HP_2013_05

MAINTENANCE AND RELIABILITY

速

Advances in monitoring/repair methods and new equipment designs increase unit availability

SAFETY HydrocarbonProcessing.com | MAY 2013

Investigation determines root cause for flare header failure

PETROCHEMICAL DEVELOPMENTS More efficient energy usage increases profitability for olefins production

TM

SENTRON LD 5000 Field Tested. Field Proven.

Petro-Canada is a Suncor Energy business TM

Trademark of Suncor Energy Inc. Used under licence.

Better performance, less downtime, better profitability – everyone approves of that.

Call 1-866-335-3369 or visit sentron.ca to begin your trial.

Select 56 at www.HydrocarbonProcessing.com/RS Select 76 at www.HydrocarbonProcessing.com/RS

MAY 2013 | Volume 92 Number 5 HydrocarbonProcessing.com

78

38 SPECIAL REPORT: MAINTENANCE AND RELIABILITY

39 Avoid hidden costs of suction-specific speed in pumping 43 49 53 57 61 69 75

J. Bailey and S. Bradshaw Rerating rotating equipment optimizes olefins plant performance D. Renard Maximize steam unit performance with precise torque monitoring T. Mayne, M. Ellul and D. Phillips Investigate power limitations in a large steam turbine N. Ghaisas Seal safety may require going beyond typical standards S. Shaw Prevent methane hydrate formation in natural gas valves A. Glaun and J. Shahda Consider both actual and virtual spare parts inventory H. P. Bloch Improve design for pump suction nozzles M. G. Choudhury, A. Kulkarni and D. Koranga

BONUS REPORT: PETROCHEMICAL DEVELOPMENTS

79 Consider novel CGC and front-end depropanizer system 85

for olefins production J. Fu, C. Zhao and Q. Xu Enhance operation and reliability of dividing-wall columns J. Shin, J. Lee, S. Lee, B. Lee and M. Lee

TERMINALS AND STORAGE—SUPPLEMENT 92 Overflow systems are the last line of defense M. Toghraei 94 Terminals and storage news

REFINING DEVELOPMENTS

DEPARTMENTS

4 8 11 17 110 113

Industry Perspectives Brief Impact Associations Marketplace Advertiser index

COLUMNS

21

Reliability Manage your time constructively

25

Integration Strategies Asset information management improves life cycle benefits for equipment

29

Boxscore Construction Analysis Jurong Island—Asia’s preemptive LNG trading hub?

35

Viewpoint Consider integral-gear compressors in CO2 services

114

Engineering Case Histories Case 72: Interaction between disciplines when troubleshooting

99 Optimize vacuum ejector operations T. Temur, M. Haktanir, F. Uzman, M. Karakaya and A. K. Avci

SAFETY/LOSS PREVENTION

105 Flare header failure: An investigation J. Tharakan

HP ONLINE EXCLUSIVE Innovations Cover Image: Eastern Petrochemical Co. (SHARQ), a joint venture between SABIC and a Japanese consortium headed by Mitsubishi, is a world-scale polyethylene (PE) complex. The facility produces 800,000 tpy of high-density PE (HDPE) and linear-low-density PE (LLDPE), in addition to the extrusion and automatic bagging and palletizing lines. Linde Engineering Dresden GmbH fulfilled the construction contract for both PE processing operations (HDPE and LLDPE ), reactors and extrusion lines. Photo courtesy of Linde Engineering Dresden GmbH, Dresden, Germany.

www.HydrocarbonProcessing.com

Industry Perspectives US-based ultra-low-sulfur gasoline rule released In late March, the US Environmental Protection Agency (EPA) released the long-awaited ruling for ultra-low-sulfur gasoline, also known as the Tier 3 rule. This proposed rule lowers the sulfur content of gasoline by 60% from the present level of 30 ppm to 10 ppm by 2017. The goal of Tier 3 is to prevent 2,400 premature deaths per year and 23,000 cases of respiratory ailments in children, all totaling $8 billion and $23 billion in health-related benefits. It supports efforts by states to reduce harmful levels of smog and soot sourced from vehicle emissions. The new gasoline sulfur rule has support from state governors, public health groups, environmentalists and the auto industry.

The US Tier 3 rule follows a similar 10-ppm specification set by the European Union’s Euro V enacted in 2009. EPA contends that the refining industry can achieve these quality specifications with available technologies and at reasonable costs.

Counterpoint. The American Petroleum Institute (API) and the American Fuel and Petrochemical Manufacturers (AFPM) offer different opinions. EPA’s proposed Tier 3 fuel regulations could raise refiners’ costs, provide little or no environmental benefit, and actually increase carbon emissions, according to API. “There is a tsunami of federal regulations coming out of the EPA that could put upward pressure on gasoline prices. EPA’s proposed fuel regulations are the latest example. Consumers care about the price of fuel, and our government should not be adding unnecessary regulations that raise manufacturing costs, especially when there are no proven environmental benefits. We should not pile on new regulations when existing regulations are working,” said to API Downstream Director, Bob Greco.

According to API, EPA’s Tier 3 proposal would increase the cost of gasoline production by up to 9¢/gal based on an analysis by the energy consulting firm, Baker & O’Brien. If EPA adds a vapor-pressure reduction requirement in a separate regulation, it would push the cost increase up to 25¢/gal, according to Baker & O’Brien.

“Tier 3 rulemaking that targets trace amounts of sulfur in gasoline is not worth the direct threat to our domestic fuel supply, consumer cost at the pump and American jobs,” according to AFPM President Charles T. Drevna. US refiners have already spent billions of dollars to achieve a 90% reduction in sulfur levels, but Tier 3 will require another $10 billion in new infrastructure and another $2.4 billion/yr in operating costs.

Balance. Tier 3 specifications will increase greenhouse gas (GHG) emissions to produce the 10-ppm sulfur gasoline. The goal of Tier 3 is to improve air quality. Yet, the path to Tier 3 will generate more GHGs to process crude oil into ultra-lowsulfur transportation fuels.

An expanded version of Industry Perspectives can be found online at HydrocarbonProcessing.com.

4 MAY 2013 | HydrocarbonProcessing.com

PUBLISHER

P. O. Box 2608 Houston, Texas 77252-2608, USA Phone: +1 (713) 529-4301 Fax: +1 (713) 520-4433 Editorial@HydrocarbonProcessing.com Bret Ronk Bret.Ronk@GulfPub.com

EDITORIAL Editor Reliability/Equipment Editor Managing Editor Technical Editor Online Editor Associate Editor Director, Data Division Contributing Editor Contributing Editor Contributing Editor

Stephany Romanow Heinz P. Bloch Adrienne Blume Billy Thinnes Ben DuBose Helen Meche Lee Nichols Loraine A. Huchler William M. Goble ARC Advisory Group

MAGAZINE PRODUCTION Vice President, Production Manager, Editorial Production Artist/Illustrator Graphic Designer Manager, Advertising Production

Sheryl Stone Angela Bathe David Weeks Amanda McLendon-Bass Cheryl Willis

ADVERTISING SALES See Sales Offices, page 113.

CIRCULATION Director, Circulation

Suzanne McGehee +1 (713) 520-4440 Circulation@GulfPub.com

SUBSCRIPTIONS Subscription price (includes both print and digital versions): Print—One year $239, two years $419, three years $539. Digital format—One year $239.

Airmail rate outside North America $175 additional a year. Single copies $35, prepaid.

Because Hydrocarbon Processing is edited specifically to be of greatest value to people working in this specialized business, subscriptions are restricted to those engaged in the hydrocarbon processing industry, or service and supply company personnel connected thereto.

Hydrocarbon Processing is indexed by Applied Science & Technology Index,

by Chemical Abstracts and by Engineering Index Inc. Microfilm copies available through University Microfilms, International, Ann Arbor, Mich. The full text of Hydrocarbon Processing is also available in electronic versions of the Business Periodicals Index.

ARTICLE REPRINTS If you would like to have a recent article reprinted for an upcoming conference or for use as a marketing tool, contact Foster Printing Company for a price quote. Articles are reprinted on quality stock with advertisements removed; options are available for covers and turnaround times. Our minimum order is a quantity of 100.

For more information about article reprints, call Rhonda Brown with Foster Printing Company at +1 (866) 879-9144 ext. 194 or e-mail rhondab@FosterPrinting.com.

Hydrocarbon Processing (ISSN 0018-8190) is published monthly by Gulf Publishing Company, 2 Greenway Plaza, Suite 1020, Houston, Texas 77046.

Periodicals postage paid at Houston, Texas, and at additional mailing office.

POSTMASTER: Send address changes to Hydrocarbon Processing, P.O. Box 2608, Houston, Texas 77252.

Copyright © 2013 by Gulf Publishing Company. All rights reserved.

Permission is granted by the copyright owner to libraries and others registered with the Copyright Clearance Center (CCC) to photocopy any articles herein for the base fee of $3 per copy per page. Payment should be sent directly to the CCC, 21 Congress St., Salem, Mass. 01970. Copying for other than personal or internal reference use without express permission is prohibited. Requests for special permission or bulk orders should be addressed to the Editor. ISSN 0018-8190/01.

President/CEO Vice President Vice President, Production Business Finance Manager

John Royall Ron Higgins Sheryl Stone Pamela Harvey

Part of Euromoney Institutional Investor PLC. Other energy group titles include: World Oil and Petroleum Economist Publication Agreement Number 40034765

Printed in USA

)/(;,7$//,& 6$)( ,6 025( 7+$1 $ 352*5$0 ,7¡6 $ :$< 2) '2,1* %86,1(66 ,7 ,6 285 '(',&$7,21 72 ,1129$7,9( 0$7(5,$/6 &86720 (1*,1((5,1* $1' 75$,1,1* $33/,(' 72 6($/,1* 62/87,216 7+(6( 6833257 $ +,*+(5 /(9(/

Ă H[LWDOOLF FRP

ITâ&#x20AC;&#x2122;S SAFE

2) 3527(&7,21 )25 <285 :25.(56 &20081,7< $1' 7+( (19,5210(17

innovate/customize/educate Select 93 at www.HydrocarbonProcessing.com/RS

Conference Delegate Bag Sponsor:

Technical Program Sponsor:

NEW DELHI, INDIA | 9–11 JULY

IRPC 2013: At The Forefront of New Technology and Operations Advancements at Refineries and Petrochemical Plants

Refining Track Sponsor:

Petrochemical Track Sponsor:

Join your peers and leaders from the HPI industry in a high-level discussion and look at the latest advances in technology and operations at refineries and petrochemical plants.

This year, advancements in technology and operations will be explored, including the unique opportunities found in greater refinery and petrochemical integration. During the two days of technical sessions, well-respected industry speakers will give case studies, examples and insight into technology and trends that are revolutionizing the HPI.

Program Print Sponsor:

Agenda Highlights Include: Speaker Gift Sponsor:

Conference Lanyard Sponsor:

Conference Host:

KEYNOTE SPEAKER:

KEYNOTE SPEAKER:

R.K. Ghosh, Director (Refineries), Indian Oil Corporation Limited (Refineries Division)

Dr. Ajit Sapre, Group President, Research and Technology, Reliance Technology Group

TRACK 1 HIGHLIGHTS: REFINING

TRACK 2 HIGHLIGHTS: PETROCHEMICALS

Profit Improvement Program in KNPC Refineries - Abbas Shamash, Kuwait National Petroleum Company

Optimization of Olefin Plants - Vera Varaprasad, Indian Oil Corporation Limited

Challenges in Design and Engineering of High Pressure Hydrotreaters and Avenues for Energy Optimization - Mr. K. Sudhaker, L&T Chiyoda Ltd.

Case Study: Design, Development, and Deployment of Energy Management System (ISO-50001:2011) at an Integrated Petrochemicals Complex - Mayur Talati, Reliance Industries Ltd.

Petrobras ULSD Revamps - Silvio Jose Vieira Machado, Petrobras

Challenges and Emission Reduction in Integrating Refineries and Petrochemical Plants, Romel Bhullar, Fluor Corporation; Anil Rajguru and Les Antaffy, Fluor Enterprises

IRPC 2013

EVENT

8:30-9:15 a.m.

Agenda

Day 2 | Thursday , 11 July

HPIRPC .com

CONTIN ENTAL BREAK FAST OPENIN G REMAR KS: Stephany Romanow, Romano KEYNO Editor, Hydrocarbo TE SPEAK n Processing ERS Dr. Ajit Sapre, Group President, Research and Technology Technolog COFFE EE , Reliance Technology E BREAK Group; ExxonMobi TRACK l Singapore K 1 - REFINI (invited) NG SESSIO N Session Chair: 7 TRACK JULY K.Venkatara K.Vennkatar aramanan, 2 - PETRO manan, INDIA | 9–11 n, CEO and Managing NEW DELHI, CHEMICALS Director, Directo Larsen SESSIO Petrobras ULSD N Toubro Revamps Session Chair: 8 Silvio Jose Vieira John Solutions InternationBaric, Licensing Technology Machado, Petrobras Petro Advanced Techniques al B.V. Manager, Shell Phyrophoric Global Technip Stone ues for Enhancing Hydrogen Hazard & Webster Industry - Renato of Catalyst Handling ebs bsterr Proc Availability Process ocess Techn in Refining - Sanjiv Ratan, Technology Benintendi hnology and Petrochem , Foster Wheeler Economic Catalysis ical Improve Profitabilit Solution for Hydrogena y by Flexibility Acetylene and CRI/Criteria BREAKFAST tion Hydrocrack n Marketingg MAPD Selective CONTINENTAL Company a Clariant Group - Shankhaneel Borah, ing Technologi Asia Asia Pacific 8:30-9:15 a.m.

Pacifi Sud Chemie and CEO, Gulf Publishing ific Pte. LTD Company es - Li Hui Ng Recovery of LTTD India Limited, John Royall, President Valuables - Extending rom Refinery OPENING REMARKS: Margins - Siddartha from the Performan Off 9:15-9:30 a.m. ce of Maximum - Vipan Goel, Murkerjee, kerjee, Air LLiquideGases to Increase Profit Division) Indiate Private W.R. Grace Propylene Catalyst iquide GGlobal KEYNOTE SPEAKERSIndian Oil Corporation Limited (Refineries Wood Mackenzie Limited 1-2 p.m. lobal E&C Solutions and Additives Comparative 9:30-10:45 a.m.

(Refineries), Suresh Sivanandam, Study of Conventiona R.K. Ghosh, Director LUNCH for Refining in Asia Bio Opportunities Based and l Petrochemic (Bio-Ethylen 2:00 - 2:30 Future Challenges als (Ethylene e Glycol) Methodology COFFEE ICALS and Foot Printing Production with the ApplicationGlycol) With the & DESSE Tools - Sharma COFFEE BREAK TRACK 2 - PETROCHEM RTS - EXHIBI 10:45-11 a.m.

Rajeev Kumar, of Life Cycle TRACK India T HALL Glycols 2 1 Limited - REFINI 2:30-4:30 SESSION TRACK 1 - REFINING NG Marketing Director, Axens p.m.

G Session Chair: Eric Benazzi, FlexibilitySESSION 9 SESSION 1 and Associates Enhances Olefins Producers’ Session Chair: 11 a.m.-1 p.m.

Poddar, President, Poddar TRACK B.K. Namdeo, Catalytic Olefins Technology KBR Session Chair: Syamal 2 - PETRO & Supplies, Executive Singal, CSIR-Indian Institute and Economics - Sourabh Mukherjee, utiv ive Director-In Hindustan Petroleum Directo CHEMICALS Use in Vehicles - SK tor-Intern ternationa rnationall Trade SESSIO Diesel from Waste Plastics: Corporatio orppora Energy Consumpti Plants - Vera Varaprasad, N rationn Limited Optimization of Olefin Session Chair: 10 on Update in of Petroleum Center (CRP) Limited Chakrapany Choudhari, Aker Powergas ay Refinery Group - Daniel Reyes, Amuay Indian Oil Corporation Manoharan – New Horizons - Atul of Paraguaná the Phenyl-methyl PDVSA , Director-Re Liquid Fuel from Coal Profit Improveme Through Optimizing Refining finery, Essar Maximizing pX Production EDC Pyrolysis LLC Oil, Ltd. PvtLtd, Mumbai National Petroleumnt Program in KNPC GTC Technologies US Furnace Radiant of Bio-Jet Fuel Process: Refineries - Joseph C. Gentry, - Mr. K. Ramesh, Company Produced as Byproduct Section Abbas Shamash, Hydroprocessed Diesel Reliance Industries Tube Failure-Cas CSIR-Indian Institute Kuwait Kuwai e Study Metallocene Engines - M.O. Garg, Temperature A Superior Fuel in IC and Second Generation A Study of Dependent Mechanical Polymer Process Developments Catalysts: Optimizing with Advanced Inc. Failure of Petroleum Chamber Paul, P.E., SK E&C USA, Processing Opportunity S. Saha, ENGG of Spent Acid Regenerati Performance Catalysis - Dr. Howard Daily ThermetricTemperature Measureme Carbon Footprint while DIV. RPTL (JEC). and ROI on Combustion nt Systems s Corporatio Minimizing Impact on Refinery Operation: India Limited KT Reliance elia Technip a Walter n ncee Refin Refinery, Tijmes, nery, Jam The Same Jamnagar, Crudes - Tanmay Taraphdar, mnagar, India Controlling Corrosion & Scheduling in Process – Aurelio Ferrucci,Tool Optimizess Both Packages 4:30-5 p.m.

Mid Term Planning Amey Majgaonka Refrigeration Systems PROMETHEUS LUNCH AFTER and Gas Compressio r, Kirloskar S.r.l.

S.r NOON Best Practices 1-2 p.m.

Pneumatic BREAK n - EXHIBIT HALL for ICALS Co. Ltd. - K. Sudhakar, Mitigation of Corrosion COFFEE & DESSERTS TRACK TRACK 2 - PETROCHEM in Hydrocarbo L&T Chiyoda 1 - REFINI 2-2:30 p.m.

5-6:30 p.m.

Limited n Processing NG Industry Walter Tosto S.p.A SESSIO SESSION 4 TRACK 1 - REFINING N 11 Fossataro, General Manager, Session Session Chair: Giacomo Chair:andStephany Refineries T TRACK SESSION 3 2 - PETRO Reduction in Integrating andRomanow, Editor, Managing Director, MRPL 2:30-4:30 p.m.

Challenges and Emission Bhullar, Fluor Corporation; Anil Rajguru CHEMICAL Hydrocarbo Session Chair: P.P. Upadhya, bon Romel onn Pro A Focus on Revamping SESSIO S Processing rocess ssing Analysis of FCC Petrochemical Plants, N Distillate Hydroprocessing: Reactor Cyclone Session Chair: 12 Ses Axens Solutions for Middle - Stefania Archambeau, Axens Reliance Industries, Les Antaffy, Fluor Enterprises Flow - Narasimham Panipat A.K. Purwaha, Engineers Eng Integration - Sanjiv Singh, Ltd. Chairman and and New Catalyst Technology urthy Bontu, India, Ltd. Managing Director, Sour Water Limited Economics-Refinery/Petrochemical Stripper Units Complex, Indian Oil Corporation Virtual Reality Virtu Complex Production Fluor Daniel Integrity andwith High Cyanides Refinery and Petrochemical as Effective Integrated Refinery an Plant of India HMEL Verifying Decisions: Decis Tool Pvt. Contents - Vikas Badithela, Implementation for Training Experiment An Efficient Tool for Coke India Limited Ltd. and Challenges - Srinivas Kapoor, Results - Alessandro Field Operators and Dynamic Simulation: Drum Reconciliation: Benefits E-Learning E-Lear - Sheo Raj Singh, Engineers Unheading - Elango ROIC Altamura, Technip Making and Universal Control System Design Narendran, Gas Processing Return of Invested Capital Simulation Italy S.p.A. DeltaValve Mr. Santosh 6:30-8 p.m.

San Joshi Effect of Reliability on Sealing on Black Powder-Generated for Competenc , GSE EnVision Valero Optimum Isolation Valve CLOSIN y Developme and - Logan Anjaneyulu, Regaining Regainin Operating G RECEP Aramco nt Pressure Hydro Treaters - Omar Al Amri, Saudi TION and Engineering of High Ltd. - Michael Michae A. Taube, Excellence Through Challenges in Design Sudhaker, L&T Chiyoda Enhanced Training S&D Consulting Optimization - Mr. K. Avenues for Energy LLC ICALS BREAK AFTERNOON TRACK 2 - PETROCHEM 4:30-5 p.m.

9:15-9:30 a.m.

9:30-10:45 a.m.

10:45-11 a.m.

11 a.m.-1 p.m.

Media Sponsors:

T H E AU T H O R I T Y O N E N E R G Y

HPIRPC.com

IRPC 2013 Agenda

5-6:30 p.m.

day, 10 Day 1 | Wednes

TRACK 1 - REFINING Ltd. Petroleum Corporation SESSION 5 Managing Director, Chennai Session Chair: A.S. Basu, Tripathi, Production - Adarsh Eco-friendly Gasoline New Solutions from RRT Global Reactors Drop Limitations in HydroprocessingLtd. Overcoming Pressure Corporation Hindustan Petroleum Mahendranadh Desu, Optimum Refinery for Saxena, KBR

6:30-8 p.m.

July

Changing Feed and

CLOSING RECEPTION

Product Demands -

- Exhibit Hall

Samir

Ivanhoe Energy SESSION 6 Cabrera, Executive Chairman, Plant Session Chair: Carlos by the Use of a Digital Efficiency and Safety Maximizing Operational Sloane Whiteley, AVEVA of Energy Management - Mayur Development and Deployment Case Study: Design, Petrochemicals Complex 2011) at an Integrated System (ISO-50001 : Ltd. Talati, Reliance Industries Tool to Improve the Reliability Dynamics as an Emerging Computational Fluid Technip KT India Limited - S. Sathish Kumaran, of the Plant Operations

V View the full list of speakers and the complete conference agenda online at HPIRPC.com co

EARLY BIRD SPECIAL Register by 31 May to Save!

Register at HPIRPC.com by 31 May to SAVE UP TO 20% Take part in this innovative, industry-leading forum. Register today and benefit from: • More than 40 technical sessions given by speakers representing companies like Indian Oil Corporation Limited, Valero, Saudi Aramco, Hindustan Petroleum Company, Reliance Refinery, PDVSA, Technip and more • Join international HPI professionals from around the world, representing operators, refineries and petrochemical plants like BP plc, Chevron Lummus Global LLP, ConocoPhillips Ltd, eni, ExxonMobil Research & Engineering Company, Linde Gas, Lukoil and Total • Ample networking opportunities between sessions allow you to connect with old and new business contacts • Explore and learn more about the latest developments in operations and technology • Get a local and global perspective on the refining and petrochemicals industries

Conference registration includes: • Pre-conference tour of Indian Oil’s Panipat Refinery & Petrochemicals Complex (9 July 2013)* • Two-day conference program (10-11 July 2013) including keynote addresses, general presentations and panel discussions • Breakfasts, lunches and refreshment breaks • Access to exhibition floor throughout conference activity *Refinery tour registration is offered on a first-come, first-served basis. Space is limited.

2013 Conference Fees: Early Bird Fee (by 31 May)

Regular Fee

Individual

$945

$1,045

Team of Two

$1,700

$1,875

Group of Five

$4,250

$4,700

For more information about IRPC 2013, please contact Melissa Smith, Events Director, Gulf Publishing Company, at +1 (713) 520-4475 or Melissa.Smith@GulfPub.com.

Supported by:

* IRPC 2013 is currently the only refining and petrochemical event that will be held in India during 2013. Attendees will have the opportunity to take part in ane exclusive tour of Indian Oil’s Panipat Refinery & Petrochemicals Complex, located 23 kilometers from Panipat and approximately 100 kilometers from New Delhi.

Since its original construction, the Panipat refinery has undergone numerous upgrades and unit additions, to arrive at its current capacity of 15 MMtpy.

| Brief Centrica signs long-term North American LNG export contract with Cheniere Centrica has an agreement with Cheniere Energy Partners to purchase 89 billion cubic feet of annual liquefied natural gas (LNG) volumes for export from Cheniere’s Sabine Pass liquefaction plant in Louisiana. This amounts to approximately 1.75 MM metric tpy, and is the equivalent of the annual gas demand of around 1.8 MM UK homes. The contract is for an initial 20-year period, with the option for a 10-year extension, and the target date for first commercial delivery is September 2018. Under the terms of the agreement, Centrica will purchase LNG on a free on board (FOB) basis, giving it destination rights for the cargoes, for a purchase price indexed to the Henry Hub natural gas price plus a fixed component. Centrica will export gas from the fifth LNG train at Sabine Pass, on which preliminary engineering work has already begun.

Upon receiving news of the agreement, UK Prime Minister David Cameron said, “I warmly welcome this commercial agreement between Centrica and Cheniere.

Future gas supplies from the US will help diversify our energy mix and provide British consumers with a new long-term, secure and affordable source of fuel.”

The natural gas Killingholme Power Station, located in North Lincolnshire, UK, will start using LNG from the US Gulf Coast sometime in 2019.

BILLY THINNES, TECHNICAL EDITOR / Billy.Thinnes@HydrocarbonProcessing.com

Brief

The Center for Chemical Process Safety (CCPS) of the American Institute of Chemical Engineers (AIChE)

has signed an agreement with India’s Oil Industry Safety Directorate (OISD). The agreement calls for the groups to cooperate in advancing the practice of process safety in the petroleum sector in India and around the world. The signing ceremony took place in March during a CCPS workshop on recognizing hazardous incident warning signs in Goa, India.

The agreement broadens the sharing of lessons learned from process safety incidents, and includes information on the latest developments in the practice of process safety.

At the request of the National Institute of Standards and Technology (NIST), representatives from the

Automation Federation participated in the first NIST meeting for developing a US cyber security program, as requested by US President Barack Obama. The meeting was held at the US Department of Commerce in Washington, DC. It is an important step in establishing the cyber security framework within President Obama’s executive order announced in his State of the Union address to confront the growing threat of cyberattacks on the US’ critical infrastructure. The framework will include “standards, methodologies, procedures and processes that align policy, business and technological approaches to address cyber risks,” and “help owners and operators of critical infrastructure identify, assess and manage cyber risk.” The next NIST cyber security framework meeting is scheduled for May 29 at Carnegie Mellon University in Pittsburgh, Pennsylvania.

US-based oil and gas industry professionals are the sixth best rewarded in the world, according to a

global salary guide produced annually by Hays Oil and Gas and Oil and Gas Job Search. The guide reveals that industry professionals working in the US earn an average of $121,400 per year. This is slightly behind Canada at $123,000 per year.

Australia is now the oil and gas pay leader, with an average industry salary of $163,600 per year. Expat industry workers are enjoying the highest levels of pay in Australia, the Philippines and Trinidad. Employer confidence in increasing staffing levels during 2013 remains high, with 75% of respondents predicting continued hiring increases. The guide bases its figures on more than 25,000 industry professionals across the globe.

ExxonMobil Pipeline Co. reported a leak in its Pegasus pipeline on March 29, resulting in the release of 3,500

to 5,000 barrels of crude outside of Mayflower, Arkansas. The pipeline—which runs from Patoka, Illinois to Nederland, Texas—was carrying Canadian Wabasca heavy crude at the time of the leak. The company says progress continues in the cleanup areas. Approximately 640 people are responding to

the incident in addition to federal, state and local responders.

Cleanup has continued 24 hours a day since the spill was first detected. ExxonMobil also indicated that nearby Lake Conway remains oil free and stated that a comprehensive containment system using a boom has been deployed as a precaution.

Regarding the pipeline, an excavation and removal plan for the affected portion of the pipeline is being developed for review by the US Department of Transportation. ExxonMobil has also received a corrective action order from the US Pipeline and Hazardous Materials Safety Administration.

PetroChina recently announced 2012 earnings of RMB 115.3 billion, which was a drop of over RMB 17 billion

compared with 2011 earnings. The company’s earnings decline could be attributed to high natural gas import prices, as well as government caps on domestic fuel prices that eroded refining margins. PetroChina saw strong upstream growth in 2012, but the downstream side of its operations operated at an RMB 43.5 billion loss. Still, this was a slight improvement over 2011 downstream losses, which were RMB 61.9 billion. The continued downstream sector losses were due to PetroChina’s inability to shift the burden of rising oil costs to its consumers, as Chinese state policy mandates a cap on gasoline and diesel prices. In 2012, PetroChina’s refinery division produced 91 MM tons of gasoline, diesel and kerosine, an increase from the 87.2 MM tons in 2011. The company also produced 6 million tons of synthetic resin in 2012 (a rise of 7% year over year) and manufactured 3.7 MM tons of ethylene (up 6.4% from 2011).

The Organization of the Petroleum Exporting Countries (OPEC) sent a letter of condolence to acting Venezuela

President Nicolás Maduro Moros following the death of the country’s President, Hugo Chávez. The president of the OPEC Conference, Hani Abdulaziz Hussain, who is also Kuwait’s Minister of Oil, referred to President Chávez’s “long and brave battle with cancer” and, on behalf of the conference, conveyed his “sincerest condolences” to President Maduro and, through him, to the people of Venezuela. He added: “There can be no doubt that President Chávez will be long remembered for his strong support to OPEC.” Shell plans to sell its refinery in Geelong, Australia.

The proposed sale of the 120,000-bpd refinery is in

line with Shell’s global strategy to concentrate investment on large-scale sites, such as the company’s Pulau Bukom refinery in Singapore. The announcement further underpins Shell’s Australia strategy to grow its retail and bulk fuels business, along with terminals and pipelines. The company aims to conclude the sales process by the end of 2014. “This announcement will be difficult for refinery employees, but Shell will support them,” a Shell executive said.

Hydrocarbon Processing | MAY 2013 9

This injector sprays liquid into gas and is just one of many used for gas cooling, water wash, desuperheating, steam quench, slurry backflush and more.

SUPERIOR SPRAY. SERIOUS RESULTS. Whether you need to cool gas, dissolve salts in an overhead line or inject chemicals to prevent corrosion, we can help optimize injector performance. Here's how: • Assistance with nozzle selection, spray direction and injector placement. There are dozens of factors to consider before choosing a spray nozzle, determining whether to spray co- or counter-current and identifying the proper placement of an injector in a vessel. We can help you evaluate your process conditions and then design an injector to provide optimal performance • Design validation using Computational Fluid Dynamics (CFD) and Fluid Structure Interaction (FSI).

We use powerful modeling tools to simulate your environment, confirm the injector will provide the expected spray performance and withstand process conditions such as thermal stresses, heat transfer, vortex shedding and more • Proven track record. Companies like Technip, Mustang Engineering, Bechtel, Shell and many others rely on us to manufacture B31.1 and B31.3 code-compliant injectors and conduct radiographic, hydrostatic, ferrite tests and more

Learn More. Call 1.800.95.SPRAY or visit spray.com/injectors

CFD MODEL ILLUSTRATES PERFORMANCE BASED ON INJECTOR PLACEMENT

WIDE RANGE OF HYDRAULIC & GAS ATOMIZING NOZZLES INCLUDING UDING CLOG-RESISTANT STANT STYLE STYLES

DOZENS ZENS OF INJECTOR DESIGNS SIGNS AVAILABLE

Unmatched Global Engineering, Manufacturing & Technical Support Nozzles | Control Systems | Headers & Injectors | Research & Testing 1.800.95.SPRAY Select 66 at www.HydrocarbonProcessing.com/RS

BILLY THINNES, TECHNICAL EDITOR / Billy.Thinnes@HydrocarbonProcessing.com

Impact

Global oil flows shift from West to East By 2020, crude oil trade flows west of Suez will drop by 4.2 MM bpd, while crude flows east of Suez will rise by 4.7 MM bpd (FIG. 1), according to a recently published report from ESAI Energy. The primary global implication of the growth in US shale liquids and Canadian oil sands is significant change in crude oil trade flows. These changes will fundamentally alter the relationship between OPEC countries and the consuming countries of North America, Europe and Asia-Pacific. Relative to the US and Europe, the countries of Asia-Pacific will become far more dependent on OPEC countries for oil supplies. This will force the countries of Asia, especially China, to take a more deliberate role in responding to Middle Eastern conflicts and instability in producing countries, which, in turn, will alter the relationship of these countries to the US, the current guarantor of the sea lanes.

The report outlines some of the potentially most significant changes to crude flows, including the reduction of North American imports from Latin America, Africa and the Middle East by 3 MM bpd between 2012 and 2020, and a drop in European imports from the same regions of another 2 MM bpd. As this happens, crude oil producers will increasingly focus on Asia-Pacific customers. FSU, African and Latin American exports to the region will rise by 1.7 MM bpd, theoretically leaving the remaining market for Middle Eastern supplies of 3 MM bpd or an annual increment of as much as 375,000 bpd.

“There are many implications of these initial trends in global oil trade,” said Sarah Emerson of ESAI Energy. “US and European interest in committing resources and personnel to the security of the Persian Gulf may face renewed political resistance. Competition for the AsiaPacific market is bound to weaken crude prices, and China’s disproportionate de-

pendence on imported oil will hasten efforts to improve energy security, including the inevitable development of shale.”

UK consumers buying less fuel at the pump In the last 10 years, average UK fuel prices have almost doubled, rising from 73.68 p/l to 136.26 p/l for unleaded gasoline, and 75.57 p/l to 142.39 p/l for diesel. The latest retail marketing survey, conducted by the Energy Institute (EI), shows there are more registered UK vehicles on the road than ever before, yet total fuel sales have dropped by 6% since 2002.

A cutback in fuel sales suggests improvements in engine performance and fuel economy, combined with changes in driver behavior. This is supported by diesel sales outperforming gasoline for the

second year running. The number of gas stations in the UK stood at 8,693 at the close of 2012. This is compared to 1967’s all-time high of 39,958. Key findings of the survey show: • Gasoline sales fell marginally to 13.42 MM tons by the close of 2012, down from 13.86 MM at the end of 2011 • Diesel sales totaled 13.86 MM tons by year end, down slightly from 13.91 MM tons in 2011 • Total 2012 road fuel sales fell slightly to 35.35 MM tons • By the close of 2012, unleaded gasoline prices had averaged 136.26 p/l (vs. 133.6 p/l in 2011), while diesel prices closed the year at an average price of 142.39 p/l (vs. 138.90 p/l in 2011) • Registered UK vehicles once again broke records, rising from 34.67 MM in 2011 to reach 36.71 MM by the end of

0.1 to 0.1 4.2 to 5.0

0.1 to 0.1

Europe East Europe

0.5 to 0.4

North America

1.0 to 0.1 0.5 to 0.5

2.7 to 1.8

Latin America

FSU 0.3 to 0.4

2.7 to 1.4

Africa

1.1 to 1.7

1.9 to 1.1

Middle East

12.2 to 15.2

Asia-Pacific

2.2 to 2.8

2.2 to 1.0 1.1 to 1.6

2012 flows west of Suez: 16.1 MMbpd 2020 flows west of Suez: 11.9 MMbpd

2012 flows east of Suez: 16.6 MMbpd 2020 flows east of Suez: 21.3 MMbpd Source: ESAI Energy, March 2013

FIG. 1. Approximate changes in net crude, 2012 to 2020 (MMbpd).

