containment would be adequate -under t h e worst conditions e
c
c
ORnnr;. 8-2608
Fig. 22.1
Schematic of MSRE Secondary Containment Showing Typical Penetration Seals and Closures
362
The primary system was pressure-tested annually and t h e c e l l a i r ac-
t3
t i v i t y w a s continuously monitored as an indication t h a t no leak had developed during operation.
The secondary system l e a k r a t e w a s checked annually
at a p o s i t i v e pressure and was continuously monitored at normal c e l l pressure (+ p s i g ) during operation.
I n addition t o t h i s , a l l primary and
secondary system block valves w e r e t e s t e d annually t o assure acceptable leak rates. 22.2.1
Primary System
Radiographs, dye checks, and other inspections were c a r e w l y reviewed A l l flanges had metal "0" rings
during construction of t h e primary system.
which were remotely leak-detected by pressurizing t h e r i n g groove t o 100 psig w i t h helium.
(See Section 1 4 of t h i s report. )
Although t h e system
was opened 33 t i m e s f o r f u e l additions, graphite sampling, and maintenance as indicated i n Table 22.1, there was no leakage of any flange above the allowable
loo3
cc/min while t h e reactor w a s operating.
I n 1965 before s t a r t i n g nuclear operation, a l l of t h e block o r check
valves were t e s t e d i n place o r removed and bench-tested.
This w a s p a r t of
t h e containment (primary and secondary) check l i s t , Section 4E of t h e MSRE Operating Procedures.22
The s p e c i f i e d maximum l e a k rate on t h e primary
system valves was 1 t o 2 cc/min at 20 psig. helium cover-gas l i n e s had excessive leakage.
A nurnber of valves i n t h e
This w a s due t o damaged "0"
rings and/or they had foreign p a r t i c l e s i n them, usually metal chips from machining.
After cleaning and i n s t a l l i n g new "0" rings they had no detect-
able leakage.
A f t e r r e i n s t a l l a t i o n , t h e l i n e s were pressurized and t h e
f i t t i n g s o r welds were checked with a helium leak detector. These tests were repeated annually.
The r e s u l t s a r e given i n Table
22.2 along w i t h results from t e s t s of t h e secondary block valves which a r e described later. I n addition t o t h e above, an annual strength t e s t was run on t h e p r i -
mary system. This w a s normally done by pressurizing t h e f u e l system, including d r a i n tanks, t o 60 p s i g with f l u s h s a l t c i r c u l a t i n g at normal opera t i n g temperature (1200OF). *No abnormalities were noted during any of these t e s t s .
(cd
363 Table 22.1
Opening of MSRE Primary Containment Flanges
No. of Times
Location
Primary Reasons
Off-gas l i n e at f u e l pump Off-gas l i n e i n vent house
6 7 4
Fuel pump vent l i n e
2
Overflow tank vent
2
Plugged valve, access t o a lower line
Drain tank No. 2 vent
1
Fuel addition
Drain tank No. 2 access
6
Fuel addition, samples, inspection
Drain tank No. 1 access
1
Inspection
DT t o FP equalizer
1
Plugged c a p i l l a r y
FP upper off-gas l i n e FP r o t a r y element FP l e v e l reference
1
T e s t o i l catch tank
1 1
Remote p r a c t i c e , inspection
Reactor graphite sampler
Line r e s t r i c t i o n Valve, f i l t e r , p a r t i c l e t r a p .Plugged valve, modify leak detector
Restricted a f t e r o v e r f i l l
33
Tot a1
Table 22.2
Water A i r or C e l l A i r
Simples, core inspection
3-5 cc/min
Tests of Block Valves
3 64 -
There were no leaks i n t h e primary system u n t i l a f t e r t h e f i n a l drain.
b) >
A t t h i s time a leak w a s indicated by an increase i n c e l l - a i r a c t i v i t y .
Subsequent investigation showed t h a t t h i s w a s i n t h e v i c i n i t y of one of t h e drain freeze valves (FV-105). 22.2.2
See Section 5 and Reference 25*
Secondary Containment
The c r i t e r i a f o r t h e secondary Containment was not as r i g i d as f o r t h e primary containment.
Small leaks could be t o l e r a t e d but it w a s s t i l l
necessary t o assure t h a t containment was adequate f o r t h e worst conditions The methods used a r e described below. Only one strength t e s t was made on t h e r e a c t o r and drain tank c e l l s , This was i n 1962, soon a f t e r construction 'of t h e c e l l s was completed,
The
tops of t h e c e l l s were closed temporarily by s t e e l membranes and t h e penet r a t i o n sleeves were temporarily blanked o f f .
The c e l l s were then hydro-
s t a t i c a l l y t e s t e d separately at 48 p s i g (measured a t t h e t o p of t h e c e l l s ) . A leak rate acceptance t e s t w a s a l s o made i n 1962.
The leak r a t e was ac-
ceptable, however, none of t h e penetrations had been i n s t a l l e d by t h i s
t i m e e 56 P r i o r t o power operation, t h e containment block and check valves were t e s t e d i n place or-removed and bench-tested as described i n t h e containment s t a r t u p check list (4E of t h e Operating Procedures) 22 The s p e c i f i e d maxi-
mum leak r a t e at t h e t e s t pressure of 20 p s i g varied according t o t h e location of t h e valve but usually was set a t 1 t o 2 scc/min f o r helium cover gas (primary containment.), 5 t o 10 scc/min f o r other gases and 3 t o 5 cc/min f o r l i q u i d s .
A considerable number of leaks were found and repaired.
Leaks which occurred i n t h e water system were mostly i n t h e check valves.
These were t h e r e s u l t of foreign p a r t i c l e s , apparently washed out
of t h e system and trapped between t h e f i x e d and moving p a r t s of t h e valves.
Two hard-seated valves .in t h e water system had t o be replaced with s o f t seated valves even though t h e seats were lapped i n an e f f o r t t o get them t o seal.
One w a s a swing check valve i n a water l i n e between t h e surge
tank and t h e condensate tank.
The o t h e r was a spring-loaded, hand-operated
vent valve on t o p of t h e surge t p k . The instrument-air block valves with few exceptions were found t o be
satisfactory.
There were n h e r o u s leaking tube f i t t i n g s i n t h e a i r l i n e s .
LJ
365
u
Severa1,quick disonnects on a i r l i n e s i n s i d e t h e r e a c t o r c e l l and draintank c e l l were found t o be leaking when checked wi$h leak-detector soluThese did not c o n s t i t u t e a leak i n secondary containment since each
tion,
l i n e has a block valve outside t h e c e l l ; but a leak here would a f f e c t t h e c e l l leak r a t e indication when a i r pressure w a s on t h e l i n e t o operate t h e . valve. The b u t t e r f l y valves i n t h e 30-inch l i n e used f o r v e n t i l a t i n g t h e c e l l s d w i n g maintenance operations were checked by pressurizing between them. The leakage measured by a flowmeter w a s excessive, and t h e valves had t o be removed from t h e system t o determine t h e cause. d i r t on t h e rubber s e a t s , and one was cut. repaired.
There was considerable
These seats were cleaned and
It was found t h a t t h e motor drive u n i t s would s l i p on t h e i r
mounting p l a t e s by a small amount, thus causing a slight e r r o r i n t h e indic a t e d , p o s i t i o n of t h e valve. t o prevent t h i s .
Dowels were i n s t a l l e d i n t h e mounting p l a t e s
Small leaks were a l s o found around t h e pins which f a s t e n
the b u t t e r f l y t o t h e operating .shaft.
LJ
These leaks were repaired w i t h epoxy
resin. The l i n e from t h e thermal-shield rupture d i s c s t o t h e vapor-condensing systems had numerous threaded j o i n t s t h a t leaked badly when pressurized with nitrogen.
Each j o i n t was broken, t h e threads were coated w i t h epoxy
r e s i n , and t h e j o i n t was remade.
All j o i n t s were l e a k t i g h t when rechecked
w i t h leak-detector solution.
The c e l l pressure w a s then increased i n increments and extensive soapchecking and leak-hunting and t h e c e l l s were press with leak-detector s o l u t
done. Alternate t o p blocks were i n s t a l l e d , o 1 p s i g t o leak-check all membrane welds
c e l l s at 1 psig, all pen
leaks were found i n t h e welds. With t h e 6 , pipe j o i n t s , tube f i t t i n g s , and mineral-
insulated ( M I ) e l e c t r i c a l c
s e a l s subjected t o t h i s pressure were
checked w i t h leak-detector s o f i t t i n g s and M I cable s e a l s .
I
n.
.
Numerous leaks were found i n tube
of t h e leaks were stopped by simply
tightening t h e threaded p a r t s of t h e s e a l .
However, t h e leak r a t e w a s
s t i l l about 4500 ft3/day, indicating some major leak t h a t had not been found.
L.i
366 A l l s h i e l d blocks were i n s t a l l e d , t h e c e l l s were p r e s s u r i z e d t o
and leak-hunting continued.
Three l a r g e leaks were'located.
5 psig,
*iiJ
One w a s
through t h e sleeve t h a t surrounds t h e f u e l off-gas l i n e from t h e reactor c e l l t o a v e n t i l a t e d p i t i n t h e vent .house; t h e sleeve was 'supposed t o have beep welded t o t h e l i n e i n t h e r e a c t o r c e l l but t h i s had been overlooked. Therefore, t h e sleeve w a s closed at t h e vent house,.where it w a s accessible. Another leak was i n an instrument air l i n e t o an i n - c e l l valve (HCV-523), and t h e t h i r d w a s from t h e vapor-condensing system t o t h e drain-tank.steam domes and out t o t h e north e l e c t r i c service a r e a t h r p a g h a l i n e t h a t was temporarily open.
A l l penetrations, tube f i t t i n g s , and M I cable seals were
again checked with leak-detector solution.
Many MI. cable s e a l s which had
not leaked at 1 p s i g were found t o be leaking, and some of those which-had been tightened and sealed a t 1 p s i g now leaked.
A s a reskt, several of t h e gland nuts s p l i t and soldering
were tightened.
was required.
Again, many of t h e s e a l s
A l l l a r g e leaks were sealed o r g r e e t l y reduced, and many of
t h e small leaks were stopped. Leak-hunting and r e p a i r s were continued at 10, 20, and 30 psig.
The
MI-cable seals as a group accounted f o r a l a r g e percentage o f t h e remaining
leaks.
u
To s t o p small leeks which may not have been located, all MI-cable
seals were coated with epoxy r e s i n at t h e seals outside of t h e cells'. This was done with t h e c e l l pressure at -1.5 p s i g so t h a t t h e epoxy would be drawn i n t o t h e s e a l s . Teflon tape, used extensively on MI-cable seals, other threaded pipe, and tube f i t t i n g s , did not perform s a t i s f a c t o r i l y i n providing a gas s e a l . During t h e 30-psig t e s t , l a r g e leaks were located i n both componentcoolant-pump-dome audible.)
flanges using leak-detector solution.
(One of them w a s
The i n l e t and o u t l e t valves t o t h e s e domes were then closed and
the' domes were opened for work on t h e gaskets. rubber gaskets were found t o be undamaged.
The,broad, f l a t , *Viton-
The leakage w a s a t t r i b u t e d t o
inadequate loading pressure on the broad gaskets, and after they were narrowed, no f u r t h e r leaks were observed. Leak-rate data was then taken at 30, 20, 10, and -c! p s i g with t h e component-coolant-pump domes valved back i n t o the system,
These a r e plot-
t e d i n Fig, 22.2 along with curves which r e l a t e t h e allowable leakage at various pressures t o t h e allowable leakage a t '39 p s i g (MCA pressure) f o r
Ld
367
ORNL-DWG 66-4753
500
400
300
200 W 0
a Y a W
-I
IO0
0
\
-io0
-@ @
ORIFICE FLOW
@
TURBULENT FLOW
--@ 0
-200
-
ASSUMED LINEAR RELATIONSHIP
LAMINAR FLOW
1
EXPERIMENTAL DATA
368
I
various f l o w regimes.
Also shown i s t h e highly conservative l i n e a r re-
l a t i o n s h i p which has no physical b a s i s . The allowable leak rates based on t h e o r i f i c e flow curve ( t h e most conservative r e a l i s t i c curve) along with t h e measured leak r a t e s during these first tests i n 1965 are given i n Table 22.3.
Table 22.3
C e l l Leak Rates
~~
Test Allowable Pressure Leak Rate b i g ) (scfd)
30* * 20
*
* The **
1965
360
130
290
125 40
195
10
Observed Leak Rate (scfd)
1966
1967
1968
1969
58
150
.
35
~
65
instrument air block valves were closed during these t e s t s .
+
The leak rates at p s i g are those calculated f o r t h e periods following t h e pressure tests. The instrument a i r block valves were open during these periods.
nuaJ checks o f t h e secondary containment consisted of: (1)t e s t i n g and r e p a i r i n g all block valves, ( 2 ) checliing t h e leak rate at
'some elevated pressure (usually 20 p s i g ) , and (3) checking t h e leak r a t e while operating with t h e c e l l s at -i! psig.
The number of valves found
leaking d u r i n g t h e s e tests a r e shown i n Table 22,2.
The measured leak r a t e s
a r e given i n Table 22.3. During operation of t h e r e a c t o r , t h e c e l l s were maintained at . .
+' p s i g
On t h r e e occasiops, and . t h. e leak rate was monitored somewhat continuously. the' r e a c t o r was shut down due t o indicated high c e l l leak r a t e s . Further
64 I
369
bd
investigation showed t h a t leakage would not have been excessive during an accident on any of these occasions.
These are described below.
I n t h e l a t t e r p a r t of May 1966, a c e l l leak r a t e of 100 scf/day at 4 p s i g w a s calculated.
The r e a c t o r w a s shut down and subsequent investi-
gation, including a 20-psig pressure t e s t , disclosed a high leakage r a t e through t h e thermocouple sheaths f r o m t h e pressurized headers i n t o t h e c e l l s . Since no s i g n i f i c a n t leakage w a s â&#x20AC;&#x2DC;detected from tkie thermocouple headers outside t h e r e a c t o r c e l l , a flowmeter was i n s t a l l e d i n t h e nitrogen supply This flow w a s included i n t h e leak-rate calculations.
line.
Prior t o t h i s
tlme t h e headers had been maintained between 5 and 50 p s i g by periodic pressurization.
To decrease t h e purge t o a minimum, t h e procedures were changed
t o maintain t h e pressure at 5 psig.
A containment block valve w a s i n s t a l l e d
i n t h e supply l i n e . In November 1966, t h e indicated c a l l leak r a t e increased t o 300 s c f d at 4 2 psig.
Therefore, t h e r e a c t o r was drained on November 20.
Leaks were
found i n two air supply l i n e s and one vent l i n e used f o r in-cell air-operated valves.
bi
One of t h e supply l i n e s was capped since it supplied a valve which
w a s only o p e r a t e d p e r i o d i c a l l y .
A rotameter w a s i n s t a l l e d i n t h e other
supply l i p e and t h e vent l i n e w a s connected t o t h e c e l l . A f t e r measuring a c e l l leak r a t e of 45 scf/day during a 10-psig pres-
sure t e s t , Run 10 w a s started.
Early i n thy run t h e new rotameter i n t h e
a i r l i n e indicated t h a t t h e i n - c e l l leak had increased.
Rotameters were
a l s o i n s t a l l e d on t h r e e more a i r l i n e s which were found t o be leaking i n The c e l l .
Later, t h e known leakage increased t o 3500 scfd.
Although t h e
c d c u l a t e d c e l l leak r a t e appeared t o remain a t about 50 scfd, Run 10 w a s terminated i n January 196
se t h e possible e r r o r i n t h e measured purge
rates i n the leaking i n s t
a i r l i n e s exceeded t h e permissible leak
rate
.
During t h e shutdown, t
i r - l i n e leaks were t r a c e
quick discon-
nects i n which neoprene se
ad become embrittled.
with elastomer s e a l s ; eight
connects, a l l near t h e center of t h e reactor
c e l l , were leaking.
of t h e s e were replaced w i t h s p e c i a l adaptors
Sevente
0
f
18 disconnects
sealed at one end by an alu$num gasket and a t t h e other by a standard
hi
metal-tubing compression f i t t i n g , (One disconnect w a s not replaced because it w a s on a l i n e t h a t i s always a t c e l l pressure.) No similar
370
d i f f i c u l t i e s were encountered.
A f t e r s e a l i n g t h e c e l l , t h e leak rate
was 50 scfd. Discussion of C e l l Leak Rate Determinations
22.3
There were at least t h r e e ways of determining t h e c e l l leak rate, These were:
-
(1)a material balance plus compensation f o r pressure and tem-
perature using changes of t h e d i f f e r e n t i a l pressure between t h e c e l l atmos.
phere and an i n - c e l l reference volume; ( 2 ) a material balance.plus compens a t i o n f o r absolute temperature changes i n t h e c e l l s , and (3) an oxygen balance The first method w a s used f o r c a l c u l a t i n g a l l o f f i c i a l leak r a t e s at t h e MSRE.
A discussion of each method follows.
Method 1
-- Material Balance P l u s
Compensation Based on t h e D i f f e r e n t i a l Between t h e C e l l Pressure and t h e Reference Volume Pressure
I n order to,determine t h e c e l l leak r a t e i n a reasonable time, very accurate indication of c e l l pressure changes was necessary.
The instrument
used t o determine t h e pressure changes w a s e s s e n t i a l l y a "U" tube manometer (Fig. 22.3) c a l l e d a "hook gage," .
L
Pointed shafts attached t o micrometers
'
protruded through "0" r i n g s e a l s i n t h e bottom of t h e chamber.
Readings
taken by adjusting t h e point of t h e s h a f t u n t i l it w a s at t h e l i q u i d surface and then reading t h e micrometer.
Thus changes i n water l e v e l corre-
1
sponding t o c e l l j p r e s s u r e changes could be accurately measured. n a l gage had a range of 2 inches of .water.
The o r i g i -
A l a t e r model was obtained i n
which t h e reference chamber could be moved a measured amount which gave a range of 12 inches of water, The changes i n c e l l pressure were measured r e l a t i v e t o t h a t of an inc e l l reference volume.
The reference volume was d i s t r i b u t e d throughout t h e
reactor c e l l and drain-tank c e l l i n an attempt t o compensate f o r c e l l temperature changes.
It was found t h a t very small leaks i n t h e l i n e s t o t h e
reference volume could give l a r g e e r r o r s i n t h e leak-rate data.
Welding
these l i n e s eliminated t h i s d i f f i c u l t y . Approximately 55 of t h e contained volume was located outside of t h e reactor and drain-tank c e l l s and had no temperature compensation.
This
U
371
ORNL DWG. 67-2689
1 * , d
NOOK C A G E
Fig. 22.3
Schematic of Hook Gage and Piping for Determining Pressure Change of Contained V Relative to the Reference Volume
372
included t h e 30-inch-diameter
reactor c e l l v e n t i l a t i o n duct ( l i n e 930) i n
t h e coolant drain tank c e l l and t h e component-coolant-pump domes i n t h e special equipment room.
These areas were sealed o f f as well as possible
t o minimize temperature changes.
A t power t h e r e w a s s t i l l a considerable
amount of a i r leakage past these from t h e main blowers which caused daynight temperature variations up t o 20째F during winter months.
The e f f e c t
of these temperature fluctuations was magnified by t h e presence of w a t e r vapor i n t h e c e l l air.
(Since June 1967, t h e r e was a s m a l l continuous
water leak (<1 gal/day) i n t o t h e reactor c e l l . )
This water evaporated i n
t h e c e l l s and condensed i n t h e cooler sections of t h e containment ( l i n e 930 and t h e component-coolant-pump gas cooler) e from t h e system.
This was periodically drained
I n colder weather more water would condense.,
Reactor power a l s o had an e f f e c t on t h e indicated c e l l leak r a t e due t o c e l l temperature changes.
When t h e reactor w a s taken f'rom zero t o f u l l
power, gamma heating i n the thermal s h i e l d and other equipment i n t h e c e l l caused t h e average c e l l temperature t o increase 4 t o 8OF and indicated a high c e l l leak r a t e f o r approximately 24 hours.
These changes were not
adequately compensated f o r by t h e pressure reference volume.
LA
Af'ter t h e c e l l s were closed following each i n - c e l l maintenance period,
t h e containment w a s purged with dry B2 t o remove oxygen, s o - t h e c e l l gas s t a r t e d out dry.
A s t h e water evaporated, t h e i n i t i a l c e l l leak r a t e was
usually high (75-130
scf/day has been calculated) b u t gradually decreased
t o an equilibrium value i n 5 t o 7 days at which time condensate s t a r t e d t o form and was drained from t h e system.
Duringsthis i n i t i a l period u n t i l -
\
condensate apped-ed, we relied on results of leak-testing a t p o s i t i v e pres-
sure before t h e C e l l w a s evacuated and purged.
Due t o t h e s c a t t e r i n t h e
data and t h e r e l a t i v e l y small change i n pressure o r temperature t h a t represents a l a r g e leak r a t e ,
*
considerable time was required t o obtain data
from which r e l i a b l e leak r a t e could be determined.
To minimize t h e t e m -
perature and other e f f e c t s , i n i t i a l and f i n a l data w a s usually taken at i
*A
I
change i n c e l l pressure of 0.3 inches of water (0.01 p s i ) per day o r a change i n the average temperature of O b 0 k O ~w i l l change t h e indicated leak rate by about 10 scfd.
W
373
L1
t h e same time of t h e day and a t t h e same operating conditions.
Calcula-
t i o n s of i n t e r v a l s of several days were more consistent than s h o r t e r periods.
Figure 22.4 i s a p l o t of t h e data used t o determine t h e leakage
r a t e at 30 p s i g i n 1965.
1
I
Typical flow r a t e s f o r t h e material balances were:
9 scfd t o t h e sump
bubblers, 13 scfd purge f o r t h e thermocouple header, and 0 t o 1000 scfd evacuation flow.
Since t h e evacuation flow was normally t h e l a r g e s t and
was changed more often, i t s s i g n a l was sent t o t h e computer which i n t e grated it and each s h i f t typed out a value f o r t h e t o t a l volume evacuated. Method 2
-- Using t h e Absolute
C e l l Pressures and Temperatures
This method w a s not used because t h e absolute c e l l pressure indication was not as accurate as t h e indication of d i f f e r e n t i a l pressure by t h e hook gage and compensating f o r temperature changes using absolute temperatures
*
did not give as consistent r e s u l t s as those obtained using t h e reference volume. Method 3
-- Using an Oxygen Balance
The MSRF: containment was purged with N2 t o keep t h e 02 concentration
<5% primarily t o eliminate t h e danger of an explosion i f an o i l leak should develop i n t h e fuel-pump l u b r i c a t i n g system.
This made keeping an oxygen
balance on t h e c e l l appear as an a t t r a c t i v e method of measuring t h e c e l l leak rate when t h e c e l l was at a negative pressure.
The precision t o which w e 'could read t h e 02 analyzer w a s only 50.1% which i s equivalent t o $60 f't3 of a i r i n t h e c e l l .
With a leak r a t e nor-
mally $20 f't3/day, it became apparent t h a t , t h e only accurate calculation . would be over very long time periods.
I n comparison t o Method 1 ( c e l l leak
r a t e using t h e c e l l pressure change and a flow balance), t h e oxygen balance leak rate was consistently low by $15 scf/day.
I n f a c t , when t h e leak rate
calculated by Method 1 w a s 4 5 scf/day, t h e 02 data indicated a negative leakage.
*Ten
LiJ
ambient thermocouples were located i n t h e r e a c t o r c e l l , 6 i n t h e drain-tank c e l l , and 1 i n t h e s p e c i a l equipment room.
374
I
422.2
QI.0 89.2
0
Fig. 22.4
Containment-Conditions During Leak Rate Test at 30 psig
375
ii
Errors involved i n t h e pressure method of calculating t h e leak r a t e would probably be associated with t h e rotameters; but t h i s would imply a 25% e r r o r i n rotameter c a l i b r a t i o n .
All of t h e rotameters involved were
disconnected and bench-calibrated at l e a s t twice.