TABLE 1. US corrosion inhibitor demand (MM dollars) % Annual growth Item

2007

2012

2017

2007–2012

2012–2017

Corrosion inhibitor demand

2,070

2,030

2,485

–0.4

4.1

Petroleum refining

500

540

640

1.6

3.5

Utilities

375

343

390

–1.8

2.6

Chemicals

361

310

385

–3

4.4

Oil and gas production

182

338

475

13.2

7

Other

652

499

595

–5.2

3.6

Hydrocarbon Processing | MAY 2013 11

Impact 2012, with each gas station servicing an average of 3,993 vehicles.

The survey also took care to break down the differences in gas station sites by sector for 2012: • Oil company sites decreased by 151 to 5,159 • Main retailer sites increased by 423 to 1,233 • Supermarket sites increased by seven to 1,317

• Smaller retailer sites increased by two to 62 • Other unbranded sites increased by 37 to 922.

The four largest oil company operations by number of branded gas stations were (2011 figures in brackets): • BP, 1,220 (1,178) up 42 • Shell, 1,028 (845) up 183 • Esso, 907 (890) up 17 • Texaco, 787 (840) down 53.

Complete Level Measurement for Harsh Environments

VEGA offers a complete line of level measurement instruments using radiation-based technology for accuracy in the most extreme environments.

VEGA detectors supply the following benefits: ϶ Non-contact measurement is unaffected by corrosive process properties and internal obstructions ϶ Patented flexible detector follows vessel contours and avoids obstructions ϶ External mounting reduces both short and long term maintenance costs

www.vega-americas.com americas@vega.com 1-800-FOR-LEVEL

12

Select 151 at www.HydrocarbonProcessing.com/RS

US demand for corrosion inhibitors to reach $2.5 billion US demand for corrosion inhibitors is forecast to rise 4.1% per year to $2.5 billion in 2017, with volume demand approaching 1.7 billion pounds (TABLE 1). Growth will be driven by higher oil and natural gas output, particularly from shale formations, as well as by increasing chemical production and an expanding economy. Additionally, robust increases in construction spending will support demand for corrosion inhibitors used in cement and concrete, industrial coatings and metal applications. Value growth will also be aided by the introduction of new hybrid products that have functions in addition to corrosion protection. These and other trends are presented in a new study from The Freedonia Group.

The oil and gas industry’s continued expansion of horizontal drilling and hydrofracturing well stimulation in shale formations will drive increases in corrosion inhibitor demand going forward, especially organic inhibitors. Increasingly caustic water produced by existing oil wells will support higher organic inhibitor usage rates, as will efforts to reuse and recycle water to avoid additional freshwater use. The availability of relatively cheap natural gas will spur faster growth in chemical production, leading to advances in corrosion inhibitor demand in both water treatment and process additive applications.

The fastest growth in corrosion inhibitor demand, albeit from a small base, will occur in concrete and cement additives due to a rebound in construction spending. Nitrites will benefit due to their popularity for protecting metal rebar in reinforced concrete. Higher construction spending will also support demand for corrosion inhibitors in industrial coatings, particularly as improvements in state and local finances allow for greater spending on infrastructure maintenance and modernization.

In a number of more mature markets such as petroleum refining, metals and utilities, moderate growth will be supported by an expanding economy.

Water treatment corrosion inhibitors account for the greatest share of demand in these markets, though in most cases process and product corrosion inhibi-

Hello Tomorrow

Dubai inspires new connections -S` ,TPYH[LZ [V V]LY KLZ[PUH[PVUZ HYV\UK [OL ^VYSK 4HRL HU \UMVYNL[[HISL Z[VW PU +\IHP HSVUN [OL ^H` HUK KPZJV]LY H WSHJL ^OLYL PUZWPYH[PVU [Y\S` [HRLZ ÅPNO[

emirates.com/usa Select 68 at www.HydrocarbonProcessing.com/RS

ÓÓÊ`iÃÌ >Ì ÃÊ>VÀ ÃÃÊ vÀ V>ÊUÊΣÊ`iÃÌ >Ì ÃÊ>VÀ ÃÃÊ Ã >ÊUÊ£äÊ`iÃÌ >Ì ÃÊ>VÀ ÃÃÊ ` >ÊUÊ£nÊ`iÃÌ >Ì ÃÊ>VÀ ÃÃÊÌ iÊ `` iÊ >ÃÌ

StarLNGTM The leading small-to-mid-scale standard LNG plant Linde Process Plants, Inc., a world leader in cryogenic technologies, has now translated the idea of product standardization into the LNG industry. These modularized plants cover about 90% of real life boundary conditions, and provide improved, simple and robust technology with high HQHUJ\ HIÄžFLHQF\ StarLNGâ&#x201E;˘ delivers a pre-engineered process design, standard documentation and modularized plant layout for shortest delivery time with minimum on-site construction.

Consider Linde Process Plants, Inc. for your next project.

Select 85 at www.HydrocarbonProcessing.com/RS

A member of The Linde Group Linde Process Plants, Inc. 6100 South Yale Avenue, Suite 1200, Tulsa, Oklahoma 74136, USA Phone: +1.918.477.1200, Fax: +1.918.477.1100, www.LPPUSA.com, e-mail: sales@LPPUSA.com

Impact tor demand will rise at a faster pace. Organic inhibitors will be the primary beneficiaries as companies look to develop new products that offer more protection at lower treat rates, and that are more cost-effective than existing alternatives.

Replacing molybdates, where possible, will remain a top priority as molybdate prices remain comparatively high, and some concerns about their environmental impact have arisen.

consensus among US citizens about how to optimize domestic energy production while preserving the environment.

US citizens overall and across political and socioeconomic groups generally are most likely to call for more emphasis on solar and wind power, but these potential future sources of energy have a long way to go in terms of technology and affordability before they can significantly affect overall US domestic energy production. Ameri-

cans are also sharply divided politically over achieving greater domestic energy production using more traditional energy sources such as oil, coal and nuclear power.

This leaves natural gas, which 59% of Democrats, 62% of independents and 79% of Republicans say should have more emphasis in the US. The technology exists and is being implemented to allow natural gas to become a more significant contributor to US domestic energy production.

Poll says US citizens want more solar, wind and natural gas No fewer than two in three US citizens want the country to put more emphasis on producing domestic energy using solar power (76%), wind (71%) and natural gas (65%). Far fewer want to emphasize the production of oil (46%) and the use of nuclear power (37%).

Least favored is coal, with about one in three respondents wanting to prioritize its domestic production. These are the results of an opinion poll administered

by Gallup, based on telephone interviews conducted March 7â&#x20AC;&#x201C;10, with a random sample of 1,022 adults, aged 18 and older, living in all 50 US states and the District of Columbia.

Democratsâ&#x20AC;&#x2122; and independentsâ&#x20AC;&#x2122; top choice is solar power, while natural gas places first among Republicans. Republicans and Democrats disagree most on the priority that should be given to oil as a future energy source, with 71% of Republicans wanting more emphasis placed on it, compared with 29% among Democrats.

Republicans are also much more supportive than Democrats of coal (51% vs.

21%) and nuclear power (49% vs. 30%).

Where people live in the US makes a difference in their views about which sources of domestic energy they want the country to emphasize more. People in the South tend to be more supportive of traditional energy sources such as oil and coal than are those in other regions.

Still, for respondents in every region, including the South, solar power is the top choice, or is tied for the top spot, among the energy sources tested. The US has a great opportunity to accelerate its economic growth over the next several years by emphasizing and using its enormous energy resources to produce domestic energy. But there has been no Select 152 at www.HydrocarbonProcessing.com/RS

15

All paths lead to

Lewis Pumps ®

Lewis® pumps are the world standard for pumps and valves in the sulphur chemicals industry. Offering a family of steam-jacketed sulphur pumps, outstanding reliability in hightemperature sulphuric acid, and new designs for molten salt energy transfer, Lewis continues its long tradition of superior products and services.

Standard replacement parts are always available.

Emergency service or parts to any major airport worldwide within 72 hours. No matter what your application, when you need a superior product with exceptional service... all paths lead to Lewis.

Customers in over 100 countries can’t be wrong.

LEWIS® PUMPS Vertical Chemical Pumps 8625 Grant Rd. St. Louis, MO 63123 T: +1 314 843-4437 F: +1 314 843-7964 Email: sales@lewispumps.com www.weirminerals.com

Excellent Minerals Solutions

Expertise where it counts.SM Select 94 at www.HydrocarbonProcessing.com/RS

ADRIENNE BLUME, MANAGING EDITOR Adrienne.Blume@HydrocarbonProcessing.com

Associations

EMGC 2013 explores Eastern Mediterranean resource potential Gulf Publishing Company held its inaugural Eastern Mediterranean Gas Conference (EMGC) at the Hilton Cyprus in Nicosia, Cyprus, from April 8–10. The event surpassed expectations in terms of both attendance and interest, and gave executives from operating, service and technology companies the chance to share insight into development opportunities for this exciting resource area.

The conference featured speakers from top companies operating in the region, including Noble Energy Inc., Hyperion Systems Engineering Group, Cyprus National Hydrocarbons Co., Total E&P Research and Technology USA LLC and others.

Monday workshop. The conference opened with the “Doing Business in the Mediterranean” workshop on Monday, and continued with two days of technical presentations, networking events and a gala dinner on Tuesday. The workshop, hosted by EMGC media sponsor Deloitte, drew over two dozen attendees seeking to share knowledge and gain information about economic conditions and tax and customs laws in Cyprus and elsewhere in the region.

Speaker Paul Mallis, Cyprus Oil and Gas Tax Leader for Deloitte, noted that Cyprus’ low corporate tax rate of 10% (which will move to 12.5% after the recent, €10 billion agreement with the Troika), is an attractive reason for doing business in Cyprus.

Day 1: Offshore, LNG and national opportunities. EMGC continued on

Tuesday with the first day of technical presentations. Executive speakers focused on the regional implications and potential of the Eastern Med’s new energy resources. Delegates from the Israeli and Cypriot governments, along with industry executives, examined offshore oil and gas op-

portunities and challenges for Cyprus, Israel and the EU as a whole.

The LNG question. Several speakers expressed the need for up to three liquefied natural gas (LNG) export trains, with a total capacity of 15 million tons per year, to deliver Eastern Med gas to other European nations. Cyprus’ Deputy Director of Energy Services for the Ministry of Industry, Commerce and Tourism, Constantinos Xichilos, noted that Noble Energy Inc. and partner Woodside Petroleum Ltd. are studying a possible LNG terminal site at Vasilikos in Cyprus.

Noble Energy’s take. Noble Energy Chairman and CEO Charles Davidson spoke at length on Tuesday morning about the discoveries his company has made in the region, and the implications these resources have for the Eastern Med’s energy future.

“The region is evolving into a major energy player in the world, not only for countries in the region, but also for the countries they will be working with,” said Mr. Davidson. The massive gas resources available in the Eastern Med will provide enough clean, low-cost energy to displace all of the fuel used by automobiles in Israel for 14 years, Mr. Davidson noted.

“We have to work cooperatively with all of our partners, especially [the regional] governments,” to make these projects work, Mr. Davidson said.

The Israeli perspective. AJM Deloitte Partner for Energy and Resource Advisory Services, Robin Mann, noted that Israel could be self-sufficient in gas by 2016. Gas from the Tamar and Leviathan fields will supply power for the country’s electricity needs, and excess gas from these fields could be sent as LNG to Asia or Europe. Alternatively, the gas could be piped through an undersea pipeline through Cyprus or Turkey and then onto Europe. Another option would be to set up a floating LNG (FLNG) facility to process and transport the gas.

Emerging markets. Tuesday afternoon featured presentations on the market for the new resources, with executives

from DNV, KBR Inc., and GE Oil and Gas Turbomachinery offering perspectives on natural gas mega-projects, including proposed LNG and FLNG projects.

Concluding the afternoon session was a panel discussion on the impact of the new energy resources, featuring panelists Rony Halman, founder and Chairman of Israel Opportunity Oil & Gas Co.; Wafik Beydoun, President and CEO of Total E&P Research and Technology USA LLC; and Philip Hagyard, Senior Vice President of Gas Monetization at Technip.

Tuesday evening gala. Day 1 concluded with a heavily-attended gala dinner at the Hilton Cyprus sponsored by Deloitte Ltd. The dinner featured a keynote speech by Noble Energy’s Charles Davidson (FIG. 1), who said that the Eastern Med’s gas resources would soon be moving to the export stage, which would inevitably involve energy exchange with Middle Eastern nations.

Mr. Davidson emphasized that the work being done in the Eastern Med by Noble Energy and its partners is a joint effort. “These aren’t only multi-billiondollar projects; they’re multi-decade projects,” Mr. Davidson said. For such projects to be successful, stable tax environment and investment climates are needed, along with a streamlined infrastructure process.

FIG. 1. Noble Energy Chairman and CEO Charles Davidson addressed attendees at Tuesday evening’s gala dinner.

Hydrocarbon Processing | MAY 2013 17

Now Accepting

Challenging Projects Sure we can do the standard tower open/clean/inspect/close work but it’s those tough and challenging jobs that have helped us earn our stripes.

We recently completed a revamp on one of the largest vacuum towers in the western hemisphere, much to our customer’s satisfaction. We’ve mastered a resection method that is an excellent and cost saving alternative when footprints are tight.

Not long ago we tackled a 114-vessel project at a natural gas processing plant. Our team blinded, opened, cleaned, inspected and repaired all of the vessels in just 14,642 man-hours; less time than had been scheduled and well under budget.

When performing any number of services, we don’t overlook anything from external pipe flanges, complicated vessel internals, feed/draw arrangements, section replacements, nozzle and strip lining installation/repair. We strive for “Zero Injury.” As for quality, well, that’s why our customers invite us back again and again and give us their annual maintenance/service contracts.

We thrive on accepting challenges then exceeding expectations but we are just as agile with a single tower project as we are with plant-wide turnarounds. We are quick on our feet and can mobilize swiftly for emergencies.

Challenge us today and we’ll have your towers productive tomorrow.

TIGER TOWER SERVICES 3012 Farrell Road, Houston, TX 77073 Phone: 281.951.2500 • Fax: 281.951.2520 www.tigertowerserivces.com Select 75 at www.HydrocarbonProcessing.com/RS

Associations

BORSIG

Mr. Davidson concluded his speech by saying, “Our goal is to build on what we’ve done already, and we hope we can look back a few years from now and say that we created a forum for discussion [with this conference] for what is happening in this region.”

Day 2: Infrastructure, resource development and beyond. The second day of

EMGC’s technical program addressed regional infrastructure developments in the morning, while the afternoon focused on resource development and the future of the Eastern Mediterranean.

FLNG outlook. Victor Alessandrini, FLNG Business Development Manager for Technip’s Offshore Business Unit, spoke about the market and future developments for FLNG (FIG. 2). The technology is a cost-optimized, mobile and environmentally non-invasive way of processing gas into LNG and transporting it to various locations, Mr. Alessandrini said.

Some technical challenges of FLNG include incorporating a full LNG plant onto a floating production, storage and offloading (FPSO) unit; gas processing facilities must be adapted to a marine environment. Space, weight and stability management are other important considerations when planning an FLNG project, Mr. Alessandrini noted.

Cypriot growth aspirations. Dr. Symeon Kassianides, Chairman and CEO of Hyperion Systems Engineering Group, next examined Cyprus’ readiness to address the challenges and opportunities before it. Dr. Kassianides noted that both the public and private sectors are preparing the country for an upswing in work opportunities in the oil and gas sectors. He also confirmed that FEED studies are being performed for an LNG terminal in Cyprus.

According to Dr. Kassianides, longterm potential projects in Cyprus include a methanol plant, an ethylene facility, a possible gas-to-liquids (GTL) plant and other facilities. “We strongly believe in Cyprus as an upcoming energy hub for the future of the Eastern Mediterranean,” Dr. Kassianides said.

Israel’s energy goals. The Wednesday afternoon session kicked off with a presentation by Jay Epstein, Business Development Manager for Israel Natural Gas Lines, on the infrastructure needed for the development of Israel’s natural gas sector. Afterward, Chairman of Israel’s Dor Chemi-

FIG. 2. Technip’s Victor Alessandrini discussed the global outlook for FLNG developments on Wednesday morning.

cals Ltd., Gil Dankner, spoke to attendees about the use of natural gas and its derivatives as alternative transportation fuels.

The future of the Eastern Med. During EMGC’s eighth and final session, the CEO of Cyprus National Hydrocarbons Co., Dr. Charles Ellinas, noted that the establishment of an LNG terminal is vital, as it will enable Cyprus to access markets in Europe, the Far East and other global regions.

Concluding the session was Noble Energy’s Director of Operations for the Eastern Mediterranean, Terry Gerhart. Mr. Gerhart detailed Noble Energy’s development plans for the Tamar, Leviathan and Cyprus A fields. Additionally, an LNG plant to process this gas could begin operations near the end of the decade and accommodate up to three trains, Mr. Gerhart said.

SERVICES for Pressure vessels and heat exchangers Power plants Mechanical Engineering Plant and pipe construction Ball valves and valves

Inaugural EMGC a success. During his

closing remarks on Wednesday afternoon, Gulf Publishing Company President and CEO John Royall thanked attendees for helping make the inaugural EMGC a success. Mr. Royall said that the conference’s networking, forum-building and awareness-raising objectives had exceeded original expectations.

Major companies such as Noble Energy, Eni, Total, Woodside Petroleum and others have made significant investments in the region, Mr. Royall noted. “My advice to the industry and to all of you in this room is to look where the experts are investing their money,” he said. “And they’re investing it here, in the Eastern Mediterranean.”

Gulf Publishing Company plans to hold its second annual Eastern Mediterranean Gas Conference in Tel Aviv, Israel, in 2014.

Compressors Membrane Technology BORSIG - Always your first choice

www.borsig.de BORSIG GmbH Phone: ++49 (30) 4301-01 Fax: ++49 (30) 4301-2236 E-mail: info@borsig.de Egellsstrasse 21, D-13507 Berlin/Germany

Select 153 at www.HydrocarbonProcessing.com/RS

l

l

Select 83 at www.HydrocarbonProcessing.com/RS

l

Reliability

HEINZ P. BLOCH, RELIABILITY/EQUIPMENT EDITOR Heinz.Bloch@HydrocarbonProcessing.com

Manage your time constructively We often view ourselves as very efficient reliability implementers. Yet, as professional engineers, we take action to save time, accomplish more and reduce stress by applying better time-management skills. Being productive requires effort.

Likewise, resourcefulness is needed to improve individual performance.

More important, we must do a better job of balancing our personal and professional lives; rest and relaxation from the harried duties from work is a positive.

In all of our professional endeavors, we should strive to be better contributors instead of mere consumers.

Set realistic goals. Serious contribu-

tors often set daily goals. Why? Goals are important. Meeting daily goals provides a sense of accomplishment and an opportunity to add value to our companies. Value-adders set and readjust their priorities every day. Many tasks require concentration and organization. Are your tools close at hand? A hammer is a tool, but so is a micrometer, a checklist or a good technical text.

Plan your day. If phone calls are necessary, strive to make them when you have the best chance of reaching your desired person. E-mailing your contact with a brief and complete message is a more time-efficient action. Likewise, send e-mails with written summaries to the appropriate distribution list for action.

Meetings. In planning meetings, keep them brief and stay on schedule. To improve everyone’s productivity, start on time and finish on time, and stick to the agenda.

When attending a meeting, never miss out on the opportunity to say nothing. Repeating what has already been said, or what can be confirmed in a written summary, takes away valuable time from others.

Delegate when possible. Delegating tasks can accomplish more with the present staff and provide training opportunities. Also, it conveys to the team members

that they are valued. Ask for “buy-in” from those to whom you delegate a task, and firmly determine a mutually acceptable delivery schedule.

Beware of paper shuffling. Productive professionals strive to handle a piece of paper only once. They resist the temptation of moving papers among temporary piles.

For example, a freelance writer once spent eight weeks massaging a four-page article.

The freelancer’s staccato work pace adversely affected the efficiency of others.

Divide and conquer onerous tasks.

Some days appear to be an endless list of tasks that add more dread to your job and attitude. One solution is to compile a complete list of work and project items to be handled; put everything on the written list. As new tasks materialize or are assigned, add them to the list. Visually crossing off completed items or tasks definitely improves your mood and attitude as the “things to be done” are completed.

Many professionals keep the “to do” list visually close by, such as on a computer screen. It is easier to stay on schedule with such queues refocusing the individual.

Set realistic priorities. Priorities are often an issue. Let your boss assist you in setting and resetting priorities. Break down large tasks into small segments; this will remove intimidation from the total endeavor, and provide attainable and achievable goals. Be realistic in timing task durations.

Of course, you can assign priorities to the tasks and activities according to importance. It is difficult to clearly distinguish between “urgent” and “important.” There may even be some off time within your day. This is the opportunity for reliability professionals to broaden their knowledge base. They can work on tool making, repair tasks or specification updates. It is an opportunity to develop technical articles and reports that will add value to others, or to scan some technical articles for later use.

Never procrastinate. Remember to manage your activities via their priorities on your task list. Stay in control and keep bosses informed of schedule changes. Task lists should be flexible. The objective is to maintain control so that the tasks accomplished each day are done by choice, and not by chance.

Maintain focus. Seasoned pros do not rush from job to job or worry about doing everything that they have listed. Timemanagement consultant Alan Lakein stresses that one rarely reaches the bottom of a “to do” list. He remindes his students that it’s not completing the list that counts, but making the best use of one’s time. We should strive to accomplish the bulk of tasks that are truly important and warrant our skills. When possible, delegate the unfinished items to others or transfer the project to tomorrow’s list. Ask if finishing the job produces significant benefits. If not, it may not be a high-priority task.

What is ‘urgent’ or ‘important?’ Some tasks yield better results than others. When looking over a list of duties, consider the results that each one will bring. True, at first glance, everything on the list seems urgent.

Still, we should ask if urgent matters are always important, deserving a major time investment. Michael LeBoeuf, a professor of time management at the University of New Orleans, makes this observation: “Important things are seldom urgent, and urgent things are seldom important. The urgency of fixing a flat tire when you are late for an appointment is much greater than remembering to pay your auto insurance premium, but its importance [the tire] is, in most cases, relatively small.” Then he laments: “Unfortunately, many of us spend our lives fighting fires under the tyranny of the urgent. The result is that we ignore the less urgent but more important things in life. It’s a great effectiveness killer.”

It is more rewarding to work at something that yields important results than it is simply to be busy at whatever activity is Hydrocarbon Processing | MAY 2013 21

Reliability at hand. Try to direct your efforts to activities that result in true accomplishments.

80/20 rule. A number of time-manage-

ment experts believe that we can narrow the top-priority items down to about 20%.

These experts cite, as a guide, the 80/20 rule. This principle was formulated by the 19th-century Italian economist Vilfredo Pareto; it states that only about 20% of the causes produce about 80% of the results.

But how can the 80/20 rule be applied to your time management? Analyze the items on your “to do” list. Perhaps, you can be 80% more effective by accomplishing 2 out of the 10 items listed. If so, those two items are important on your list. Also, analyze a project before diving in. How much of it is truly important to your objective? What part of the job will produce the most significant results? This portion of the task is a priority.

Time-management consultant Dru Scott, after discussing Pareto’s principle, explains how to make it work for you. She says: “Identify the vital ingredients necessary to achieve your objective. Do these things first. You will get the most results in the least amount of time.”

Example. One refinery’s statistics from the 1970s established that 7% of its 3,200 process pumps experienced a disproportionate share of pump outage events. Slightly over 60% of the money spent on pump maintenance was allocated on a chronic 7% pump population.

These pumps were quite obviously problematic machines; they received more attention than the average refinery pumps.

Applying the discussed principles, what percentage of your day’s activities would you expect to categorize as top priority? Of course, that will depend upon your specific responsibilities.

Enjoy the benefits. Now, we can appreciate that being the master of your time is not a matter of being preoccupied with never wasting a minute or rushing from crisis to crisis. Rather, effective time management means selecting the appropriate task for the present conditions. It means discerning what activities yield the best results and then spending time on those tasks.

There are no fixed rules for personal organization. Remember: Be flexible, experiment and adapt. Discover what works best for you. By gaining better control of your time, you will find more sense of accomplishment each day. Although more will remain for tomorrow, there will be satisfaction in directing your efforts to the most important things. There is enough time to complete the tasks that matter. Do not be a victim of hectic circumstances; be the master of your time.

HEINZ P. BLOCH resides in Westminster, Colorado.

His professional career began in 1962 and included long-term assignments as Exxon Chemical’s regional machinery specialist for the US. He has authored over 500 publications, among them 18 comprehensive books on practical machinery management, failure analysis, failure avoidance, compressors, steam turbines, pumps, oil-mist lubrication and practical lubrication for industry. Mr. Bloch holds BS and MS degrees in mechanical engineering. He is an ASME Life Fellow and maintains registration as a Professional Engineer in New Jersey and Texas.

22

Select 154 at www.HydrocarbonProcessing.com/RS

Scheduled maintenance, inspections, emergency responseâ&#x20AC;¦Team delivers

LEAK REPAIRS

FIELD HEAT TREATING

FIELD MACHINING

VALVE INSERTION

HOT TAPS / LINE STOPS

TECHNICAL B OLTING

VALVE REPAIR

NDE/NDT INSPECTION

T

eam is a world-class service company with the right people, technology and experience needed to keep your plants online and in production. Our highly skilled technicians work to earn your continued trust and conf idence one job at a time. Å&#x2026; 6DWLVI\LQJ FXVWRPHU QHHGV VLQFH Å&#x2026; 7UXH DYDLODELOLW\ Å&#x2026; 0RUH WKDQ ZRUOGZLGH ORFDWLRQV Å&#x2026; WUDLQHG DQG FHUWLI LHG VHUYLFH VSHFLDOLVWV Å&#x2026; :RUOGZLGH VWRFN RI LQ KRXVH VHUYLFH HTXLSPHQW Å&#x2026; %XQGOHG DSSURDFK KHOSV \RX UHDOL]H FRVW VDYLQJV Å&#x2026; ,PSURYHG SURGXFWLYLW\ PHHWV RU H[FHHGV \RXU JRDOV | www.teamindustrialservices.com

Select 95 at www.HydrocarbonProcessing.com/RS

EMISSIONS CONTROL

PIPE REPAIR SERVICES

TURNAROUND SERVICES

PIPELINE SERVICES

relentless

UOP has been serving catalyst customers for decades, and constant innovation ensures we’ll be there for decades to come.

Pilot plants. A true commitment to fast, thorough research and development. A vast team of experienced, dedicated engineers and scientists. The reasons to choose UOP catalysts are many, but perhaps the biggest of all is that UOP never quits innovating. When you use UOP catalysts, you’ll know you’re using products specifically designed to help optimize your operation. UOP invented the refining and petrochemical technology used in most operations today, and that knowledge and expertise comes through in every catalyst we develop. You’ll always get the high stability and global support your business needs, and you’ll be working with a company that will keep you on the leading edge year after year.

For more information about UOP catalysts, visit www.uop.com/catalysts. © 2013 Honeywell International, Inc. All rights reserved

Integration Strategies

RALPH RIO, CONTRIBUTING EDITOR RRio@arcweb.com

Asset information management improves life cycle benefits for equipment Traditionally, the various production management and asset management systems in the hydrocarbon processing industry (HPI) and other heavy industries resided in silos with separate management, architecture and technologies. Asset performance management (APM) involves collaboration among production and maintenance groups to improve execution with higher productivity, reduced risk and improved return on assets (ROA).

The asset and production management systems have large amounts of data and information that can be used for optimization across these systems. Result: Improved performance for both maintenance and operations. Asset information management (AIM) provides the means for optimization across the APM spectrum. AIM for APM offers new opportunities for achieving key performance indicators (KPIs) and C-suite objectives.

APM and equipment life cycles. APM applies to the long-

term, “operate and maintain” phase of an asset’s life cycle, which defines revenue, margin and profits. An APM strategy involves integrating production management (making the product) with maintenance (ensuring the capability to produce). This enables both production and maintenance management to align and meet objectives with higher productivity, reduced risk and improved ROA. APM includes sharing asset management information with collaborative production management (CPM) or manufacturing execution system (MES) applications. Such collaboration provides visibility for new opportunities to improve asset availability. It includes an enhanced understanding of risk, with fact-based risk assessment. Also, APM enables organizations to uncover opportunities to balance operational constraints and improve ROA. A range of applications do come together in the APM domain. Some have traditionally been associated with asset management, including enterprise asset management (EAM), predictive maintenance (PdM), reliability and condition monitoring. Others come from production management, including MES, CPM, quality management, laboratory-information management systems, and historians.

AIM for APM. Good information management leverages smart assets for better operations. Smarter assets include remote diagnostics, service contracts for suppliers of complex equipment and higher uptime for the users. Smarter operations aids improved production scheduling (including maintenance), yield/quality (via value-added performance) and asset longevity (reducing equipment stresses). Together, smarter assets and

67%

Improve uptime (MTBF, MTTR)

25%

57%

Extend asset longevity Maintenance cost control

54%

Improve quality or yield

51%

Visibility to management

46%

Safety and risk management

43% 29%

Limited previous software

22%

Capital project plan and execute

19%

Project cost capitalization 0%

20%

92%

35%

93%

34%

88%

34%

86%

43% 37% 28%

80%

57%

40%

62%

44%

63%

40%

89%

60%

High Medium Total 80%

100%

Source: ARC survey with 134 respondents January 2013.

FIG. 1. Business drivers for EAM.

operations provide opportunities for managing the entire plant including maintenance costs and energy management.

Asset management systems require information about the plant equipment to function. Managing this information is critical. This asset (equipment) information includes data about the asset over its entire life cycle.

Managing asset information involves the business processes and technology for: • Organizing the structure to classify the information for consistent, accurate record-keeping and extraction for analysis and reporting in a robust, consistent manner • Creating information entry and storage • Controlling access to information according to role-specific user needs and authority • Utilizing change of management to modify information with proper authorization and timing, which includes recording additions, changes and deletions; it is an audit trail • Auditing the asset-related data.

AIM information types. AIM information pertains to both

structured and unstructured data: Documents contain reference materials including drawings and standard operating procedures. Examples include drawHydrocarbon Processing | MAY 2013 25

Integration Strategies TABLE 1. Comparing key maintenance and executive KPIs EAM KPIs

Financial reported affected C-suite KPIs

Uptime

P&L and balance sheet

Revenue increase and less inventory

Asset longevity Balance sheet

Cash conservation

Cost control

P&L

ProďŹ tability

Safety

Annual report

Risk management

ings from the design-and-build phase of the asset life cycle or upgrades during the operate-and-maintain phase.

Transactions have a predetermined set of data fields, which are transferred concurrently to provide an activity record. For asset management, examples include work orders for maintenance and inspections in the EAM system.

Process data includes real-time process values and timestamped recorded values. This data provides the basis for scheduling PdM. It also provides the input for making reports.

Reports, both online and offline analysis, can include incident reports, KPIs and analysis for assessments of problems or improvements.

Usually, each organization manages its own information. This leads to independent technology, applications and structures.

Inconsistencies in organizing, creating, controlling, change management and asset hierarchies inhibit collaboration and under-

mine information sharing information. These inconsistencies all weaken an APM strategy.

Why change for APM? ARCâ&#x20AC;&#x2122;s annual survey of end users regarding maintenance management systems consistently yields high ranks for these key metrics: uptime, asset longevity, cost control, quality/yield and safety. With the increasing speed of business, visibility into current status has grown in importance (FIG. 1). This requires that asset information is managed with rolebased visibility such that only the most current and pertinent information is visible to each user. Achieving and exceeding these metrics require optimization across the various systems for maintenance and operations.

These goals relate directly with C-suite metrics, which are in the P&L statement and balance sheet (TABLE 1). Executes respond to their metrics much like you do. When you have a positive effect on these metrics, you get and retain C-suite attention. This helps with obtaining the resources needed to be successful.

RALPH RIO, research director for Enterprise Systems, has been with ARC since 2000. Prior to ARC, he served in various marketing management roles at GE Fanuc Automation (GEF), Intellution, Digital Equipment Corporation, and Codex Corporation, as well as manufacturing engineering roles at Texas Instruments and General Electric. He holds a BS degree in mechanical engineering and an MS degree in management science from Rensselaer Polytechnic Institute in Troy, New York.

GE 0HDVXUHPHQW &RQWURO

6DIHW\ )LUVW

PanaFlow HT 6,/ XOWUDVRQLF Ă&#x20AC;RZ PHWHU

In hazardous environments safety comes first 6DIHJXDUGLQJ WKH SHRSOH ZKR ZRUN LQ KD]DUGRXV UHÂżQHU\ HQYLURQPHQWV LV QDWXUDOO\ \RXU ÂżUVW FRQFHUQ $W WKH VDPH WLPH \RX QHHG WR DYRLG FRVWO\ SODQW GRZQWLPH ZKLOH JHWWLQJ WKH PRVW RXW RI \RXU SURFHVV 3DQD)ORZ +7 LV WKH ÂżUVW XOWUDVRQLF Ă&#x20AC;RZ PHWHU ZLWK 6,/ DFFUHGLWDWLRQ 7KDW PHDQV XOWUD VDIH RSHUDWLRQ LQ KLJK WHPSHUDWXUHV ZLWK XOWUDVRQLF DFFXUDF\ VSHHG DQG QR PDLQWHQDQFH ,W¡V D VDIHW\ ÂżUVW

ZZZ JH PFV FRP SDQDĂ&#x20AC;RZ

26 MAY 2013â&#x20AC;&#x201A;|â&#x20AC;&#x201A;HydrocarbonProcessing.com

Select 155 at www.HydrocarbonProcessing.com/RS

SIL

Cer tified

International Headquarters 11920 East Apache Street Tulsa, Oklahoma 74116 United States

T: +1.918.234.1800 F: +1.918.234.2700 johnzinkhamworthy.com

Emissions+, the FREE emissions and conversions calculator

by John Zink Hamworthy Combustion. Available in the iTunes App store.

Š2013 John Zink Company LLC. JOHN ZINK HAMWORTHY COMBUSTION is a trademark of John Zink Company LLC. JZ, KEU, KALDAIR and TODD are registered trademarks of John Zink Company LLC. HAMWORTHY COMBUSTION is a registered trademark of Hamworthy Combustion Engineering Ltd. COEN is a registered trademark of Coen Company, Inc. AIROIL-FLAREGAS is a trademark of Hamworthy Combustion Engineering Ltd. HAMWORTHY PEABODY is a trademark of Coen Company, Inc.

Select 80 at www.HydrocarbonProcessing.com/RS

Select 102 at www.HydrocarbonProcessing.com/RS

Boxscore Construction Analysis

LEE NICHOLS, DIRECTOR, DATA DIVISION Lee.Nichols@GulfPub.com

Jurong Island—Asia’s preemptive LNG trading hub?

In the South Pacific, the artificial island of Jurong Island is the heart of Singapore’s energy and chemicals industry. It houses over 100 leading global petroleum, petrochemicals and specialty chemicals companies. With a rich history as an oil trading hub, Singapore is on course to become a regional liquefied natural gas (LNG) trading center with the construction of its first LNG terminal.

Developed by Singapore LNG, the Singapore LNG terminal is being constructed on a 30-hectare plot at the Meranti Seafront on Jurong Island. The $1.7-billion (B) facility will be used for importing LNG, reloading and regasifying, and storage. It is the first terminal constructed with bidirectional operations and storage. The new terminal will allow companies to unload LNG cargoes, store them, and then ship them at a later date. It will also aid in the redistribution of LNG supplies to regional destinations that cannot build large import facilities or that have ports too small to handle mega-sized LNG vessels.