They were found t o be
accurate t o w e l l within 5%. The containment c e l l was nominally at
+ psig,
but j u s t downstream of
t h e component-cooling pumps, t h e pressure w a s +6 psig.
It was thought t h a t
i f gas were leaking out here and i n t o t h e c e l l at a proportionately higher
rate, we might account f o r t h e anomaly.
Simultaneous solution of t h e leak-
rate equations f o r t h i s s i t u a t i o n showed t h a t although a solution w a s mathematically possible, t h e in-leak would have t o be a t %3%02 and t h e out-leak a t ' 2 0 % 02 which i s not p r a c t i c a l . These r e s u l t s caused us t o t h i n k t h a t something i n t h e c e l l w a s chemic a l l y consuming some of t h e oxygen.
Removal from t h e c e l l air of C2.5
scf/day of pure 02 would account f o r t h e leak-rate discrepancy. One suggested mechanism f o r chemically removing t h e 02 w a s by o i l decomposition.
If some of t h e o i l from t h e component-cooling pumps were con-
t a c t i n g hot pipe (say at FV-103 o r i n control rod thimbles) it would probably at l e a s t p a r t i a l l y decompose according t o t h e approximate r e l a t i o n :
Cell a i r samples were taken t o determine t h e C02 content and they showed
QO.l%
C02.
This was at l e a s t t h r e e times higher than t h e C02 content i n
air, but w a s less than one would expect t o s e e i f all t h e 02 were being consumed by o i l decomposition. The c o n t a i w e n t c e l l and support s t r u c t u r e w a s composed primarily of carbon s t e e l so another prime suspected oxygen depletion mechanism w a s r u s t . If all t h e missing 02 were being consumed t o form FepOg, it would t a k e
about 120 g/day of Fe.
Since t h e r e i s abou;t 5000 f t 2 of carbon s t e e l sur-
face a r e a i n t h e c e l l , t h i s would correspond t o a corrosion r a t e of only 0.5 mils/year (neglecting t h e support s t r u c t u r e and piping of component
coolant system).
;Li
376
This analysis l e d t o t h e conclusion t h a t enough oxygen w a s being re-
moved chemically by o i l decomposition and r u s t i n g t o produce t h e discrepancy t h a t existed between t h e d i f f e r e n t methods of c e l l leak-rate calculation. 22.4
Vapor Condensing System
An accident can be conceived i n which molten s a l t and water could
simultaneously leak i n t o t h e r e a c t o r o r drain-tank c e l l s .
The quantity of
steam produced could be such t h a t t h e pressure i n t h e c e l l s w b u l d increase above design pressure.
A vapor-condensing system w a s provided t o prevent
the steam pressure from r i s i n g above t h e 39-psig design pressure and t o r e t a i n t h e non-condensable gases. vapor-condensing system.
A 12-in. l i n e connected t h e c e l l s
$0
the
This l i n e contained two rupture d i s c s i n p a r a l l e l
(a S i n . d i s c with a bursting pressure of 15 p s i g and a 10-in. ilisc with a bursting pressure of 20 psig). The l i n e from t h e rupture d i s c s went t o t h e bottom of a 1800-ft3 v e r t i c a l tank which contained %1200 f t 3 of water t o condense t h e steam.
The non-condensable gases went from t h e t o p of t h i s
tank t o a 3900-ft3'retention tank. The tanks were pressure-tested by t h e vendor t o 45 psig.
A f t e r being
connected t o t h e reactor c e l l , t h e system w a s leak-tested a t 30 p s i g during i n i t i a l t e s t i n g of the reactor and drain-tank c e l l s i n 1965. During the annual c e l l leak t e s t s at pressure, t h e vapor-condensing system was a l s o pressurized.
This protected t h e rupture d i s c s and provided
an i n t e g r i t y t e s t of t h e vapor-condensing system. The water used i n t h e v e r t i c a l tank contained potassium n i t r i t e potassium borate as a corrosion i n h i b i t o r . i n h i b i t o r addition was made.
Based on annual s&les,
one
There was no increase i n t h e . i r o n concentra-
t i o n which indicated l i t t l e o r no corrosion, I n i t i a l l y 4 l e v e l probes were provided t o assure proper w a t e r l e v e l . However, a f t e r one of t h e l o w level probes failed, a bubbler-type l e v e l instrument w a s i n s t a l l e d f o r measuring t h e l e v e l periodically, Since t h e r e has been no unplanned increases i n t h e pressure of t h e reactor and drain-tank c e l l s , t h e vapor-condensing system has not been used.
u
377
22.5
Recommendations
Leak-checking t h e numerous valves required disconnecting many l i n e s , p a r t i c u l a r l y tube f i t t i n g s and autoclave connections.
Disconnecting nu-
merous f i t t i n g s g r e a t l y increases t h e probability that one o r more w i l l leak when connections are remade.
For t h i s reason, an e f f o r t should be
made i n t h e design of t h e piping t o minimize t h e number of disconnects
necessary f o r leak-checking.
For example, nitrogen l i n e s could be con-
nected as shown i n Fig. 22.5 t o f a c i l i t a t e checking both t h e s a f e t y valve and the l i n e inside t h e c e l l .
All block valves should be of high q u a l i t y and soft-seated.
Accessi-
b i l i t y t o t h e items t o be checked and t h e points used i n checking them should be considered.
F l e x i b i l i t y should be provided i n t h e l i n e s which
must be disconnected. Leaks from i n - c e l l a i r l i n e s cause e r r o r s i n leak-rate calculations. When disconnects are necessary, t h e e f f e c t of f l u x on materials on con-
bi
s t r u c t i o n should be considered. MI-cable s e a l s of t h e type used at t h e MSRE (brass and s t a i n l e s s s t e e l ) should be modified o r another type used.
Substituting f e r r u l e s of
t e f l o n , graphite-impregnated asbestos, or other r e l a t i v e l y soft material f o r the brass f e r r u l e s may be s u f f i c i e n t . t o q r e v e n t leakage through them. However, r a d i a t i o n damage and sheath temperature must be considered.
Stan-
dard pipe-threaded connections f o r s t a i n l e s s steel. t o s t a i n l e s s s t e e l
j o i n t s should be avoided where possible. ed on threaded connections f o r a gas seal
Teflon tape should
nor on small l i n e s where fâ&#x20AC;&#x2122;ragments-can e n t e r t h e l i n e and clog it o r prei
vent check valves from s e a t i n g properly. should-be cleaned i n t e r n
All pipe, tubing, and valves .
r i o r t o installation.
MI-sheathed thermoc
are preferable t o t h e type used.
if cost o r other reasons
However,
seals at t h e ends) preclude using them, t h e
i n i t i a l design should provide f o r pressurizing t h e terminal headers and measuring t h e leakage.
Sol
ed j o i n t s i n t h e header system a r e recom-
mended where practical; Gas a n d y z e r s used t o continuously monitor t h e c e l l atmosphere should
be located i n a w a r m area t o prevent moisture from condensing i n them.
378
379
Fluids used i n manometers o r s i m i l a r gages which a r e continuously connected t o t h e contained volume such as t h e hook gage should have a low vapor pressure t o avoid a s i g n i f i c a n t l o s s during operation.
The piping
t o t h e gage should be pitched t o prevent accumulation of water and should not be connected t o t h e bottom of temperature compensating o r reference volumes. Temperature compensating volumes should represent t h e e n t i r e containment.
The 30-in.
c e l l vent
i n e and t h e component-cooling system enclosure
(5% of c e l l volume) had no reference volume.
w e temperature reference
volume should be d i s t r i b u t e d throughout t h e c e l l s .
The MSRE used 6-in.
v e r t i c a l pipes on a weighted volume b a s i s i n t h e r e a c t o r and drain-tank cells.
Smaller diameter piping with t h e same t o t a l volume might have given
a more repqesentative temperature compensation.
The reference volume system e x t e r n a l t o t h e c e l l s must have a minimum volume and should be located i n an area h e l d at as constant a temperature as possible.
S u f f i c i e n t c e l l temperature thermocouples should be located
i n all sections of t h e containment t o enable calculation of average temperature and temperature changes as accurately as possible. If a s e n s i t i v e c e l l pressure measuring device similar t o t h e hook gage
i s used i n future reactor containment, it should be designed t o meet secondary containment requirements so t h a t block valves a r e not required. However, i f block valves are needed t o i s o l a t e t h e instrument, a separate c i r c u i t from t h e other block valves should be used.
Pressure readings a r e
required above t h e pressures a t which t h e other block valves a r e closed. Safety jumpers could be used t o keep t h i s c i r c u i t open during cell-pressure testing.
,
23.
U
BIOLOGICAL SHIEISING AND RADIATION LEVELS
T. L. Hudson The c r i t e r i o n f o r t h e MSRE b i o l o g i c a l s h i e l d design w a s t h a t t h e dose
rate would not exceed 2.5 mrem/hr during normal operation at any point on t h i e s h i e l d e x t e r i o r t h a t i s located i n an unlimited access area, . Since the MSRE had t o f i t within an e x i s t i n g r e a c t o r containment c e l l and build-
ing, t h e s h i e l d design allowed f o r addition of shielding as needed t o reduce radiation .level at l o c a l i z e d hot spots o
The shielding calculations
are reported i n Reference 57. The reactor vessel was completely surrounded by a water-cooled,
and-water-filled
thermal shield.
The thermal s h i e l d and f u e l c i r c u l a t i n g
1oop.were located i n t h e reactor c e l l . layers of' concrete blocks.
steel-
The t o p of t h e r e a c t o r c e l l had two
An annular space f i l l e d w i t h magnetite sand and
w a t e r provided t h e shielding f o r t h e sides and bottom.
See Fig. 23.1.
When t h e reactor was not i n operation, t h e f u e l was drained t o one o r both drain tanks which were located i n the drain-tank c e l l .
Magnetite con-
u
c r e t e w a l l s faced a l l accessible areas and t h e t o p consisted of two layers of concrete blocks. 23.1
Radiation Surveys -Approach
t o Power
Extensive health-physics surveys o f t h e r e a c t o r a r e a w e r e performed during t h e i n i t i a l approach t o power. l i g h t s h i e l d inadequacies,
These surveys were made t o b r i n g t o
With t h e exception of t h e areas discussed i n
t h e following paragraphs, t h e shielding w a s found t o be adequate. 23.1.1
Coolant Drain Tank C e l l
A t 1 kW t h e s c a t t e r i n g of fast neutron and gamma rays from t h e reactor c e l l i n t o t h e coolant drain tank c e l l through t h e 30-in.
reactor c e l l ven-
t i l a t i o n l i n e 930 caused radiation readings of 8 mrem/hr gamma and 50 mrem/hr fast neutron near t h e e x i t of the l i n e .
A t 25 kW very high readings
were found again a t l i n e 930 (70 mrem/hr gamma, 600 mrem/hr fast neutrons and 30 mrem/hr thermal neutrons).
6 in. of borated polyethylene was coolant drain tank c e l l .
A w a l l of
16 i n . of concrete blocks and
b u i l d ad3acent t o t h e 930 l i n e i n t h e
I n addition, t h e reactor off-gas l i n e 522, i n t h e
LilJ
y S f l 1fl NNQ
"XX,
N01133S
11QM 1133-
A
L
1
snoef
NMOa-OlOH
NQld
T8€
382
u
coolant drain tank c e l l was found t o be giving a high background t o t h e A t t h e end of t h e 25-kW run, t h e reading w a s 100 mR/hr at 1 i n . from
area.
t h e S i n . t h i c k lead s h i e l d around t h e 522 l i n e .
Later a t higher power
levels it w a s determined t h a t part of t h e r a d i a t i o n was from t h e drain-tank vent l i n e 561. Therefore t h i s l i n e was shielded with 3 i n . of lead. Even though shielding w a s added i n s i d e t h e coolant drain tank c e l l , t h e r a d i a t i o n
at t h e door (500 mR/hr gamma, 150 mrem/hr fast neutrons, 75 mrem/hr thermal neutron) and halfway up t h e access ramp (22 mR/hr gamma, 3 mrem/hr fast neutrons, 32 mrem/hr thermal neutron (was high during full-power operation. This- area was c l e a r l y marked with r a d i a t i o n zone signs a t t h e entrance t o t h e ramp and t h e door at t h e bottom of t h e ramp w a s locked during nuclear operation t o prevent entry i n t o t h e c e l l . 23.1-2
North E l e c t r i c Service Area
When t h e reactor power w a s r a i s e d t o 1 MW i n A p r i l 1966, t h e r a d i a t i o n l e v e l i n t h e Morth E l e c t r i c Service Area (NESA) w a s .found t o be high: mR/hr on t h e balcony and 8000 mR/hr a t t h e west w a h .