The original contract to construct and operate the terminal was assigned to PowerGas in 2007. In 2009, Singapore’s Energy Market Authority (EMA) took over control of the project and created Singapore LNG (SLNG) to continue development. In 2010, SLNG awarded the engineering, procurement and construction (EPC) contract to Samsung C&T. TABLE 1 lists additional awards.

The first phase is scheduled to be completed by mid-2013. This phase involves the construction of one jetty and two storage tanks, each with a capacity of 188,000 cubic meters (m3) and the ability to process 3.5 million tons per year (MMtpy) of LNG. The primary jetty (FIG. 1) is being constructed to handle the newest and largest Q-max vessels and LNG carriers. A third, 188,000-m3 storage tank and two additional jetties will be commissioned in 2014, raising total storage capacity to 6 MMtpy. A fourth

tank is scheduled to come online in 2017 at a cost of $500 MM, raising site capacity to 9 MMtpy. The SLNG terminal’s master plan provides for seven storage tanks and a peak capacity of 20 MMtpy.

With LNG demand increasing in Asia, the government of Singapore is contemplating the construction of a second LNG terminal. In November 2012, the EMA issued a tender for a consultant to conduct a six-month feasibility study.

The second terminal could double Singapore’s existing storage capacity.

SLNG received the first LNG cargo from Qatar Operating Co. Ltd. (Qatargas) in the first quarter of 2013. This load will be used to commission the SLNG terminal before full operations begin in the second quarter. Singapore has confirmed BG Singapore Gas Marketing (BGSGM), a BG Group subsidiary company, as the LNG aggregator for train one. BG’s exclusive license allows it to import LNG and sell regasified LNG in Singapore up to 3 MMtpy, or until the year 2023. BG will supply the terminal from its exten-

TABLE 1. Project awards for the SLNG terminal GDF Suez and PowerGas

Front-end engineering and design (FEED) study.

Samsung C&T

EPC, awarded February 2010. Also awarded contract for third storage tank and Secondary Berth Project.

Fluor

Engineering and related management services.

Foster Wheeler Asia Pacific

Project management consultancy services.

WorleyParsons

Basis for design and FEED design. Development of EPC contract tender documents along with subsequent assessment and recommendations.

FIG. 1. View of the Singapore LNG terminal’s primary jetty. Photo courtesy of Singapore LNG Corp.

Hydrocarbon Processing | MAY 2013 29

Boxscore

Upgrade & Revamp

REEal improvement of your compressors

Ever worried about the reliability, energy efficiency and

is fundamental if plants are to meet modern

und

demands in terms of run cycle, energy

so

s

iro nm

ts

REE

v

8

En

9

The continuous improvement of compressors

e s Effici bl

en

10

Relia

environmental soundness of your compressors? Not any more!

y entall

efficiency and legal compliance. In order to

7

do this, clear and accurate data on the real

6

performance of your compressor is required.

5 4

The REE Assessment carried out on-site by

3

HOERBIGER senior compressor experts

2

assesses your equipment using the REE index.

1

It allows you to see where there is room for

0

improvement and where savings can be made, helping you to choose the right technology and make the right business decisions.

For more information: info-hkth@hoerbiger.com

www.hoerbiger.com

30

Select 156 at www.HydrocarbonProcessing.com/RS

sive LNG portfolio and supply positions in Trinidad, Egypt, Nigeria, Equatorial Guinea and the proposed Queensland Curtis LNG facility in Australia. This baseload demand will start the LNG project with additional supplies coming from a combination of LNG and piped natural gas, short-term and long-term contracts, and spot cargoes.

BGSGM has already signed over 1.5 MMtpy of gas sales contracts with a variety of customers in Singapore. The commitments include six large-scale powergeneration companies including Senoko Energy, PowerSeraya, Tuas Power Generation, SembCorp Cogen, Keppel Merlimau and Island Power Co. The additional gas supplies by SLNG will aide in domestic use. Gas-fired plants produce 80% of Singapore’s electricity needs. This percentage is expected to increase to 90% within the next few years.

Unable to fulfill demand, Singapore has been purchasing gas supplies from Indonesia and Malaysia. The SLNG terminal should cover the gap between supply/ demand and lessen natural gas supplies via pipelines from neighboring countries.

Benefits. The International Energy Agency (IEA) forecasts Asian LNG consumption to annually increase to 790 Bm3 by 2015. Within that time, Southeast Asia is scheduled to construct a dozen import terminals resulting in over 35 Bm3 of additional capacity. At present, no real LNG trading hub has been established within the region. It is difficult for companies to purchase LNG at larger (cheaper) quantities. Imports are linked to the price of oil; these prices are five times higher than those in North America, which are based on Henry Hub pricing. An Asian LNG trading hub would allow pricing to better reflect supply and demand fundamentals.

Singapore’s strategic location, businessfriendly environment and bidirectional LNG terminal and storage facilities make it a prime candidate to become Asia’s LNG trading hub.

Location. Singapore’s greatest benefit is its central location to the Asian market. It is nestled between LNG demand centers in Northeast Asia and LNG supply sources in Southeast Asia, the Middle East and Australia. It also sits at the southern tip of the Strait of Malacca, one of the most important shipping lanes in the world and a

Boxscore Construction Analysis gateway to Southeast Asia. This strategic location allows LNG traders to centrally store their LNG cargoes and ship them to heavy consumers such as China, Japan and South Korea.

Singapore’s business-friendly environment is also a major selling point. The government has enacted various incentive programs and tax breaks to attract major oil and gas companies. The Global Trader Programme (GTP) was instituted in 2001 to encourage companies to establish regional/global operations headquarters in Singapore. Companies can benefit from concessionary tax rates of 10% on their qualifying trade income. In 2007, the government adopted a 5% concessionary corporate tax rate for LNG trading income with the intent to spur the development of a LNG trading hub.

In response to these benefits, LNG companies have flocked to Singapore.

Within the last five years, the number of major LNG players has gone from zero to 14; they include BG, BP, ConocoPhillips, Gazprom, GAIL, GDF Suez, Shell and Statoil.

Challenges. Although Singapore benefits from its central location to LNG-hungry Asian markets, a corporate-friendly atmosphere and a new bidirectional LNG terminal facility, it still faces regional and global challenges before it can emerge as Asia’s LNG trading hub.

Singapore is in stiff competition with other Asian LNG players that are actively trying to establish regional LNG hubs.

The stiffest competition stems from Malaysia; this nation is constructing its own bidirectional LNG terminal at Pengerang in Southern Johor. The first construction phase of the $1.3-B Pengerang Independent Deepwater Petroleum Terminal (PIDPT) is set for completion in 1Q 2014. Located within the Pengerang Integrated Petroleum Complex (PIPC), PIDPT is being developed by the Johor state government, Netherland’s Royal Vopak and Malaysia’s Dialog Group. It will have a total storage capacity of 5 MMm3 and allow for storage, loading and regasification of LNG for trading and domestic use. PIDPT’s construction will coincide with Petronas’ Refinery and Petrochemical Integrated Development Project (RAPID). RAPID is a $20B integrated refinery and petrochemical complex located at the PIPC. The proj-

ect will consist of a 300,000-bpd refinery capable of producing 9 MMtpy of petroleum products, and the state-of-the-art petrochemical complex will produce 4.5 MMtpy of downstream petrochemicals.

The RAPID project aims to transform Southern Johor into a new Asian petrochemical hub.

Petronas is also constructing a ninth LNG production train at its Bintulu complex. The new LNG train will add 3.6 MMtpy of capacity. Once completed, the Petronas LNG Complex will have a combined capacity of over 27 MMtpy— creating one of the world’s largest LNG production facilities at a single location.

Malaysia is also set to commission the world’s first LNG regasification unit located on an island jetty. The 3.8-MMtpy Melaka terminal is located approximately 3 km offshore Sungai Udang Port. Two floating storage units (FSUs) will receive and store LNG; the facility has subsea and onshore pipelines connected to the Peninsular Gas Utilization pipeline network. Melaka is expected to receive its first LNG cargo load in August 2013.

South Korea, Japan and Indonesia are also possible Asian LNG hubs. Large energy-consumer countries—South Korea and Japan—already have the necessary facilities and storage tanks to trade LNG but lack the centralized location. Indonesia is striving to increase LNG export capacity with BP’s Tangguh LNG expansion project and Mitsubishi Corp., Pertamina, and Medoc’s joint venture Donggi-Senoro LNG project. Other Asian countries are embracing the use of floating liquefied natural gas (FLNG) and floating storage and regasification (FSRU) units in lieu of pricey onshore terminals capable of handling large LNG vessels.

Singapore’s plans are challenged by the recent development of US LNG exports.

With the Panama Canal expansion to be completed in 2015, it is unknown if US LNG exports will be routed through an LNG hub like Singapore or sent directly to Asian consumers such as Japan, Thailand, the Philippines and South Korea.

The Government of Singapore Investment Corp. has even made an investment in Cheniere’s Sabine Pass LNG export

Advanced Inspection and Assessment Solutions for Process Fired Heater Coils ƒ Improve heater reliability ƒ 0LQLPL]H RIÀLQH VWDWXV

ƒ Minimize heater downtime ƒ ([WHQG UXQ WLPH

Learn more about FTIS™ through animation on our website: www.QuestIntegrity.com/FTIS 1200 Process Heaters in 35 Countries Since 1997

+1 253 893 7070 A TEAM Industrial Services Company Select 157 at www.HydrocarbonProcessing.com/RS

31

Boxscore Construction Analysis terminal to take advantage of cheap US gas supplies. Whether or not that gas will be sent to Singapore is unknown.

An Asian LNG hub could reduce natural gas prices dramatically throughout this region. Singapore’s combination of geographic, economic and infrastructure advantages will facilitate its efforts as the preemptive Asian LNG hub. However, competition from other Asian countries, primarily Malaysia, could curtail these plans. Regardless, Singapore continues to attract significant LNG players eager to make large investments in its emerging LNG sector.

New Database. Hydrocarbon Processing will unveil the new, enhanced Construction Boxscore Database in May. For more than 60 years, the Construction Boxscore Database has provided oil and gas professionals with real-time informa-

tion on refining, petrochemical and gasprocessing construction projects from around the globe.

Boxscore has deep roots in the hydrocarbon processing industry since its inaugural publication in Petroleum Refiner, the forerunner to HP, in August 1947. In the 1950s, “Box score,” as it was called, was one page with a list of 80 projects.

The latest-generation Boxscore contains over 3,500 projects in over 125 countries.

New, enhanced search functions allow users to search broadly or pinpoint specific projects by location. The new project data page provides users with detailed project information on the operating, engineering, licensing and construction companies associated with each project, as well as the project’s name, status, location, cost, capacity, completion date, scope and project history. Users can access contact information for key project personnel.

For more information, visit www. constructionboxscore.com. Learn why Boxscore has been used for over 60 years by engineers, contractors, marketing professionals and business developers to identify construction projects around the world for lead generation, market research, trend analysis and planning.

LEE NICHOLS is director of Gulf Publishing Company’s Data Division. He has five years of experience in the downstream industry and is responsible for market research and trends analysis for the global downstream construction sector.

At present, he manages all data content and sales for Hydrocarbon Processing Construction Boxscore Database, as well as all corporate and global site licenses to World Oil and Hydrocarbon Processing.

Detailed and up-to-date information for active construction projects in the refining, gas processing, and petrochemical industries across the globe | ConstructionBoxscore.com

HERMETICALLY SEALED CENTRIFUGAL PUMPS Today’s Application: LNG-PROCESSING, HANDLING AND STORAGE HERMETIC design features Q

100 % leakage free

Q

Low life-cycle-costs

Q

Low noise level

Q

High reliability

Q

Customized design – adapted to your process requirements

9 9 9 9 9

Customer‘s technical specification

HERMETIC-Pump

/6+6

Type CAMTV 52

Q

Capacity:

120 m³/h

Q

Head:

1400 m

Q

Pressure rating: PN 100

Q

Motor power:

HERMETIC-Pumpen GmbH info.hp@hermetic-pumpen.com · www.hermetic-pumpen.com

Select 158 at www.HydrocarbonProcessing.com/RS

32 MAY 2013 | HydrocarbonProcessing.com

370 kW

9 9 9 9

Manually inputting the control signal feels pretty primitive.

I need to get back in automatic mode for better efficiency.

YOU CAN DO THAT You can automate your control signal using the Fisher ® Control-Disk™ valve from Emerson. Process control loops containing butterfly valves are often placed in manual mode due to poor control performance. This results in operators constantly monitoring and adjusting the control signal, significantly reducing efficiency.

With a control range comparable to a segmented ball valve, the Control-Disk valve enables control closer to the target set point. This allows you to leave your control loop in automatic mode, regardless of process disturbances. With low maintenance requirements and sizes up to NPS 36, it’s time to put the Control-Disk valve in your loop. Visit www.Fisher.com/automatic to watch an animation video or download a brochure.

The Emerson logo is a trademark and service mark of Emerson Electric Co. ©2013 Fisher Controls International LLC. D352200X012 MZ8

GE Works to redefine pump efficiencies.

In processing industries, GE’s SPS pumping systems work to increase efficiency, reduce environmental impact and provide versatile pumping solutions. For example, our multi-stage centrifugal SPS pumps are a widely-used, efficient solution for circulation, transfer boosting and disposal applications. Our SPS systems provide enhanced pumping efficiencies by lowering noise and vibration levels, decreasing construction lead times and delivering improved reliability and extended runtimes. With GE, you have a worldwide support system ready to provide engineering, field service or sales support.

Take a look at our results, and see how GE can work for you.

Call +1 713 260 6731 or e-mail sps.inquiry@ge.com

www.geoilandgas.com/sps

Select 86 at www.HydrocarbonProcessing.com/RS

Viewpoint

ROBERT L. DE MARIA Maintenance Engineering Technical advisor, Dakota Gasification Co.

Consider integral-gear compressors in CO2 services In 1998, the author began researching suitable technology for compressing 95 MMSCFD of “bone dry” (–100°F dew point) carbon dioxide (CO2) from 16.7 psia to 2,710 psia. The compressed CO2 is delivered via a 205-mi pipeline to operations near Weyburn, Saskatchewan, Canada. The oil fields use the gas in enhanced oil recovery (EOR). As an added benefit to the environment, virtually all of the injected CO2 is expected to remain permanently sequestered in the depleted oil fields long after these fields are abandoned.

Options. Three compression options

ROBERT L. DE MARIA is the maintenance engineering technical advisor at Dakota Gasification Co. in Beulah, North Dakota.

He provides technical support on plant equipment reliability through inspections and monitoring, troubleshooting of equipment problems as well as failure analysis, corrective engineering and ensuring plant compliance on the mechanical integrity portion of the OSHA 1910 PSM Std. Mr. De Maria has held previous positions in plant reliability, project management, plant engineering, maintenance and utilities management. He has 43 years of experience in the energy, petrochemical and food-processing industries. His expertise is in rotating equipment reliability, which includes design, audit, selection, troubleshooting, modifications and monitoring. His present focus is on special projects (urea) and total plant reliability through asset data and information management. He received a BS degree in mechanical engineering from Stevens Institute of Technology and has authored numerous papers on rotating equipment reliability.

were closely considered for this project: Option 1: Motor + gear increaser + low-pressure (LP) compressor + medium-pressure (MP) compressor + highpressure (HP) compressor Option 2: Motor + gear increaser + LP compressor + MP compressor and motor + pump Option 3: Motor + integral-gear compressor with 4 pinions (8 stages of compression) with a maximum pinion speed exceeding 26,000 rpm.

Selection process. The manufactur-

ers’ capability, experience and reputation were considered during the selection of the compressor train supplier. The operating company took steps to ensure project success in applying an untested design. Budgetary and personnel resourcing were allocated to achieve high availability. A reliability design audit, lifecycle cost analysis and sub-supplier preference reviews were conducted. During this comprehensive audit process, the shaft seals received considerable attention, and the long-term reliability upgrades were the focus of many discussions. An agreement was reached on the final design.

Cost considerations finalized on using carbon-ring seals for all stages. However, the seal housing design would be capable of accepting dry-gas seals in the event that the carbon-ring seals were proven unsatisfactory. A factory-performance test with CO2 was done to confirm that aerodynamic performance was achieved.

Of course, mechanical test-run data were captured and closely analyzed.

Key design elements. Other mechani-

Implementing Option 3. In 1999, the

Option 3 system was installed (FIG. 1).

This unit has been operating for over 12 years; thus, significant data are available.

The Dakota Gasification Co. confirmed that the integral-gear compressors in dry CO2 service could achieve several important benefits: • Requiring the lowest capital cost • Reducing footprint of compressor • Highest total efficiency • Providing simplest design, with high unit availability • Using a bull-gear-driven main lube oil pump simplifies the lube oil system by eliminating oil-rundown tank • Installing two 50% capacity 20,000 hp machines allows partial continuous production • Using inlet guide vanes provides the highest compressor flow turndown.

cal features under scrutiny included state-of-the-art “flexure pad” bearings, hydraulically fitted coupling hubs and diaphragm-spacer couplings at the driver connection. The motors are synchronous across-the-line starting design with liquid cooling. An original equipment manufacturer-designed control system manages

FIG. 1. The integral-gear compressor in service for the Dakota Gasification Co. Hydrocarbon Processing | MAY 2013 35

Viewpoint speed and continuously monitors machine condition. The control system is integrated into the plant’s digital control system (DCS), along with shaft vibration, thrust position and bearing-temperature monitoring—all are arranged for automated alarm and shutdown.

Within the control system, the design needed to ensure that no liquid will form below the 1,100-psig compression sections during startup and normal opera-

tion. Air-flooded interstage gas coolers are used. A glycol/water solution is circulated through the oil cooler and motor for cooling. Heat is rejected from the glycol/water solution through an air exchanger.

In 2006, an identical third integralgear compressor was placed in service to support additional CO2 sales. All three machines were converted to synthetic lubricating oil; mineral-base lubricants had been used previously. The changeover was

Problem Solving Synthetic Lubricants Reduce overheating, oxidation, excessive bearing and mechanical seal wear Improve your pump reliability, extend MTBO*, reduce downtime and energy consumption with Summit Syngear SH®-7000 Series and Barrier Fluid Series. These synthetic lubricants are resistant to rust, oxidation, corrosion and improve wear protection. They have excellent low temperature fluidity and high temperature stability. Summit Syngear SH®-7000 Series and Barrier Fluid Series are compatible with most process fluids being pumped and commonly used seal materials. These PAO based lubricants come in a wide range of viscosities.

Summit Barrier Fluids are NSF H1 Registered. *mean time between overhaul

prompted when it was discovered that mineral oil was being contaminated by certain sulfur constituents in the impure CO2 stream. Further evidences showed that the carbon-ring seals allowed gas to leak into the lubricant. However, the owner’s reliability focus prompted rigorous inservice testing and comparison of different oils under varying load and operating conditions. With the synthetic oil, a minimum efficiency gain of 2% was realized. In addition, the new oil demonstrated higher resistance to degradation. The incremental cost of the synthetic oil was paid back within a few months. Moreover, the synthetic oil has now been in service for over six years with no replacements necessary.

The continued suitability of the synthetic oil is verified by periodic oil analysis.

Seal service life. The carbon-ring seal service life on HP stages 7 and 8 has met the manufacturer’s warranty. However, a measure of unscheduled downtime and recompression of higher-than-anticipated rates of CO2 has occurred. The events indicate a leak back to suction due to seal wear. The original seals did not reach the intended goal of 20 years between seal repairs. However, the carbon-ring seals for stages 1 through 6 have proven satisfactory. In 2010, the Dakota Gasification Co., in partnership with John Crane and MAN Turbo and Diesel, began the design and development of dry-gas seals for this application. These seals were designed, built and tested for the seventh and eighth stages. Installation is planned for June 2013.

Overall, these eight-stage compressors have met and exceeded expectations. The cost savings exceeded 10% as compared to the initial investment from the nextbest option. The machines have achieved the lowest operating cost, along with 96% availability. With increased interest in CO2 sequestration projects, EOR and urea production, this technology presents merits for further consideration.

Summit 9010 CR 2120 Tyler, TX 75707

36

Industrial Products 800.749.5823 www.klsummit.com

Select 159 at www.HydrocarbonProcessing.com/RS

Made in USA

FIG. 2. Impellers for eight-stage compressor.

Since 1968

Always a Revolution Ahead!

www.hytorc.com info@hytorc.com 1-800-FOR-HYTORC

VISIT US AT OTC - BOOTH #4929 Select 54 at www.HydrocarbonProcessing.com/RS

| Special Report MAINTENANCE AND RELIABILITY Hydrocarbon processing facilities are multi-billion-dollar industrial complexes designed with service lives exceeding 30 years. To keep these plants operating as designed, companies conduct regular maintenance and reliability programs to replace worn equipment. Maintenance and reliability efforts are “insurance policies” to maintain safe operations and yield high-quality products over the service lives of these facilities and processing units. Such efforts reap numerous benefits.

Retrofit of valves for the tank farm operated by Turkish Petroleum Refineries Corp.’s (Tüpraş’) refinery at Izmit, Turkey. Photo courtesy of Rotork Controls, Bath, UK.

Special Report

Maintenance and Reliability J. BAILEY and S. BRADSHAW, ITT Goulds Pumps, Seneca Falls, New York

Avoid hidden costs of suction-specific speed in pumping More than 30 years ago, a landmark study observed a relationship between suction-specific speed and the probability of pump failure that changed the way pumps are selected. This observation created the common perception that a lower suction-specific speed value equals higher pump reliability, with a value below 11,000 established as a common benchmark for good reliability.

This perception has remained relatively unchanged, even though pump design and manufacturing methods have advanced significantly. Pump purchasers continue to rely on a specification limit that derives from one study conducted by one company several years ago. If purchasers order the wrong design or more pump than is needed, then end users, suppliers and plant designers all experience economic consequences.

Here, suction-specific speed, which is one of the most commonly mentioned (yet least understood) terms in oil and gas pumping, is broken down and analyzed.

The start of the specification limit. Referred to as “Nss” in the US or “S” in Europe, suction-specific speed originally helped pump designers predict and compare pump performances. Used since centrifugal pump theory was first developed, Nss is not a speed at all, but rather a simple measure of a pump’s suction. It is based on the net positive suction head required (NPSHR) of a pump, and it should be calculated only at the pump’s maximum diameter and best efficiency point (BEP) flow: N SS =

RPM Q NPSHR 0.75

pipe near the pump. If this effect is strong enough, it can cause cavitation (FIG. 1).

Numerous technical papers and articles reported the problems caused by suction recirculation, but none quantified the problems until the aforementioned refinery—Amoco’s Texas City, Texas refinery—reported the results of a five-year study in 1982. The study involved nearly 500 centrifugal pumps, all of which were designed in the 1960s or earlier. The findings suggested that, when pump Nss values exceeded 11,000, reliability halved.

From this, a commonly understood cap on Nss value was born. The 11,000 value was never promoted as an industry standard, but it became widely accepted by end users, engineering contractors and others as a specification to help ensure pump reliability.

Modern design and manufacturing techniques. Much

work has been done in pump design and manufacturing to improve pump performance since the origin of the Nss limitation: • More robust construction standards, as set forth in API 610, 8th edition, reduce the vibration effects of “off-best efficiency point (BEP)” operation • Modern impeller design methods limit the need to increase the impeller eye diameter to achieve lower NPSHR • The increased use of investment casting, ceramic core techniques and better mold washes have improved component surface finish, accuracy and repeatability.

(1)

Where RPM = rotating speed of the pump, and Q = flow. Note: For double-suction pumps, divide Q by 2.

For a number of years, pump users sought to lower the elevation of tanks to reduce piping costs and the expenses associated with higher-elevation equipment. This required pump manufacturers to strive for lower NPSHR values. Based on the formula in Eq. 1, as NPSHR decreases, Nss increases.

The relationship between higher Nss and lower reliability comes from the way designers traditionally achieved lower NPSHR—by increasing the diameter of the impeller eye. As the impeller eye grows larger, it results in increased suction recirculation. Pump capacity is further reduced and the intensity of the circulation increases, causing a reversal of flow at the suction

Impeller cross-section

d1 Recirculation

FIG. 1. As the diameter of the impeller eye grows, it results in increased suction recirculation.

Hydrocarbon Processing | MAY 2013 39

Maintenance and Reliability Most importantly, widespread use of computational fluid dynamics (CFD) allows pump designers to better define hydraulic passages and improve vane profiles, instead of simply enlarging the impeller eye and hoping for the best outcome.

CFD is a sophisticated, computer-based modeling tool that can be used in the pump design process to simulate various designs, identify flow problems, develop solutions and evaluate operating strategies. As such, CFD is a cost-effective alternative to physical modeling. With modern manufacturing techniques and materials, the companies that make pumps are betNPSH 3% head drop performance 0.35

Cavitation number, ␴3

0.3 0.25

Parabola CFD Parabola test data Ellipse CFD Ellipse test data Circular CFD Circular test data Blunt CFD Blunt test data

0.2

0.15 0.1 0.7

0.8

0.9

1.0 Q/QBEP

1.1

FIG. 2. CFD test results vs. physical test results.

1.2

1.3

ter able to produce the designs developed using CFD, enabling real-world tests on working pumps that are the ultimate test of any design technique.

A recent study sought to demonstrate the validity of one company’s CFD modeling approach. The results ultimately showed how a single, hard-and-fast limit on CFD can be counter-productive. The study investigated how the suction performance of a pump was improved simply by changing the leading-edge profile of the impeller vane while holding all other parameters constant. Four profiles were created—parabola, ellipse, circular and blunt—using rapid prototyping techniques.

As shown in FIG. 2, CFD results matched physical testing fairly well; the experience was almost exactly as predicted by the CFD models. Variations only occurred at 120% best efficiency flow, which means that assumptions around the analysis tend to break down when there is extreme overload on the impeller.

It is important to note that CFD studies are not a “magic bullet”—they offer only a rough approximation of reality. The methodology should always be verified before CFD studies are used to select a pump. When choosing a vendor, several questions must be asked about CFD studies: • How many cells and what type of mesh (i.e., hexahedral or unstructured) were used?

• What turbulence model was used, and what conditions were applied to it?

• What were the inlet/outlet condition assumptions?

• What was the residual error (root mean square and peak) in the converged solution?

• What distribution of Y+ values was obtained?

• To what level of accuracy is the vendor willing to commit?

TABLE 1. Predicted impeller life based on impeller specifications NPSH 3%, ft (m)

Nss /S, US units

Predicted impeller life* for Cast CA6NM (Bhn 262) hours

Blunt

36.8 (11.2)

10,386

15,208

Circular

33.4 (10.2)

11,170

16,631

Impeller profile

Ellipse Parabola

30 (9.1)

12,104

22,971

28.3 (8.6)

12,644

32,023

*Estimated service life of impeller pumping fresh water at 25°C with 23 ppm of dissolved gas content: BEP flow with ␴ = 0.27 and inlet eye velocity = 30.4 m/s (99.7 ft/s).

16,000 15,000

Attainable pump Nss

14,000 13,000 12,000

Not attainable with acceptable performance

Between bearing designs

11,000

Attainable and acceptable performance

10,000 9,000 8,000 0

1,000

2,000 Pump Ns

3,000

FIG. 3. Performance attainable using modern design techniques.

40

Select 160 at www.HydrocarbonProcessing.com/RS

4,000

Maintenance and Reliability • Has the CFD methodology been validated in lab tests?

In general, the use of CFD allows pump manufacturers to produce smarter solutions with higher reliability. Ultimately, these improvements benefit customers through lower cost and reduced risk related to pump station construction and operation.

Industry costs of relying on old specs. Despite these advancements, persistent application of the Nss limit is slowing innovation. Purchasers continue to exclude more economic and reliable pumps from consideration because they are not compliant with the 11,000 Nss value.

For instance, the results of the aforementioned study showed that the parabolic leading-edge profile has the best cavitation performance. In fact, the expected impeller life is doubled between the parabola and the circular profile. For this application, a customer who specifies an Nss value of less than 11,000 would get a pump with poorer NPSHR performance and shorter life, with no improvement in vibration or mean time between failures (MTBF) compared to the design with a higher Nss number (see TABLE 1).

End users take on increased lifecycle costs from purchasing larger, slower and less efficient pumps. Plant designers suffer from reduced optimization because of bigger piping requirements and higher tank elevations. The point is not that Nss is irrelevant, but that, with modern designs, manufacturers can produce reliable pumps with higher Nss numbers than were possible before.

A modern role for Nss in pump selection. When select-

ing pumps, customers should work closely with consultants and pump vendors to understand the design options and the reliability performance of pumps for each application. If done properly, CFD simulations can provide good predictions of pump performance.

FIG. 3 shows the performance attainable and compliant with HI/API 610 vibration limits, using modern design techniques. While suction-specific speed is still a relevant measure for pump manufacturers to assess, setting a hard limit for this specification is no longer appropriate. Doing so places an unnatural constraint on the industry that limits pump users, suppliers and plant designers alike.

JOHN BAILEY is the oil and gas global product marketing manager for ITT Goulds Pumps, responsible for marketing its full range of products that serve customers in the oil and gas industry. Before joining ITT, Mr. Bailey spent 12 years in marketing and management roles for global companies serving industrial and technical markets.

He completed GE Corporate’s Hands-On MBA program, and also holds an MBA degree from the University of Connecticut and a degree in mechanical engineering from the University of Illinois at Champaign.

SIMON BRADSHAW is the director of API product development and technology for ITT Goulds Pumps. He has more than 20 years of engineering experience in the pump industry. Mr. Bradshaw also has supported the Hydraulic Institute in the development of pump standards and best-practice guides. He holds a BS degree in mechanical engineering from Heriot-Watt University, and is a registered chartered engineer in the UK and a member of the Institute of Engineering Designers.

Select 161 at www.HydrocarbonProcessing.com/RS

Hydrocarbon Processing | MAY 2013 41

A new name, a long history.

Selas Fluid Processing is now Linde Engineering North America Inc. For over 60 years, we’ve been there for our reļning and petrochemical customers. Linde Engineering North America Inc. offers single source responsibility for technology, engineering, procurement and construction.

• • • • •

Selas Fluid reľnery and petrochemical ľred heaters Oxidation/incineration technologies Engineered revamps and rebuilds Hydrogen and synthesis gas plants Air separation plants

We’re local. We’re global.

And we’re proud to be both.

Head Ofľce: Five Sentry Parkway East, Suite 300, Blue Bell, PA 19422 USA 610-834-0300 Texas: 3700 West Sam Houston Pkwy. South, Suite 425, Houston, TX 77042 USA 281-717-9090

LENAsales@linde-le.com www.lindeus-engineering.com Select 73 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability D. RENARD, Elliott Engineered Solutions, Jeannette, Pennsylvania

Rerating rotating equipment optimizes olefins plant performance Refrigeration service. Refrigeration compressors are sin-

gle-casing configurations with multiple side streams or side loads and/or extraction streams, as shown in FIG. 3. Propylene compressors are generally larger volume than the corresponding ethylene compressor. However, both applications require unique analysis when evaluating side-stream mixing performance to ensure that required interstage pressures and flows can be met. Refrigeration compressors have multiple nozzle connections that make the ability to reuse the casings a project objective.

2,500 2,000 Plant size, thousand tpy

The operators of mega ethylene plants are under constant pressure to meet enhanced performance specifications and more stringent environmental regulations, while, at the same time, reducing energy costs and improving feedstock flexibility. As ethylene plants have evolved in response to market demands (FIG. 1), technical advances in rotating equipment have boosted operating ranges and performance parameters for the centrifugal compressors and steam turbines—both are at the heart of ethylene plant operations. When considering the effect of changing market conditions on plant performance, particularly feedstock flexibility, producers must first evaluate changes to gas composition and process parameters in light of the plant’s existing turbomachinery.

Rerating installed compressors and steam turbines can be a cost-effective, time-saving solution for increasing throughput without investing in new equipment. Advances in flow path design, stage performance, aerodynamics, manufacturing technology and materials science make it possible to achieve new process parameters within the existing casings, with minimal changes to foundations, piping and other connections.

With careful planning, from review of process parameters through turnaround execution, equipment rerates can be accomplished during a normal maintenance shutdown.

The presented case study will discuss typical cracked-gas (CG) and refrigeration services within an ethylene plant, highlighting technical advances that make it possible to achieve new process parameters with rerated equipment. The case study focuses on a CG equipment train installed in an ethylene plant in the late 1960s that is still in operation due to multiple rerates to meet new processing requirements.

1,500 1,000 500 0 1980

1990

2000 Year

2010

2020

FIG. 1. Growth trend in ethylene plant capacity.

3 casings

Cracked-gas service. CG turbomachinery configurations

vary from installation to installation. CG trains usually consist of large-volume capacity compressors driven by high-power steam turbines. Generally, CG trains have two, three or four compressor casings and sometimes a speed-increasing gear between the casings, as shown in FIG. 2. The most common arrangement is three casings. Three-compressor trains include a double-flow compressor for the low-pressure (LP) section.

Three- and four-compressor arrangements provide more flexibility when considering process changes than a two-compressor train. In most cases, interstage pressures can be modified to accommodate the change in casing flow and head to optimize total performance.

ST driver

MD

ST driver

LP 1st section

LP double flow

HP

4 casings LP 2nd section

MD

2 casings ST driver

LP

Gear

MP/HP

MD Key ST–Steam turbine LP–Low-pressure compressor MP–Medium-pressure compressor HP–High-pressure compressor

FIG. 2. Typical CG train arrangements.

Hydrocarbon Processing | MAY 2013 43

Maintenance and Reliability finite-element analysis (FEA), solids modeling and rotor dynamic analysis. These tools enable design engineers to model a three-dimensional view of the aerodynamic flow path and the effects that design will have on overall performance.

CFD analysis can be used to create impeller stage ratings. Higher and lower flow stage ratings Rerating installed compressors and steam are derived from the tested components to form a family of stages. Within each family, impeller geomturbines can be a cost-effective, timeetry is fixed. Blade heights are varied for higher and saving solution for increasing throughput lower flows. Stage analysis results are continuously checked and verified against actual aerodynamic without investing in new equipment. performance and field tests. This allows the design engineer to match impeller performance and stationary diaphragm performance to achieve optimum overall stage performance. By creating and extending impelrange within a given compressor model. Since the 1960s, imler families, the application engineer can now select from sevpeller efficiencies have improved from 65% to nearly 90%, eral impeller designs to optimize stage-to-stage performance and the range of flow has doubled. This broadened flow range, throughout the compressor aerodynamic flow path. coupled with the development of improved compressor stage Refrigeration compressors usually have multiple sidehead and efficiency, has dramatically expanded a given comstreams that require accurate prediction methods. CFD analypressor’s operating range, as illustrated in FIG. 4. sis has enhanced the designer’s ability to accurately optimize Compressor design technology has advanced through the the mixing of two flows while minimizing pressure drops for use of tools such as computational fluid dynamics (CFD), more reliable performance prediction. Propylene compressors represent a particular challenge when rerating an existing unit. In many cases, with multiple side streams, there may only be a single impeller in a specific section, for which the end user provides a pressure tolerance. The ability to select from various impeller families improves the likelihood that a solution can be achieved.