20
Investigation showed
t h a t t h e r e w a s radioactive gas i n t h e l i n e s through which helium i s added t o t h e drain tanks.
Two check valves i n each l i n e prevented t h e
LlJ
gas from
g e t t i n g beyond t h e secondary containment enclosure, but t h e enclosure, of 1/2-in.
s t e e l , provided l i t t l e gama shielding.
The pressure i n t h e f u e l
system at t h a t ‘time w a s controlled by t h e newly i n s t a l l e d pressure control .
valve with r a t h e r coarse trim, and t h e pressure fluctuated around t h e cont r o l point (normally 5 psig) by k %2%. These pressure f l u c t u a t i o n s caused f i s s i o n product gases t o diffuse more rapidly i n t o t h e drain tanks and back through t h e 1/4-in.
l i n e s through t h e s h i e l d i n t o t h e MESA.
The r a d i a t i o n
level was lessened by i n s t a l l i n g a temporary means of supplying an i n t e r -
-
mittent purge t o t h e gas-addition l i n e s t o sweep t h e f i s s i o n product gases back i n t o t h e drain tanks.
During t h e June shutdown, a permanent purge
system was i n s t a l l e d t o supply a continuous helium purge of 70 cc/min t o each of t h e t h r e e gas-addition l i n e s .
This was proved successful by sub-
sequent full-power operation i n which t h e general background i n t h e MESA
w a s <1mR/hr. Transmitter Room
- w i n g t h e r a d i a t i o n survey a t
5 MW a hot spot of
”‘
30 mR/hr gamma w a s found on t h e southwest corner about 4 f t above t h e floor. _-
383
bd A 2-ft by
4-ft by 1-in. t h i c k l e a d sheet was attached t o t h e w a l l over t h e
hot spot and a radiation zone was established. t h e i n i t i a l approach t o full power, stacked con-
Vent House -During
c r e t e blocks were added t o t h e f l o o r area of t h e vent house, over t h e charcoal beds and between t h e vent house and t h e r e a c t o r building t o keep dose
rates low. bricks.
Very narrow beams coming from cracks were shielded with l e a d
Even so, t h e background radiation l e v e l i n t h e vent house w a s
%7 mR/hr at f'ull power.
The vent house was established as a radiation
zone area.,
Water Room
- Induced
a c t i v i t y i n t h e t r e a t e d water rose t o an unex-
pectedly high level during Run
.
The a c t i v i t y proved t o be 12.4 h r 42K
produced i n t h e corrosion i n h i b i t o r .
A survey of possible replacements
f o r potassium l e d t o t h e choice of lithium, highly enriched i n t h e 'Li tope t o minimize tritium production.
iso-
See Water System (Section 12) of t h i s
report f o r additional d e t a i l s Top of Reactor C e l l
bi
- It was expected t h a t additional shielding would
be required d i r e c t l y ab0 between t h e shield block
he reactor, where t h e r e a r e cracks (Q/2 i n . ) w i n g a full-power run i n J u l y 1966, two very
narrow beams of y with neutron radiation were found between c e l l blocks f o r t h e first time. tron.
These read up t o 10 mR/hr gamma and 60 mrem/hr fast neu-
They were properly marked. 23.2
Radiation Levels During Operation
, the
During operation at
narrow beams were noted from time t o
t i o n a l areas were acceptabl time.
r a d i a t i o n l e v e l s i n a l l opera-
These were mainly
house.
Due t o induced a c t i v i t y i n t h e
t r e a t e d water, areas ne
pment such as t h e surge tank and heat
exchanger were t r e a t e d as
zones.
Periodic r a d i a t i o n surveys have
not disclosed any appreciable changes i n r a d i a t i o n l e v e l s . Some t y p i c a l r a d i a t Table 23.1.
Changes i n
s i n s i d e t h e shielding a r e given i n evels following a shutdown from power are
shown i n Figs. 23.2, 23.3, and 23.4.
384
Table 23.1 Radiation Dose Rates i n Various Areas .During and Following Full-Power Operation
Gamma Dose Rate (R/hr) At
LOCATION
Reactor Cell
.
Drain Tank C e l l Coolant C e l l
d Fuel Sampler-Enricher d Off-gas Sampler
7 MW
7
x 104 4.2 x 103
b React o r Drained
10 k@
5.4 x
103
4.2 x 103
2.4
103
2.6 x
104
React o r Drained andc Flushed
2
103
100
1000 500
-
‘After 5-hOUrS operation a t 10 kW which followed sustained operation at f’ull power. bImmediately following t h e ‘above 5-hr operation a t 10 kW. C
Two days af’ter t h e above f u e l drain.
%nside t h e sampler shielding during sampling.
385
ORNL-DWG 73-608
0 EAST WALL REACTOR CELL - RM 6000-1
DAYS DECAY
ion Level in Reactor Cell
386 ORNL-DWG 73-609
5
2
5
2
d 0
1 2
3 4
5 6
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 DAYS DECAY
Fig. 23.3
Gamma Radiation Level i n Drain Tank C e l l (Rm 6000-6 between FD-1 and FD-2)
387 ORNL-DWG 73-610
2
5
2
Id 5
2
101
aj
5
2
lo0 5
0
1
2
3
4
5
6
7
8
9
10
11
DAYS DECAY
Fig. 23.4
Gamma Radiation Level i n Coolant C e l l (Rm-6010)
12
13
388
23.3 Conclusions The b i o l o g i c a l shielding was adequate as designed except for the few , I
,
l o c a l i z e d areas discussed previously.
Periodic radiation surveys have not
indicated any shif't or deterioration i n any o f t h e shielding.
389
24.
INSTRUMENTATION R. H. Guymon
24.1
Introduction
It i s not wi,--in t h e scope of t h i s report t o cover i n d e t a i l t h e performances of all instrumentation,--An
attempt has been made t o review it
from an operational viewpoint and'report s i g n i f i c a n t items. on instrumentation which
en given i n previous sections on t h e perforndividual systems and components w i l l not be
mance o f t h e e n t i r e plant repeated here.
Information
The per
of t h e on-site computer i s covered i n
Reference 58. 24.2
Description
Most of t h e MSRE instrumentation w a s of conventional type found i n Emphasis w a s placed on assuring ade-
other reactors o r chemical plants.
quate containment and reducing r a d i a t i o n damage. signed t o f a i l safe. 24.2.1
A l l c i r c u i t r y w a s de-
A very b r i e f description follows.
Nuclear Instruments
a l l of t h e nuclear instruments were located i n a 36-in.-dia.
water-filled thimble which extended from t h e high bay
through t h e r e a c t o r c e l l t o t h e v i c i n i t y of t h e reactor Fig. 24.1.) Three uncompensated i o n
d t h e i r associated f a s t - t r i p com-
parators provided s a f e t y i (High r e a c t o r o u t l e t tempe i n a two-out-of-three
f o r scramming t h e c o n t r o l rods.
so caused a rod scram.)
configuration. ovided.
These chambers
and t h e e f f e c t of t t o r e d i n t o a s i g n a l whic recorder.
They wer
.'control interlocks.
6d
These were used d automatic PO-
r position were
orded on a 10-decade logarithmic power rod i n h i b i t , rod reverse, and other
390
Q(NL-UUG 644-625
OUTER SLEEVE ASSEMBLY (WITH EXMNSION JOINT)
1\48-in00
-EXf! JOINT
--
Fig. 24.1
,
&-CONTAINMENT
PENETRATION EL. awn
s&
in.
Elevation View of Nuclear Instrument Penetration
6.’
391
6d
Two compensated ion chambers were used f o r servo control of t h e regul a t i n g rod and f o r some interlocks. The power w a s recorded on a l i n e a r recorder.
Seven decades we
covered by means of manual range s e l e c t o r
switches. One high s e n s i t i v i t y BF3 chamber was provided f o r power indication during f i l l i n g of t h e reactor. 24.2.2
Process Radiation Instruments
Two kinds of detectors, ion chambers and Geiger Mueller tubes, were used t o i n d i c a t e t h e l e v e l of a c t i v i t y i n various process streams and t o provide necessary interlocks. 24.2.3
Health physics Monitoring
Gamma r a d i a t i o n w a s monitored by seven monitrons located throughout
t h e building.
The a i r contamination w a s monitored f o r beta-gamma emitting
p a r t i c l e s by seven constant a i r monitors.
The building evacuation system
operated when two o r more monitrons o r two o r more constant a i r monitors from a s p e c i f i c group of instruments detected a high l e v e l of r a d i a t i o n o r
air contamination.
br
24.2.4
Stack Activity Monitoring
The containment s t a c k a i r w a s checked f o r beta gamma p a r t i c u l a t e s by passing a s i d e stream through a f i l t e r paper and monitoring t h e a c t i v i t y with a GM tube.
A f t e r passing through t h e f i l t e r paper, t h e gaseous sample
passed through a charcoal t r a p which w a s monitored by another GM tube t o detect iodine.
A second side stream passed through another f i l t e r paper.
This was monitored f o r al
a t e s using a thallium-actuated zinc
sulfide screen detector,
1-alumel thermocouples (some of which were i n s t a l l e d spares) wer t h e salt thermocouples exc
A l l of even were located on t h e outside of t h e l i n e s f o r temperature indication.
The seven
ermocouple wells were located as follows:
one i n t h e r e a c t o r neck;
d two spares i n t h e r a d i a t o r i n l e t l i n e ;
and, one and two spare
radiator outlet line.
o r equipment.
The more important
o r multipoint recorders o r indicators.
u
Readout of others w a s accomplished by means of a scanner system which allowed t h e signals from up t o 100 thermocouples t o be s e n t t o a r o t a t i n g
392
mercury switch and subsequently t o an oscilloscope f o r display.
A switch
allowed s e l e c t i o n of any one of f i v e groups of 100 signals. 24.2.6
Pressure Indicators
Most of t h e remote indicating pressure and dp instruments were pneumatic o r e l e c t r i c force elements.
When containment w a s required, t h e vents
were referenced t o atmospheric pressure through r o l l i n g diaphram seals. S t r a i n gage and Bourdon pressure gages were used extensively.
Small changes
i n reactor c e l l pressure were determined using a "hook gage".
This was es-
s e n t i a l l y a water manometer with a micrometer f o r accurately reading changes i n water l e v e l . 24.2.7
Level and Weight Indicators
Bubbler type l e v e l instruhents were .used f o r measuring t h e s a l t l e v e l s i n the f u e l pump, coolant pump, and overflow tank. strument was a l s o i n s t a l l e d i n t h e coolant pump.
A float-type l e v e l in-
The drain tanks were sus-
pended by pneumatic weigh c e l l s t o determine t h e amount of salt t h a t they contained.
I n addition t o t h i s , two resistant-type single-point l e v e l
probes were provided i n each drain tapk.. Sight glasses, f l o a t s , bubblers, dp c e l l s , e t c . , were used i n t h e a u x i l i a r y systems. 24.2 8 Flow MeasurementNo flow instrument wm-provided i n t h e f u e l s a l t loop, however a ven-
turi meter w a s i n s t a l l e d i n t h e coolant salt loop. This w a s a standard venturi with a NaK-filled dp c e l l . Flows i n a u x i l i a r y systems were measured by o r i f i c e s , c a p i l l a r i e s p i t o t tubes, rotameters and matrix type flow elements 24.2.9 .
.
Miscellaneous
A semi-continuous mass spectrometer was used t o monitor t h e coolant
air stack f o r beryllium.
Other instrumentation included t h e following:
pulse-type speed elements, ammeters, voltmeters , and wattmeters f o r t h e salt pump motors; potentiometer and synchro position i n d i c a t o r s f o r t h e
control rods, f i s s i o n chambers, and r a d i a t o r doors; o q g e n and moisture analyzers f o r t h e helium cover-gas system; and an oxygen analyzer f o r t h e c e l l atmosphere.
393
24.3
I n i t i a l Checkout and Startup Tests
A comprehensive functional checkout of t h e control, s a f e t y , and alarm instrument w a s made by Instrument and Controls.personne1 p r i o r t o operation of each system.
Although some design and wiring e r r o r s were found, these
were of minor nature and were e a s i l y corrected. i n s t a l l a t i o n w a s excellent.
I n general, t h e q u a l i t y of
The following were included i n t h i s checkout:
t h e s e t p o i n t s of all switches were adjusted t o t h e proper values; continu i t y and r e s i s t a n c e checks of all thermocouples were made; t h e location of each thermocouple w a s determined by heating t h e thermocouple and measuring t h e voltage at t h e patch panel; and t h e continuity of all c i r c u i t s w a s checked and all recorders were put i n t o operation.
Most of t h i s w a s done
as construction was completed o r as t h e instruments were put i n t o service. P r i o r t o t h e first c i r c u l a t i o n of flush and coolant salts, a complete operational check was made of a l l ' instruments Bcheduled t o be used. This w a s done following t h e instrument s t a r t u p check l i s t . 2 2 This involved: (1)a complete checkout of a l l c i r c u i t s .
6.r
This was done by changing t h e
variable o r i n s e r t i n g a f a l s e s i g n a l at t h e prilsary element and assuring t h a t a l l c i r c u i t r y functioned at t h e proper setpoints; ( 2 ) a complete check of t h e patch panel t o assur
at all thermocouples - were . connected t o t h e at t h e operational records were up t o date;
proper readout instrument (3) a complete test of a1.1
equipment t o assure t h a t it would start
if needed and would ftmctio
t h a t no switches were i n a c t
r l y ; (4) a complete inspection t o assure o r jumpers i n s t a l l e d ; and ( 5 ) t e s t s t o
assure t h a t c r i t i c a l equipmen
ctioned properly, such as rod drop times.
This instrument start year.
Early t e s t s revealed
and o t h e r instrumen k
cated s e t p o i n t s which
l i s t w a s repeated approximately every r r o r s i n wiring, s e t p o i n t s which had drif'ted, t i o n properly.
Later tests indi-
and occasionally a malfunctioning
instrument. Most of t h e t r o u b l s t a r t u p check l i s t s adequate f o r s t a t i s
covered by doing t h e instrumentation iately.
Therefore records a r e in-
394
,
24.4
6
Periodic Testin&
Due t o t h e importance of some instruments o r c i r c u i t s , they were a l s o tested periodically during operation.
A s indicated by t h e sections which
follow, these tests did not reveal many serious troubles w h i c h needed correction.
They did provide assurance t h a t t h e c i r c u i t s should function i f
needed,
I n determining t h e amount and frequency of t e s t i n g , c a r e f u l con-
sideration should be given t o t h e above weighed against t h e harm done t o equipment due t o r e p e t i t i v e t e s t i n g and t h e p o s s i b i l i t y of interruption of operation.
Rod scrams, load scrams, and r e a c t o r drains occurred at t h e
MSRE due primarily t o t e s t i n g of equipment o r instrumentation. The periodic t e s t s made at t h e MSRE and t h e r e s u l t s of these t e s t s a r e given below. â&#x20AC;&#x2122;
24.4.1
Nuclear Instruments
A complete check of a l l nuclear instruments w a s made each month during
nuclear operation.
These were done by Instrument and Controls personnel
using Section 8A of t h e Operating Procedures.22
All t e s t s were s a t i s f a c -
t o r y except those l i s t e d i n Table 24.1. 24 4 2
Process Radiation Monitors
The process r a d i a t i o n monitors were tested each week during operation -by i n s e r t i n g a source near t h e primary detector and noting that each cont r o l i n t e r l o c k functioned properly and t h a t all annunciations occurred. A l l tests were s a t i s f a c t o r y except those l i s t e d i n Table 24.2,
Some e a r l y
d i f f i c u l t y was encountered i n t h a t t h e hole i n l e a d shielding f o r i n s e r t i n g t h e source was not properly placed. 24.4 e 3 Personnel Radiation and Stack Activity Monitorinq Periodic t e s t s of t h e h e a l t h physics monitors and stack a c t i v i t y instruments were o r i g i n a l l y done by MSRE personnel per Operating Procedure
8C,22 This was l a t e r taken over by o t h e r ORNL groups.
The o r i g i n a l sche-
dule f o r t h e h e a l t h physics monitors w a s t o m a k e a source check of each
instrument weekly, a matrix check monthly, and a complete evacuation t e s t semiannually.
As confidence i n t h e instruments w a s established, t h e fre-
quency of these tests was reduced t o monthly, quarterly, and semiannually.
395
Table 24.1
Date
3/22/66
Results of Periodic Tests of Nuclear Instruments
Repairs Necessary Replaced condenser i n t h e power supply of l i n e a r power channel No. 1. Replaced period balance module of nuclear safety channel No. 1.
6/16/67
Changed out high voltage supply of nuclear s a f e t y channel No. 3.
7/14/67
Repaired s c a l e r of wide-range counting channel No. 1.
8/12/68
Replaced relay of nuclear s a f e t y channel No. 3.
1/7/69
Replaced fast t r i p comparator of wide-range counting channel No. 2.
2/3/69
Replaced chamber of wide-range counting channel No. 1.
9/11/69
Replaced s c a l e r of wide-range counting channel No. 2.
10/21/69
Replaced operational amplifier of wide-range counting channel No. 1.
10/20/69
Replaced s c a l e r of wide-range counting channel No. 1.
396
Table ,24.2 Results of Periodic Tests ' O f Radiation Monitors
Troubles Detected
Date
4/13/66 5/10/66 1/9/68 1/26/68 7/21/68
3/3/69 10/10/69
arm
did
not occur, RM-596.
Instrument would not calibrate, RM-557. Two i n d i c a t o r lights burned out, RM-565. Indicator l i g h t burned out, RM-675. Indicator l i g h t burned out, RM-827. Indicator l i g h t burned out, RM-565. Indicator l i g h t burned out, RM-565.
397
cli
The stack monitors were o r i g i n a l l y t e s t e d with a source each week. Due t o t h e i r excellent performance record, t h i s w a s l a t e r reduce monthly tests. Troubles encountered were repaired immediately.
24.4.4
Safety C i r c u i t s
Periodic checks were made of all c i r c u i t s and instruments designated by t h e Instrument and Controls group as being safety.
These ipcluded rod
scram c i r c u i t s , f u e l pump and overflow tank pressures and l e v e l s , helium supply pressures, emergency f u e l drain c i r c u i t s , r e a c t o r c e l l pressures, coolant pump speeds and flows, r a d i a t o r temperatures and sampler-enricher interlocks.
These tests involved simulating a f a i l u r e apd checking as much
of t h e c i r c u i t r y as possible without i n t e r r u p t i n g operations.
A t first
these were performed weekly, then a l l except t h e rod scram checks were changed t o a monthly b a s i s .
A l l instruments functioned s a t i s f a c t o r i l y ex-
cept those indicated i n Table 24.3.
24.5
.Performance' of
t h e Nuclear Safety Instrumentation
These instruments proved t o be very r e l i a b l e . an abnormal number of spurious t r i p s occurred.
There were periods when
Many of these were believed
t o have origfnated i n f a u l t y , vibration-sensitive reltays- fn commercial elec~
t r o n i c switches which provided t h e high temperature t r i p . signals.
Another
possible source w a s the chattering of t h e r e l a y s which-change t h e sensit i v i t y of t h e f l u x amplifi c h a t t e r i n g and elimination tem reduced t h e fiequen The fast t r i p comparat e n t l y l a r g e s i g n a l w a s applie t r i p comparator modules t o e Operation of t h e r e l a y
n t h e s a f e t y c i r c u i t s . .Correction of t h e producing components elsewhere i n t h e . . e t o a very t o l e r a b l e - l e v e x .
found t o be inoperative i f a s u f f i c i e input.
A diode w a s added t o t h e fast
e t h i s difficulty. x i n t h e nuclear system generated con-
s i d e r a b l e noise, making it d i f f i c u l t t o reset t h e safety-system channels. The resistor-diode combination f o r damping t h e voltage induced by r e l a y operation was replaced by a Zener diode and a diode combination t h a t was
bd
more s a t i s f a c t o r y .
398
Table 24.3
Results of Periodic Tests of Safety C i r c u i t s
Date
Troubles Detected
5124165
Noted t h a t PR-522 and PI-522 d i d not agree.
6/9/65
Two s a f e t y channels t r i p p e d when t e s t i t i g one channel, scram setpoint was at 120% i n s t e a d of 150% and low current t e s t did not function properly.
1/15/66
Fuel pump pressur? switch s e t p o i n t needed r e s e t t i n g .
4/10/66
Reactor c e l l pressure switch s e t p o i n t needed r e s e t t i n g .
5/20/66
Two sampler-enricher pressure switches s e t p o i n t s needed resetting.
6/12/66
Sampler-enricher pressure switch needed r e s e t t i n g .
6/15/66
,
Rod motion was jerky.
9 119/66
Two sampler-enricher access door l a t c h e s d i d not function properly.
11/2/66
A s a f e t y channel would not r e s e t .
12/16/66
Reactor c e l l pressure switch s e t p o i n t needed r e s e t t i n g .
11/27/67
Burned-out i n d i c a t o r l i g h t i n s a f e t y channel.
5/13/69
Defective solenoid c o i l on r e a c t o r c e l l b l o c k valve header.
9/8/69
Reactor o u t l e t s a f e t y i n t e r l o c k would not c l e a r .
11/24/69
Sampler-enricher pressure switch s e t p o i n t needed r e s e t t i n g .
399 I n t h e summer of 1967, t h e l-kW 48-V-dc
t o 120-V-ac i n v e r t e r , which
supplies ac power t o one of t h e t h r e e s a f e t y channels, f a i l e d during switching of t h e 48-V dc supply. power t r a n s i s t o r s .
It was repaired by replacement of two
During t h i s same period t h e output of t h e ion chamber
i n s a f e t y channel-2 decreased d r a s t i c a l l y during a nonoperating period, and t h e chamber w a s replaced p r i o r t o resumption of operations.
The trouble
proved t o be a f a i l u r e i n a glass s e a l t h a t allowed water t o e n t e r .the mag, nesia i n s u l a t i o n i n t h e cable, which i s & i n t e g r a l p a r t of t h e chamber. .A period s a f e t y amplifier f a i l e d when l i g h t n i n g struck t h e power l i n e t o
t h e r e a c t o r s i t e , and a replacement amplifier f a i l e d as it w a s being installed.
The f i e l d - e f f e c t t r a n s i s t o r i n t h i s type of'amplifier was sus-
c e p t i b l e t o damage by t r a n s i e n t voltages, and it was found t h a t under some conditions, damaging t r a n s i e n t s could be produced when t h e amplifier was removed from o r i n s e r t e d i n t o t h e system. signed, t e s t e d , and installed:
A protective c i r c u i t was de-
The module replacement procedure w a s modi-
f i e d t o reduce the p o s s i b i l i t y of damage incurred on i n s t a l l a t i o n of t h e module.
b,
Two r e l a y s i n t h e s a f e t y r e l a y matrices f a i l e d , both with open
coil circuits.
A chattering contact on t h e fuel-pump motor current r e l a y
a f e t y channel 2 t o t r i p s e v e r a l times before t h e problem w a s overcome by p a r a l l e l i n g two contacts on t h e same relay. t h e core o u t l e t temper
A defective switch on
e a l s o caused several channel t r i p s and one re-
a c t o r scram before t h e trouble was i d e n t i f i e d and t h e switch w a s replaced. A wiring e r r o r i n a s a f e t y c i r c u i t was discovered and corrected.
Interlocks had recently been added i n t h e "load scram" channels t o drop t h e load when the control rods scram.
A wiring design e r r o r r e s u l t e d i n these interlocks
being bypassed by a s a f e t y
er.