Improving the performance of the entire flow path also requires consideration of all of the stationary components including the diaphragms, seals and casing volutes, as shown in FIG. 5. For example, diaphragms are milled and bolted to eliminate rough surface finishes inherent with older cast technology. Interstage sealing is accomplished with abradable and deflection-tolerant interstage seals to maintain efficiencies for longer periods. These advancements in stage performance make it possible to reuse existing compressor casings with minimum impact on the equipment train driver. ImproveAdvances in compressor technology. Centrifugal com-

pressors, by design, cover a wide range of flow capacities. The development of impeller “families” has enabled a wider flow

FIG. 3. Refrigeration compressor with multiple side streams.

Impeller efficiencies (approximate)

0.9 Polytropic efficiency, %

0.85 0.8

0.25 Mid-1990s ––> Today 1980s Mid-1960s and 1970s 1950s and 1960s

0.7

0.65 0.6 0.0

0.05

0.1 Flow coefficient

FIG. 4. Evolution of impeller performance.

44 MAY 2013 | HydrocarbonProcessing.com

0.15

0.2 FIG. 5. Typical compressor component upgrades.

Maintenance and Reliability timize both the rotating and stationary turbine components.

Additional improvements include replacement of labyrinth seals with brush-type seals and tip seals (FIG. 6).

ments in manufacturing technology have also contributed to improved compressor design performance. For example, five-axis milling techniques allow high-performance impeller blades to be used for higher flow and/or head to expand a compressor’s operating range.

Advances in steam turbines. Significant plant capacity

increases that result in higher compressor flow—and, therefore, power—cannot be achieved without a correspondingly significant increase in steam flow. Steam turbine drivers must use the existing steam operating conditions to match the rerated compressor train speed. Casing size limitations provide unique challenges to reconfiguring the existing steam flow path to achieve the desired power and speed.

CG and propylene drivers are usually high-power units that may require significant changes when the compressor power has increased. Even with improvements in stage efficiencies over the years, an increase in power usually requires an expanded steam flow area inside the turbine, which may or may not be possible within the physical limits of the existing casing. Additional modifications to the existing aerodynamic flow path, such as removing stages, are usually required to increase the steam flow area.

CFD analysis and other analytical tools have advanced performance and reliability in turbine component design, specifically rotating blades and stationary diaphragms. CFD analysis on high-pressure (HP) and LP staging is used to op-

Case study. This case study references an ethylene plant that was originally built in the late 1960s with a two-body CG compressor train, as shown in FIG. 7. Over the years, the LP compressor was rerated three times prior to this project, and the HP compressor and steam turbine driver were rerated

FIG. 6. Typical steam turbine component upgrades

We Process Your Energy TECHNOLOGIES AND EQUIPMENT FOR OIL & GAS TREATMENT YOUR EFFICIENCY, OUR FLEXIBILTY Because Oil & Gas projects p j require highly reliable and cost effective solutions we timely deliver sinc since over 35 years fully guaranteed modular processequipment packages tailored ta to meet specific customer’s requirements; for all kind of Oil & Gas pro processing applications.

Because each project is unique and highest quality design is paramount to success we offer a uniq unique portfolio of proprietary gas treatment technologies licensed by Prosernat. Backed by our strong investment and partnerships in R&D, with IFP Energies Energ nouvelles and Total, we are proud to bring to the market the most advanced ad technologies in the following areas: Natural Gas Sweetening, Natural Natu Gas Dehydration, Sulfur Recovery, Hydrocarbon Dewpointing, Separation Separatio and CO2 Capture from Flue Gases.

Select 163 at www.HydrocarbonProcessing.com/RS

Hydrocarbon Processing | MAY 2013 45

Maintenance and Reliability To meet the required increase in flow and to accommodate the feedstock revision, the LP and HP compressors required all new rotors and stationary components, but with a reduced number of stages in each unit.

Advanced high-performance impeller technology was used to achieve the expanded flow requirements while limiting the overall train power to an increase of approximately 35%. This allowed the steam turbine driver to be rerated for the third time. The steam turbine also required a new rotor and new diaphragms, with a reduced number of stages.

The end user was able to achieve its operating objectives for capacity increase and feedstock flexibility while minimizing site work and reusing the existing casings, with minimal investment in new turbomachinery hardware. From the original plant installation in the late 1960s, the total train power has increased by more than twice the original design. Application of the latest in compressor and steam turbine technology has enabled the existing compressors and steam turbine to remain in operation for nearly 50 years.

Before

ST driver

LP

Gear

HP

After

ST driver

Booster

LP

Gear

MP/HP

FIG. 7. Rerate case study: Before and After.

twice. These earlier rerates increased plant operating flow and power by 55% over the original installation specifications.

Recently, the end user decided to change the plant’s feedstock and to expand the plant capacity by an additional 58%.

The owner asked the compressor’s original equipment manufacturer (OEM) to conduct a feasibility analysis. The OEM worked closely with the end user, the engineering contractor and the ethylene process licensor to evaluate process conditions and develop a solution.

Early in the analysis, it became clear that such a large flow increase would require installing a booster compressor to reduce the volume flow into the existing units. Reducing the volume flow to the LP and HP compressors allowed these units to be rerated yet again for a fourth and third time, respectively.

DAN RENARD, commercial operations manager retired, spent 28 years at Elliott Co. in various management positions in the compressor and turbine application engineering and marketing departments. He holds a BS degree in mechanical engineering from the Pennsylvania State University. Until his retirement in 2012, he was the marketing manager for the Engineered Solutions Group located at Elliott’s US headquarters in Jeannette, Pennsylvania, leading a team responsible for the rerating of turbomachinery equipment worldwide.

M

S 59

hield

T

Finally, a Safety Lifecycle g Management Solution...

WITH the experts to implement it.

D

ES

IG

N,

IMP

LEMENT, &

S CON

TR

UC

T

aeSShiel Safety Lifecycle Management aeShield™ System Sys stem is a comprehensive platform for executing exe ecuti a sustainable risk management pro ogram through automation of the safety program lifecycle life ecycle process. The system provides a complete comp solution by maintaining rela ation relationships among the risk reduction ttargets, ta rrgets design verification calculations, innspecti and test plans for integrity inspection maanage management, and actual historical data. a SShiel tracks and analyzes PSI, providing ae aeShield a eerts and al a reports on process safety health alerts r ti in real time. aeShield facilitates work w an flow and compliance with ISA84.00.01/ C 61511 615 and the related requirements IEC OSH of O OSHA 1910.119.

Powered By

aesolns.com 46 MAY 2013 | HydrocarbonProcessing.com

Select 165 at www.HydrocarbonProcessing.com/RS

ADVERTORIAL

Maintenance & Reliability: Connectivity of Data for IPL & SIF Testing. Are you making the best use of your data?

MIKE SCOT T / Mike.Scott@aesolns.com

Hazardous events pose serious threats to operating companies by negatively impacting reputation and commercial performance as well as endangering people or the environment. Appropriate test plan strategies provide an important validation that safety systems and protective systems are functioning correctly. Test documentation can be time consuming to maintain, inefficient for technicians to work with, and difficult to connect test results back to the original drivers. aeShield™ provides enhanced connectivity between risk assessment and testing strategies to allow operating facilities to make the best use of limited resources. Improved connectivity between data allows for better decision making on test strategies and improved tracking of test performance. aeShield optimizes strategies for Safety Instrumented Function (SIF) test plans including online device level testing and offline full SIF level testing. Test plan template management within aeShield reduces tedious efforts of updating testing plans manually and ensures current test plans include accurate SIF functionality. The Look Ahead feature illustrates which tests need to be completed in upcoming months to improve decisions about when it is most effective to conduct those tests. By emphasizing drivers for testing and documenting required SIF functionality, aeShield provides maintenance groups a cohesive platform for managing test plan strategies.

Test Plan Look Ahead TODAY’S DATE

SELECTED DATE

April 3, 2013

April 26, 2013

7

Execution Status SIF-LEVEL TEST PLANS

April 2013 W T

M

T

1

2

3

F

S

4

5

6

Advance Month >

S

M

T

May 2013 W T

F

S

1

2

3

4

8

9

10

11

12

13

5

6

7

8

9

10

11

14 15

16

17

18

19

20

12 13

14

15

16

17

18

21 22

23

24

25

26

27

19 20

21

22

23

24

25

28 29

30

26 27

28

29

30

31

dĞƐƚ WůĂŶƐ ΀^/&΁Ͳ΀ ƐƐĞŵďůLJ΁Ͳ΀/> ZĂƟŶŐ΁ 1. [AE-101-100]-[PT-100, PT-200, PR-300]-[3] 2. [AE-101-117]-[TT-100, TT-200]-[2] 3. [AE-101-096]-[FF-100]-[1] 4. [AE-101052]-[LT-100]

IL RATING

1

2

3

Ύ ůŝĐŬ ŽŶ ĚŝīĞƌĞŶƚ ĚĂLJ ƚŽ ĐŚĂŶŐĞ ůŝƐƚ

FIG. 1. Test plan look ahead reporting.

DEVICE-LEVEL TEST PLANS

Complete

Complete

Incomplete

Incomplete

OVERALL STATUS Incomplete

Complete

25%

Early

SIF Testing

S

from the HAZOP/LOPA to the IPL instrumentation model, enabling automatic notification when additional IPL tests are needed.

75%

50%

On-Time

Late

FIG. 2. Easily identify overall status of facility test plans.

Test Data Collection

aeShield offers multiple ways to collect and maintain electronic data. aeShield queries existing asset or calibration systems already being used to store test results and links the test results from the external system to design information. By linking test results and design information back to the initial requirement data, aeShield provides the mechanism to compare realworld test data to assumptions made in front end loading and design activities.

The aeShield import utility enables users to batch import data from a flat file, manually enter data directly into aeShield, or use both mechanisms. This flexibility enables less time spent entering data and improves data availability.

By establishing connectivity between risk assessment, instrumentation design models, and real world data collection, aeShield improves decision making for testing strategies and provides tracking of test performances. Organizations can select from a variety of implementation options and services that best fit their needs. As an industry leader in process safety and SIS design and implementation, aeSolutions is prepared to assist you at every step of the Process Safety Lifecycle, helping achieve compliance and sustainability.

IPL Testing

aeShield increases visibility of how assumptions made during the HAZOP/LOPA process have significant effects on testing requirements. Independent Protection Layers (IPLs) must be validated to ensure accuracy of selected SIL targets. aeShield links the data

At Risk

Use your QR code reader to learn more about aeShield – Safety Lifecycle Management or visit www.aesolns.com.

Select 165 at www.HydrocarbonProcessing.com/RS

KOBELCO Oil-Free Screw Compressors –

A Brilliant Solution for Dirty Gases Flare and Vapor Recovery Service Kobelco oil-free screw compressors are the clear solution for heavy, complex, corrosive, unpredictable gases. They don’t need oil in the compressor chamber, so there’s no risk of contamination or breakdown in viscosity. Best of all, they compress any gas and deliver years of continuous, uninterrupted operation.

High-Capacity Oil-Free Screw - The world’s most sophisticated Oil-Free screw compressor, with the largest capacity – up to 65,000 CFM (110,000 m3/hr). Ideal for refinery flare gas recovery and petrochemical polymer forming gas.

Advanced Oil-Free Screw - Compact design conserves valuable space on offshore platforms and FPSO’s and in refineries. Handles VRU, Flash Gas, LP and MP flare services, accommodating fluctuating gas compositions with heavy hydrocarbons, H2S and water.

Ask Kobelco! The Best Solution for Any Gas Compression.

Kobe Steel, Ltd.

Kobelco Compressors America, Inc.

Tokyo +81-3-5739-6771 Munich +49-89-242-1842

Houston, Texas +1-713-655-0015 sales@kobelco-kca.com

www.kobelcocompressors.com Select 82 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability T. MAYNE and M. ELLUL, Qenos Olefins Australia, Melbourne, Australia; and D. PHILLIPS, Emerson Industrial Automation, Hanover, Maryland

Maximize steam unit performance with precise torque monitoring

with pickup teeth installed on a torsionally soft spacer and intermeshed at a central location. Two monopole sensors 180° apart are mounted on the coupling guard. As the coupling rotates, the ferromagnetic teeth create an AC voltage waveform in the sensor coil, which is digitally processed using known calibration parameters. Due to the intermeshed pickup teeth, the system is referred to as a single-channel phase displacement system, producing two independent torque measurements (FIG. 1).

The system will output torque, power, speed and temperature, which can be easily integrated with any DCS system (FIG. 2).

At the olefins plant, the operating cycle of the steam-driven, CGC train is 7–8 years. During this cycle, the plant reaches production limitations because this compressor train encounters a power limit. To determine the cause of the power limit as “turbine fouling” or “compressor fouling”—or a combination of both—was not confidently possible with the instrumentation installed.

One option was investigated to add more power by upgrading the turbine power rating from 7.5 MW to 9 MW. This re-

4 Sensor 1 Sensor 2

3 2

Voltage

1 0 -1 -2 -3 -4 Time

FIG. 1. The torque-meter coupling produces two independent torque signals.

60,000 50,000 40,000

60 Torque Power Speed Temp

50 40

30,000

30

2,0000

20

10,000

10

0 16:37:11 16:40:42 16:44:12 16:47:44

Temperature, °C

Torque meter installation. The meter consists of two rings

quired a capital investment of $2 million. The plant elected to defer this investment and, instead, a torque meter was installed during the major eight-year shutdown.

The installation involved replacing the existing coupling spacer and flexible halves with the “drop-in” torque meter’s integral flexible elements. The torque meter assembly was dynamically balanced to API standards, so it was not necessary for the user to return any coupling components for the retrofit. The coupling guard was modified so that the two variable-reluctance sensors could be installed, completing the mechanical installation (FIGS. 3–5).

Value (N-m, kW, rpm)

All turbomachinery is subject to degradation that, over time, will affect the system’s efficiency and operational performance. Precise monitoring of turbomachinery performance with continuous torque-monitoring systems can be used to identify gradual efficiency loss, allowing for the development of a more focused maintenance scope to return the system to its optimum operation and efficiency.

Torque monitoring based on heat balance, energy balance and other methods utilizes numerous parameters such as pressure, temperature, flowrate, gas composition, etc., which require precise instrumentation to measure with low uncertainty.1 However, phase displacement technology can be used to accurately measure torque directly at the coupling to within 1% of full-scale torque, a combination of all electrical and mechanical sources of error. This accuracy provides the lowest amount of uncertainty when computing efficiency, compared to alternative methods.

A torque-monitoring system was recently installed on a cracked-gas compressor (CGC) train at Qenos Olefins in Australia to determine the causes of a power limitation. The torque-meter coupling utilizes phase displacement technology for long-term reliability, eliminating the need for recalibration.

0 16:51:12 16:54:43 Time

16:58:13

17:01:44

FIG. 2. Typical output from the torque-meter coupling.

Hydrocarbon Processing | MAY 2013 49

Maintenance and Reliability Results. On restarting the plant and having completed a number of compressor efficiency improvements, the torque meter clearly showed that the 7.5-MW turbine did not require an uprate and that the major power losses were coming from the

Precise monitoring of turbomachinery performance with continuous torquemonitoring systems can be used to identify gradual efficiency loss, allowing for the development of a more focused maintenance scope.

CGC. The torque meter also allowed online tuning of the seal gas system of the compressor to establish the lowest power draw from the recycles that the seal system introduces. An additional 200 KW of power was reduced from the turbine load, with the manual adjustments made on the seal gas system.

The torque meter is now being used to monitor turbine steam-fouling issues and process-related compressor fouling

so that corrective online washing can be activated as soon as issues arise.

The historical data collected from the torque meter will also provide a baseline of mechanical loading through the drivetrain of the CGC over time. This data will be used to determine if increases in the maximum continuous operating speed rating of the compressor and the turbine can be accomplished at minimal costs. This would achieve increases in the operating envelope of the compressor.

Furthermore, the value of the torque meter justified the installation of a second system for the olefins plant’s second steam-cracking plant turbine/ compressor train in October 2012.

1

LITERATURE CITED Kurz, R., K. Brun and D. Legrand, “Field performance testing of gas turbine-driven centrifugal compressors,” Proceedings of the 28th Turbomachinery Symposium, Turbomachinery Laboratory, Texas A&M University, College Station, Texas, pp. 216–220, 1999.

DANIEL PHILLIPS is the field service engineering manager for Emerson Industrial Automation’s Kop-Flex brand of couplings in Baltimore, Maryland. He assists users with installation, commissioning and troubleshooting of power transmission products. Mr. Phillips has extensive experience with applying torque-monitoring solutions to increase the reliability and efficiency of equipment in the metals and oil and gas industries. He has a BS degree in mechanical engineering from the University of Maryland, Baltimore County, and he has 10 years of experience in the mechanical engineering field.

MARK ELLUL has worked in Qenos Olefins Australia’s olefins refinery for 30 years as an instrument and electrical specialist.

He has coordinated field maintenance activities and worked in the process and control applications group. Mr. Ellul has also been assigned to major rotating machinery instrument upgrade projects.

TREVOR MAYNE is the lead machinery engineer for Qenos Olefins Australia’s olefins refinery. He has worked with rotating equipment in the olefins refinery and in the plastics and synthetic rubber plants over the last 20 years. Mr. Mayne has held positions in reliability and in field maintenance, both at Qenos Olefins Australia’s Altona plant and in Saudi Arabia with ExxonMobil.

FIG. 3. Completed mechanical installation at Qenos Olefins.

FIG. 4. Torque-meter coupling retrofit at Qenos Olefins plant.

50 MAY 2013 | HydrocarbonProcessing.com

FIG. 5. Existing coupling arrangement (top) and retrofitted torque-meter coupling (bottom).

The world of synthetic lubricants just took three giant leaps forward.

Challenging times and changing technology call for forward-thinking solutions. That’s why we’ve taken our proven Mobil SHC™ synthetic lubricants — the standard-setting oils and greases for more than 40 years — into the future, unleashing the next generation of productivity with three new advances. Each delivering overall balanced performance with substantial energy-efficiency benefits.

Mobil SHC™ 600 Series — The enhanced formulation improves viscosity and low-temperature properties as it delivers outstanding performance across a wide range of circulating and gear applications.

Mobil SHC™ Gear — This supreme gear oil was reengineered to deliver optimum equipment oil life in gearboxes, even under extreme conditions, with significant reduction in energy consumption.

Mobil SHC™ Gear OH Series — Our customized formulation for the Off-Highway sector features dependable technology with excellent low-temperature performance.

Discover the advanced technology of Mobil SHC. To see how four decades of synthetic innovation just jumped a generation ahead, visit mobilindustrial.com.

The energy efficiency design is a trademark of Exxon Mobil Corporation. Energy efficiency relates solely to the fluid performance when compared with conventional reference oils of the same viscosity grade in gear applications.

The technology used allows up to 3.6% efficiency compared with the reference when tested in circulating and gear applications under controlled conditions. Efficiency improvements will vary based on operating conditions and applications. © 2013 Exxon Mobil Corporation.

All trademarks used herein are trademarks or registered trademarks of Exxon Mobil Corporation or one of its subsidiaries.

Select 71 at www.HydrocarbonProcessing.com/RS

I N D U S T R I A L K N O W-H O W

GEARED TOWARD PROCESS GAS APPLICATIONS

FLOW EQUIPMENT LEADERSHIP

$BNFSPO T JOUFHSBMMZ HFBSFE DPNQSFTTPST CSJOH ZFBST PG SFMJBCJMJUZ UP QSPDFTT HBT BQQMJDBUJPOT Integrally geared compressors may be new to some, but weâ&#x20AC;&#x2122;ve manufactured over 12,000 since 1955. Theyâ&#x20AC;&#x2122;ve proven their reliability, ďŹ rst for industrial gas applications, now for natural gas compression and process applications, including small-scale LNG. The simplicity of design means fewer working parts, more uptime and lower all-around costs. Find out more at www.c-a-m.com/cs.

Cameronâ&#x20AC;&#x2122;s Process Gas Advantages: t .FFUT "1* 4UBOEBSET

t 0JM 'SFF 1SPDFTT (BT

t (SFBUFS 3FMJBCJMJUZ

t 4NBMMFS 'PPUQSJOU

t 0QUJNVN $POUSPM

t 0QUJNJ[FE &Gm DJFODZ

t -PX .BJOUFOBODF

t -PXFS -JGF $ZDMF $PTU

t 1SPWFO 5SBDL 3FDPSE

AD00523CC

RAISING PERFORMANCE. TOGETHERâ&#x201E;˘ Select 55 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability N. GHAISAS, PE, Fluor Canada Ltd., Calgary, Alberta, Canada

Investigate power limitations in a large steam turbine An investigation was carried out to rectify power limitations for a steam-turbine driver of a cracked-gas compressor (CGC) in a world-class ethylene plant.

Description. The turbine is a nine-stage, impulse-type, multi-valve, straight-condensing turbine. Five years after the initial start up, the first-stage nozzles of the turbine were replaced with a new set of nozzles supplied by the original equipment manufacturer (OEM). The replacement was done to yield higher power output now required for olefins production. The new nameplate-rated power of the turbine is 54,000 hp at 3,927 rpm. The revamp excluded any modifications to the compressor drive and couplings due to adequate design margins that could accommodate higher flowrate, pressure and torque.

Steam is supplied to this turbine from a high-pressure header at 700 psig at 900°F through a 24-in. nozzle. An inverted, oil-operated trip and throttle valve is mounted just upstream of the turbine. The top half of the steam chest contains five venturi-type, single-seated governing valves and valve seat nozzles. Each valve has a nominal diameter of 4 in. The valves are sequentially lifted by a bar mechanism. The I/P (current to pressure) converter of the actuator receives a 0 mA–20 mA signal from the turbine’s electronic governor in response to speed changes.

For nearly three years after the revamp, the turbine was in continuous operation and exhibited normal performance.

However, first signs of deviation from steady-state conditions became apparent when the turbine was no longer able to deliver power required by the CGC train. Turbine speed could not be maintained and had gradually reduced by as much as 5% from the rated speed of 3,927 rpm. Consequently, the setpoint for first-stage suction pressure of the compressor had to be changed to increase the pressure. The servo-motor piston in the turbine governing system was at a fully retracted position, indicating that all five of the bar-operated valves were lifted.

The situation became critical because the CGC’s capacity reduction directly impacted ethylene production, and, ultimately, negatively impacted earnings for this facility.

Source of power limits. Two possible reasons for the turbine’s power limitation were initially investigated and included: • Deposit buildup on the rotor from steam impurities • Detached row(s) of blades.

An analysis of the steam quality had been done on a monthly frequency. Test results did not indicate that impurities were present in live steam. Turbine vibrations were low (15 mi-

crons p-p on both radial bearings, and there was no cognizable changes in the thrust position of the rotor. Thus, the listed root causes were ruled out. Instead, as suggested by the author, a nozzle-bowl pressure survey was conducted to identify any abnormalities inside the steam path.

With the governor valve lift indicator showing a fully open lift bar, steam-inlet pressure, nozzle-bowl pressures and firststage pressure (also known as nozzle-ring pressure) were recorded. TABLE 1 lists details from the nozzle-bowl survey. Reference was also taken to the “steam-throttle flow vs. first-stage pressure” and “steam-throttle flow vs. valve lift” charts supplied by the turbine manufacturer.

Interpretation of the measured data revealed that: • Pressures in nozzle bowls 1, 2 and 4 were nearly equal to the steam inlet pressure, but pressures in bowls 3 and 5 were found to be equal to first-stage pressure, i.e., the downstream pressure. FIG. 1 shows the arrangement of governing valves.

TABLE 1. Measured parameters Measurement

Value

Steam inlet pressure, barg

46.1

First-stage pressure, barg

39.8

Nozzle bowl No. 1 pressure, barg

46

Nozzle bowl No. 2 pressure, barg

45.8

Nozzle bowl No. 3 pressure, barg

39.6

Nozzle bowl No. 4 pressure, barg

46

Nozzle bowl No. 5 pressure, barg

39.6

Steam chest internal

Governor valve

4

Double nut

2

1

5

Threaded end

3 Inlet steam

Valve lifter bar Steam chest Venturi nozzle

Inlet nozzle

Valve seat

FIG. 1. Side view of the governing valves for the turbine.

Hydrocarbon Processing | MAY 2013 53

Maintenance and Reliability • Since the pressure in nozzle bowls 1, 2 and 4 were in the proximity of steam-inlet pressure, the possibility of restricted steam flow due to a clogged strainer in the trip and throttle valve was eliminated. Similarly, since the governor lift rod had traversed to full lift position, binding or interference in governor linkage was not considered as a contributing factor to the problem.

blocking the nozzle passages. This explained the reduced steam flow to the turbine and the resulting power limitation.

The venturi nozzle for valves 3, 4 and 5 were found to be loose in the steam chest. The nozzles were originally set in bored openings in the steam chest with a tap fit and secured by one set screw.

During the shutdown, a boroscopic inspection of the rotor was conducted to confirm that there was no deposit buildup on the blades.

Following inspection, all five governing valves Operation of the cracked-gas compressor were replaced with a set of spare valves. Loose nozis a major factor governing the profitability zles were secured into place and locked with three set screws to prevent them from coming off. Tack of any ethylene facility. Limitations on the welding of nozzles was not carried out due to the turbine driver for this compressor can be possibility of thermal distortion of the steam chest.

After repairs and calibration checks of the governthe difference between profitability ing system, the turbine was warmed up. This was or doom for any olefins site. followed by a over-speed-trip test and then the turbine was coupled to the compressor string. FIG. 3 is the calibration curve for the actuator.

• These results inferred that the power limitation was Further evaluations were carried out to investigate why the attributable to valves 3 and 5, in that these valves had likely two governing valves had broken. These valves are the closest detached from the lifting bar and blocked the venturi nozzles to the steam-inlet opening in the steam chest. Thus, they are underneath. first valves to take full steam pressure. At the time of revamp, Following the pressure survey, the turbine was shut down only the first-stage nozzle ring was replaced, but, apparently, no to open the steam chest top cover for internal inspection mechanical design check was done by the manufacturer on the (FIG. 2). Upon disassembly, the stems of governing valves 3 upstream steam path. It was determined from calculations that the steam velocity at the turbine inlet nozzle, corresponding and 5 were found to have sheared along the valve stem threads. to post-revamp conditions, was in the proximity of 150 ft/sec.

The broken valves were sitting on the venturi nozzles, thus Although it is not an absolute value, good engineering guides recommend flow velocity through nozzles to be less than 150 ft/sec for reasonable pressure drop and flow distribution. Continuous, prolonged operation at high-steam velocity and the associated energy of trapped steam can create turbulence in the steam chest. Such turbulence must have loosened the securing nuts on governing valves 3 and 5.

The steam velocity worked its way through the weakest link, which is the valve threads, and then sheared the threads.

The findings and evaluation were communicated to the turbine manufacturer. The manufacturer concurred with the analysis and asked that the broken valves be returned for metallurgical investigation. In the end, the new set of replacement valves was nitrided at the manufacturer’s facility to create a casehardened surface on ASTM A410 valve material. In the next FIG. 2. Steam chest under investigation. available opportunity, these valves were installed in the steam chest. Since this changeout, the turbine has operated normally 100 and is able to meet the process conditions demand.

Lift, %

60

20 0

10

40

Signal, %

FIG. 3. Calibration curve for the actuator.

54 MAY 2013 | HydrocarbonProcessing.com

60

80

100

NEETIN GHAISAS is a Fluor Fellow for rotating equipment and a director, design engineering in Fluor’s Calgary, Alberta, Canada, office. He holds an MS degree in mechanical engineering and is a registered practicing Professional Engineer in the province of Alberta, Canada. Mr. Ghaisas possesses over 31 years of professional experience, especially in the specification, selection, application and troubleshooting of rotating equipment. Mr. Ghaisas is a subject matter expert for Fluor Corp. on compressors, steam turbines, reliability-centered maintenance and root-cause-failure analysis. Further, he has a number of years of experience in machinery vibration diagnosis including transient and steady-state analysis. In Fluor’s Calgary office, he serves as a group leader for rotating equipment engineers. Mr. Ghaisas is a member of the task force member for several API standards and is also a member of Machinery Function Team for Process Industry Practices.

Johnson Screens designs and manufactures a wide range of vessel internals for media retention.

Among these innovative products is a line of internals VTFE JO EPXO æPX QSPDFTTFT UP SFUBJO UIF PGUFO DPTUMZ media and to provide a collection area for the process æPX BDSPTT UIF FOUJSF WFTTFM EJBNFUFS PS MFOHUI #FDBVTF PG UIFJS TUSFOHUI EVSBCJMJUZ æPX DIBSBDUFSJTUJDT +PIOTPO 4DSFFOT TVQQPSU HSJE BOE EJTUSJCVUPS USBZ TZTUFNT BSF XJEFMZ VTFE JO q IZESPUSFBUFST q desulfurizers q IZESPDSBDLFST q ESZFST q TBOE åMUFST q gas sweeteners q PUIFS BCTPSQUJPO TZTUFNT

CONTACT US!

www.johnsonscreens.com

"6453"-*" "4*" 1"$*'*$ &6301& .*%%-& &"45 "'3*$" 5&-

5&-

TBMFTBTJBQBDJÃ¥D!KPIOTPOTDSFFOT DPN TBMFTFVSPQF!KPIOTPOTDSFFOT DPN JAPAN 5&-

TBMFT KQO!KPIOTPOTDSFFOT DP KQ

/035) 4065) $&/53"- ".&3*$" 5&-

IQJTBMFTOTB!KPIOTPOTDSFFOT DPN Select 91 at www.HydrocarbonProcessing.com/RS

Solutions That Fit

SM

Compliance Systems

Safety Personnel

Total Safety has the people, programs and processes to help mitigate risks during critical operations. Our skilled safety professionals have expertise in designing custom solutions that fit your specific needs, ensure safe operations, and protect your people and assets. Look to our dependable safety and compliance solutions to create the ideal working environment for your workers worldwide.

Watch our new Centralized Confined Space Monitoring System video on our website at TotalSafety.com.

Professional Services 888.448.6825 | TotalSafety.com Select 99 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability S. SHAW, AESSEAL plc, Rotherham, South Yorkshire, United Kingdom

Seal safety may require going beyond typical standards In 1984, a single, large-scale emission of methyl isocyanate (MIC) led to the loss of thousands of lives at Bhopal, India.

The event became one of the most consequential process safety failures for the manufacturing industry. Follow-up releases and articles have documented how machinery failures played a role in the Bhopal disaster.1 Incremental changes made to the original process design involving pumps and seals ultimately contributed to this catastrophic event.

A comprehensive examination of current seal standards indicates that possible inconsistencies and weaknesses are present in these guides. Such deficiencies were confirmed in leakage testing conducted by a leading seal manufacturer in mid-2012. There is consensus among respected industry professionals that users must reach beyond typical standards in applying safer mechanical seals when hazardous and toxic chemicals are involved.

Pumps, seals and what happened in Bhopal. In the first major change from original plant design, engineers and managers at the Bhopal facility adopted an alternative fluid-transfer method. The operators pressurized the MIC storage tank and reversed the flow directly into the derivatives unit, thus avoiding the failure-prone MIC-transfer pumps. According to important cross-references released to the public record in early 2012, this alternative practice was adopted because it “minimized the potential for transfer-pump seal failures and exposing employees to the lethal process.”1 The second major change involved a “circulation pump” processing the MIC through a fluorocarbon-based refrigeration system. Refrigeration was used to keep the MIC temperature near 0°C, thus greatly reducing the risk of a thermal runaway reaction. On January 9, 1982, there was a severe incident in which a mechanical seal face had shattered. A significant MIC release occurred, and 25 employees were sent to hospitals with serious injuries. The refrigeration system was no longer used after January 12, 1982.

The reasons for making process changes were identical. Both operational changes were implemented to avoid future pump and seal failures that could expose employees to lethal chemicals.

In May 1984, an independent audit team arrived at the Bhopal MIC facility. The team made several recommendations, one of which was to install dual seals on the centrifugal pumps.

Dual seals (FIG. 1) and appropriate seal-support systems or flush-plan arrangements can effectively control fugitive emis-

sions. In this case, the dual seals and support systems were not installed before the tragic release event on December 2, 1984.

Prevailing industry standards. In all hazardous services, sealing safety and equipment reliability interact. Present practices for sealing volatile compounds are linked to American Petroleum Institute (API) 682/ISO21409, which is the international standard most often used. The standard specifies the requirements and offers best-practice recommendations for sealing centrifugal and rotary-positive displacement pumps. The hydrocarbon processing industry considers this standard of critical importance in hazardous, flammable and/or toxic services.

It is widely understood that shaft sealing systems conforming to API 682/ISO21409 must meet four stated reliability objectives, which are: 1. All seals should operate continuously for 25,000 hours without the need for replacement.

2. Containment seals (wet- and dry-mechanical seals) should operate for at least 25,000 hours without the need for replacement at any containment seal chamber pressure equal to or less than the seal leakage pressure switch setting (not to exceed a gauge pressure of 0.07 MPa/0.7 bar/10 psi), and for at least 8 hours, at the seal chamber conditions.

3. All seals should operate for 25,000 hours without the need for replacement while either complying with local emis-

FIG. 1. With a dual mechanical seal, the space between the sleeve and inside diameter of the two sets of seal faces is filled with a pressurized barrier fluid.

Hydrocarbon Processing | MAY 2013 57

Maintenance and Reliability sions regulations, or exhibiting a maximum screening value of 1,000 ml/m3 (1,000 ppm by volume), as measured by the US EPA Method 21, whichever is more stringent.

4. The minimum performance requirements and permitted leakage detailed in Section 10.3.1.4.1 of the standard specifies an average liquid leakage rate of less than 5.6 g/hr per set of seal faces.

The fourth point is very significant. Because 5.6 g/hr is the maximum acceptable leakage rate given in this standard, and seal users expect compliance. Many users base their safety risk analyses on this information. However, mechanical seals can enter the supply chain and vastly exceed the presumed maximum leakage rate of 5.6 g/hr.

Essential testing protocol and results. Testing for liquid

leakage by actually using liquids is not always convenient. Testing with clean air is feasible as long as the equivalencies are established by calculation, calibrated tests and measurements. With this goal in mind, API 682/ISO 21409 allows seal manufacturers to follow an air-testing protocol. The protocol is described in Section 10.3.4.2 of the standard and is summarized in TABLE 1.

However, while meeting the air-test criteria, a mechanical seal can utterly miss the liquid leakage limit given in objective 4, from Section 10.3.1.4.1 of the standard. That objective, as stated above, is to limit allowable liquid leakage from new mechanical seals to TABLE 1. Key items of API/ISO protocol for mechanical seal air tests Each sealing section shall be independently pressurized with clean air to a gauge pressure of 0.17 MPa (1.7 bar or 25 psi). The volume of each test setup shall be a maximum of 28 l (1 ft3).

Isolate the test setup from the pressurizing source and maintain the pressure for at least 5 minutes.

Maximum pressure drop during the test shall be 0.014 MPa (0.14 bar or 2 psi).*

5.6 g/hr. This potentially far-reaching safety issue prompted a thorough examination by calculations. The accuracy of the calculations was then validated by air-testing in accordance with the API protocol. Air tests were followed by water tests.

TABLE 2 summarizes the air-test results. Using the criteria from TABLE 1, air leakage through a 0.008-in. orifice in a pressurized 26.5-l loop would pass the test. But the tests also confirmed that seals in a pressurized lower-volume loop (1.5 l) fitted with the same orifice would not meet the stipulations of TABLE 1. In a 1.5-l loop, the pressure loss was 22 psi in 5 minutes. Result: The seal failed the test. On the positive side, the test descriptions and results of TABLE 2 indicated that a well-designed and properly manufactured mechanical seal is virtually airtight. It will experience no pressure decay in either the small (1.5 l) or large (26.5 l) test loop.

Testing for liquid leakage. For testing with water in the seal loops, the loops were fitted with a 0.2-mm (0.008-in.) orifice.

Water was added to the loop and connected to a pressurized volume of nitrogen. Three different nitrogen pressure settings were used for the water-test sequence, as listed in TABLE 3. All water tests showed that the liquid-leakage rates exceeded the maximum permissible limit of 5.6 g/hr: • Filled with water and pressurized to a gauge pressure of 0.2 MPa (2 bar, or 29 psig), a mechanical seal actually leaked 1,500 g/hr—270 times the allowable rate.

• When pressurized to 7 barg (102 psi), the leakage rate jumped to 50 g/min or 3,000 g/hr (6.6 lb/hr)—535 times the allowable rate.

• Finally, when pressurized to 40 barg (580 psi), which is close to the maximum pressure limit for a “Type A” seal under API 682/ISO21049, the leakage rate is 125 cc/min (4.2 fl.oz/ min) or 7,500 g/hr (16.5 lb/hr)—1,300 times the allowable rate.

None of the leakage rates in TABLE 3 complied with the design intent of Section 10.3.1.4.1 in API 682. Moreover, such extreme leakage rates will obviously not meet the reasonable expectations of reliability and safety-focused users.*

TABLE 2. Summary of air-test results Test description

Pressure drop

26.5-l Closed-loop system with virtually airtight mechanical seal and no orifice (control test)

Zero pressure drop—Pass

26.5-l Closed-loop no leakage system with 0.008-in. orifice added

2-psi pressure drop—Pass

26.5-l System with initially airtight 2-psi pressure drop—Pass mechanical seal and 0.008-in. orifice added 1.5-l System with initially airtight 22-psi pressure drop—Fail mechanical seal and 0.008-in.orifice added 1.5-l System with virtually airtight mechanical seal and no orifice

Zero pressure drop—Pass

TABLE 3. Liquid leakage rates at various pressures with 0.008-in. (0.2-mm) orifice Test pressure

Leakage

Actual vs. allowable leakage

2 barg (29 psig)

1,500 g/hr

270 * the maximum

7 barg (102 psig)

3,000 g/hr

535 * the maximum

40 barg (580 psig)

7,500 g/hr

1,300 * the maximum

58 MAY 2013 | HydrocarbonProcessing.com

ANSI/API RP 754 and safety guidelines of the Baker Panel. The safety-critical implications of the air test vs. actual

liquid-leakage discrepancies are staggering. Seals that enter the market based on passing the present API/ISO air test can, in the most extreme case, leak 7,500 g/hr of flammable, toxic or hazardous product.

The significance of this issue becomes clearer when we examine another API standard, ANSI/API RP 754. This document was created following the 2005 explosion at the BP refinery in Texas City, Texas. It emphasizes recommendations, which following the Texas City event, were made by the BP U.S. Refineries Independent Safety Review Panel (the Baker Panel) and the US Chemical Safety Board. The stated aims of ANSI/API RP 754 are to: • Indicate changes in company or industry performance, to be used to drive continuous improvement in performance • Perform company-to-company or industry segment-tosegment benchmarking • Serve as a leading indicator of potential process safety issues, which could result in a catastrophic event.

A significant focus of ANSI/API RP 754 is “loss of process containment” (LOPC) events. LOPC events are categorized in

EARLY BIRD SPECIAL

HPIRPC.COM

Register by 31 May to Save!

NEW DELHI, INDIA | 9–11 JULY

Register by 31 May to SAVE up to 20% Conference registration includes: • Pre-conference tour of Indian Oil’s Panipat Refinery & Petrochemicals Complex (9 July 2013)* • Two-day conference program (10-11 July 2013) including keynote addresses, general presentations and panel discussions • Breakfasts, lunches and refreshment breaks • Access to exhibition floor throughout conference activity *Refinery tour registration is offered on a first-come, first-served basis. Space is limited.

2013 Conference Fees: Early Bird Fee (by 31 May)

Regular Fee

Individual

$945

$1,045

Team of Two

$1,700

$1,875

Group of Five

$4,250

$4,700

For more information about IRPC 2013, please contact Melissa Smith, Events Director, Gulf Publishing Company, at +1 (713) 520-4475 or Melissa.Smith@GulfPub.com.

❑ Individual

❑ Team of Two

❑ Group of Five

Company _______________________________________________________________________________________ Address ____________________________________________________________________________________________________ City ______________________________________________________________

State _________________________________

ZIP _______________________________________________________________

Country _______________________________

Phone ____________________________________________________________

Fax ___________________________________

Name ____________________________________________________________

Title __________________________________

Signature _________________________________________________________

Date __________________________________

Enclose with payment to: 2 Greenway Plaza, Suite 1020, Houston, Texas 77046 USA or Register online at

HPIRPC.com

View the Agenda at

HPIRPC.COM

NEW DELHI, INDIA | 9–11 JULY

Early Bird Special Register by 31 May to SAVE up to 20% IRPC 2013 Advisory Board Members: ANDREA AMOROSO Vice President, Process Technology eni - Refining & Marketing Division

GIACOMO FOSSATARO General Manager Walter Tosto S.p.A.

A.K. PURWAHA Chairman & Managing Director Engineers India Limited

A. K. ARORA Director General Petroleum Federation of India

DR. MADHUKAR ONKARNATH GARG Director CSIR-Indian Institute of Petroleum

JOHN BARIC Licensing Technology Manager Shell Global Solutions International B.V.

RAJKUMAR GHOSH Director (Refineries) Indian Oil Corporation Limited (Refineries Division)

GIACOMO RISPOLI Executive Vice President, Research & Development and Projects eni - Refining & Marketing Division

ANINDYA SUNDAR BASU Managing Director Chennai Petroleum Corporation Ltd

CHAKRAPANY MANOHARAN Director-Refinery Essar Oil Ltd.

ERIC BENAZZI Marketing Director Axens

SAMIK MUKHERJEE Country Manager Technip India Limited

CARLOS CABRERA Executive Chairman Ivanhoe Energy

B. K. NAMDEO Executive Director, International Trade & Supplies Hindustan Petroleum Corporation Limited

DR. CHARLES CAMERON Head of Technology, Downstream; VP, Formulated Products Technology BP plc

SYAMAL PODDAR President Poddar & Associates

STEPHANY ROMANOW Editor Hydrocarbon Processing DR. AJIT SAPRE Group President Research and Technology Reliance Technology Group MICHAEL STOCKLE C Eng FlChemE Chief Engineer - Refining Technology Foster Wheeler MR.K.VENKATARAMANAN Chief Executive Officer & Managing Director Larsen & Toubro S VENKATRAMAN Director (Business Development) Gail (India) Limited

IRPC 2013 Sponsor Organizations: Technip

Carpenteria Corsi s.r.l

Petroleum Economist

KBR Technology

Walter Tosto

OilCareers

GTC Technology

Axens

IndianOil

Elliott Group

Gulf Publishing Company

Petrofed

Maintenance and Reliability four tiers by severity, with the most extreme LOPC events being labeled as “Tier 1” and defined as “LOPC events of greater consequence.” To be categorized as a “Tier 1” hazard, one of two conditions must be met: 1. LOPC results in a lost-time injury, fatality or fire; or 2. Any LOPC above a defined size limit, regardless of the consequence.

For an incident to be considered “most serious,” the release must be classified as an “acute release.” By definition, an acute release must meet a threshold value or quantity limit within one hour. TABLE 4 lists the threshold quantities for a Tier 1 event.

From TABLE 3, a mechanical-seal assembly containing a 0.008-in. (0.2-mm) orifice at a typical pressure (7 barg or 102 psi) was capable of leaking 3,000 g/hr (6.6 lb/hr). Also, it passes the present API air test. If the duty was a TIH Zone A material, leakage rates are so severe that, by ANSI/API RP 754 definitions, the seal would be in a constant state of the most serious reportable “Tier 1” acute leakage. This leakage would be considered a separate new Tier 1 acute LOPC process safety incident each and every hour that the seal was used to contain the process.

TABLE 4. Threshold quantity limits for a Tier 1 LOPC hazard Threshold release category Material hazard classification

Threshold quantity

1

TIH Zone A materials

5 kg (11 lb)

2

TIH Zone B materials

25 kg (55 lb)

3

TIH Zone C materials

100 kg (220 lb)

4

TIH Zone D materials

200 kg (400 lb)

5

Flammable gases or liquids 500 kg (1,100 lb) with initial boiling point ≤ 35°C (95°F) and flash point < 23°C (73°F) or other packing group I materials excluding strong acids/bases

6

Liquids with initial boiling point > 35°C (95°F) and flash point < 23°C (73°F) or other packing group II materials, excluding moderate acids/ bases

7

Liquids with flash point ≥ 23°C 2,000 kg (4,400 lb) (73°F) and ≤ 60°C (140°F) or 14 bbl or liquids with flash point > 60°C (140°F) released at a temperature at or above flash point or strong acids/ bases or other packing group III materials or Division 2.2 nonflammable, nontoxic gases (excluding steam, hot condensate, and compressed or liquefied air)

Use seals that satisfy common-sense safety demands.

When it comes to sealing issues, industry must apply the lessons from Bhopal and other disasters. It is obvious that a test that can allow up to 1,300 times the intended leakage should be revised. By comparing the test results to API’s own hazardous release thresholds, it appears that safety should be sufficient grounds to amend the current standard.

The world’s best petrochemical and oil refining plants are not using API or other industry specifications as stand-alone criteria.

The leading users of, say, centrifugal pumps, recognize that specification supplements or amendments are needed. The various cautionary clauses given in the typical introductory statements to different standards are well-known best-practices companies. For example, one of these clauses reads: “The purchaser may desire to modify, delete or amplify sections of this standard.” Another preface is worded: “Standards are not intended to inhibit purchasers or producers from purchasing or producing products made to specifications other than those of API.” Appropriately, the API expressly declines any liability or responsibility for loss or damage resulting from the use of its standards, or for the violation of any regulation with which the standards publication may conflict.

Reliability and safety professionals have to manage risks effectively, and standards can indeed be extremely helpful. Testing, however, has shown beyond all doubt that not all standards are harmonious and noncontradictory. Invoking an inconsistent standard will not sufficiently mitigate risk. When it comes to sealing hazardous, flammable and/or toxic products, engineers should understand that just because a seal meets the present levels that it does not mean this seal will perform safely in service. The recommended action would be for the test loop volume in the API/ISO protocol for mechanical seal air tests to be changed to 1.5 l. This recommended change is reflected in the first point of TABLE 5.

We are encouraged by reliability professionals on several continents who are now taking tangible steps to reduce mechanicalseal failure risks. Also, because of the weakness resulting from the presently allowed large-volume protocol of TABLE 1, at least one major mechanical-seal manufacturer is now voluntarily

1,000 kg (2,200 lb) or 7 bbl

TABLE 5. Proposed key items of API/ISO protocol for mechanical-seal air tests Each sealing section shall be independently pressurized with clean air to a gauge pressure of 0.17 MPa (1.7 bar or 25 psi). The volume of each test setup shall be a maximum of 1.5 l.

Isolate the test setup from the pressurizing source and maintain the pressure for at least 5 minutes.

Maximum pressure drop during the test shall be 0.014 MPa (0.14 bar or 2 psi).

subjecting its API 682-qualified seals to the far more stringent “low-volume” air-testing described earlier. TABLE 5 restates the “low-volume” air-testing protocol. Finally, users can reduce the risk of installing flawed mechanical seals and, thereby, increase safety and reliability by specifying, buying and installing only seals with initial liquid leakage rates not exceeding 5.6 g/hr.

NOTES * Footnote: A demonstration video, which includes footage of each stage of testing detailed in this article, has been released into the public domain. The author actively encourages interested readers and safety-focused seal users to review the readily available video: http://www.sealsuccess.com/api-682-mechanical-seals-leaking/.

1

LITERATURE CITED Bloch, K. and B. Jung, “The Bhopal Disaster: Understanding the impact of unreliable machinery,” Hydrocarbon Processing, June 2012.

STEPHEN SHAW, CEng, FIMechE, CMOSH, is a chartered engineer and a Fellow of the Institution of Mechanical Engineers. In addition, he is a Chartered Safety and Health Practitioner. Since 2008, he has been the chairman of AESSEAL plc. He was appointed group engineering director of AES Engineering Ltd., in 2011.

Hydrocarbon Processing | MAY 2013 59

SEPARATION ANXIETY? C O ME T O B O O T H 10 0 5 AT O T C T O L E A R N H O W T HI S S K ID D E D S E PA R AT O R C A N HE L P!

You can relax.

Pentair Separation Systems designs and manufactures advanced technologies for the high performance separation of solids, liquids and gases.

Dramatically decrease or eliminate contamination like: ɒ )RXOLQJ ɒ )RDPLQJ ɒ &RUURVLRQ ɒ 6KXW GRZQV

Reduce your anxiety.

Increase your efficiency.

SEPARATION SYSTEMS

(936) 788-1000 www.pentairseparations.com Select 64 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability A. GLAUN and J. SHAHDA, GE Oil & Gas, Avon, Massachusetts

Prevent methane hydrate formation in natural gas valves Gas flow across a control valve is considered a classic “throttling” process that is defined by energy not being added or extracted from the process gas as it traverses the valve.

Therefore, total enthalpy is preserved, entropy increases and the process is thermodynamically irreversible.

The consequences of this process are that many real gases experience a drop in temperature while following the constant enthalpy line as the pressure drops across the valve. This effect was first described by William Thomson and James Joule, and it now bears their names. The JouleThomson effect is leveraged in the production of cryogenic fluids such as liquid oxygen, nitrogen and argon, and it is the principle of operation behind most air conditioners and refrigeration in use today.

Natural gas production, storage and transmission usually take place close to ambient conditions, where a small change in temperature can induce the formation of methane clathrates (hydrates). Once formed, methane hydrates can block valves, fittings and pipelines. Newer facilities are using higher transmission pressures, causing the temperature inside the valve to approach or drop below 0°C, with the risk of icing on the outside of the valve.

The discussion here focuses on the thermodynamics involved and on the requirements for a successful natural gas valve application in which the incidences of hydrate formation and icing of the valve are reduced. Computational fluid dynamics (CFD) studies are also presented showing the JouleThomson effect in a real-world valve application.

How do methane hydrates form? Hydrates form in natu-

ral gas pipelines when the local fluid temperature drops below the hydrate-formation temperature at a specific pressure. This temperature drop can occur when the natural gas flows through a control valve, or when gas travels through transmission pipelines under cold ambient conditions or through any other piece of process equipment where the flow is restricted or accelerated in such an orifice plate. This phenomenon of temperature drop with pressure drop in a real gas is known as the Joule-Thomson effect. Note: Hydrates can form at temperatures well above the freezing point of water (FIG. 1).

Hydrate-formation temperature is difficult to predict and is the subject of many academic papers. Prediction depends on temperature and pressure, water concentration and the composition of the natural gas, where small concentrations of heavy hydrocarbons and other gases such as O2 , N2 , H2 S and CO2 can affect the formation temperature. Software programs are available to help the user predict the formation temperature, but the only way to know for certain is to test a sample of the gas in question.

How is the gas temperature drop calculated? Flow across a control valve is considered a throttling, constant enthalpy (is40 35

Methane clathrate, stable

30

drates (also known as methane ice) are crystalline water icelike particles, where methane molecules are trapped inside hydrogen-bonded water molecules. Under the right conditions of pressure and temperature, these form semi-solid particles that tend to agglomerate, building up inside pipelines, valves and other process equipment.

Why worry about methane hydrates in valves? Hydrate ice particles may clog flow passages in control valves and, in particular, valves with noise attenuation trim (small drilledhole cages, labyrinth passage stacks, etc.). This sometimes causes a major reduction in the flow across the valve, badly affecting system operation. Severe hydrate formation may even clog large passages of the valve body and pipeline.

Pressure, MPa

25

What are methane hydrates? Natural gas/methane hy-

20 15 10 5 0 175

200

225

250 Temperature, K

275

300

325

FIG. 1. Stability curve showing that methane hydrate is stable at 0.1 MPa (1 bar) if temperatures are low enough, and that it is stable far above the melting point of ice (H2O) if pressures are high enough.

Data courtesy of Lawrence Livermore National Laboratory.

Hydrocarbon Processing | MAY 2013 61

Maintenance and Reliability enthalpic) process. This implies that the process occurs over a very short period, making it adiabatic (no heat is lost or gained during the process); enthalpy is preserved, and the process is irreversible (i.e., entropy increases and cannot be recovered).

For a real gas flowing through a control valve, this process gives a lower downstream temperature. Additional lowering of the downstream temperature may occur due to high downstream velocity of the expanded gas (Eq. 1). For natural gas and reasonable downstream velocities of less than 0.3 Mach number (Ma), the velocity terms in Eq. 1 are two orders of magnitude smaller than the enthalpy and can usually be ignored.

V12 V2 = h2 + 2 (1) 2 2 Where: h = Specifc enthalpy V = Fluid velocity 1, 2 = Upstream and downstream conditions, respectively.

Two common methods exist to calculate the temperature drop of natural gas for a given pressure drop across the valve.

The first method is to determine the enthalpy at the inlet presh1 +

Methane throttling process 200

350

Con stan t en thal py

330 Temperature, K

140

120

1

340

320 310

nt sta Con

290

80

2

Hydrate form ation

280

100 90

bar re, u s s pre

B

300

270 3.5

160

line

A 3.7

3.9

Entropy, kJ/kg

4.1

4.3

sure and temperature and then to determine the outlet temperature at the same enthalpy and outlet pressure. Software programs and web-based calculators can give this data, but the Mollier chart for methane can also be used, assuming an isenthalpic process in the valve from Eq. 1.

A Mollier chart, at minimum, displays properties of pressure, temperature, enthalpy and entropy on one diagram, allowing the user to define a state using only two properties and reading off the other properties (FIGS. 2–4). By definition, this is an accurate method of determining the downstream temperature; it is only limited by the accuracy of the Mollier chart and by the user’s ability to graphically interpolate the chart. Using software may be more precise, but the authors believe that a Mollier chart gives the user a visual sense of how the values are changing and leads to a better understanding of the thermodynamics.

After determining the inlet condition on the chart, the user follows the lines of constant enthalpy until the downstream pressure line is reached. The temperature now can be read at this new position. The caveats to this method are that the assumption of constant enthalpy is just that—an assumption. In reality, there is some heat transfer across the valve/pipe boundary, and the process is never precisely a true throttling process.

These “inefficiencies” will result in lower temperatures than the ideal determined above.

The second method is a general rule used in the natural gas industry where, for every 100-psi pressure drop, there is a corresponding 7°F temperature drop; however, this rule is limited to a maximum valve inlet pressure of 1,000 psi. Using the Mollier chart for methane at room temperature, the accuracy of this rule can be evaluated. It varies from 5.5°F/100 psi for inlet pressures of approximately 300 psi, to 6°F/100 psi for inlet pressures of approximately 1,000 psi.

The rule takes into account inefficiencies and is somewhat conservative. However, for high inlet pressures and small pressure drops, the rule is very conservative. For example, from a

4.5

FIG. 2. Temperature drop inside a single-stage trim valve (Line A) and a multi-stage trim valve (Line B).

FIG. 3. Single-stage contoured plug valve (Line A).

62 MAY 2013 | HydrocarbonProcessing.com

FIG. 4. Multi-stage, expanding-area trim valve (Line B).

Maintenance and Reliability

Pressure, psi

drop of 1,000 psi to 800 psi, the temperature drop is 4.5°F/100 Valve icing. Under high pressure-drop conditions, the outlet psi per the Mollier chart. temperature in the valve may fall below the freezing point of The temperature downstream of the valve is a concern, but water. This may not cause hydrate formation inside the valve the lowest temperature inside the valve trim must also be calculated. In fact, the pressure and temperature Hydrate-formation temperature is difficult at the trim vena contracta (smallest area of flow in the trim) are usually lower than the pressure and temperato predict and is the subject of many ture downstream of the valve. The lower temperatures academic papers. Software programs are inside the valve are due to sudden accelerations of the available to help the user predict the gas inside the valve trim and typically are thermodynamically isentropic (reversible) in nature. formation temperature, but the only way These internal pressure drops can be very large for to know for certain is to test a sample single-orifice valves and less so for multi-stage control valve trims that drop the pressure over a number of of the gas in question. controlled pressure-drop stages; they can be visualized if the reader follows the constant entropy line on the Mollier chart. Fortunately, this vena contracta low temperabecause inhibitors such as monoethylene glycol (MEG) can be ture is not a permanent change of state, and the temperature due used with gas at −10°C. Even so, condensation and freezing on to this effect recovers after the gas passes through the valve trim, the outside of the valve body and pipeline can have serious efleaving only the vena contracta and adjacent areas cold. fects. For example, coastal gas fields on the Saudi Arabian peninsula are notoriously humid and prone to ice buildup.

Extremely thick layers of ice can build up, preventing access How can hydrate formation be avoided? There are several to the valve body or pipe wall. These layers of ice can add sigsolutions to reduce the incidence of hydrate formation in valves: nificant weight to the valve and pipeline, with the possibility of 1. Appropriate inlet temperature. The natural gas inlet structural and/or vibration problems. The valve bonnet may temperature can be chosen so that, when the pressure drops become iced, thus seriously impacting the valve stem packing across the valve, the resulting downstream temperature of the and raising the potential for leakage. natural gas is always above the hydrate-formation temperature.

Gas temperatures are normally determined by the gas field, so external heating of the inlet gas prior to entering the valve A real-world problem. A natural gas producer was flowing gas may be the only option. This is, however, expensive in terms of through a control valve with the following winter conditions: heating equipment and fuel costs. The required inlet tempera• Upstream: 975 psia, at 57°F ture can be determined using the Mollier chart or the 7°F/100 • Downstream: 180 psia, with icing on the valve and pipe. psi rule, by starting at a known safe outlet temperature and The icing was unacceptable to the plant operator, and the then working backward. only line heaters available were rated at 350 psia and could not 2. Inhibitor injection. Inhibitors can be injected upstream be used to heat the inlet gas. For the purposes of this example, of the control valve to prevent the gas from reaching the hydratethe natural gas is assumed to be methane. formation temperature, thereby preventing the formation of hyAn isenthalpic analysis showed that the downstream temperadrates. The most common inhibitors are methanol and ethylene ture reaches 14°F (FIG. 5, points 1–2). Using a gas industry genglycol; these typically can be recovered from the gas and recireral rule, the downstream temperature could reach 1.3°F (FIG. 5, culated. However, inhibitor injection and recovery can be costly. points 1–3). The end user required that the downstream temper3. Valve trim design. As noted earlier, even if the valve ature be no less than 40°F to prevent icing and hydrate formation. outlet is above the hydrate-formation temperature, the internal valve trim temperature may not be, and hydrate formation M Heat input 59 kJ/kg 1,200 tem in. ga is possible within the valve. If this is the case, then selecting a p. 5 s 7°F multi-stage valve that gradually lowers the pressure across the 1,000 Inlet pressure 1 5 6 valve trim will help the situation. Note: Trim selection cannot prevent downstream hydrate formation if the downstream 100°F 40°F 800 temperature is below the hydrate-formation temperature. The Con 20°F stan Joule-Thomson effect is a “state” condition from upstream to t te 600 60°F mp . downstream, and changing the valve trim will not affect this.

80°F 0°F The red lines in FIG. 2 show the properties of methane as 400 the fluid travels through the control valve from upstream (1) to Original–isenthalpic Original–7°F/100 psi downstream (2). The long, dashed red line labeled “A” repre200 Solution 1–isenthalpic 2 4 Outlet pressure 3 sents a single-stage control valve where the temperature drops Solution 1–7°F/100 psi 0 below the hydrate-formation line (blue line), making it pos770 790 810 830 850 870 890 sible for hydrates to form inside the trim. The dotted red line Enthalpy, kJ/kg labeled “B” represents a multi-stage control valve where the FIG. 5. Heating required at inlet pressure to keep outlet temperature temperature does not drop below the hydrate-formation line, above 40°F. thus preventing hydrates from forming inside the trim.

Hydrocarbon Processing | MAY 2013 63

Maintenance and Reliability Solution 1: Heat gas at the valve inlet. A common solution to this type of icing problem is to use pipeline heaters just upstream of the valve. The fuel for the heaters is usually the flowing natural gas itself. However, this solution is costly in terms of lost gas and the expense of high-pressure heaters.

Referring to FIG. 5 and using the 40°F minimum outlet temperature requirement (dashed blue line), point 4 can be located and the temperature can be back-calculated to maintain 40°F at the outlet of the valve. The gas inlet temperature should be above 95°F (point 6) to avoid falling below 40°F at the outlet.

Using an isenthalpic analysis, the inlet temperature should be above 84°F (point 5) to avoid falling below 40°F at the outlet.

The difference between points 5 and 6 is quite substantial, and, as discussed earlier, the authors believe that, for high pressures, the 7°F/100-psi rule is overly conservative. The isenthalpic analysis is ideal for this measurement; the true value lies somewhere in between.

Note: The addition of hydrate inhibitors might lower the end user’s specification of 40°F minimum temperature at the outlet of the valve. In this case, point 4 would move to a lower value to the left and the analysis would be repeated, thereby lowering the minimum required inlet temperature to prevent hydrate formation.

The rate of energy input required to heat the gas can be read directly from FIG. 5 by subtracting the enthalpy at point 6 from the enthalpy at point 1. If this value is multiplied by the mass flowrate in kg/s, then the answer is the rate of energy input in kJ/s or kW. As mentioned before, this analysis is independent of the type of trim in the valve. If the analysis shows that the temperature at the outlet is low enough to form hydrates, then changing to a multi-turn or multi-stage trim will not alter the conditions at the outlet. A multi-stage valve will, however, limit very low temperatures inside the valve trim.

Solution 2: Use available low-pressure heaters. The first option considered was to save the customer from having to buy new equipment by using the existing, 350 psia-rated line heaters (FIG. 6). This method required staging the pressure drop by placing another valve in the line. The first pressure drop occurred from 975 psia to the heater maximum pressure of 350 psia, and then down to the outlet pressure of 180 psia.

In the methodology of this solution, the outlet drop should not fall below 40°F, which allows point 4 to be located. An isM tem in. ga p. 5 s 7°F Inlet pressure 1,000

600 400

1

100°F

80°F

3 5

100°F 5 3

2

80°F 60°F

830 850 Enthalpy, kJ/kg

870

FIG. 6. Minimum inlet temperature when using the existing low-pressure line heaters.

64 MAY 2013 | HydrocarbonProcessing.com

600

40°F 732 psi Con 20°F stan 6 t te mp .

0°F

1

200

4 810

800

400

2

Heater max. pressure 200

Outlet pressure 0 770 790

6

60°F

40°F

C 20°F onstan t te mp .

0°F

Heat input, 56 kJ/kg

1,200

Pressure, psi

Pressure, psi

800

M tem in. ga p. 5 s 7°F Inlet pressure 1,000

Heat input 16 kJ/kg

1,200

enthalpic analysis is used to back up to the heater pressure of 350 psia, which gives points 3 and 5, respectively. Point 2 is located on the 40°F minimum line, and an isenthalpic analysis is used to back up to the inlet pressure of 975 psia, giving points 1 and 6, respectively. The heat input is calculated from points 2–5. The addition of heat at 350°F reduces the minimum inlet temperature to 84°F from 95°F, with no heat addition.

Note: As with solution 1, the addition of hydrate inhibitors might lower the end user’s specification of 40°F minimum at the outlet of the valve. In this case, point 4 would move to a lower value and the analysis would be repeated, thereby lowering the minimum required inlet temperature to prevent hydrate formation.

Solution 3: Apply new low-pressure heaters. If the answers from the first two solutions are inadequate, the next step is to examine the lowest-pressure-rated line heaters that can be used and still operate year-round at the minimum inlet temperature of 57°F. This requires a slightly different methodology than that used previously.

Referring to FIG. 7 and starting at the minimum inlet temperature at point 1, the pressure must then be determined for when 40°F is reached. This gives point 6, which is at 732 psia.

Knowing that the endpoint is point 4, one can work backwards, using isenthalpic analysis, to arrive at point 5. The enthalpy difference between points 5 and 6 is the resulting heat input required. Comparing the result of solution 3 to solution 1, a small reduction in heat input is required. Note that the heat input found when using the general rule is identical to that found when using the isenthalpic analysis.

At this point, it becomes a question for the end user of economics and complexity. Solution 1 appears to be less complex, since it requires only one control valve; however, a large pressure drop across one valve results in a severe service application with low internal valve trim temperatures, possibly requiring an expensive multi-turn or multi-stage valve. High-rated pressure-line heaters also must be purchased, and significant heat must be added to the upstream gas.

Solution 2 does not appear to be useful since an additional valve would need to be added to the line to accommodate the pressure drop from 350 psia to the outlet pressure of 180 psia, and the system would not be able to run unless the ambient

890

Outlet pressure 0 770 790

4 810

830 850 Enthalpy, kJ/kg

870

890

FIG. 7. Heat input at lowest line heater pressure for preventing hydrates at minimum inlet temperature.

Maintenance and Reliability (inlet) temperature reached 84°F. However, there are some geographical locations where this might not be such a burden.

Solution 3 requires the complexity of an additional valve, but the pressure drop is broken up into two reasonable steps, resulting in two less severe applications and warmer internal valve trim temperatures. Lower-rated pressure-line heaters would need to be purchased, and significant heat would need to be added to the upstream gas.

Numerical analysis for valve trim temperature. CFD can be used to model the flow through the trim of the valve.

An accurate CFD analysis to capture Joule-Thomson effects is only possible if advanced real gas formulations are used. A real gas model takes into account non-ideal compressibility effects, whereas an ideal gas CFD analysis will only predict localized drops in temperature resulting from increases in velocity and reductions in local pressure due to the acceleration of the fluid as it negotiates turns in the valve trim.

The proprietary CFD program used in this case has a real gas model that uses the Redlich-Kwong formulation to predict the fluid properties, taking into account the non-ideal compressibility of the working fluid. The program predicted 54°F at the outlet of the trim, which compares favorably to an isenthalpic analysis using a Mollier chart, which predicts 54.5°F. FIG. 8 shows a representative 22-turn trim (pictured is a half-symmetry model of one flow channel) with an inlet

pressure of 975 psia and an outlet pressure of 180 psia. The results show the even, gradual, staged pressure drop through the valve trim. This style of trim serves two main purposes: One is to lower the outlet jet Ma, producing a quiet valve, and the other is to reduce the temperature drop inside the valve trim to minimize hydrate formation and icing.

FIG. 9 shows the temperature results. The plot clearly shows the Joule-Thomson effect of a permanent temperature drop from inlet to outlet. It also shows areas inside the valve trim

FIG. 8. CFD pressure plot of a representative 22-turn, multi-stage valve trim.

Oil Mist Lubrication. Worldwide.

Need oil mist lubrication expertise?

New Systems ᔢ Accessories ᔢ Installation ᔢ Repairs ᔢ Support ᔢ

For any oil mist application – in any location – we’ve got you covered.

+1 713-682-3651 TFHudgins.com sales@tfhudgins.com ©T.F. Hudgins, Incorporated. All rights reserved.

Select 166 at www.HydrocarbonProcessing.com/RS

Hydrocarbon Processing | MAY 2013 65

Partner with the Best

Maintenance and Reliability

Visit us at OTC Houston 6 to 9 May 2013 Hall B, Booth 4717-5

FIG. 9. CFD temperature plot of a representative 22-turn, multi-stage valve trim. A real gas solver formulation allows for the solving of Joule-Thomson effects.

where the temperature can drop below the outlet temperature, although areas are localized and the temperature recovers.

Even accepted global standards for valve sizing, such as IEC 60534-2-1, do not take this real gas effect into account and base the sizing exclusively on upstream temperature, assuming an ideal gas where interstage and downstream temperature equals the upstream temperature. (Note: IEC 60534-2-1 does warn that compressibility of real gases should be taken into account if an accurate upstream density is to be calculated.)

IEC control valve noise prediction standard 60543-8-3 explicitly states in its scope statement that ideal gas laws are assumed, and it uses the upstream temperature to determine downstream density, velocity and Ma. For this specific problem, the downstream velocity can be under-predicted by 8%.

With over 50 independent subsidiaries and more than 220 engineering and sales offices spread across the world, SAMSON ensures the safety and environmental compatibility of your plants on any continent.

To offer the full range of high-quality control equipment used in industrial processes, SAMSON has brought together highly specialized companies to form the SAMSON GROUP.

A01120EN

Select 167 at www.HydrocarbonProcessing.com/RS

SAMSON AG · MESS- UND REGELTECHNIK Weismüllerstraße 3 60314 Frankfurt am Main · Germany Phone: +49 69 4009-0 · Fax: +49 69 4009-1507 E-mail: samson@samson.de · www.samson.de SAMSON · www.samsongroup.net 66 GROUP MAY 2013 | HydrocarbonProcessing.com

Takeaway. Hydrate formation and icing in natural gas pipelines and valves can be greatly reduced or even prevented entirely if a detailed study of the thermodynamics of the system is undertaken. An intimate knowledge of the process gas is essential so that properties, such as hydrate-formation temperature, can be accurately determined. Also, using real gas analysis, internal valve trim temperatures can be calculated, leading to a better understanding of the type of valve trim required to inhibit hydrate formation.

ASHER GLAUN is a senior engineer and technologist for Masoneilan Control Valves at GE Oil & Gas. He has worked in the control valve industry for over 12 years. Prior to his work with GE Oil & Gas, Mr. Glaun was employed for 11 years at Bird Machine Co. in the design of high-speed centrifuges. His work at GE involves leading new technology development specializing in fluid dynamics, CFD, structural analysis/FEA and valve acoustics.

Mr. Glaun graduated with a BSc degree in mechanical engineering from the University of Cape Town, South Africa, and he obtained an MS degree in mechanical engineering from Northeastern University in Boston, Massachusetts.

JOSEPH SHAHDA is a senior applications engineer for Masoneilan Control Valves at GE Oil & Gas. He has over 16 years of experience in the control valve industry, with a focus on applications engineering and delivering control valves solutions to customers worldwide. Mr. Shahda holds an MS degree in mechanical engineering from Northeastern University in Boston, Massachusetts.

Bringing energy and the environment into harmony.®

RELIABILITY ISSUES NEVER SLEEP. WE MAKE SURE YOU CAN. At Dresser-Rand, we know any unscheduled interruption of a client’s downstream operation is simply unacceptable. That’s why we make engineering reliability a primary focus. From one of the world’s largest hydrogen compressors (pictured at right) to our single-stage steam turbines, Dresser-Rand downstream solutions are easy to maintain. But for us, it’s not just about selling superior products, it’s about providing peace of mind as well. So, when you do need help, rest assured we’re there for you.

Visit us at OTC2013, Reliant Arena Hall, Booth 9659, Houston, Texas, May 6-9 The Americas: (Int’l +1) 713-354-6100 / EMEA: (Int’l +33) 2-35-25-5225 Asia-Pacific: (Int’l +60) 3-2093-6633 / info@dresser-rand.com Compressors–Turbo & Recip / Steam Turbines / Gas Turbines / Engines / Control Systems / Expanders Select 62 at www.HydrocarbonProcessing.com/RS

» www.dresser-rand.com

Select 67 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability H. P. BLOCH, PE, Consulting Engineer, Westminster, Colorado

Consider both actual and virtual spare parts inventory Globally, hydrocarbon processing industry (HPI) facilities operate hundreds of thousands of fluid machinery assets. The various service-specific internal components of these assets are supported by many millions of spare parts, all of which must be properly preserved and catalogued.