Although t h e c i r c u i t s were wired t h i s way f o r a time before being discovered, t h e scram i n t e r l o c k s were always operative during power operation, s It
Operate" mode when any s a
L a t e i n 1965, t h e pow
2623 r e l a y safety ele-
ments w a s changed from 115-V-ac t o 32-V-dc. ac pickup on t h e other mod routed.
hi
This w a s done t o eliminate
twough which t h e r e l a y c o i l current was
The 115-V-ac relays were not changed a t this time.
These f b c tioned s a t i s f a c t o r i l y u n t i l t h e summer of 1967 when two of t h e 1 5 relays failed. By t h e end of 1967, seven had f a i l e d , a l l w i t h open c i r c u i t s o r
400
s a f e conditions. f o r 32-V-dc
I n April 1968, all 15 were replaced with relays designed
operation.
W
Soon afier i n s t a l l a t i o n , 3 of these f a i l e d due t o
contact welding (unsafe condition).
This was apparently due t o e a r l y
There were no more f a i l u r e s during extensive
failure of defective relays.
tests made at t h i s time o r during subsequent operation.
S t a r t i n g Septem-
ber 30,. 1968, d a i l y tests were conducted on t h e e n t i r e rod scrams r e l a y matrix t o detect s i n g l e failures.
(The matrix had been t e s t e d weekly be-
~
fore t h a t . )
Moise suppressors were i n s t a l l e d across some o f t h e non-
s a f e t y contacts of these relays t o alleviate t h e noise which had sometimes caused d i f f i c u l t y during in-service tests. 24.6
Performance of t h e Wide-Range Counting Channels
I n i t i a l c r i t i c a l i t y tests disclosed t h a t t h e neutron f l u x attenuation i n the instrument penetration did not follow an i d e a l exponential curve. The deviation w a s t o o l a r g e t o be adequately compensated f o r by the verni-
stats i n t h e wide-range counting channels.
This i s i l l u s t r a t e d by curve
A of Fig. 24.2 which shows f i s s i o n counter response (normalized count rate)
vs w i t h d r a w a l from t h e lower end of t h e penetration i n guide tube 6.
W
It
w a s reasonable t o conclude from t h i s , and from similar curves, t h a t t h e excess neutrons responsible f o r t h e d i s t o r t e d p a r t of t h e curve were ent e r i n g t h e penetration along i t s length.
The count r a t e s i n other guide
tubes nearer t h e upper h a l f of the penetration were even more d i s t o r t e d than curve A.
Since a flux f i e l d w i t h attenuation p e r Curve A precluded
successful operation of t h e wide-range counting instrumentation, s h i e l d s of sheet cadmium were i n s e r t e d i n guide tubes 6 and 9 t o s h i e l d t h e f i s s i o n chambers from s t r a y neutrons.
Curve B of Fig. 24.2, normalized count rate
vs distance, shows t h e improvement f o r guide tube 6. Reactor period information from t h e wide-range, counting channels inh i b i t e d rod withdrawal and caused rod reverses.
Reactor period s i g n a l s
from counting channels operating at l o w input l e v e l s a r e characterized by slow response.
This inherent delay produced a problem w i t h servo-controlled
rod withdrawal during start.
I n a servo-controlled start, t h e demand sig-
n a l caused t h e regulating rod t o w i t h d r a w u n t i l t h e period-controlled "with-
d r a w i n h i b i t " i n t e r l o c k operated.
1
If t h e period continued t o decrease, t h e
W
c
c
c . .
V
REACTOR SHIELD-.
t
MIDPLANE OF =RE-
REACTQR VESSEL
CADMIUM SHIELDING SUBASSEMBLY-TYPICAL IN GUIDE TUBES 6 AND 9 SECTION A-A LOCATION OF GUIDE TUBES OUTER SHELL REMOVED FOR CLARITY- SHADED WEDGES ARE CADMIUM -UNSHADED WEDGES ARE ALUMINUM
Fig. 24.2
Guide Tube Shield in the MSRE Instrument Penetration
402 I1
reverse" interlock acted t o i n s e r t the rods i n d i r e c t opposition t o the
servo demand.
LJ
These "withdraw inhibit" and "reverse" t r i p points were
originally established at periods of +20 and +10 sec respectively.
The
delayed low-level response of t h e "inhibit" interlock allowed s u f f i c i e n t incremental rod withdrawal t o produce a 10-sec period and thus cause a reverse.
The s i t u a t i o n was aggravated by coasting of t h e shim-locating motor
i n the regulating rod l i m i t switch assenibly,. . To correct t h i s , t h e "withdraw inhibit" and "reverse" period t r i p points were changed t o +25 and +5 sec, respectively , an electro-mechanical clutch-brake was inserted i n t h e s h i m locating motor-drive t r a i n and dynamic braking c i r c u i t r y w a s i n s t a l l e d f o r the regulating rod drive motor. Throughout operation, d i f f i c u l t y w a s experienced with moisture penet r a t i o n i n t o t h e f i s s i o n chamb-ers.
The cause w a s diagnosed as excessive
s t r a i n and flexing of t h e tygon tubing sheath on the e l e c t r i c a l cables. The average l i f e t i m e was about 6 months p r i o r t o July 1968. A t t h i s time the type of tygon tubing used t o cover the cable and t h e method of sealing - - - ..
the chamber connections were changed.
This seemed t o improve t h e moisture
resistance. I n early 1967, a failure occurred due t o a short i n t h e cable t o the preamplifier.
6iiJ
The drive tube u n i t was modified t o provide f o r controlled
cable bends and there were no reoccurrences of t h i s type of d i f f i c u l t y . 24.7
Performance of the Linear Power Channels
The l i n e a r power channels and rod servo instrumentation performed very well throughout the operation.
When the f u e l was changed t o 233U, t h e rod-
control servo was modified t o allow an increase i n t h e servo dead band t o compensate f o r t h e more rapid flux response of t h e reactor. The compensated ion chambers use a small e l e c t r i c motor t o change compensation.
Early i n 1966 one of these motors had t o be replaced.
A water leak occurred i n one. of t h e compensated ion chambers i n the summer of 1969. When it w a s examined, numerous leaks were found along t h e seam weld i n t h e 321 s t a i n l e s s s t e e l bellows sheathing t h e cable, and the aluminum can at the outer support r i n g was honeyconibed by corrosion.
(The
water i n the instrument shaf't contained lithium n i t r i t e - b u f f e r e d with boric acid f o r inhibition of corrosion. )
LJ
403 24.8 Because o
B F a Nuclear Instrumentation
.
very unfavorable geometry, t h e strongest pract,a
-e neu-
t r o n source would not produce 2 counts/sec * f r o m t h e f i s s i o n counters i n t h e wide-range counting channels u n t i l t h e core vessel was approximately h a l f full of f u e l s a l t ; neither would it produce 2 counts/sec with f l u s h s a l t i n t h e core a t any level.
This was t h e minimum count r a t e required
t o obtain t h e permissive "confidence" i n t e r l o c k which allows f i l l i n g t h e core vessel and withdrawing t h e rods.
Therefore a counting channel using
a s e n s i t i v e B F 3 counter was added t o e s t a b l i s h "confidence" when $he core
vessel was l e s s than h a l f - f i l l e d with f u e l salt. i n 1966.
This was i n s t a l l e d e a r l y
The chamber had t o be replaced i n t h e summer of 1967 due t o mois-
ture leakage i n t o t h e cable. 24.9
No other troubles were encountered.
Nuclear Instrument Penetration
The nuclear power produced at the MSRE was determined by an overall
u
f
system heat balance.
All nuclear power instruments were c a l i b r a t e d t o agree
with t h i s primary standard.
During extended runs at higher powers, t h e
nuclear instruments indicated 15 t o 20% higher than t h e heat balance.
This
w a s found t o be due t o a r i s e i n water temperature i n t h e nuclear i n s t r u ment penetration which apparently changed t h e attenuation c h a r a c t e r i s t i c s of t h e water.
I n June 1966, a heat exchanger system was placed i n operation
to cool the water which reduced t h e temperature change f r o m zero t o full
power t o % 1 8 째 ~ ~ f r o 72 m 'F
and reduced t h e difference i n t h e two power meas-
urements t o about 5%.
elements had t o be replace p a i r s were necessary on t h e e l
ing t h e e n t i r e operation. onics.
Occasional re-
One rod and load scram r e s u l t e d
f r o m a f a l s e s i g n a l from one of these monitors (RE-528).
404 .
W
24.11 Performance of t h e Personnel Radiation . Monitoring and Building Evacuation System During e a r l y t e s t i n g it was found t h a t t h e r e were some areas where
Two additionalâ&#x20AC;&#x2122;horns
the building evacuation horns could not be heard. -.
.
.
and four additional beacon a l a r m lights were i n s t a l l e d .
Other than oc-
casional &nor r e p a i r s , t h e system has functioned s a t i s f a c t o r i l y . 24.12
Performance of the. Stack Monitoring System
The stack monitoring system was very reliable.
The manual range
switching made it d i f f i c u l t t o i n t e r p r e t data from t h e recorder charts and t h e Rustrak recorders were very inconvenient t o use. I
24 e 13 Performance of Thermocouples- and,the. . . Temperature IZtadout-and-f3ontrol Inshrumentation- -
A t o t a l of 1071 thermocouples were i n s t a l l e d a t t h e MSRE.
O f these,
866 were on salt systems (351 on t h e c i r c u l a t i n g loops and 515 on t h e drain tanks, drain l i n e s , and fieeze valves). i n f i v e years of service. maintenance.
O f t h e 1071, only 12 have failed
Five others were damaged during construction and
A breakdown of t h e f a i l u r e s i s given i n Table 24.4.
*
Table 24.4
..
....
.
.
.
Failufes Among t h e 1071 MSRE .Thermocouples
. Nature of Failure
.-
Number
Damaged during construction
3
Damaged during maintenance Lead open during operation
2
Abnormally l o w reading (detached? )
3
Unknown reason
3
.. .. .. .. ......
LE4
Total
6
17
h4
405 \
bd
Only three thermocouple wells were provided i n t h e c i r c u l a t i n g salt systems, one each i n the coolant radiator i n l e t and o u t l e t pipesâ&#x20AC;&#x2122;and one
i n t h e reactor neck. o r vessel walls.
The remaining thermocouples were attached t o t h e pipe
The thermocouples on t h e radiator tubes were insulated
t o protect them from the e f f e c t s of the high-velocity a i r t h a t flows over them during power operation; the others were not insulated and thus were
,
subject t o e r r o r because of exposure t o heater shine and t o thermal coni
vection flow of t h e c e l l atmosphere within t h e heater insulation.
In
March 1965, with t h e f u e l and coolant systems c i r c u l a t i n g salt at isothermal conditions, a complete s e t of readings was taken from a l l t h e thermocouples t h a t should read the temperature of t h e c i r c u l a t i n g s a l t .
data was taken i n June 1967 at t h e start of R u n 12. s e t s of measurements are shown i n Table 24.5.
A similar s e t of
The r e s u l t s of t h e two
Comparison of t h e standard
deviations f o r t h e radiator thermocouples with those f o r t h e other thermocouples shows the e f f e c t of insulation on reducing t h e s c a t t e r .
Comparison
of t h e sets of data taken over two years apart shows very l i t t l e change, certainly no greater s c a t t e r .
Figure 24.3 shows t h a t t h e s t a t i s t i c a l dis-
t r i b u t i o n of t h e deviations on individual thermocouples from t h e mean also changed l i t t l e i n the two years. P
Table 24.5 Comparison of Readings of Thermocouples of S a l t Piping and Vessels Taken with the S a l t Isothermal
Thermocouple
Radiator tubes Other
1206.73 12.3 1102.3 f 10.6
1207.4 k 9.8
406
ORNL- DWG 67-41786
400
90
I
W
I
I-
DATA TAKEN MARCH i965
5 60 r
I-
I
0
-
DATA TAKEN JUNE 4967 1
v)
(3
$ a W
40
a
8
30
k
z
w
2 w
20
n
40
0
.-60
-40
-20
0
20
40
60
DEVIATION FROM AVERAGE (" F) Fig. 24.3
Comparison of MSRE Thermocouple Data fromMarch 1965 and June 1967
407
b
The s c a t t e r i n t h e various thermocouple readings w a s reduced t o an acceptable l e v e l by using biases t o correct each reading t o The o v e r a l l average measured while both f u e l and coolant systems were c i r c u l a t i n g s a l t These b i a s e s were entered i n t o t h e computer and
at isothermal conditions.
were axtomatically applied t o t h e thermocouple readings.