Operating companies and consultants often consider developing systems to unify or standardize spares maintained in plant stores or inventories. Standardization initiatives can provide economic benefits. Proper and unified part identification efforts usually precede standardization of the spare parts inventory. Both identification and standardization can help reduce the number of spare part units held in inventory and reduce plant costs. Also, access to spares will be facilitated in instances where central part depots are involved.

Dual-purpose units. Some spares could be manufactured with “multi-purposes” in mind. Multi-purpose parts could be produced with oversize dimensions, which, upon finishmachining, will suit the specific requirements of a particular machine. Certain bore dimensions (for example, in coupling and impeller hubs) could be stocked with undersized and prebored concentricity-accurate pilot holes. On an as-needed basis, these bores could be finish-machined to the precise dimension on site. A single maximum diameter impeller, as shown in FIG. 1, could be the standardized spare part for several pumps. This impeller could be trimmed for use in a variety of pumps presently operating at a particular site and used in different pumping services. Similarly, an oversized shaft could be placed in the storeroom for future finish-machining and used on a range of similar pumps or other fluid machines.

Consider everything. The concepts of shared capital spares and/or requiring the vendor to stock spare parts are not new.1 Both concepts—shared spares and vendor stocking—warrant more investigation. Major turbo-machinery (fluid machinery) rotors can be jointly owned by a group of potential users, call it an “ownership pool.” The financial details for being able to use and replace co-owned spares should be well defined in advance as part of the shared ownership pool. Vendor stocking arrangements place the vendor under contractual obligation to have certain machines or parts available on very short notice. In return, the potential user of these parts or machines accepts the contractual obligation to purchase vendor-stocked assets at a predefined premium cost.

Case for stocking only upgraded parts. Many spare parts

kept in storage at HPI facilities belong to fluid machines that are not optimized in terms of hydraulic efficiency, permissible range of operation, or component reliability. For decades, capturing efficiency gains and avoiding equipment failures through proper upgrading have been the priority concerns of industry. Moreover, maintenance-cost reductions have been achieved through selective upgrading of existing parts and machinery. This upgrading is usually planned by the user or equipment owner, and such efforts receive considerable input from competent vendors. Cost justifications are developed jointly and, if the results warrant, are implemented during the next repair or scheduled downtime.

Whenever upgrading is done, the issue of spare parts adequacy should be investigated. Fortunately, as of 2013, there are truly advanced options that go well beyond simply buying new physical spare parts. These options should be investigated and evaluated by forward-thinking of the plant and equipment operation.

Such exercises would engage assistance from design engineers or consulting contractors. Before standardizing old, existing parts, consider if these parts will be used again. Gain input on this question before moving forward. Understand that the future of the spare parts business is a new direction for HPI facilities.

Prioritize candidate machines for upgrading. A facil-

ity’s or corporation’s list of failure frequencies is one primary way to determine where future upgrading will be most costeffective. These failure records often reveal weak links or repeat failures that should be examined in greater detail. Risk-prone components deserve to be upgraded whenever life-cycle costs (LCCs) are favorable.

There are many reasons why upgraded components for fluid machinery will often differ from the old original parts.

Modern or state-of-the-art designs may incorporate advanced metallurgies or adjusted hardness of existing metallurgical compositions. Hardness adjustments are often based on wear amounts observed in the fluid environment of a particular process unit. Actual hydraulic and wear-related performance data are needed to impart value to a new spare part. These data are supplied either by site engineers or a designated, experienced, consulting entity.

There can be desirable changes to a more favorable ductility of ferrous parts, with upgrade opportunities using advanced stainless steel compositions. These changes are particularly advantageous in oil sand projects and other apHydrocarbon Processing | MAY 2013 69

Maintenance and Reliability plications where abrasive wear is prevalent. Upgrading to a more suitable impeller (FIG. 1) has sometimes allowed shutting down one pump in fluid loops which previously had two pumps operating in parallel.

Fluid machines older than 20 years (especially, process pumps) are in the top tier of candidates for upgrades. One proceeds by studying such operating parameters as head vs. flow at best efficiency performance (BEP), avoiding excessive loads on bearings; transient conditions at startup and shutdown, and the establishment of best sealing alternatives. These projects will include flush plans selected for seal reliability and for better power efficiency. The work requires involvement of experienced personnel. Whoever is given the upgrade task should receive the full support of plant engineering personnel.

In well-managed corporations with extensive computerized maintenance management systems (CMMS), one would first investigate the extent to which relevant failure data are avail-

D'

FIG. 1. A single impeller with diameter, D, could be site-trimmed to the more exact diameter, Dˇ. Doing so would allow it to suit the needs of a particular operating point or pumping service.

Note: The side plates will remain at the maximum diameter for maximum preservation of energy (high efficiency).

D

able from these systems. As-built data sheets are sometimes available from a user’s own central engineering group. In all instances, the as-running or operational performance of fluid machines must be ascertained before standardization studies can be expected to yield maximum benefit.

Partner with advanced reverse-engineering capabilities.

Competent upgrade vendors view every repair event as an opportunity to upgrade. These vendors should be considered for partnership discussions. The best upgrade vendors use contour mapping and measuring machines for reverse-engineering of existing parts. Their expertise is important. One reference claimed to have found 14% of in-stock spares at a petrochemical plant to be unusable.1 Not to be outdone, the machine shop superintendent of a major US oil refinery claimed that 30% of the spare parts at his location were incorrectly dimensioned or simply unserviceable for a variety of reasons.

When working with a competent upgrade company’s nearest vendor shop, the shop, regardless of location, will be supported by the upgrade company’s home engineering organization. Also, the best-equipped and most promising upgrade providers have highly modern fluid-machinery testing facilities, as shown in FIG. 2.2 The local shop will usually start its work with automated contour mapping of the parts to be replicated or improved. All dimensions are stored in a computer. However, based on the possibility of dealing with deficient parts, the upgrade company will use its considerable experience and judgment to monitor the accuracy of these parts.

Simply reverse-engineering without further establishing the correctness of such efforts could replicate certain vulnerabilities. Impeller or blade contour adjustment studies are made on some parts and must be catalogued. Available “tweaking options” are sometimes evident only to the best upgrade providers. The LCC assessment of upgraded parts becomes a key ingredient of further decision-making.

While equipment owners may wish to select a particular vendor as the single-source supplier, this upgrade provider will have to demonstrate full competence in understanding many different machines, models, brands and configurations. But this is not where the provider’s demonstration of competence ends; it is, rather, where it begins.

The 21st century environment demands technology-optimized methods and procedures. In this environment, an upgrade provider must demonstrate the ability to create a virtual spare parts storehouse. Actual physical spare inventories are used in limited cases; these are situations where they still make economic sense.

Advantages with a virtual inventory. In the future, spares

FIG. 2. Partial view of a modern pump test stand with variable speed drives. Source: Hydro, Inc., Chicago, Illinois.

70 MAY 2013 | HydrocarbonProcessing.com

parts will not be sitting in an HPI facility’s warehouse. The future is in “make it as you need it.” Major multinational equipment upgrade companies use best-available technology throughout.

In general, these upgrade experts “paint” existing parts with a scanner. A three-dimensional (3D) model is constructed from the scanned data. The work flow goes from an electronic library (a virtual inventory) to a printed sand-mold process in which no pattern making is involved. The entire process can take only a fraction of the total time that it would have needed for conventional methods.

Maintenance and Reliability The actual process uses 3D printer technology to print a sand mold directly onto a print bed. No drawings are created, and this true state-of-the-art process eliminates the need for hard tooling. The process also supports geometries that were once restricted to production-quantity investment castings.

Even complex cores and undercuts are now readily obtained with modern 3D printer technology. The process is obviously ideal for spare parts that, by definition, are needed in small quantities. Surface finishes and mechanical properties are virtually identical to traditional sand castings, and tolerances are superior to many traditional processes. Sand molds can often be produced in less than a single week. So, if a pump designated as your â&#x20AC;&#x153;Aâ&#x20AC;? pump needed repairs after 20 months of operationâ&#x20AC;&#x201D; the mean-time between failure (MTBF)â&#x20AC;&#x201D;and it took 20 days before it could be properly repaired, realize that there is a 97% probability that the â&#x20AC;&#x153;Bâ&#x20AC;? pump will operate flawlessly during those 20 days. Chances are that management is willing to take the 3% probability or risk an outage that exists for the parallel or â&#x20AC;&#x153;spareâ&#x20AC;? pump. A rapid repair with an old part would also take time, and result in â&#x20AC;&#x153;old,â&#x20AC;? inefficient or failure-prone operation with parts that really deserved to be upgraded.

Final considerations. Consider fluid-machinery capital

spare pooling and vendor stocking programs. Both are considerations that deal with standardization and cost savings.

Spare-parts cataloguing should, initially, concentrate on highvalue or repeat-failing parts; upgrading should be part of the

standardization efforts. Electronic libraries and virtual electronic inventories exist today at some forward-looking locations. They are the future.

While creating a virtual parts inventory of upgraded parts can be a massive challenge, accomplishing the task will be worth it. The ultimate savings will be huge. Enlist a competent equipment upgrade company, one with international presence. Pick a group with the most modern tools and with equipment testing capability.2 Allocate funds for a demonstration project. Work with a consulting company; with their input, let the upgrade company produce the upgraded components wherever cost justifications can be shown. Work toward a target return on investment (ROI). If only a somewhat lesser ROI is achieved, then there will still be improvement.

So, to get things off to a manageable start, identify possibly 10 repeat-failure or low-MTBF process pumps. Ask the consulting company and parts provider to actually produce the upgraded components wherever cost justifications can be shown.

Report on tangible achievements and then expand your base to include more assets. Do not make the mistake of making standardization an unwieldy open-ended task. Closely scrutinize the process, as illustrated in FIG. 3, and, if needed, use it to overcome an operations departmentâ&#x20AC;&#x2122;s objections. The operations department may be averse to running a process pump without having a spare (or standby) pump sitting on the adjacent foundation. Explain to the operations staff that there may be incre-

EFFICIENCY MAT TERS. RESTORE ONLINE STATUS QUICKLY AND SAFELY.

Managing turnarounds on time and on budget can present many challenges.

Cudd Energy Services helps you meet these challenges head on. Our Ă&#x20AC;HHW RI SXPSLQJ WUDQVSRUW DQG VWRUDJH YHVVHOV DFFRPPRGDWHV D ZLGH UDQJH RI Ă&#x20AC;RZ UDWHV IRU +3+7 RSHQ Ă&#x20AC;DPH HQYLURQPHQWV WKDW JHW \RX EDFN RQOLQH VDIHO\ DQG HIÂżFLHQWO\ :LWK D PLOOLRQ &6) FDSDFLW\ WKH 4XHHQ 6WRUDJH WDQN UHGXFHV IUHTXHQW GHOLYHULHV WKDW FDXVH FRQJHVWLRQ DQG FDQ EH VDIHO\ UHSOHQLVKHG ZLWKRXW LQWHUUXSWLQJ SXPSLQJ RSHUDWLRQV VDYLQJ \RX WLPH DQG PRQH\ (TXLSSHG ZLWK HPHUJHQF\ VKXW GRZQ GHYLFHV WKH 'XDO 0RGH 3XPS IHDWXUHV D KHDW UHFRYHU\ V\VWHP WKDW UHGXFHV IXHO FRVWV DQG LWV (3$ 7LHU & $ 5 % emission rating helps reduce emissions.

7UXVW \RXU WXUQDURXQG WR D SURYHQ OHDGHU &KDQQHOYLHZ 7;

WWW.CUDD.COM

PROVEN EXPERIENCE. TRUSTED RESULTS.

Select 168 at www.HydrocarbonProcessing.com/RS

Hydrocarbon Processingâ&#x20AC;&#x201A;|â&#x20AC;&#x201A;MAY 2013 71

100

50

Repair time or days unavailable

The Quiet Work-Horse!

Maintenance and Reliability

10 7 5 4

R=

95%

R=

97%

R=

%

98

R=

.5% 99% R=

98

R=

.5% 99

R=

.7% 99

2 A MASI ACHE g in Beij 2013 . May 13.–16 h E18 Boot

1

1

2

3

15 20 30 4 5 7 10 Mean-time-between-repairs (MTBR), months

40 50

100

FIG. 3. The probability that a “spare” pump with an MTBR of 15.5 months will operate flawlessly in the 7 days it takes to repair the “main” pump will be 98.5%.3

ECOTROL® control valve Advantages, that should not be kept quiet! ● High reliability guaranteed by precision manufacturing processes and quality control ● Emission control and leakage conforming to the highest international standards ● Tubeless, integrated mounting of positioners acc. to VDI 3847 ● Minimal life cycle cost ● A range of awarded patents Take advantage of the most technically innovative control valve in a generation, up to DN 400 (16")!

The «State of the Art» solution!

ARCA Regler GmbH, D-47913 Tönisvorst Phone +49-2156-7709-0, Fax …-55, sale@arca-valve.com

www.arca-valve.com

72

15

3

Request Technical Information or a Quotation:

ARCA Flow Group worldwide:

20

• Competence in valves, pumps & cryogenics • Subsidiaries and partners in Switzerland, the Netherlands, India, P.R. China, Korea and Mexico!

Select 169 at www.HydrocarbonProcessing.com/RS

mental time needed for components to make their way from an electronic library to becoming physical parts. However, depending on the MTBF and days of unavailability because the spare pump is out for repair, the probability of the presently operating pump failing in that time may be acceptably low.

The next phase in the standardization and upgrading process, involves quantifying the deliverables. Perhaps an experienced consulting company, in conjunction with the upgrade provider could commit to a full review of a given number of machines per month. Throughout the process of physically upgrading components or producing a virtual inventory, remember that the quality of any job or task will only be as good as the people involved. We advocate involving experienced and well-motivated individuals in the task of identifying upgrade candidates. From there, consider progressing to a virtual library for all assets in the facility. As with everything else, you get what you pay for.

The ultimate results of these endeavors may not be immediately obvious, but they will become quite evident in time.

LITERATURE CITED Bloch, H. P., Improving Machinery Reliability, Third Edition, Gulf Publishing Co., Houston, 1998.

2 Bihler, K., D. Dominiak, B. Keith and J. Johnson, “Apply new pump software to test performance,” Hydrocarbon Processing, October 2012, pp. 91–96.

3 “HP In Reliability,” Hydrocarbon Processing, August 1992, p. 25.

1

HEINZ P. BLOCH resides in Westminster, Colorado. His professional career began in 1962 and included long-term assignments as Exxon Chemical’s regional machinery specialist for the US. He has authored over 520 publications, among them 18 comprehensive books on practical machinery management, failure analysis, failure avoidance, compressors, steam turbines, pumps, oil-mist lubrication and practical lubrication for industry. Mr. Bloch holds BS and MS degrees in mechanical engineering. He is an ASME Life Fellow and maintains registration as a Professional Engineer in New Jersey and Texas.

Overpressure risk never goes away.

That’s why we offer total pressure relief solutions.

Farris Total Pressure Relief Solutions Include: Pressure Relief Valves PSM Engineering /Audit Service & Software Asset Management Services & Repairs

The Total Solution.

It’s simple. Safety relief valves and services to support your facility’s entire lifecycle. Pressure relief systems designed right from the start. Pressure relief valves that help your plant operate safely. Global valve maintenance and repair support. And PSM software and audit services to ensure your plant operates the way it was designed to during an overpressure event.

Total Pressure Relief Solutions.

We invite you to see all that we can do for you. http://farris.cwfc.com or +1 440-838-7690

Select 56 at www.HydrocarbonProcessing.com/RS

Select 61 at www.HydrocarbonProcessing.com/RS

Special Report

Maintenance and Reliability M. G. CHOUDHURY, A. KULKARNI and D. KORANGA Reliance Industries, Mumbai, Maharashtra, India

Improve design for pump suction nozzles Several analysis methods exist to determine the effectiveness of the pump suction piping system’s first adjustable support and how that impacts nozzle loading. Parameters for consideration include how to model the trunnion dummy leg, determining the thermal load for the dummy leg and accurately predicting the first adjustable support base plate’s stiffness.

DIAGNOSING THE PROBLEM The pump piping system is a common component of most hydrocarbon processing complexes. It is essential that this system is analyzed correctly, so that future problems can be minimized. Within this system, the first support adjacent to the nozzle plays a vital role during analysis. This support is often an adjustable type that helps in maintaining the pump shaft axis in line with the nozzle axis.

A popular analysis program uses the pipe centerline modeling approach. Generally, an ambient temperature pipe trunnion from the pump centerline to the trunnion support base plate is modeled during analysis. This approach is sometimes criticized, as the trunnion pipe is welded at the main pipe bottom and the diametrical growth of the pipe is ignored. To circumvent this criticism, some analysts advocate a more refined analysis using the element from the suction pipe centerline to the bottom of the suction pipe, where the trunnion has been welded as a rigid element, with a temperature the same as that of the suction pipe. Then a trunnion with ambient temperature and without fluid density and pressure is modeled.

A commonly used analysis technique is the centerline analysis method. In this method, the dummy is modeled from the straight pipe neutral or at the mid-node of a three-node elbow, as shown in FIG. 1. The trunnion itself is assumed to be at ambient pressure and temperature. This analysis does not address the issue of pipe diametrical growth at elevated temperatures.

The rigid-element analysis addresses this issue.

Centerline modeling in straight line FIG. 1. Normal centerline analysis method.

To address the diametrical pipe growth issue, rigid-element analysis is used. In method, a rigid element is modeled from the pipe centerline to its periphery. Then the dummy with ambient temperature and pressure properties is attached to it (FIG. 2). The rigid element takes care of the radial expansion of the parent pipe. Due to this additional radial expansion, there will be a substantial rise in vertical pipe displacement, which increases nozzle loading.

CASE STUDY This sample scenario offers an opportunity to see the direct effect of both techniques. Consider a side-suction pump system at a 340°C design temperature, as shown in FIG. 3. This system is then analyzed using a recent edition of popular pipe stress analysis software.

Rigid element

FIG. 2. Rigid-element analysis method.

Centerline modeling at elbow FIG. 3. A side-suction pump system at a 340°C design temperature.

Hydrocarbon Processing | MAY 2013 75

Maintenance and Reliability Various cases in the software are then formed: a) Normal centerline method: For this case, only a dummy is modeled in front of the nozzle with a sliding base support, with FIG. 4 showing µ = 0.1. b) Rigid element method: in this instance, the rigid element is modeled from the pipe center to the parent pipe’s outer periphery (see Node A to Node B in FIG. 5) while maintaining the same conditions as the parent pipe. Then a dummy is attached at Node B, with the base support still sliding.

TABLE 1 summarizes the resulting nozzle loads from both analysis methods.

TABLE 1 demonstrates that the loads increase tremendously during the actual simulation of the pipe’s diametrical expansion. This can be attributed to substantial nozzle displacement in the vertical direction, which increases the vertical loads.

There are many ways to counter this vertical loading; the obvious option is changing the line’s routing. But this and other methods increase the overall cost to the system. However, one other way can decrease the loading on the nozzle.

The first adjustable support was closely examined for support stiffness implications on the nozzle loading. Normally in software analysis, default stiffness in the program is used. FIG. 6 shows a typical adjustable support.

Center line

Guide gap

Center line

Plan

FIG. 4. Normal centerline method.

Pipe trunnion

Guide gap

Elevation FIG. 6. The trunnion support rests on a thin stub-base plate, which is attached to the ground with the help of bolts and a series of plates.

FIG. 5. Rigid-element method.

TABLE 1. Nozzle loads tabulated FX, N

FY, N

FZ, N

MX, N-m

MY, N-m

MZ, N-m

DX, mm

DY, mm

DZ, mm

Normal centerline method

4,564

2,256

6,021

–987

–4,309

4,172

0

0

–1.936

Rigid element method

4,563

73,437

12,948

77,036

–4,308

6,507

0

0.008

–1.936

MX, N-m

MY, N-m

MZ, N-m

DX, mm

DY, mm

DZ, mm

TABLE 2. The results show reduction in the vertical loads FX, N

FY, N

FZ, N

Normal centerline method

4,564

2,256

6,021

–987

–4,309

4,172

0

0

–1.936

Rigid–element method

4,563

73,437

12,948

77,036

–4,308

6,507

0

0.008

–1.936

Rigid–element method with actual support stiffness

46,611

15,641

7,973

14,715

–4,372

4,562

0

0.002

–1.936

76 MAY 2013 | HydrocarbonProcessing.com

Maintenance and Reliability By predicting the accurate support stiffness, more accurate As FIG. 2 illustrates, the trunnion support rests on a thin analysis of the pump suction piping is possible. Even after usstub-base plate that is attached to the ground with bolts and a ing the pipe element radial growth, the nozzle loads have not inseries of stainless steel (SS), polytetrafluoroethylene (PTFE) and carbon steel (CS) plates.

The base plate material will be the same as the trunnion material. The PTFE plate inserted between the SS For high-temperature lines, diametrical and CS plates will reduce the friction loads as its fricpipe growth plays an important role in tion co-efficient is 0.1.

The most common assumption made during this assessing the nozzle loading on a pump. analysis is that the first adjustable support’s stiffness is When incorporated in a pump system the stiffness default value. In the case of this software analysis, it is 1.0E12 N/mm. This high value of stiffanalysis, this thermal loading will greatly ness corresponds to a rigid base, like solid ground or increase the loading on the nozzle. some concrete structures. Since this scenario features a support that is effectively resting on a plate 10-mm thick, the stiffness will be lower. creased. This analysis type approach may be adopted when it is necessary to do a rigorous analysis for a critical pump nozzle load.

STIFFNESS CALCULATION To predict the effects of the first support stiffness on the nozzle loads, the support stiffness must be calculated accurateFINAL ANALYSIS ly. The stub base plate’s vertical displacement must be accurate For high-temperature lines, diametrical pipe growth plays an to calculate the plate’s vertical stiffness. important role in assessing the nozzle loading on a pump. When The simulated vertical displacement results are shown in incorporated in a pump system analysis, this thermal loading will greatly increase the loading on the nozzle. To minimize the FIG. 7. For simplification, the scenario assumed that the vertical overall alteration of the system, it is advisable to use the actual bolt stiffness is sufficiently high as to assume it to be acting as support stiffness of the first adjustable support, thus reducing a vertical restraint. The plate dimensions considered were 250 the overall cost to the system. mm ⫻ 250 mm ⫻ 10 mm, with a 1,000-N external load applied on the trunnion periphery. From the stiffness equation, F = KX, for an applied load of 1,000 N, and vertical displacement of .0031512 mm, the vertical (Y) base plate stiffness is calculated as 31,7339 N/mm.

UTILIZATION By utilizing the base support stiffness, a more accurate analysis can be performed. TABLE 2 lists the results of all three analysis techniques.

TABLE 2 shows that the vertical loads are considerably reduced. This is due to the fact that the vertical displacement was reduced by utilizing the calculated support stiffness.

FIG. 7. Simulated vertical displacement results.

Select 170 at www.HydrocarbonProcessing.com/RS

77

| Bonus Report PETROCHEMICAL DEVELOPMENTS The petrochemical industry is a complex global business.

Feedstocks dominate total production costs for hydrocarbonbased products. Consumer demand for petrochemical-based end products will sustain an annual growth rate of 4% through 2018. Developing nations are the drivers for the expanding petrochemicals industry, and they will account for 70% of the global ethylene market.

Ariel view of Braskemâ&#x20AC;&#x2122;s UNIB 2 RS cracker at Triunfo located in the state of Rio Grande Do Sul, Brazil. Besides the cracker, Braskem has its Innovation and Technology Center and the worldâ&#x20AC;&#x2122;s first green ethylene plant, part of the green polyethylene (PE) production, along with three PE and two polypropylene industrial units located in the Triunfo Petrochemical Complex. Photo courtesy of Braskem America.

Petrochemical Developments

Bonus Report

J. FU, C. ZHAO and Q. XU, Lamar University, Beaumont, Texas

Consider novel CGC and front-end depropanizer system for olefins production In many liquid-cracking ethylene facilities, the multistage charge-gas compressor (CGC) is used with a front-end depropanizer (DeC3 ) process. The optimal design and operation of an integrated CGCDeC3 system can provide energy savings and other benefits. In the presented case study, an innovative design for an integrated CGC-DeC3 system introduces part of the heavier condensate from the early compression stages directly into the highpressure (HP) depropanizer. Result: The compression work and CGC stripper loading are reduced. This design also lowers heat duty for part of the heat-exchanger network. Using rigorous simulation, this new conceptual design can offer significant energy-saving potential.

process is used predominantly by earlierconstructed ethylene plants. It can be applied to all feedstocks from ethane to gasoil (GO). Other processing schemes are preferred when considering high efficiency and when feedstock flexibility to process heavier feeds such as propane, naphtha and GO are a priority.l In the presented study, the DeC3 system will be evaluated; the new system applies a twotower subsystem positioned immediately after the CGC and drying steps.

FIG. 1 shows the integrated CGC and front-end DeC3 system of an ethylene plant. After the naphtha feedstock is cracked, it is sent to the oil-quench and water-quench towers. In the quench sys-

tem, the cracked gas is cooled and partially condensed. The quench-tower overhead vapors are sent to a four-stage CGC section for compression and drying before being directed to the DeC3 subsystem.

The DeC3 subsystem is used to separate C3 and lighter components from the charge gas. It consists of two distillation columns: a HP depropanizer (HP DeC3 ) operated at 12 bar, and the low-pressure depropanizer (LP DeC3 ) operated at 7 bar. The C3 and lighter components are removed from the top of HP DeC3 , while C4 and heavier components exit from the bottom of the LP DeC3 .

By applying the HP and LP DeC3 subsystems, the olefin-unit operational flex-

Minimum flow bypass

BACKGROUND Steam cracking is the dominant olefin-production method. It includes several basic processing processes: cracking feedstocks, quenching charge-gas effluents, compressing cracked gas combined with drying (deacidification and dehydration), chilling the charge gas, hydrotreating acetylene and methyl acetylene/propadiene (MAPD), and separating of various components, such as ethylene, propylene, methane, propane and hydrogen. There are three types of olefin plants as defined by the recovery sequence, first separation step and position of the acetylene hydrogenation; these process methods include the: • Front-end demethanizer (DeC1 ) system • Front-end deethanizer (DeC2 ) system • Front-end depropanizer (DeC3 ) system.

The front-end DeC1 system is the most commonly adopted scheme. The

CGC 1st

CGC 2nd

CGC 3rd ER-50

ET-12

Feed from quench

EH-01

EH-11 EV-06 EH-03

EH-02

EV-04

EV-12

EV-02

EV-01

EV-03 ER-60

To water quench CGC 4th EV-16

EH-52

ET-16 To water quench EH-16

To DeC4 EP-16

EV-42

Anti-surge recycle

EH-51 ER-05

ER-32

Vapor to DeC1 Anti-surge recycle

EH-53

EH-58 EH-56

ER-31 Liquid to DeC1

EV-41

EH-61

ET-10 EH-50

ET-11

EV-40

EH-59

EP-20 EH-60

To DeC4

FIG. 1. Flow diagram of the integrated CGC and front-end DeC3 system.

Hydrocarbon Processing | MAY 2013 79

Petrochemical Developments ibility can be improved when heavy feedstock, such as naphtha, is used. The system is considered cost-effective because most of the C3 and lighter components can be removed from the feed stream and sent to the LP DeC3 tower. This design significantly reduces the refrigeration duty and minimizes fouling problems, which are experienced in a one-tower system.

In addition, there is no need for external hydrogen supply for the acetylene reactor, since the overhead stream from the HP DeC3 contains enough hydrogen for hydrogenation. Note: The fourth stage of the CGC is actually located after the overhead drum of the HP DeC3 . It only compresses the C3 and lighter components; this also reduces the compressor loading as compared to a back-end DeC1 process, which compresses everything before the cracked gas enters the demethanizer.

Process description. As shown in FIG. 1, the CGC system consists of four stages with intercooling, liquid separation, pumps and various auxiliaries. The cracked gas from the quench is fed to the first-stage compressor through a suction drum, and its pressure is raised from 1.28 bar to 3.36 bar. An interstage cooler chills the compressed cracked gas to 20°C. Similarly, each of the compression stages

consists of a suction drum, i.e., EV-01, EV-02, EV-12 and EV-42, a compressor (first stage to fourth stage), and an aftercooler, i.e., EH-01, EH-02, EH-03 and EH-51. The pressure of the cracked gas is raised to 7.76 bar after the second compression, 14.87 bar after the third, and 39.7 bar after the fourth. In the CGC flash drums, the cracked-gas mixture is separated into two phases: oil-liquid and vapor; or into three phases: water, oilliquid and vapor. Condensates from the second-stage discharge drum (EV-03), the third-stage suction drum (EV-12) and the dryer knockout drum (EV-06) are sent back to the second-stage suction drum (EV-02), where the condensed liquid is fed to the CGC stripper (ET-16) and the vapor goes to the second-stage compressor. Condensate from the thirdstage discharge drum (EV-04) is recycled to the third-stage suction drum (EV-12).

Other than the listed units, a caustic wash tower (ET-12), and a series of dryers and dehydrators are included in the CGC section. Between the second and third stages, the cracked gas is treated by a caustic tower to remove acid gases— carbon dioxide (CO2 ) and hydrogen sulfide (H2S)—to mitigate corrosion and freezing problems in the cold-separation section. The liquid hydrocarbon from the

Minimum flow bypass CGC 1st

CGC 2nd

CGC 3rd ER-50

ET-12

Feed from quench

EH-01

EH-03

EH-02 EV-02

EV-01

EH-11 EV-06 EV-04

EV-12 EV-03

ER-60

To water quench CGC 4th EV-16

EH-52

ET-16

EH-16

To DeC4

Anti-surge recycle

To water quench

Anti-surge recycle

EV-42

EH-51 ER-05

ER-32

Vapor to DeC1 EH-53

EH-58 EH-56

EP-16

ER-31 Liquid to DeC1

EV-41 Cooling water Propylene refrigerant LP pressure steam Quench oil Quench water Electricity HP steam Extra pipeline

EP-add Water

EH-61

ET-10 EH-50

ET-11

EV-40

EH-59

EP-20 EH-60

FIG. 2. New design of the integrated CGC and front-end DeC3 system.

80 MAY 2013 | HydrocarbonProcessing.com

To DeC4

second-stage suction drum (EV-02) is fed to the CGC stripper (ET-16), where the overhead is sent back to the waterquench tower, while the bottoms are pumped to the debutanizer tower. After the third-stage discharge drum (EV-04), a dryer feed cooler (EH-11) chills the cracked gas to 13.5°C. Next, the dryer knockout drum (EV-06) removes water, and the heavy hydrocarbons are sent back to the second-stage suction drum. The cracked-gas dryer (ER-50) and crackedgas dehydrator (ER-60) further eliminate water from the cracked gas before the gas enters the HP DeC3 tower (ET-10).

The cracked gas from CGC is cooled to –14°C by the HP DeC3 feed chiller (EH-50) prior to feeding to tray 21 of HP DeC3. The overhead stream from HP DeC3 is sent through the fourth-stage compressor to reach a pressure of 39.7 bar. A guard bed (ER-32) removes arsenic from the cracked gas. A series of acetylene reactor feed/effluent heat exchangers (EH-51 and EH-52) are arranged to warm the feed to the reaction temperature; the acetylene reactor (ER-05) hydrogenates acetylene from the HP DeC3 overhead.

Effluent from the acetylene reactor is routed back through the acetylene-reactor effluent cooler (EH-53), and sent to a molecular-sieve-dryer guard bed (ER31) to remove any traces of remaining water. The dryer effluent is further cooled in the HP DeC3 reflux chiller (EH-56), and then partially condensed by C3 refrigerant to approximately –33.5°C in the HP DeC3 condenser (EH-58), before entering the HP DeC3 reflux drum (EV-41).

The condensed HP DeC3 overhead leaves the reflux drum and is sent to the top tray of the HP DeC3. The vapor overhead from the reflux drum, containing C3 and lighter components from the cracked gas, is sent to the DeC1 and cold-box section.

To avoid excessive bottom temperatures, some C3 components are allowed in the bottom of the HP DeC3. High bottom temperatures tend to induce fouling of the reboiler and tray due to polymerization of C4 and C5 dienes. The HP DeC3 is reboiled by quench oil in the HP DeC3 reboiler (EH-59).

The bottoms from the HP DeC3 are sent to tray 24 of the LP DeC3 (ET-11), where the overhead is totally condensed to 3°C against a tertiary refrigerant in the LP DeC3 condenser (EH-61). The condensed overhead is sent to the LP DeC3

Petrochemical Developments reflux pumps (EP-20) and is split to provide reflux for both the HP and LP DeC3s.

The larger portion of the LP DeC3 reflux is sent to the top tray of the LP DeC3. The remainder is routed to tray 6 of the HP DeC3. The LP DeC3 is reboiled against LP steam in the LP DeC3 reboiler (EH-60).

The outlet material from LP DeC3 bottom, containing the C4 and heavier components, is sent to the debutanizer section.

NOVEL DESIGN This study focused on energy savings by improving the design of an integrated CGC-DeC3 system. To study the integrated system behavior, four stages of compressors with suction drums, caustic wash tower, cracked-gas dryer, HP and LP depropanizers with reboilers and condensers, C2 convertor and other related process units are included in the case study model.

New design configuration. The gen-

eral idea of the new design is to introduce heavy streams from the early compression sections directly into the fractionation section. Results: Loading of most processing units can be reduced, thus achieving energy savings for compression work, heating/cooling duty and electricity. However, some processing requirements must be satisfied. First, the heavy stream must be acid-gas free before entering the fractionation sections. Second, the streams must be dried before entering the DeC3 column.

Third, certain pressure increments for the heavy streams may be needed to make them viable as feed to the DeC3 columns.

FIG. 2 is one innovative design, where condensate streams from the third-stage suction drum EV-12 and discharge drum EV-04 are sent into the HP DeC3 directly.

A caustic wash (ET-12) ensures no H2S or CO2 contamination. To carry out the new design, several major changes were considered: • One pump (EP-ADD) is added to increase the pressure of the first heavy stream from 7.73 bar to 14 bar.

• One dryer (EV-ADD) is required to knock out water from the heavy stream.

• The feed tray of the heavy stream to the HP DeC3 must be identified and optimized.

• Operating conditions of both LP and HP DeC3 columns, e.g., reflux ratio, boil-up ratio, are optimized.

Based on this design, several improvements are expected:

• A significant amount of compressing work on heavy components in the second and third stages can be saved.

• Operation loading of the CGC stripper can be reduced due to a lower inlet flowrate, including a work reduction in the pump.

• The cooling duty for the CGC third-stage aftercooler can be reduced.

• Improvements in other units, including pumps and heat exchangers, are achievable.

Simulation and optimization results.

The optimization work of the new design is conducted with commercially available simulation software.a By establishing a steady-state model of the integrated CGCDeC3 system, the operation parameters are tested. Likewise, the separation specifications and safety requirements are monitored before the optimal results are obtained. For example, in the CGC section, the inlet/outlet pressure and temperature of each compressor stage are carefully handled to maintain a defined range.

TABLE 1 summarizes the operating constraints of the DeC3 subsystem. There are specifications for the top temperatures and pressures of the two columns. More importantly, both towers are designed to operate with a bottoms temperature less than 82°C. This low temperature re-

duces fouling in the trays and reboilers.

The calculated results from this study are 74.62°C of the HP DeC3 bottom, and 78.54°C of the LP DeC3 bottom; both are within the specs.

The DeC3 subsystem is the first separation unit of the whole fractionation section. It removes C3 and lighter components from the heavier ones, which will cause critical impacts to downstream processing if not operated properly. The C4 fraction in the top of HP DeC3 is limited to less than 100 ml/m3, which in this study, is reduced to 10 ml/m3. Some C3s are allowed in the bottoms of both columns to maintain low temperatures, thus mitigating fouling problems. The specification of C3s in the HP DeC3 bottom is less than 37 mol%, and the calculated value is 14.29 mol%. It requires less than 1,095 ml/m3 of C3s in the bottom of LP DeC3. Study results indicate 914.61 ml/m3. Other than listed specifications, the C2 fraction in the bottom of HP DeC3 is limited to less than 10,951 ml/m3. The result from this study is 3,580.20 ml/m3, which meets processing requirements. In summary, the calculated data are all qualified with respect to the specifications listed in TABLE 1.

TABLE 2 indicates the energy usage results with the new and the old designs, along with possible energy reductions.

According to different utilities, seven

TABLE 1. Operating constraints of HP DeC3 and LP DeC3 columns Column

Description

Model data

Comment

Top temperature, –30±3 °C

–31.51

Bottom temperature, 71±5 °C

74.62

Qualified

12.2

Qualified

10

Qualified

14.29%

Qualified

3,580.2

Qualified

2.9

Qualified

Pressure, 12±2 bar

C4 in top, < 100 ml/m3 HP DeC3 (ET-10) C3 in bottom, < 37 mol%

C2 in bottom, < 10,951 ml/m3

1,3-Butadiene

3.52

Isobutylene

6.29

n-Butane

0.19

Propadiene

0.94%

Propylene

13.07%

Propane

0.28%

Acetylene

166.08

Ethylene Ethane

2,424.31 989.81

Top temperature, 7±5 °C Bottom temperature, 75±6 °C LP DeC3 (ET-11)

Pressure, 7±0.8 bar Propadiene 3

C3 in bottoms, < 1,095 ml/m

Qualified

78.54

Qualified

6.5

Qualified

914.61

Qualified

909.36

Propylene

3.46

Propane

1.79 Hydrocarbon Processing | MAY 2013 81

Petrochemical Developments TABLE 2. Energy consumption optimization results summary Unit

Description

Utility, MMBtu/hr

EH–53

C2 Reactor effluent cooler

Cooling water

EH–01

CGC 1st-stage aftercooler

Cooling water

Previous design

New design

Saving

Saving, %

–9.73

–9.76

–0.03

–0.3

–23.63

–26.43

–2.8

–11.8

EH–02

CGC 2nd-stage aftercooler

Cooling water

–9.93

–8.98

0.95

9.6

EH–03

CGC 3rd-stage aftercooler

Cooling water

–18.08

–13.98

4.1

22.7

EH–11

Dryer feed chiller

Cooling water

–0.63

–0.79

–0.18

–25.6

EV–12

CGC 3rd-stage suction drum

Cooling water

–0.33

–2.53

–2.2

–668.1

EH–50

HP depropanizer feed chiller

C3 refrigerant

–8.54

–6.85

1.69

19.8

Total savings

–0.1 (0.2%)

EH–58

HP depropanizer condenser

C3 refrigerant

–13.51

–13.54

–0.03

–0.2

EH–56

HP depropanizer reflux chiller

C3 refrigerant

–12.35

–12.44

–0.09

–0.8

EH–61

LP depropanizer condenser

C3 refrigerant

–6.56

–6.48

0.08

1.3

4.1 (10.0%)

EH–52

Acetylene reactor feed heater

LP steam

4.49

4.50

–0.01

–0.2

EH–60

LP depropanizer reboiler

LP steam

6

5.88

0.12

2.0

EH–59

HP depropanizer reboiler

Quench water

5.72

6.80

–1.08

–18.9

–1.2 (21.5%)

EH–16

CGC stripper reboiler

Quench oil

3.22

1.87

1.36

42.1

2.4 (75.3%)

CGC 1st

CGC stage 1

HP steam

11.94

11.94

0.0

0.0

2.8 (26.9%)

CGC 2nd

CGC stage 2

HP steam

10.37

9.97

0.40

3.9

CGC 3rd

CGC stage 3

HP steam

7.59

7.09

0.50

6.6

0.87 (2.1%) CGC 4th

CGC stage 4

HP steam

11.28

11.31

–0.03

–0.3

EP–16

CGC stripper pump

Electricity

0.024

0.016

0.01

33.3

EP–20

LP depropanizer reflux pump

Electricity

0.019

0.018

0.001

5.3

EP–Add

Added pump

Electricity

0.00

0.011

–0.01

–

different colors are used to represent cooling water, C3 refrigerant, LP steam, quench oil, quench water, HP steam and electricity. Detail locations of each utility usage in the integrated CGC and DeC3 system are indicated in FIG. 2.

Results in the “Saving %” column of TABLE 2 show that almost half of the energy consumption units can improve performance via reduced energy requirements.

For example, the heating duty of the CGC stripper reboiler can be reduced by 42.1%, and the work required by the CGC stripper pump can be reduced by 31.3%. The cooling duty of the CGC second- and third-stage aftercoolers can have a reduction of 9.6% and 22.7%, respectively.

In the CGC, the most significant improvements are achieved by the secondand third-stage compressors, which can reduce needed work by 0.4 MMBtu/hr and 0.5 MMBtu/hr. There is a negligible increment of 0.1 MMBtu/hr on the cooling water duty, a small amount of additional heating duty, 1.2 MMBtu/hr from quench water, and a tiny increment of 0.002 MMBtu/hr on electricity.

The energy savings are significant with the new design as compared with 82 MAY 2013 | HydrocarbonProcessing.com

previous designs. For example, the total amount of saved cooling duty from C3 refrigerant is 4.1 MMBtu/hr—a 10% reduction. This suggests significant reductions in the workload for the C3 refrigerant system. Meanwhile, the heating duty required for LP steam can be reduced by 2.8 MMBtu/hr, which is a 26.9% savings as compared to the previous design. In addition, a 2.4 MMBtu/hr reduction saves 75.3% heating duty for the CGC stripper reboiler under the new design. Finally, the amount of work needed for the compressors can be reduced by 0.87 MMBtu/hr in total, which saves 2.1% from the previous consumption of CGC work.

ACKNOWLEDGMENTS This research work was supported in part by the Texas Air Research Center, Texas Hazardous Waste Research Center and the Graduate Student Scholarship from Lamar University. a

1

NOTES The optimization work of the new design is carried out with Aspen Plus 7.3, Aspen Technology, Inc., 2011.

LITERATURE CITED Falqi, F., The Miracle of Petrochemicals—Olefins Industry: an In-Depth Look at Steam-Crackers, Universal Publishers, Boca Raton, Florida, 2009.

–0.002 (4.7%)

JIE FU is a PhD candidate with the Dan F. Smith Department of Chemical Engineering at Lamar University. He holds a BS degree from the East China University of Science and Technology and was an exchange student at the University of Houston. His research interests include process design, dynamic simulation and production scheduling for energy saving and emission reduction in industries.

CHUANYU ZHAO is a PhD candidate with the Dan F. Smith Department of Chemical Engineering at Lamar University. She received her BS degree from the East China University of Science and Technology, and was an exchange student at the University of Houston. Her research areas include process modeling, planning, scheduling and optimization, process simulation, process safety and process synthesis with particular applications in petrochemical, refining and electroplating industries.

QIANG XU is an Associate Professor of the Dan F. Smith Department of Chemical Engineering and 2012 University Scholar at Lamar University. He holds BS and MS degrees along with a PhD from Tsinghua University, China, all in chemical engineering. His research involves process modeling, scheduling, dynamic simulation and optimization, industrial pollution prevention and waste minimization, and chemical process safety and flexibility analysis. His research work has been extensively supported by industries and by federal and State of Texas funding agencies.

THE EXPECTED. We provide solutions. Smith & Burgess is the expert in flare and relief system design.

Our methodology is cost-effective and we pride ourselves on using realistic assumptions that meet industry standards, yet don’t burden the owners with excessive concerns. We are committed to our customers. At Smith & Burgess we are passionate about protecting your assets. We insist on quality. If that’s beyond what you expected, we did our job right.

www.smithburgess.com Select 72 at www.HydrocarbonProcessing.com/RS

BETE...

Retractable Spray Lance

your one-stop resource for drop-in spray lances and quills.

MaxiPass provides a full cone spray pattern with even distribution

Performance Through Engineering CFD Analysis

BETE is your complete single-source supplier of custom spray lances, quills, and chemical injectors. Our in-house operation includes all aspects of design, fabrication, and performance testing, ensuring a seamless process from RFQ to delivery.

BETE’s welders have decades of extensive experience in welding dissimilar metals and exotic alloys and are fully qualified to Section IX. BETE’s integrated engineering, quality, and manufacturing departments combine to meet a wide variety of code, testing, and inspection requirements.

ASME B31.3 fabrications are our specialty.

Distribution

Request a lance quote...

BETE’s Application Engineers are ready to take your sketch/inquiry and produce a recommendation and drawing of our proposed solution. We’ll use our experience to work with you to improve your process with the right spray nozzle coupled with custom fabrications designed for your application.

BETE. Your strategic partner for engineered spraying solutions.

Evaporation Made in the USA MaxiPassTM Nozzles The ultimate in clog-resistance with the largest free passage available in a full cone nozzle

SpiralAirTM Nozzles A unique three-stage process that uses atomizing air more efficiently than conventional designs, resulting in better atomization at high liquid flow rates

Spray Lances Customized spray nozzle lance assemblies are used for scrubbing, quenching, injection and washing that meet the strictest quality and performance criteria

Select 58 at www.HydrocarbonProcessing.com/RS

BETE Fog Nozzle, Inc. T +1(413)772-0846 F +1(413)772-6729 www.bete.com

Bonus Report

Petrochemical Developments J. SHIN, J. LEE and S. LEE, LG Chem Ltd., Daejon, South Korea; B. LEE, AMT Pacific Co. Ltd., Seoul, South Korea; and M. LEE, Yeungnam University, Gyongsan, South Korea

Enhance operation and reliability of dividing-wall columns Dividing wall columns (DWCs) are an attractive option when the goals are to reduce energy consumption and capital investment for distillation processes. A major component in a DWC system is the reflux splitter, which separates the liquid products from the column overhead into two liquid streams: one to the prefractionation section and the other to the main column section of the DWC. A more reliable reflux-splitting system stabilizes the entire column, thus minimizing unit/process downtime. Several novel developments for a reflux-splitting system can improve the operation and reliability of DWCs. Background. DWC technology has been successfully applied

in various petrochemical processes.1,2 A DWC can reduce energy consumption by removing the remixing phenomenon within a conventional two-column system for a ternary mixture. DWC systems can minimize capital investment by intensifying a conventional two-column system into a single column with a simple dividing wall.

The concept of DWC was proposed in 1949; however, it took a long time to adopt this principle.3 Availability of proper design

tools is a major obstacle in developing more complex separationcolumn systems as compared to a conventional column. Since the 1980s, powerful CAD software enables the design and commercialization of DWCs for a range of petrochemical processes.

Once the structure of a DWC is fixed for a given product specification, the most important optimizing variables remaining are the liquid and vapor split in the dividing wall section, as shown in FIG. 1. These variables have a significant effect on the total separation performance in the DWC.4 FIG. 2 shows the effect of internal flow distribution on the energy consumption in a typical DWC. This figure illustrates the existence of an optimal internal flow distribution that generates the lowest energy consumption. Generally, the internal liquid and vapor flows into the prefractionator and main dividing wall section are the most crucial design factors. Both impact the total energy consumption and separation efficiency. The energy efficiency of DWC can drastically deteriorate by a small deviation in the internal flows from the optimal conditions. Conversely, only the liquid split can be adjusted during operation. The vapor split cannot be manipulated arbitrarily once the column is

6.0

Liquid split

5.4

Main column

Dividing wall

Reboiler heat duty, MW

4.8 4.2 3.6 3.0 2.4 1.8 1.2

Prefractionation section

0.6 0.0

Vapor split

FIG. 1. Schematic diagram of a typical DWC.

25 Preva 30 por in 35 terna l flow 40 rate, 45 kgmo le/hr 50

15 hr ole/

55 40

35

10

20 25 te, kgm a r ow id fl

30 iqu Prel

FIG. 2. Typical effects of liquid and vapor split on energy efficiency in DWCs. Hydrocarbon Processing | MAY 2013 85

Petrochemical Developments constructed. For this reason, in most commercial DWCs, there should be a device to split the liquid coming from the column overhead into sections divided by a wall, i.e., a prefractionation section and main column section.

Several reflux splitters have been adopted in commercial DWC installations. A simpler and more reliable reflux-splitting system is the optimum choice for better operability and reliability of the total column system. A novel reflux-splitting system can provide better operation and a more energy-efficient DWC. Internal reflux splitters for DWC. Conceptually, the split-

ting of a liquid stream can be achieved by a simple arrangement of a pump, valves, pipes and vessel, as shown in FIG. 3. Nevertheless, it still requires additional plot area, equipment and instruments. The increased equipment inventory within the system should be minimized. For this reason, this concept is barely used in practical applications of DWCs. The ideal reflux-splitting device for DWC applications should have these technical features and requirements, considering its purpose and mounting location: • Design does not include a pump and a moving/rotating unit.

• Height and volume of the column should be minimized.

• System can accurately manipulate and control the split ratio.

• Unit is simple, durable and reliable.

Moving bucket reflux splitter. First-generation reflux-splitting systems are shown in FIG. 4.5,6 These early systems consist of a casing that is subdivided into three chambers. The feed chamber is located above where the liquid is directed onto either the prefractionation or main column chamber depending on the positioning of the dividing body. The split ratio is controlled by adjusting the timing of the hollow inner bucket into the prefractionation or the main column chamber. The dividing body is actuated by magnetic coupling, which allows a pressure- and vacuum-tight design. The exterior drive is a pneumatically driven rotary motor.

Manual reflux splitter. The moving elements in the first-gen-

eration reflux splitter can be damaged by the constant movement of the mechanical parts. In a petrochemical plant, the splitter should operate at least 8,400 hr/yr without mechanical problems. For example, the naphtha cracking center (NCC) has more severe requirements to address possible equipment malfunction because it needs to have run length of at least three to four years without regular scheduled plant maintenance. A more reliable liquid-splitting device should be used in such cases. For this purpose, a manual-type reflux splitter that requires no moving element was devised, as shown in FIG. 5, and it was successfully installed on several commercial DWCs in LG Chem plant sites.7 The internal structure of the manual reflux splitter appears complex, but the basic principle is quite simple. As shown in FIG. 5, the splitter is divided into several sections that receive the liquid drawoff from the column overhead. Note: Each section has a liquid distributor with a different number of holes to adjust the liquid flowrate.

An on/off type of automatic valve is mounted to each section. If the valve is closed, the received liquid will overflow into the main section, which is connected directly to the main column without a valve. By combining the opening of each valve, the split ratio can be controlled by predetermined values. The control range of the split ratio can be determined

Vessel

FC FIG. 3. Conceptual design of liquid splitting in a DWC.

FIG. 4. Reflux splitters with moving elements.5, 6

86 MAY 2013 | HydrocarbonProcessing.com

FIG. 5. Cutaway view of a manual reflux splitter.

Petrochemical Developments from a sensitivity study of the DWC column for possible operation scenarios, such as variations in feed compositions and product quality. The performance of the reflux splitter can be verified before mounting it into the DWC column using a water-run test.

In-the-column reflux splitter. The latest develop-

umn. As shown in FIG. 7, the reflux-split ratio can be controlled accurately by closing and/or opening the valves. LG Chem’s experience from several petrochemical applications suggests that,

A new internal reflux-splitter system with no moving part can achieve more reliable operation of DWCs. A DWC can reduce energy consumption by eliminating the ‘remixing’ phenomenon with a conventional two-column system into a single column.

ment in the reflux-splitter system increases unit reliability with a simple modification of the conventional distributor and collector tray, as shown in FIG. 6.8 The in-the-column reflux splitter consists of a modified collector tray, collector box and two transfer lines.

The collector tray gathers the liquid from the top section and transfers it to the collector box located below.

Holes in the middle of the collector tray (see FIG. 6B) divide the liquid into four sections in the collector box with predefined fractions. As shown in FIG. 6C, the splitting ratio can be adjusted by the on/off operation of two valves, which are located on the transfer lines (see FIGS. 6A and 6C) at the outside of the column. If the two valves are open, then the liquid stream from the middle sections goes to the pre-section via liquid transfer lines, which are connected to the distributor for the presection. Thus, it will increase the liquid flow to the presection. If the two valves are closed, the liquid flow to the main section will be increased by the liquid overflow to the main section. Four combinations in the on/off valve are possible, and, thus, the four split ratios are available. In the design phase, the required split ratios can be predefined considering possible operational scenarios.

The developed reflux splitter satisfies all the technical requirements or specifications described earlier; it can reduce fabrication and installation costs. This splitter type is simpler and cheaper than the manual reflux splitter, as well as the bucket splitters. In addition, it can reduce the column height, thus providing the opportunity to install more trays or packing sections with a longer length.

The operating range of this splitter can be predetermined and the performance can be verified before installation in the col-

in most cases, only four combinations are sufficient to control the split ratio for expected operational variations.

Evaluation. Several types of internal reflux splitters were analyzed in terms of reliability, structural simplicity and ease of maintenance. TABLE 1 summarizes the results of the study comparing three different reflux splitters.

Recent developments in the internal reflux-splitting system make the DWC application more reliable with a simpler structure. The reflux splitter without a moving element may not cover a wider range of split ratios, but it can provide more robust and reliable operation with less maintenance. Overall, TABLE 1. Comparisons of several reflux-splitter systems Splitter type

Operating window/control

Reliability/ maintenance

Moving bucket Wide/continuous Fair/overhaul needed

Cost/structure High/complex

Manual

Narrow/discrete

Excellent/regular Moderate/simple valve inspection

In-the-column

Narrow/discrete

Excellent/regular Low/simple valve inspection

3.5 Design Actual, min. flow Actual design Actual, max. flow 3.0

2.5

2.0

1.5 1.5

FIG. 6. In-the-column reflux splitter: A) overall, B) collector tray and C) collector box and transfer lines.

2.0

2.5

3.0

3.5

FIG. 7. Results of the water-run test for the in-the-column reflux splitter.

Hydrocarbon Processing | MAY 2013 87

Petrochemical Developments system specific design with a modeling study for possible operational scenarios is recommended to overcome the flexibility issues of the newly developed splitter.

ACKNOWLEDGMENT This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012012532).

LITERATURE CITED Asprion, N. and G. Kaibel, â&#x20AC;&#x153;Dividing wall columns: Fundamentals and recent advances,â&#x20AC;? Chemical Engineering Progress, Vol. 49, 2010, pp. 139â&#x20AC;&#x201C;146.

2 Shin, J., S. Lee, J. Lee and M. Lee, â&#x20AC;&#x153;Manage risks with dividing-wall column installation,â&#x20AC;? Hydrocarbon Processing, June 2011, pp. 59â&#x20AC;&#x201C;62.

3 Wright. R. O., â&#x20AC;&#x153;Fractionation Apparatus,â&#x20AC;? US Patent 2,471,134, May 1949.

4 Lee, S., M. Shamsuzzoha, M. Han, Y. Kim and M. Lee, â&#x20AC;&#x153;Study of structure characteristics of a divided wall column using the sloppy distillation arrangement,â&#x20AC;?

Korean Journal of Chemical Engineering, Vol. 28, 2011, pp. 348â&#x20AC;&#x201C;356.

5 http://www.montz.de/sites/products/reflux.fr.html.

6 http://www.nascentprocess.net.

7 Lee, B., G. Kim, M. Lee, J. Lee, J. Shin, and S. Lee, â&#x20AC;&#x153;Liquid splitter,â&#x20AC;? Korea patent pending, 10-2010-0120467, November 2010.

8 Lee, B. and G. Kim, â&#x20AC;&#x153;Liquid splitter,â&#x20AC;? Korea patent pending, 10-2012-0030223, March 2012.

1

JOONHO SHIN is a process systems engineer at LG Chem Ltd., in South Korea. In 1997, he began his professional career as a design and control specialist with SK Engineering & Construction.

Dr. Shinâ&#x20AC;&#x2122;s industrial experience has focused on modeling, optimization and control of chemical and petrochemical industrial plants. He holds a BS degree in chemical engineering from Korea University, and an MS degree and PhD in chemical engineering from the Korea Advanced Institute of Science and Technology (KAIST).

SUNGKYU LEE is a process systems engineer at LG Chem Ltd. in South Korea. He holds BS and MS degrees in chemical engineering from Chungnam National University. He has worked on modeling, optimization and control of chemical and petrochemical plants since 2002.

JONGKU LEE is the vice president of LG Chem Ltd. at Research Park, South Korea. He holds a BS degree in chemical engineering from Seoul National University, and an MS degree and PhD in chemical engineering from KAIST. Since joining LG Chem in 1994, he has worked on numerous process modeling and optimization projects. Dr. Lee is in charge of LG Chemâ&#x20AC;&#x2122;s process modeling and solutions group and is performing research in the areas of energy saving and sustainability in chemical plants.

BYEONGKYEOM LEE is a chief technology officer of process engineering, mass-transfer equipment design and application for AMT Pacific Co. Ltd., in South Korea, since 2000. He holds a BS degree in chemical engineering from Sungkyunkwan University. He began his professional career as a process engineer at Kolon Engineering Inc. in 1989, and worked in SK Engineering & Construction Co. Ltd. from 1991 to 2000 as a specialist of mass-transfer equipment design and application. He has over 20 years of experience in distillation column design and applications for refinery, petrochemical and chemical plants.

MOONYONG LEE is a professor at the school of chemical engineering at Yeungnam University in South Korea (http://psdc.yu.ac.Kr). He holds a BS degree in chemical engineering from Seoul National University, and an MS degree and PhD in chemical engineering from KAIST. Dr. Lee worked in SK Energyâ&#x20AC;&#x2122;s refinery and petrochemical plants for 10 years as a design and control specialist. His current areas of specialization include modeling, design and control of chemical processes.

Â&#x2021; 5LJRURXVO\ WHVWHG IRU RYHU \HDUV IRU H[FHSWLRQDO SHUIRUPDQFH Â&#x2021; 5HVLVWDQW WR KDUVK FKHPLFDOV KLJK WHPSHUDWXUHV DQG KLJK SUHVVXUHV Â&#x2021; /RQJ VHUYLFH OLIH Â&#x2021; *OREDOO\ DYDLODEOH

88 MAY 2013â&#x20AC;&#x201A;|â&#x20AC;&#x201A;HydrocarbonProcessing.com

Select 171 at www.HydrocarbonProcessing.com/RS

Select 92 at www.HydrocarbonProcessing.com/RS

UPSTREAM, MIDSTREAM OR DOWNSTREAM, THEY ALL FLOW THROUGH

AVONDALE World-Class Engineering and Modular Construction for Energy Infrastructure Projects At Avondale, our highly skilled workforce is energized and set to deliver innovative solutions for upstream, midstream and downstream energy infrastructure projects. Backed by our vast facilities, unmatched capabilities, and 75-year legacy of success, weâ&#x20AC;&#x2122;re harnessing our energy to build solutions today that will power the world for years to come.

Visit us at www.hii-avondale.com

504-654-5000 | 5100 River Road Avondale, LA 70094 Select 77 at www.HydrocarbonProcessing.com/RS

Special Supplement to

TERMINALS AND STORAGE Overflow systems are the last line of defense T–92 Terminals and storage news T–94 CORPORATE PROFILES InduMar Products, Inc. T–95

CB&I T–97

2013

TERMINALS AND STORAGE

OVERFLOW SYSTEMS ARE THE LAST LINE OF DEFENSE M. TOGHRAEI, Engrowth Training, Calgary, Alberta, Canada

The overflow system is the last line of defense against the excessive container filling. An overflow system consists of an overflow nozzle with internal and/or external piping and, in some cases, a siphon breaker. In this context, the container could be an atmospheric tank or a non-flooded vessel. However, not all the containers need an overflow system. If siphoning the liquid out of the container is not affordable, or if the other protective layers against overfilling provide sufficient coverage, an overflow system may not be necessary. In general practice, atmospheric tanks—but not vessels—have overflow systems.

Overflow nozzle location. Orienting the overflow nozzle

tends to not be significant from a process perspective, unless the tanks possess complexities such as compartments. The overflow nozzle orientation can be decided based on plot plan preference (FIG. 1).

In a very basic design, the overflow nozzle may be located high on the side of the tank, usually above high high liquid level (HHLL) and below the roof seam. Subsequent research revealed that this design causes liquid splashing during the tank overflow, which is hazardous if the liquid in question is an “aggressive” liquid, meaning it is flammable, irritable or corrosive.

To fix this problem, the overflow nozzle can be piped down to the bottom of the tank, where an elbow can prevent the tank foundation from washing out due to the overflow liquid (FIG. 2).

If a tank is blanketed at the top, such an overflow arrangement could provide an opportunity for the blanket gas to escape from the tank. To prevent this, the overflow nozzle could be extended internally to the lower part of the tank. One option is to lead this Orientation

Elevation

Plan view

Side view

FIG. 1. Plot plan preference

pipe down to below the low low liquid level (LLLL) in order to ensure the presence of a liquid at the inlet of the overflow, preventing the escape of the blanket gas. Care should be taken to not extend the internal pipe to the very bottom of the tank, especially if the stored liquid is not very clean and sludge may be present at the bottom of the tank, which could clog this pipe.

Additionally, there may be cases where it is required to have both an internal pipe and an external one, in which case it is important to utilize a siphon breaker. In the absence of such a breaker, the overflow stream will continue even after the liquid has dropped to below the threshold for the overflow nozzle.

In other words, the tank will continue to overflow even after recovering from an upset.

Overflow sizing. Proper overflow nozzle sizing relies on a

careful definition of its purpose within a refinery’s needs. The nozzle should provide enough area to discharge the flow out of a container to protect it from overflowing, when all other outlets to the container are blocked and all the inlets are passing. Alternatively, some companies define an overflow scenario as one where all the nozzles are blocked except for the biggest inlet nozzle. The second definition is obviously less conservative than the first one. Therefore, the first step of overflow sizing is deciding on the flow (or summation of flows) that causes a container to overflow.

To develop a methodology for overflow sizing, first observe the phenomenon. When all the outlet nozzles are blocked, the liquid level in the container starts to rise from normal liquid level (NLL). This continues until the liquid level reaches the bottom of the overflow nozzle, at which point the tank begins to overflow. Because the overflow nozzle is not sealed by flow, the overflow area is sufficient and the tank level continues to rise.

Assume for the sake of discussion that the overflow nozzle area is not adequate for overflow and that the tank level continues to rise. This is significant if and when the liquid level reaches the seam of the roof. At this point, the flow out of the overflow nozzle should be equal to the flow that comes into the tank. In such a case, the overflow nozzle fails if and when it is unable to accommodate a flow equal to inlet flow and the liquid level inside of the tank reaches the roof space.

Assume a hypothetical draining container exists at the top of the tank. The solution is: Q

FIG. 2. The overflow nozzle can be piped down to the tank bottom.

T–92

TERMINALS AND STORAGE 2013 | HydrocarbonProcessing.com

Cd A

2

2gh

(1)

Q is the inlet governing flowrate , Cd is the discharge co-efficient (which can be considered as 0.6 for a sharp-edged outlet) and h is the distance between line from the overflow nozzle to the top of the tank. From Eq. 1, the overflow area (A) and the overflow inside can be calculated (FIG. 3).

TERMINALS AND STORAGE Example 1. Determine the overflow size for a water tank with

15 m height and 24 m diameter. The governing in-flow is 850 m3/hr as shown by: Q Cd A 2 2gh (2) Assume that the height of liquid above the centerline of the overflow nozzle is 0.5 m, then Eq. 3 is: 850 0.6 A 2 2 9.8 0.5 (3) 3,600 D = 15.8 in. or 16 in.

When a pipe is connected to the overflow nozzle, the same formula can be used with a slight correction: Cd Q= ×A× 2gh (4) L 1+ f D f is the moody friction factor and L is the total actual and equivalent length of overflow pipe with D as the diameter.

Two-phase flow. Another aspect of overflow nozzle needs to

be addressed—two-phase flow. Initially, the overflow system (nozzle and downcomer) is full of air until the onset of overflowing when a partial flow will appear. If the level goes higher and flow increases, the liquid stream and air will generate a twophase flow. There is not much research on vertical downflow liquid-gas two-phase flows; however, three regimes have been identified in such an arrangement.1 At a low liquid rate, the regime will be “annular flow” for a short period, “slug flow” at a moderate rate and “bubble flow” at a high liquid rate (FIG. 4).

Because the air pressure is not high, there is a chance that in the last stage of the regime, it will be converted into a single, liquid phase. This actually is preferable, because if the regime is annular, then the capacity of the overflow system is reduced; and in the two other regimes there would be a big vibration in the overflow pipe. This favored regime conversion can happen if the liquid velocity is higher than the “sweeping” velocity. To ensure the liquid velocity is higher than sweeping velocity, it is recommended that Eq. 5 is followed:2 Fr ≥ 0.31

(5)

Fr is the liquid-phase Froude number: Fr =

2

VL ρL × g×D ρL − ρG

850 3,600 =1.82 m/sec 2 ⎛ 25.4 ⎞⎟⎤ π ⎡⎢ ⎜ × 16×⎜ ⎟⎥ ⎜⎝1,000 ⎟⎠⎥ 4 ⎢⎣ ⎦ 1.8 = 0.91>0.31 2 ⎡ 25.4 ⎤ ⎥ 9.8 ⎢16× ⎢⎣ 1,000 ⎥⎦

(8)

With Eq. 8, the air slugs should be swept away. Another method to ensure there is no entrainment in the overflow downcomer is:3 Fr 1.6

h D

2

(9)

Siphon breaker sizing. A siphon breaker is a pipe that is located

at the highest point of the inverted U and is routed to the atmosphere, outside or inside of the tank. This pipe allows the system to suck the gas from outside the system. This breaks the siphon and prevents flow out of the overflow nozzle, due to the siphon phenomenon. As a rule of thumb, this pipe is a straight pipe with a length of around 4 m to 5 m and a diameter of 2 in. to 3 in.

However, for the detailed design, it should be considered that a quick gas stream from the outside needs to “break the vacuum” by travelling through this pipe, so that there is an increase in the vacuum pressure (almost 0 kpag) to a pressure equal to the liquid column in the inverted U leg.

h

FIG. 3. Overflow nozzles can fail when the liquid level reaches

the roof space.

(6)

However, air density is negligible in comparison to liquid density and the Froude number can be reduced: Fr =

2

VL g×D

(7)

Therefore, the Froude number of overflow should be greater than 0.31. Otherwise, the overflow nozzle should be decreased, and the liquid head above nozzle should be increased accordingly.

Example 2. In Example 1, make sure there is minimum vibra-

tion during overflowing:

Annular flow

Slug flow

Bubble flow

Increasing liquid flow FIG. 4. Three liquid rate possibilities.

HYDROCARBON PROCESSING | TERMINALS AND STORAGE 2013

T–93

TERMINALS AND STORAGE

TABLE 1. Pipe friction data for clean commercial steel pipe with flow in complete turbulence Nominal size Friction factor, fr

â &#x201E;2 in.

1

0.027

â &#x201E;4 in.

1 in.

11â &#x201E;4 in.

11â &#x201E;2 in.

2 in.

21â &#x201E;2 in., 3 in.

4 in.

5 in.

6 in.

0.025

0.023

0.022

0.021

0.019

0.018

0.017

0.016

0.015

3

This flow is in an adiabatic condition because of its rapid occurrence. In a simple system, the flow through a 2-in. pipe is about 0.29 kg/sec and is meant to fill out a specific space in the bend of the inverted U to break the vacuum. This space is theoretically only a thin layer of gas in the inverted U cross-section which provides discontinuity in liquid siphon flow. However, in practical situations, it is possible that a larger space will be needed to fill out and discontinue siphon flow. A realistic approach is to assume that the destination space could equal the bend volume.

Ideal Gas Law PV = nRT can predict the required mass in this flow and these two numbers together provide the time to break the vacuum. This is the time that must elapse to stop the flow of siphon out of the overflow nozzle. Now it is up to the designer to decide if this time is acceptable or if the siphon breaker pipe should be enlarged.

Example 3. In Example 1, calculate the siphon breaker size.

The source data is atmospheric pressure; assume a pressure of 101 kpag, an ambient temperature of 20°C, and the destination point data is pressure zero. The pipe is assumed to be 4 m long with a 2-in diameter.

The solution is based on calculating adiabatic compressible flow.4 N represents all the losses in pipe and is K + f (L/D). K is considered 1 for an exit and for friction factor ( f ), a fully turbulent flow friction factor is used (TABLE 1).5 For N equal to 2.5, flux will be 143.6 kg/s.m2 or 0.29 kg/s.

This flow should fill the bend volume. If the stretched length of bend is 5 m, its volume would be 0.65 m3. If the height of the inverted siphon leg is 10 m, the pressure in the bend will be: 1,000Ă&#x2014;9.8Ă&#x2014;10 P = Ď ghâ&#x20AC;&#x2122; = = 100 KP absolute pressure (10) 1,000

PV

8â&#x20AC;&#x201C;10 in. 12â&#x20AC;&#x201C;16 in. 18â&#x20AC;&#x201C;24 in.

0.014

0.013

0.012

m RT M

100 .065

(11) m 0.008314 20 273 29

m = 1,554 gr R

0.008314 m 3 kPa mol. K

M = 29 for air T = 20°C = 293K Gives m = 1,554 gr or 1.554 kg Elapsed time would be: 1.554 m = = 5.3 sec t= 0.29 m°

(12)

During this 5.3 sec, 1.26 m3 of water will be wasted due to siphoning: Volume = 5.3 ⍝ 850/3600 = 1.26 m3 or 0.02% of total tank volume

(13)

LITERATURE CITED Barnea, B. â&#x20AC;&#x153;Flow Pattern Transition for Vertical Downward Two Phase Flowâ&#x20AC;?, Chemical Engineering Science, Vol. 37, Issue 5, 1982.

2 Coker, A. K., FORTRAN: Programs for Chemical Process Design, Analysis and Simulation, Gulf Professional Publishing, p. 182, 1995.

3 Perry, R. H., and D. W. Green, Eds., Perryâ&#x20AC;&#x2122;s Chemical Engineersâ&#x20AC;&#x2122; Handbook, 6th Ed., McGraw-Hill, New York , pp. 6-28, 1984.

4 Perry, R. H., and D. W. Green, Eds., Perryâ&#x20AC;&#x2122;s Chemical Engineersâ&#x20AC;&#x2122; Handbook, 6th Ed., McGraw-Hill, New York, pp. 6-23, 1984.

5 â&#x20AC;&#x153;Crane Technical Paper No. 410,â&#x20AC;? Crane Valves Co., New York, 1988.