"he biases were
revised at t h e beginning of each run and were checked when isothermal cond i t i o n s e x i s t e d during t h e runs. were reliable,
Generally t h e biased thermocouple readings
However i n a few cases, t h e r e were s h i f t s which caused cal-
culational errors. During e a r l y operation t h e thermocouple scanner gave considerable d i f f i c u l t y due t o 60-cycle noise pickup, poor s t a b i l i t y , and d r i f t i n g of t h e salvaged oscilloscopes.
Refinements were needed i n design t o allow b e t t e r
i d e n t i f i c a t i o n of scanner points and provide a means by which t h e operator could c a l i b r a t e t h e instruments. operated very s a t i s f a c t o r i l y .
A f t e r t h e s e were corrected, t h e system
The r o t a t i n g mercury switches l a s t e d much
longer than t h e expected 1000-hour mean l i f e .
One switch failure occurred
when t h e nitrogen purge gas w a s inadvertently stopped.
6.,
Single point E l e c t r a Systems alarm switch modules w e r e used f o r cont r o l of freeze valves and f o r other alarm and control actions.
These gave
considerable trouble during e a r l y operation due t o d r i f t i n g o r dual setpoints and general maloperation. correct these.
A number of modifications were made t o
Printed circuit-board contacts were gold-plated t o reduce
contact r e s i s t a n c e , t h e t r i m pots used f o r h y s t e r e s i s adjustment were replaced with fixed r e s i s t o r s , and r e s i s t o r values i n modules having ambig\
uous (dual s e t p o i n t s were changed t o r e s t o r e t h e proper b i a s l e v e l s ,
These
changes , together with s t a b i l i z a t i o n , by aging, of c r i t i c a l r e s i s t o r s i n t h e switch modules and more t h e performance.
A check sh
ous periodic t e s t i n g procedures improved t h a t out of 109 switch s e t p o i n t s , 832 had
s h i f t e d less than 20째F over a six-month period.
Multiple s e t p o i n t s s t i l l
occurred and various o t h e r failures were encountered.
1969, records were kept on t h e
During 1968 and
lures of i n d i c a t o r l i g h t s on these mol
dules.
There were about 50 f a i
es per year.
This w a s important because
a burned-out l i g h t bulb could cause alarm o r control action.
408
24.14
3
Performance of Pressure Detectors
Most of the pressure instruments at t h e MSRE performed very well.
Dif-
f i c u l t y was encountered with t h e d i f f e r e n t i a l pressure c e l l used t o obtain the pressure drop i n the helium f l o w tlirough t h e charcoal beds (PdT-556). The span and zero s e t t i n g s s h i f i e d badly although t h e pressure capability had hever been exceeded.
Three dp c e l l s f a i l e d i n t h i s service,
Two of
-these were removed, t e s t e d , and inspected without determining t h e cause of ithe trouble.
The last replacement functioned s a t i s f a c t o r i l y at first but It i s s t i l l i n s t a l l e d .
then gave similar d i f f i c u l t i e s . 24.15
Performance of Level Indicators
The bubbler-type l e v e l instruments used i n t h e f u e l pump overflow tank More details are given in-Sections 5.8
and coolant pumps performed well.
and 6.5.
Some d i f f i c u l t y was encountered i n controlling t h e purge flow un-
t i l t h e t h r o t t l i n g valves were replaced. A high and low level resistance type l e v e l probe was provided on each
drain tank.
During early operation t h e excitation and s i g n a l cable leads
on both probes of t h e m e 1 flush tank failed. . .
These failures were caused
by excessive temperature which caused oxidation and embrittlement of t h e copper-clad,
mineral-insulated copper-wire cables.
These cables were de-
signed on t h e assumption t h a t they would be souted i n a i r .above t h e tank insulation and t h a t t h e i r operating temperature would not exceed 200'F; however, i n the actual i n s t a l l a t i o n , t h e cables were covered with insulat i o n 'and. t-..h e temperature at t h e point of attachment t o t h e probe was probably i n excess of 800'F. Repairs were accomplished by replacing t h e copperclad, mineral-insulated copper wire excitation and s i g n a l cables and port i o n s of t h e probe head assembly with a stainless-sheathed, nickel-wire cable assembly. .. f u e l drain tank No. 1 failed. wire inside t h e c e l l .
ceramic-beaded
I n t h e summer of 1968, one of t h e probes i n The failure was found t o be an open lead
The probe was restored t o service by a cross con-
nection outside t h e c e l l t o t h e equivalent lead of t h e other probe. The' i n i t i a l instrumentation provided t o assure proper water l e v e l i n -..
the' vapor-condensing tank consisted of 4 resistance type probes spaced
.. . .
'
409
bii
4 in. apart near t h e desired t h e instrumentation s t a r t u p
er l e v e l . I n September 1966, while doing . . . . .... k l i s t , one o f t h e two high-level switches
was found t o be defective.
ss of t h i s switch caused t h e l o s s of one chan-
n e l of information needed
stablish t h a t t h e water l e v e l i n t h e tank was
Since a second switch failure might require t h a t t h e r e a c t o r be
correct.
shut down &til the switches could be repaired, and since t h e removal of t h e switches fkom t h e tank is a d i f f i c u l t operation, a bubbler-type l e v e l measuring system, which included a containment block valve and associated s a f e t y c i r c u i t s i n t h e purge supply, w a s designed and was i n s t a l l e d i n t h e reactor-cell vapor suppression tank.
This i n s t a l l a t i o n u t i l i z e d a d i p tube
which w a s included i n t h e o r i g i n a l design i n a n t i c i p a t i o n of such need. This new l e v e l system a l s o enabled t h e operator t o check t h e water l e v e l i n t h e tanks as a routine procedure. 24.16
Performance of t h e Drain Tank Weighing, Systems
The same type pneumatic
Lj
eighing devices were used on t h e f u e l drain
tanks and t h e coolant drain tank.
The coolant drain tank weight indicators
proved t o be stable, showing no long-term drifts o r e f f e c t s of external variables.
With
5756 l b of salt i n t h e tank a t about 1200°F, t h e extreme
spread of 40 indicated weights over a period of a week was k 22 l b .
This
was only k 0.4% and was q u i t e s a t i s f a c t o r y . The indicated weights of t h e t h r e e tanks i n t h e fuel system exhibited rather large unexpl
300 pounds.
The mechani
probably was due t o chan
tures of attgched pipin sure seemed t o a l s o affe
I n some cases these amountedto 200 t o using t h i s was not d e f i n i t e l y established, but forces on t h e syspended tanks as temperatank f’urnaces changed.
e weighing s y
t h e multipositlon pne complished by s e l e c t i n g l e c t o r valves.
The
ion d r i f t , d i f f i c u l t y w a s experienced with t o r switches.
Manometer readout was ac-
i c u l a r weigh c e l l channel’with pneumatic seomposed of a stacked array of individual.
valves operated by cams on t h e operating handle s h a f t .
u
valves gave false weight indications. solved t h i s problem.
ms were useful
ng and d r a i n i
i n observing t r a n s f e r s of I n addition t o t h i
Reactor c e l l pres-
Leaks ‘in these
A redesign of t h e switching device
410
P r i o r t o power operation, one of t h e weigh c e l l s f a i l e d and was replaced. '
The f a i l u r e was determined t o be due t o p i t t i n g of t h e b a f f l e and
nozzle i n t h e c e l l .
This p i t t i n g w a s apparently caused by amalgamation of
mercury with t h e p l a t i n g on t h e b a f f l e and nozzle.
How t h e mercury got in-
t o the weigh c e l l s has not been determined; however, it was believed t o have come from.the manometers and t o have been p r e c i p i t a t e d on t h e b a f f l e by expansion c'ooling of t h e air leaving t h e nozzle.
Appreciable q u a n t i t i e s
of mercury were a l s o found i n t h e tare pressure regulators on t h e control panel; however, no mercury was found i n t h e interconnecting tubing o r i n other portions of t h e system. 24.17
Performance of t h e Coolant S a l t Flowmeters
Several days a f t e r t h e start of coolant salt c i r c u l a t i o n , t h e output of one of t h e two salt.flowmeter channels s t a r t e d d r i f t i n g down s c a l e , output of t h e other c h p n e l remained Steady.
The
The t r o u b l e w a s i s o l a t e d t o
a zero s h i f t and possibly a span s h i f t i n t h e NaK-filled d i f f e r e n t i a l pressure t r a n s m i t t e r i n thk drif'ting channel.
Since t h e exact cause 8could not
be determined, a spare dp c e l l was i n s t a l l e d .
Both channels f'unctioned
s a t i s f a c t o r i l y throughout t h e remainder of t h e r e a c t o r operations.
Tests
on t h e defective u n i t were inconclusive, however,- it was determined t h a t t h e s h i f t s were temperature-induced zero s h i f t s possibly caused by incomp l e t e f i l l i n g o r a leak i n t h e s i l i c o n e o i l portion of the-instrument,
An operational inconvenience persisted'throughout'operations,
When-
ever t h e r a d i a t o r air flow was increased, a.+i r leaking through t h e insulat i o n around t h e salt l e g s t o t h e dp c e l l s caused t h e temperature t o decrease rapidly.
This necessitated adjustment of t h e heaters.
Due t o t h e discrepancy between t h e r e a c t o r power l e v e l indicated by f u e l burnup, heat balance, e t c , , it is planned t o r e c a l i b r a t e t h e dp c e l l s during t h e next f i s c a l year.
g4
.
24.18
411
Performance of Relays
The d i f f i c u l t i e s encountered with t h e control rod r e l a y s a r e described i n Section 24.5.
Experience with other relays i s given below.
A f t e r about 2 years of operation, t h e 48-V-dc-operated
considerable heat damage t o t h e i r b a k e l i t e frames.
relays showed
The manufacturer,
General E l e c t r i c , advised t h a t overheating of t h i s p a r t i c u l a r model was
a common problem i f t h e relays were continuously energized.
Early i n 1967,
twenty of t h e 139 relays were replaced with a l a t e r improved model. a f e w months some of these a l s o showed signs of deterioration.
Within
Therefore
i n June 1967, a l l 139 relays were field-modified by replacing t h e b u i l t - i n r e s i s t o r s with externally mounted r e s i s t o r s .
No trouble was experienced
a f t e r t h i s modification. I n September 1969, t h e load-scram c i r e d . t , which drops t h e r a d i a t o r doors and stops t h e blower, t r i p p e d several times.
Investigation showed
t h a t some of t h e r e l a y contacts had developed unusually high resistance
due t o oxide films.
LJ
Because of t h e way t h e contacts were p a r a l l e l e d i n
t h e matrix, t h e film was not burned o f f each time t h e contact closed, as i n a normal application.
These contacts were cleaned and no f u r t h e r dif-
f i c u l t y was encountered.
24.19 Two "on-site''
Training Simulation
r e a c t o r k i n e t i c s simulators w e r e developed f o r t h e pur-
pose of t r a i n i n g t h e MSRE operators i n nuclear s t a r t u p and power operation. 'Phe s t a r t u p simulator used t h e control rod p o s i t i o n s i g n a l s as inputs, and provided outputs of log count r a t e , period, l o g power, and l i n e a r power. The r e a c t o r ' s period interlocks, f l u x control system, and l i n e a r f l u x range s e l e c t o r were a l s o operational.
I n addition t o t h i s , t h e power l e v e l simu-
l a t o r used t h e r a d i a t o r door position and cooling a i r pressure drop signals
as inputs and provided readout of key system temperatures.
Both simulators
were s e t up on general purpose, portable EA1 TR-10 analog computers.
Much
of t h e a c t u a l MSRE hardware, such as control rods were used r a t h e r than simulated.
Thus t h e operators manipulated t h e a c t u a l r e a c t o r controls and
became used t o t h e instrumentation and controls system.
a very r e l i a b l e t r a i n i n g t o o l .
This proved t o be
412
24.20 . Miscellaneous The original mass spectrometer usedto monitor for beryllium in the coolant stack was replaced with an improved instrument near the start of power operation. Only occasional repairs or preventative maintenance was required since then. Folded charts were used on some recorders. These did not function properly due to the low chart speeds being used.
24.21 Conclusions and Recommendations Considering the quantity and complexity of the instrumentation, a minimal amount of difficulty was encountered. Since the MSRE was an experimental reactor, it was in m a y areas over-instrumented. This was probably due to hot-knowingwhat information might be needed and not taking enough credit for the ability-of the operators. This led to unnecessary difficulties in normal operations or in running special experiments. At the same time, there were areas where additional information'would have been beneficial
-
-.
hs
413 REFERENCES 1.
M. W. Rosenthal, P. Re Kasten, and R. B, Briggs, "Molten-Salt Reactors History, Status d Potential," NucZ. AppZ. Tech., E, 107 (1970)
-
.
2.
MSR Program Semiannu. Progr. Rep., Jan. 31, 1964, OWL-3626.
3.
MSR Program Semiannu.
4. MSR Program Sem&mnu.
gr. Rep,, July 31, 1967, ORNL-3708.
. Rep.,
Feb. 28, 1965, ORNL-3812.
5. MSR Program Semhanu.
. Rep.,
Aug. 31, 1965, OWL-3872.
6.
MSR Program Semiannu.
. Rep,,
Feb. 28, 1966, ORNL-3936.
7.
MSk Program Sembanu. Progr. Rep., Aug. 31, 1966, ORNL-4037.
8. MSR Program Sem&mnu. Progr. Rep., Feb, 28, 1967, ORNL-4119. 9.
MSR Program Semiannu.
g r . Rep., Aug. 31, 1967, OWL-4191.
10. MSR Frogram S e ~ a n n u .Progr. Rep., Feb. 29, 1968, ORNL-4254. 11. MSR Program Sem&mnu.
hi,
gr. Rep., Aug. 31, 1968, ORNL-4344-
12.
MSR Program Sermhnnu. Progr. Rep., Feb. 28, 1969, ORNL-4396.
13.
MSR Program Sem&mnu, Pr
14. MSR Program Semiannu. 15.
R. C. Robertson, MSRE
16.
R. â&#x201A;ŹI Guymon, .
Rep., Aug. 31, 1969, ORNL-4449. Rep., Feb. 28, 1970, OW-4548. and Operations Report, Part I -De-TM-728 (Jan. 1965). enreich, and J. R. Engel, MSRE Design and Program, ORNL=TM-911 (Nov. 1966). he MSRE Flush
Fuel Salts, ORNL-
.
Houtzeel, Spray, Mist, Bubbles 7 (June 1970) nation of the
on, Plans for 2974 (April 1970). urement and Heat-Transfer Performance
bi
21,
J. R. Engel, MSRE Design and Operations Report, Part XI-A- Test Program for U'" Operation, ORNL-TM-2304 (Sept. 1968).
I
!
!
414 References (continued) 22.
R, H. Guymon, MSRE Design and Operations Report, Part VIII, Operating Procedures, Vols, I and 11, ORNL-TM-908 (Dec. 1965).
23.
R. E. Thoma, Chemical Aspects of MSRE Operations, 0-4658
24.
R. H. Guymon and P, N, Haubenreich, MSRE Design and Operations Report, Part VI, Operating Safety Limits for the Molten-Salt Reactor Experiment, ORNL-TM-733 (3rd revision), (July 25, 1969).
25.
R. H. Guymon, R. C. Steffy, Jr., and C. H. Gabbard, Preliminary Evaluation of the Leak in the MSRE Primary System Which Occurred during the Final Shutdown, ORNL-CF-70-4-24, (April 22, 1970).
26.
P. G. Smith, Development of Fuel- and Coolant-Salt Centrifugal Pumps for the MSRE, ORNL-TM-2987 (Oct. 1970).
27.
C. H. Gabbard, Inspection of MSRE Fuel Circulating Pump after the Zero-Power Experiments (Run 31, ORNL-CF-66-8-5 (Aug. 1966).
28.
R. B. Briggs, Measurement of Strains in Heat Exchanger Nozzle and Piping of MSRE, ORNL-CF-66-2-66, (Feb. 1966).
29.
C. H. Gabbard, Thermal-Stress and Strain-Fatigue Analyses of the MSRE Fuel and Coolant Pump Tanks, ORNL-TM-78 (Oct. 1962).
30.
C. H. Gabbard, R. J. Kedl, and H. B. Piper, Heat Transfer Performance of the MSRE Heat Exchanger and Radiator, ORNL-CF-67-3-38 (Mar. 1967).
31.
C. H. Gabbard, Reactor Power Measurement and Heat Transfer Performance in the Molten Salt Reactor Experiment, O R N G W 3 0 0 2 (May 1970).
32.
H. E. McCoy, An Evaluation of the Molten Salt Reactor Experiment Hastelloy N Surveillance Specimens -First Group ORNL-TM 1997 (Nov. 1967).
33.
H. E. McCoy, An Evaluation of the Molten Salt Reactor Experiment Hastelloy N Surveillance Specimens -Second Group ORNL-TM-2359 (Feb, 1969).
34.
H. E. McCoy, An Evaluation of the Molten Salt Reactor Experiment Hastelloy N Surveillance Specimens - Third Group, ORNL-TM-2647 (Jan. 1970)â&#x20AC;&#x2122;.
35.
H. E. McCoy, An Evaluation of the Molten Salt Reactor Experiment Hastelloy N Surveillance Specimens -Fourth Group, ORNL-TM-3036 (Mar. 1971).
36.
H. E. McCoy and B. McNabb, Intergranular Cracking of INOR-8 in the MSRE, ORNL-4829, (Nov. 1972).
37.
C. H. Gabbard, Design and Construction of Core Irradiation-Specimen Array for MSRE Runs 19 and 20, ORNL-%2743 (Dec. 1969).
(Dec. 1971).
-
415 References (continued) 38. R. B. Briggs, Effects of Irradiation of Service Life of MSRE, ORNLCF-66-5-16 (May 1966). 39. R. B. Briggs, Assessment of Service Life of MSRE, ORNL-CF-69-8-3, (Aug. 1969).
40, B. E. Prince et al., Zero-Power Physics Experiments on the Molten-Salt Experiment, ORNL-4233 (Feb. 1968).
41. J, R. Engel and B. E. Prhce, Zero-Power Experiments with 2ssU in the MSRE, ORNL-TM-3963 (Dec. 1972). 42. C. H. Gabbard and C. K. McGlothlan, A Review of Experience with the MSRE Main Blowers, ORNL-CF-67-4-1, (Abg. 1967). +
0 43. C. H. Gabbard, Bearing Failures of Main Blowers MB-1 and MB-2, CF-67-4-1, Addendum-l mar. 19683.
44. R. L. Baxter and D. L. Bernhard, "Vibration Tolerances for Industry," paper presented at ASME Plant Engineering and Maintenance Conference, Detroit, Michigan, Paper No. 67-PEM-14 (Apr. lW12, 1967).
u
45. E. C. Parrish and R, W. Schneider, Tests of High Efficiency Filters and Filter Installation at ORNL, ORNL-3442, (May 1963).
46. H. R. Payne, The Mechanical Design of the MSRE Control Rods and Reactor Access Nozzle, MSR-61-158 (June 1961). 47. J. L. Crowley, Temperature and Salt Levels in the MSRE Reactor Access Nozzle - A Possible Relation to MSRE Power Blips, MSR-69-13 (Feb. 1969).
48. P. N. Haubenreich, R. Blumberg, and M. Richardson, "Maintenance of the Molten-Salt Reactor Experiment" paper presented at Winter Meeting, A N S , Washington D. C. (Nov, 1970).
49. C. H. Gabbard, Radiation Dose Rates to the MSRE Fuel-Pump Motor and Control Rod Drives, MSR-65-48. 50. R. C. Robertson, MSRE pp. 244-75 (Jan. 1965).
gn and Operations Report, 0W-m-728,
-
51. R. H. Guymon, MSRE Design and Operations Report, Part VII, Operating Procedures, ORNL-TM-908, Vol. I1 (Dec. 1965). 52. R. B. Gallaher, MSRE Sampler-Enricher System Proposal, ORNL-CF-61-5-120, (May 1961). 53.
MSRE S t a f f , MSRE Sampler-Enricher, An Account of Recent Difficulties and Remedial Action, MSR-67-73 (Sept. 1967).
il
References (cpntinued) 54.
P. N. Haubenreich and MSRE Staff, An Account of Difficulties with the Sampler-Enricher that Led t o a Second Capsule Being Left in the Pump Bowl, MSR-68-125 (Sept. 1968).
55.
S. E. Beall et al., MSRE Design and Operations Report, Part V, Reactor Safety Analysis Report, ORNL-TM-732 (Aug. 1964)
56.
R. H. Guymon, Tests of the MSRE Reactor and Drain Tank Cells, MSR62-95 (Nov. 1962).
57.
P. N. Haubenreich et al., MSRE Design and Operations Report, Part 111, Nuclear Analysis, ORNL-TM-730 (Feb. 1964)
58.
Go Bo Burger, J, R. Engel, and C, De Martin, Computer Manual for MSRE Operators, ORNL-CF-67-1-28 (Jan. 1967)
59.
P. N. Haubenreich, Safety Considerations in R6sumption of MSRE
.
.
Operation, ORNL-CF-69-8-10
(Aug. 1969).
.
60.
P. N. Haubenreich, Responses to RORC Recommendations for MSRE, MSR69-104
61.
B. F. Hitch et al., Tests of Various Particle Filters for REmoval of Oil Mists and Hydrocarbon Vapor, ORNL-TM-1623 (Sept. 1966).
62.
W. C. Ulrich, Purge of MSRE Off-gas Line 522, MSR-66-8, (Mar. 1966).
63.
R. H. Guymon and J. R. Engel, Introduction of Oxygen in Fuel-Drain Tank, MSR-67-1 (Feb. 1967).
bd
U
417 OWL-TM-3039 I n t e r n a l Distribution
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
h*,
12. 13. 14-18. 19. 20. 21. 22. 23. 24. 25. 26.
S. E. Beall S. Bettis
E. R. R. W. J. F.
Blumberg B. Briggs B. C o t t r e l l L. Crowley L. Culler S. J. Ditto J. E. Engel D. E. Ferguson A. P. Fraas C. H. Gabbard W. R. Grimes R. H. Guymon P. H. Harley P. N. Haubenreich H. W. Hoffman T. L. Hudson P. R. Kasten A. I. Krakoviak K e r m i t Laughon, AEC-OSR M. I. Lundin
27. 28. 29. 30. 31. 32. 33. 34. 35. 36-37 38. 39. 40. 41. 42. 43. 44 45. 46. 47-48. 49. 50-52. 53.
R. N. Lyon R. E. MacPherson H. E. McCoy H. C. McCurdy A. J. Miller R. L. Moore L. C. Oakes A. M. Perry M. Richardson M. W. Rosenthal Dunlap Scott M. R. Sheldon M. J. Skinner I. Spiewak D. A. Sundberg J. R. Tallackson R. E. Thoma D. B. Trauger G. D. Whitman Central Research Library Y-12 Document Reference "action Laboratory Records Department Laboratory Records (RC)
External Distribution
54-56. 57-58. 59-60. 61. 62. 63. 64. 65. 82. 83.
85.
Director, Division of Reactor Licensing, USAEC, Washington D.C. 20545 Director, Division of Reactor Standards, USAEC, Washington, D.C. 20545 N. Haberman, USAEC, Washington, D.C. 20545 D. F. Cope, AEC-ORO M. Shaw, USAEC, Was ton, DOC. 20545 David E l i a s , USAEC, Washington, D.C. 20545 A. Houtzeel, TNO, 176 Second Ave., Waltham, Mass. 02154 R. C. S t e f f y , TVA, 303 Power Building Chattanooga, Tenn. 37401 Manager, Technical Information Center, AEC (For ACRS Members) Research and Technical Support Division, AEC, OR0 Technical Informatio