1

MOHAMMAD TOGHRAEI is a consultant with Engrowth Training. He has over 20 years of experience in the field of industrial water treatment. His main expertise is in the treatment of wastewater from oil and petrochemical complexes.

TERMINALS AND STORAGE NEWS METHANOL STORAGE IN LOUISIANA

Kinder Morgan Energy Partners has entered into a long-term contract with Methanex to support the construction of methanol storage capacity near Kinder Morganâ&#x20AC;&#x2122;s Geismar liquids terminal (GLT) in Geismar, Louisiana. Kinder Morgan will build, own and operate the storage tanks and related infrastructure, including improvements to its existing dock at GLT. The assets will provide critical marine, rail and truck access in support of the 1 million tpy methanol production plant being relocated by Methanex from Chile.

The terminal infrastructure is expected to be in service during the second half of 2014, coinciding with the anticipated startup of the relocated plant. In a separate deal, Kinder Morgan also acquired Quality Carriersâ&#x20AC;&#x2122; 26-acre terminal located in Chester, South Carolina. The 19-tank facility currently provides storage for a single customer, with a capacity of 35,000 bbl. The terminal receives product by rail and distributes by truck.

Tâ&#x20AC;&#x201C;94

TERMINALS AND STORAGE 2013â&#x20AC;&#x201A;|â&#x20AC;&#x201A;HydrocarbonProcessing.com

TESTIMONY BLASTS DHS EFFORTS AT ASSESSING CHEMICAL SECURITY RISK

The US Government Accountability Office (GAO) has found that the tiering approach used to regulate high-risk chemical facilities by the Department of Homeland Security (DHS) Infrastructure Security Compliance Division fails to properly consider each of the risk elements involved in a potential terrorist attack (threat, vulnerability and consequence). This conclusion was presented on March 14 by GAO Director Stephen Caldwell during testimony before the US House Energy and Commerce Committee. A similar finding was also the basis for the International Liquid Terminals Association petition to DHS to remove gasoline facilities from regulation under the Chemical Facility Anti-Terrorism Standards (CFATS). GAO estimates that it could take seven to nine more years for ISCD to complete its first review of security plans from all CFATS-regulated sites, which began during 2011.

INDUMAR PRODUCTS, INC.

IF IT LEAKS OR CORRODES INDUMAR CAN STOP IT Since 1986 InduMar Products, Inc. has provided a unique solution with the STOP ITR PIPE REPAIR SYSTEM for pipe leak repair that has won wide acceptance.

The STOP IT PIPE REPAIR SYSTEM provides a method to restore product flow in minutes, instead of the many hours required to cut out a pipe section, fabricate and install a new piece. STOP IT’s properties allow it to stop leaks in about 30 minutes and it has a wide range of chemical resistance. STOP IT can be found in most industries including refineries, chemical plants, pulp and paper mills and power plants.

The introduction of Stop It HPT for leak repair and rehabilitation of high-pressure pipe systems operating up to 2,000 psi has expanded the company’s leak sealing capabilities far above the 400 psi of the original Stop It Pipe Repair System.

The company’s recent alignment with Amcorr’s VISCOTAQ line of anti-corrosion sealants allows us to provide the latest technology for use in the corrosive environments our customers operate in.

InduMar is continuously developing new solutions for leak and corrosion problems. Let us know about your leak or corrosion problems.

Stopit HPTM repair of methane pipeline operating at 550 psi

CONTACT INFORMATION 3355 W Alabama St Suite 110 Houston, TX 77098 viscotaqhp@indumar.com

Select 172 at www.HydrocarbonProcessing.com/RS SPONSORED CONTENT

HYDROCARBON PROCESSING | TERMINALS AND STORAGE 2013

T–95

Select 54 at www.HydrocarbonProcessing.com/RS

CB&I

CB&I PROVIDES GREATEST EXPERIENCE OF ANY TANK CONSTRUCTION COMPANY CB&I is the most complete energy infrastructure focused company in the world and a major provider of government services. Drawing upon more than a century of experience and the expertise of approximately 50,000 employees, CB&I provides reliable solutions while maintaining a relentless focus on safety and an uncompromising standard of quality.

CB&I combines proven process technology with global capabilities in engineering, procurement and construction to deliver comprehensive solutions to customers in the energy and natural resource industries. With premier process technology, proven EPC expertise, and unrivaled storage tank experience, CB&I executes projects from concept to completion.

With over 46,000 tanks built in more than 100 countries, CB&I has accumulated more storage design and construction experience than any other organization in the world. In addition to being a leader in engineering, procurement, fabrication and construction of storage tanks, we have also designed and built more than 100 storage terminals. We have the capability to design and install pipelines for these facilities, as well as other ancillary equipment.

Many customers draw upon this knowledge and extensive construction experience early in a project’s development, enabling us to provide project-specific solutions that deliver maximum long-term value, lower up-front costs, and shorter schedules.

Safety is a core value at CB&I, and we are proud to have one of the best safety records in the industry. Throughout our organization, every employee worldwide is committed to safe work practices. Our award winning safety program promotes a culture of involvement and dedication with a goal of zero incidents for everyone involved in our projects.

SPONSORED CONTENT

CONTACT INFORMATION 2103 Research Forest Drive The Woodlands, TX 77380 USA Tel: +1 832 513 1000 Fax: +1 832 513 1005 info@cbi.com www.CBI.com

HYDROCARBON PROCESSING | TERMINALS AND STORAGE 2013

T–97

www.ConstructionBoxscore.com

N O O S G N I M CO The Redesigned and Enhanced Construction Boxcore Database • Details for thousands of active, global construction projects in the refining, petrochemical, gas processing and LNG industries • Streamlined Quick Search Functionality that filters projects by region, project type, time and company • Contact information for key personnel • The Boxscore Update e-newsletter sent weekly with new projects and updates • NEW: Customized Reports Logon to www.ConstructionBoxscore.com and discover how the HPI’s most trusted source of construction data just BECAME EVEN MORE POWERFUL! For more information, contact Lee Nichols, Data Director, at +1 (713) 525-4626

Welcome to the NEW Construction Boxscore Database

Refining Developments T. TEMUR, M. HAKTANIR, F. UZMAN, and M. KARAKAYA, Turkish Petroleum Refineries Corp.

(TĂźpraĹ&#x;), Kocaeli, Turkey; and A. K. AVCI, BoÄ&#x;aziçi University, Istanbul, Turkey

Optimize vacuum ejector operations Vacuum distillation unit (VDU) designs generally involve jet ejectors due to their low capital and operating expenses, high reliability, durability and simplicity. Once the ejector is installed and commissioned, it may never require any maintenance as long as it is operated within its design limits. The major drawbacks associated with using ejectors are the dramatic performance losses with deviations from operating condition limits and low thermal efficiency.1 Ejector models, found in the literature, were developed for design purposes that estimate the cross-sectional areas of the converging and diverging sections or their respective ratios, and for determining intermediate pressures and velocities.1â&#x20AC;&#x201C;4 Upon validation against field measurements, these models can be used to evaluate the performance of installed ejectors and enhance operating efficiency.

Case history. Simulation and evaluation of ejectors and field trials have focused on a wet-fuel vacuum unit with precondensers at TĂźpraĹ&#x;â&#x20AC;&#x2122; Izmit refinery. The ejector system comprises of three-stage parallel ejectors, a precondenser, an intercondenser and an aftercondenser, as shown in FIG. 1. The ejectors at each stage are designed to accommodate one-third and two-thirds of the load. Motive steam to the ejectors is supplied directly from the low-pressure steam header at a nominal pressure of 3.5 kgf / cm2. Due to numerous branches from the refinerywide steam header, instantaneous pressure variations are unavoidable; thus, the ejectors are subjected to unregulated motive steam flow with pressure generally above the ejector design value of 3.5 kgf /cm2. A practical procedure for estimating cost savings via regulation of the steam pressure is evaluated in the presented case study. Annual savings for a 44,500-bpd vacuum unit is estimated to reach $526,000.

Procedure. Prior to the field trials, theoretical evaluations of the

ejector performance were conducted to screen the conditions at which the ejector will function properly. The principal condition is that motive steam reaches supersonic velocities (Ma > 1) in the nozzle throat.5 As shown in FIG. 2, the evaluation task is initiated by validating a mathematical model that predictsâ&#x20AC;&#x201D;along the length of the ejectorâ&#x20AC;&#x201D;the pressure and velocity profiles, either of the motive steam or of its mixture with the entrained vapor from the vacuum tower. The validation step is based on comparing steam consumption by the actual ejector system and

those calculated by the model, which is easy to apply and capable of accurate predictions. If the discrepancies between the model results and time-averaged process data are high, then the model must be revised by progressive relaxation of certain assumptions.

Basic model. The one-dimensional (1D) model assumes that

mixing the motive steam and entrained vapor occurs at constant pressure.1 The other major assumptions forming the basis for the calculations are:1 â&#x20AC;˘ The mixing pressure is equal to the saturation pressure of steam at Tmotive. â&#x20AC;˘ Flow along the ejector is adiabatic. â&#x20AC;˘ Expansion of motive steam in the nozzle is isentropic. â&#x20AC;˘ Isentropic expansion coefficient is constant. â&#x20AC;˘ Motive steam and the mixture follow ideal gas behavior. â&#x20AC;˘ Velocity of the entrained vapor is zero. â&#x20AC;˘ Outlet velocity of the mixture is negligible.

The steam flowrate through each ejector is calculated as: mmotive kg s

A1 Pmotive RTmotive

1 2

n

1 â&#x20AC;&#x201C;1

(1)

Proposed pressure control system

1st-, 2nd- and 3rdstage ejectors Precondensers

A PI

B

C

D

E

F

PI PI

PI PI

Vacuum distillation column

PI

H6 Surge drum

Steam header

Offgas

Barometric seal drum

FIG. 1. Process flow diagram of the three-stage ejector system.

Hydrocarbon Processingâ&#x20AC;&#x201A;|â&#x20AC;&#x201A;MAY 2013 99

Refining Developments where Tmotive is given by the Antoine equation (Eq. 2) for temperature: ⎤ ⎡ 3,892.7 ⎥ −273.15 (2) Tmotive ⎡⎣ C ⎤⎦ = ⎢⎢ 42.678− ⎥ ln P 1,000 −9.486 ( ) motive ⎥⎦ ⎢⎣ The compression ratio, γ, and the nozzle efficiency, ηn , are assumed to equal to 1.31 and 0.85, respectively.1,6 The crosssectional areas of the nozzle throats differ from stage to stage, but are identical for a pair of parallel ejectors. Position indicator 1 (FIG. 3) denotes the midpoint of the nozzle throat, while 2 denotes the nozzle outlet. When comparing the model results with the actual consumption data for validation, consistency leads to the next step. However, high discrepancy requires modifying the existing model, which is beyond the scope of this article.

1 A2 = Ma22 A1

⎡ 2 ⎛ γ −1 2 ⎞⎤ ⎢ ⎜1+ Ma2 ⎟⎟⎟⎥ ⎢ γ +1⎜⎝ ⎠⎥⎦ 2 ⎣

(γ +1)/(γ –1)

(3)

Eq. 3 can be solved for the Mach number at the outlet (Ma2) since both the throat and outlet diameters are known from the equipment data sheets. The velocities of the motive steam into the ejectors can be calculated using the flowrates obtained from Eq. 1. The throat velocities are then interpolated linearly between the values at the ejector inlet and nozzle outlet. If Ma1, the throat velocity, is less than the critical value of 1, the ejector shows a poor performance, which leads to loss of vacuum at the top of the tower.

Minimum steam pressure. Once the relationship between

Calculate steam consumption by the ejectors for various Pmotive (basic model) Revise the ejector model, e.g., raise the hierarchy

the motive steam pressure and nozzle throat velocity is established, in practice, it remains a trial-and-error procedure to determine the minimum motive steam pressure at which the critical throat velocity can be reached, and to realize the operational cost savings if steam from the header had been regulated prior to the ejectors. Note: A safety margin of Ma = 0.05 is allowed to prevent potential malfunctioning of the ejectors during the field trials.

Model validation

Do the model predictions compare with average process data?

No

Critical throat velocity. The ratio of the cross-sectional area of the nozzle outlet (position 2 in FIG. 3) to that of the nozzle throat (position 1) can be expressed as:

Yes Read the steam header pressure, Pmotive Pmotive = Pheader Calculate total steam consumption at Pmotive by the ejectors

Optional Determine the nozzle throat velocity in the 3rd-stage ejectors for Pmotive (CFD model)

Stop!

Minimum Pmotive has been reached. Further reduction in No motive steam pressure can lead to loss of vacuum at the tower top

Is throat velocity greater than Ma = 1?

Reduce the current trial pressure: Pmotive = Pmotive –0.1 kg/cm2

Determine the nozzle throat velocity in the 3rd-stage ejectors for Pmotive (Basic model)

Yes

Is throat velocity greater than Ma = 1.05?

Motive steam

Yes

No Field trial at this Pmotive is allowed. Calculate actual steam consumption by the ejectors Calculate the operating cost savings by reducing the motive steam pressure

FIG. 2. Procedure for determining the cost savings by working at the minimum possible motive steam pressure of the ejectors.

100 MAY 2013 | HydrocarbonProcessing.com

Entrained vapor

1.4 1.3 1.2 1.1 1.0 0.9

Velocity, Ma

Calculate the inlet velocity of motive steam into the 3rd-stage ejectors for Pmotive

2 1 0

Pressure, kgf/cm2

Throat velocity estimation

Optional rigorous path. The discussed model is based on 1D gas dynamics. The obtained results are expected to shed light on how the actual ejector system configuration should be tested in the field. Conversely, when verified by experimental data, a computational fluid dynamics (CFD) model of a system is a very useful tool for testing and analyzing complex geometric designs under a variety of operating conditions.

Apart from the basic model, the CFD model of the ejector is constructed to predict the nozzle throat velocity by providing a detailed CAD drawing, and the inlet steam velocity and outlet pressure as boundary conditions. In the ejector, the velocity and pressure profiles of motive steam and entrained vapors can be obtained over the entire domain by solving the compressible turbulent Navier-Stokes equations. Depending on the complexity of the geometry, the grid resolution and the underlying physics (e.g., high Mach number flows), solution methods for these sets of equations may demand too much

4 3 2 1 0

2

1

2

3

4

3 4 Location along the ejector

5

5

6

FIG. 3. The basic model results for the pressure and velocity profiles along the ejector H6 for different motive steam pressures.

Refining Developments

validate the model. Steam header pressure and total consumption trends from Jan. 1, 2012, to Aug. 30, 2012, are shown in FIG. 4. For an average pressure of 4 kgf /cm2, the model forecasts the actual consumption rate within 4% error. On the day of the field trial, the header pressure over a 3-hour period was 3.7 kgf / cm2, which equates to a total steam consumption rate of 13.5 tph. The model prediction was 7.5% off the actual consumption rate, which is acceptable, considering the numerous assumptions made during development and possible orifice-flow measurement errors. TABLE 1 summarizes the model findings and actual figures.

The critical velocity at the nozzle throat of ejector H6 (FIG. 1) was estimated using Eq. 3 and the iterative procedure listed in FIG. 2 for motive steam pressures of 3, 3.5 and 4 kgf /cm2.

The Mach number, which was fairly unaffected by reductions in pressure (see FIG. 3), was above the critical value of 1. Thus, it is possible to operate the ejectors at pressures around their minimum design value of 3 kgf /cm2. Decreasing the pressure below 3 kgf /cm2, however, led to throat Mach numbers less than the critical velocity, and, thus, a loss of vacuum at the top of the tower.

Findings of the basic model were verified by the CFD model, and the results are shown in FIG. 5 for motive steam pressures of 4, 3.5, 3 and 2.8 kgf /cm2. Nozzle throat velocities for the first three pressures that were above the critical Mach number of 1—while going below 3 kgf /cm2—pulled the velocity below sonic levels. The CFD model considered only motive steam flow through the ejector, since reaching supersonic speeds at the nozzle throat is the most important criterion for a successful ejector operation.5 The suction section was not included in the computational domain.

Economic gain through regulation. From the model-

ing studies and field trials, it is apparent that regulating the pressure of motive steam leads to considerable reduction in TABLE 1. Comparison of model results and actual steam consumption rates Model results Motive Model results for total Actual steam pressure, for consumption at consumption, consumption, Error, tph tph % kgf /cm2 ejector stage, tph

3.7 *

4 *

1

2

3

2

4.2

7.3

13.5

14.6

7.5

2.1

4.5

7.9

14.4

14.9

4

Year-long (2012) average of steam header pressure

5.0 4.9 4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 3.8 3.7 3.6 3.5 11/3/2011 12/23/2011 2/11/2012

16

15

14

13

Total steam consumption, tph

Case study. The time-averaged-process data can be used to

Although the motive steam pressures below 3 kgf /cm2 were anticipated by the models to be detrimental to operation, the system was technically analyzed by the engineers and operators. It was concluded that drawbacks resulting from vacuum loss for a very short time would be minimal. Therefore, the steam pressure through ejector H6 was further reduced to 2.4 kgf /cm2 by tighter restriction of valve F, which, however, led to a dramatic 16% increase in the column top pressure from 45 mm Hg to 52 mm Hg. Restriction on valve F was immediately lifted, and the pressure increased back to 3 kgf /cm2. Once again, the model prediction for the system was correct.

Steam header pressure, kgf/cm2

computational power and time.7 However, pertaining to the experience of the authors, any ejector system can be dealt within no more than a few seconds, using a now-standard desktop PC with 2 GB of RAM.

Field trials. In the Tüpraş refinery, the system considered items A through F, as shown in FIG. 1. The gate valves admit 12 steam to the ejectors from the header, and the valves cannot 4/1/2012 5/21/2012 7/10/2012 8/29/2012 10/18/2012 be used to regulate steam flow during normal operation. To FIG. 4. 2012 trends of steam header pressure and total steam prevent an irreversible upset of the vacuum system, only the consumption by the ejector system. valve (item F in FIG. 1) that is associated with the smaller of the third-stage ejectors (H6) was restricted with extreme care to reduce the steam pressure in accordance with the results of the basic and CFD models. All the other five ejectors were subjected to steam at the header pressure (3.7 kgf /cm2 on average). Upon the reduction from 3.7 kgf /cm2 to 3 kgf /cm2, the lower limit of safe operation, the column vacuum was actually increased by 3 mm Hg (FIG. 6), while total steam consumption rate decreased by 0.2 tph. The vacuum increase is explained by the relief of the aftercondenser loading as the amount of steam FIG. 5. Velocity profiles in the ejector and nozzle throat (insets) obtained with the CFD model for various motive steam pressures. condensed is reduced.

Hydrocarbon Processing | MAY 2013 101

Refining Developments TABLE 2. Operational cost savings upon regulation of the motive steam pressure Motive steam pressure, kgf /cm2

3.51 2

3

Steam consumption at ejector stage, kg/h 1

2

3

1,896

3,986

7,026

1,651

3,471

6,117

Total reduction, tph

A1 A2 mmotive Ma1 Ma2 Pmotive R Tmotive

n

Cost savings, $million/yr

NOMENCLATURE Cross-sectional area at the nozzle throat, m2 Cross-sectional area at the nozzle outlet, m2 Motive steam flowrate through the ejectors, kg/s Mach number at the nozzle throat Mach number at the nozzle outlet Pressure of motive steam, kgf /cm2 Universal gas constant, kJ/kg-K Temperature of motive steam, °C Compression ratio Nozzle efficiency

LITERATURE CITED El-Dessouky, H., et al., “Evaluation of steam jet ejectors,” Chemical Engineering and Processing, June 2002, Vol. 41, pp. 551–561.

2 Keenan, J. H. and E. P. Neumann, “A simple air ejector,” Journal of Applied Mechanics, Vol. 9, No. 2, pp. A75–A81.

3 Huang, B. J., et al., “A 1D analysis of ejector performance,” International Journal of Refrigeration, May 1999, Vol. 22, pp. 354–364.

4 Aly, N. H., A. Karmeldin and M. M. Shamloul, “Modeling and simulation of steam jet ejectors,” Desalination, January 1999, Vol. 123, pp. 1–8.

5 Lieberman, N. P. and E. T. Lieberman, A Working Guide to Process Equipment, 2003, 2nd Ed., McGraw-Hill, New Jersey.

6 Eames, I. W., S. Aphornaratana and H. Haider, “A theoretical and experimental study of a small-scale steam jet refrigerator,” International Journal of Refrigeration, June 1995, Vol. 18, pp. 378–385.

7 Nazarov, M. and J. Hoffman, “Residual-based artificial viscosity for simulation of turbulent compressible flow using adaptive finite element methods,” Numerical Methods in Fluids, No: 10.1002/fld.3663.

1

Reduction 1 2 3

245

515

909

1.67

0.5263

Design value From the basic model Based on the year-long average of cost of steam generation, $36/t

Vacuum column top pressure, mm Hg

Total steam consumption, tph

15.5 15.0 14.5 14.0

Pmotive = 3 kgf/cm2

Pmotive = 2.4 kgf/cm2

Pmotive = 3.7 kgf/cm2

13.5 13.0 54 52 50 48 46

TOLGA TEMUR is a senior process engineer and has been with Tüpraş since 2008. He is responsible for the crude units and vacuum processes. Mr. Temur also has experience in cracking and hydrodesulfurization processes. He holds a BS degree in chemical engineering and an MS degree in fuel and energy technologies from Boğaziçi University.

44 42 40 2:52 pm 3:21 pm 3:50 pm 4:19 pm 4:48 pm 5:16 pm 5:45 pm 6:14 pm 6:43 pm 7:12 pm 7:40 pm

FIG. 6. Variation of total steam consumption by the ejectors and the vacuum column top pressure by the pressure of motive steam into ejector H6.

consumption without impacting the vacuum system. From FIG. 1, installing a control valve on the steam header before it branches off to the ejectors is necessary to maintain pressure at 3 kgf /cm2.

The economic consequence of such a commonplace operation is immense compared to its simplicity. Operational cost savings for the vacuum system by the reduction and regulation of the motive steam pressure is summarized in TABLE 2. Consumption rates at 3.5 kgf /cm2 are values at the minimum design pressure, and are obtained from the equipment data sheets; whereas, rates at 3 kgf /cm2 are calculated using the basic model. The projected reduction in steam consumption for the ejector system under consideration is 1.67 tph, and its economic payoff is $526,000/yr. Remember: This systematic demonstration of the concept that even the most steadfast of the refinery equipment should be evaluated for potential performance increases and cost reductions. When the evaluation task in the field is expensive and poses risks to unit operations, even the simplest process models can assist engineers or unit operators develop a safe procedure for the specific evaluation.

102 MAY 2013 | HydrocarbonProcessing.com

MERT HAKTANIR is a senior process engineer at Tüpraş and has four years of engineering and troubleshooting experience in catalytic cracking and crude/vacuum distillation units. His current interests include upgrading and conversion of heavy crudes and shale oils, and detailed kinetic modeling of complex reaction systems. Mr. Haktanir holds a BS degree in chemical engineering from Boğaziçi University, and MS degree in chemistry from Purdue University. At present, he is pursuing a PhD in process and reactor design from Istanbul University.

FIRAT UZMAN joined Tüpraş in 2010, and is the senior R&D engineer with expertise and focus on modeling, control and optimization of refinery processes. His previous career experience includes tenure in the automotive industry. Mr. Uzman holds a BS degree in mechatronics engineering from Sabanci University, and is pursuing an MS degree in industrial engineering from Koç University.

MUSTAFA KARAKAYA, PhD, is an R&D engineer with Tüpraş.

He received BS and MS degrees and a PhD in chemical engineering from Boğaziçi University. His professional and research interests include CFD design and analysis of catalytic reactors, development of GTL catalysts, and modeling and optimization of hydrotreating processes.

AHMET K. AVCI, PhD, has received BS and MS degrees and a PhD in chemical engineering from Boğaziçi University. From 2003 to 2005, he worked as the R&D manager for Procter & Gamble. In 2006, he joined Boğaziçi University as a full-time faculty member. Assoc. Prof. Avci’s research interests are focused on catalytic hydrogen and synthesis gas production and conversion technologies, and on intensification of catalytic reactors by microchannel technology. He is the leader of numerous projects funded by governmental research institutes and the industry and is the author of more than 20 articles published in international journals.

Expert Engineering.

Proven Results.

We are never satisfied.

At Waukesha Bearings Corporation™, our forward thinking team is committed to leading the markets we serve through application knowledge, superior technology and operational excellence.

Our bearing and seal solutions are engineered to meet the unique needs of our customers, giving us the broadest application experience in oil & gas, power generation, and industrial markets. Our global businesses are customer focused – in everything we do – by delivering excellence in technology, product performance, and customer service and support.

Count on our team. We’re there when you need us.

www.waukeshabearings.com/HP Select 74 at www.HydrocarbonProcessing.com/RS

33RD ANNUAL

International Operating Conference & Trade Show

JUNE 3-5, 2013 HOUSTON, TEXAS

Hilton Americas-Houston George R. Brown Convention Center

KEYNOTE SPEAKER

Cal Ripken, Jr. • Baseball’s Iron Man and Hall of Famer

Retired from baseball in October 2001 after 21 seasons with the Baltimore Orioles, Cal Ripken, Jr. was entered in the record books as one of only seven players in history to

Develop your technical expertise in terminal operations by attending focused presentations and workshops.

CONFERENCE SESSIONS INCLUDE:

Tank Overfill Protection: Evaluating the Revised API 2350 Standard

achieve 400 home runs and 3,000 hits.

Cherished by fans around the globe as baseball’s “Iron Man,” in 1995 he broke Lou Gehrig’s record for consecutive games played, and voluntarily ended his streak in 1998 after playing 2,632 consecutive games. This is regarded as one of

Mechanical Integrity Programs: The Case for a Proactive Approach The Panama Canal Expansion and Other Critical Issues for U.S. Ports

the single greatest accomplishments in sports history. Ripken’s name has become synonymous with strength, character, endurance, and integrity. His philosophy of working hard, playing with passion,

America’s Second Industrial Revolution: The Renaissance of U.S. Chemical Manufacturing

and enjoying the game has made a tremendous impact on the sport and on fans everywhere. ★

Plus!

Mitigating Cyber Threats: Tools for Improving Computer and Network Security

13 conference presentations will be conducted in Spanish

Ensuring Due Diligence in Dealing with Vacuum Truck Hazards

info@ilta.org | +1-703-875-2011 | www.ilta.org Select 59 at www.HydrocarbonProcessing.com/RS

Safety/Loss Prevention J. THARAKAN, Suncor Energy Products, Calgary, Alberta, Canada

Flare header failure: An investigation A 60-in. flare header suffered three identical failures with buckling and cracking at the circumferential weld at the bottom of the pipe. The failures are believed to have originated during hot reliefs.

The flare header handles liquid, vapor and gas intermittently relieved from various process units. The flare header runs horizontally for a length of 3 km and has a built slope of 1 in 500 for free draining. The hottest relief is from the coker unit, which could be at a maximum temperature of 815°F. The flare header was designed to ASME B 31.3, non-severe cyclic service with a design pressure of 50 psig at 815°F. It was fabricated using 0.375-in. thick carbon steel plate with nonpost-weld-heat-treated welds. The header was also heat traced and insulated.

FIG. 1 depicts the failure locations. All failures were within about 60 ft on either side of the tie-in from the sulfur unit (sulfur lateral).

FIG. 2 shows a typical buckling and cracking failure at the bottom of the circumferential weld on the flare header.

Points to ponder. At the time of investigation, the failed areas were enclosed with leak containment devices (LCDs) installed to arrest leaks on a temporary basis. Therefore, closer observation of the failure wasn’t possible.

Buckling occurs due to compressive stress (or compressive strain) and cracking is caused by tensile stress. Therefore, it appeared strange how both could occur at the same location. Another boggling question was which occurred first, the buckling or the cracking?

The welds are, in general, stronger than the pipe. Why would the girth weld area buckle, as opposed to the pipe itself?

Why were all the failures occurring at the vicinity of the sulfur lateral? The buckling at the bottom section of the pipe could Sliding support (typ.)

Axial restraint

Sulfur relief

Vertical restraint

be associated with a bending of the pipe in the vertical plane.

The pipe is designed to move in the horizontal plane during thermal growth. Why should it move in the vertical plane?

Movement of the pipe in the vertical direction can result if there is a temperature difference between the top and bottom of the pipe. Therefore, it was suspected that there could be some insulating deposit at the bottom of the flare header near the sulfur lateral.

Onstream inspections. Gamma ray scanning ruled out the

presence of internal deposits in the flare header. Scanning results from the welds and the pipe near the sulfur lateral eliminated environmental cracking and thinning.

Visual inspection of the 60-in. header revealed: • The pipe developed an ovality due to a support reaction at the vicinity of the saddle supports

FIG. 2. Typical buckling and cracking failure at the bottom of the circumferential weld on the flare header.

Guide

Failure-1

Failure-3

To flare knockout pot

Failure-2

110 ft 60-in. flare header

From coker unit

FIG. 1. Failure locations: All failures were within about 60 ft on either side of the tie-in from the sulfur unit (sulfur lateral).

FIG. 3. Evidence of vertical movement of the pipe.

Hydrocarbon Processing | MAY 2013 105

Safety/Loss Prevention • The pipe was not touching one of the bottom supports (1in. gap), due to a permanent deformation in the vertical plane • The lone Y- stop that prevented upward movement of the line was forced open (see FIG. 3) • There was no ovalization at the long radius bends • There was no sign of excessive axial or lateral movement • There were no fretting marks on supports, that could be associated with vibration.

The first three observations led to the conclusion that the flare header must have undergone thermal movement in the vertical plane.

0.07000 0.06000

Strain limit. When a pipe is bent by the application of an ex-

0.05000

Єc

0.04000 0.03000 0.02000 0.01000 0.00000 -0.01000

Design observations. Several design features were observed during the investigation. The saddle support for the 60-in. flare header had an angle of contact of 72°. This is less than the minimum angle of contact of 120° for saddles for horizontal pressure vessels. The flare header was not designed for hydro-fill conditions. Some pipe spans that exceeded 40 ft, coupled with deficiency in saddle design, caused high local stress and ovalization of the pipe near supports, as confirmed with stress analysis.

The pipe thickness (t) was too thin for its outside diameter (D), i.e., it had a large D/t ratio of 160, as against the industry norm of ≤120. In FIG. 4, it can be seen that when the D/t ratio increases, the allowable strain limit decreases.

0

50

100 D/t ratio

150

200

FIG. 4. D/t ratio vs. strain limit, Єc.

ternal moment, it tends to develop changes in cross-section.

The outer radius develops flattening and inner radius kinks inward (buckling).

Strain limit is often used to assess the bending capability of pipe. As shown in FIG. 4, the strain limit decreases with an increase in the D/t ratio. The strain referred to here is the mechanical strain that produces stresses. Free thermal expansion produces thermal strain without stressing the material. In a piping system, restraints always limit free thermal expansion; therefore, some mechanical strain is also induced during thermal expansion.

Gresnigt’s equations are the basis of FIG. 4.1 Єc = 0.5 ⫻ t/(D-t) – 0.0025 for (D-t)/t < 120 Єc= 0.2 ⫻ t/(D-t) for a (D-t)/t ≥ 120

MOVE OVER GC

AIM

HIGHER

THE FUTURE OF ON LINE GAS ANALYSIS IS HERE

HIT YOUR FURNACE COMBUSTION TARGETS WITH THE SERVOTOUGH LASERSP

EFFECTIVE COMBUSTION CONTROL IS CRITICAL TO EFFICIENT ETHYLENE PRODUCTION.

The NEW SERVOTOUGH SpectraScan

With fast response TDLS technology that monitors closer to the burn, the SERVOTOUGH LaserSP’s unique flue gas analysis solution reduces fuels costs and CO2 emissions. You can’t get a faster response.

The SERVOTOUGH SpectraScan is a breakthrough in measurement speed for light hydrocarbons C1-C5+.

A stand-alone solution with superior technology and benefits, the SpectraScan changes everything.

FIND OUT HOW YOU COULD IMPROVE THE EFFICIENCY OF ETHYLENE PRODUCTION

LEARN MORE ABOUT THE REVOLUTION IN HYDROCARBON GAS ANALYSIS

SERVOMEX.COM/LASERSP

SERVOMEX.COM/SPECTRASCAN

DOWNLOAD THE WHITE PAPER FROM:

OR TO REQUEST ONE, EMAIL: LASERSP@SERVOMEX.COM

DOWNLOAD THE WHITE PAPER FROM:

OR TO REQUEST ONE, EMAIL: SPECTRASCAN@SERVOMEX.COM

S E RVO M E X .CO M

106 MAY 2013 | HydrocarbonProcessing.com

Select 175 at www.HydrocarbonProcessing.com/RS

A MEASURABLE ADVANTAGE

Safety/Loss Prevention

Thermal bowing. The flare header is designed for flexibility at 815°F. As per the design, it is expected to move only in the horizontal plane while undergoing thermal expansion. A movement in the vertical plane has to counter the gravity loads; however, the several conditions have led to thermal bowing of the flare header in the vertical plane: • Occasional temperature difference between the top and bottom half of the horizontal flare header • Ovalization of the pipe, which reduced the stiffness of the pipe in the vertical direction, compared to the horizontal direction • To accommodate growth, the pipe tends to bend in the direction of least stiffness.

The reason for the circumferential thermal gradient is not well understood. Within the horizontal flare header, there could be a two-phase flow of hot fluid, with the film-heat transfer coefficient being different for the liquid flow and the vapor flow.

This difference can cause a temperature gradient between the bottom and top of the pipe. Another possibility is the intermittent and partial flow of liquid, hotter or colder than the mean line temperature; it could create a circumferential thermal gradient.

FIG. 5 shows thermal bowing of a pipe due to circumferential thermal gradient. The pipe will bend with outer curvature at the hotter region. The nature and extent of stresses generated is dependent on the boundary condition. In this example, the vertical restraint opposes free thermal bowing. The region of the pipe where actual displacement is less than needed for free thermal expansion would be in compression and if reversed, in tension. Therefore, the hotter half of the pipe will be under compression and the colder half in tension.

The cracking occurred after the buckling damage. For cracking, tensile stress is required, and this must have resulted during straightening of the pipe when thermal gradients receded or when reversed.

About 150 ft on either side of the sulfur lateral was deemed as the only area where all the following conditions required for local buckling co-existed: • High temperature due to downstream coker unit relief • Restraints that opposed thermal bowing • Large D/t ratio • Pre-existing ovality due to large local stresses near saddle supports.

Repairs carried out. After thorough evaluation, repair strategy improvement began. Key improvements included: Vertical restraint

D/t is not the lone parameter limiting the bending capacity of pipe. The other factors are: • Nonhomogeneity or the presence of imperfections in the material • Initial out-of-roundness • Loading conditions • Residual stress • Strength of the material in the longitudinal and circumferential directions • Shape of the stress–strain diagram.

Cold side–tensile stress

Hot side–compressive stress Sliding support

FIG. 5. Thermal bowing of pipe due to circumferential thermal gradient.

Initial conclusions. The preliminary investigation concluded that the failure was initiated by local buckling at zones of compressive strain when thermal bowing occurred in the flare header. The vertical restraint amplified the stresses at the region of Failure 1. At failure locations 2 and 3, the sulfur lateral restrained the rotation and lifting of the 60-in. header, thereby increasing the stresses.

The large pipe D/t ratio is the root cause of local buckling.

From FIG. 4, the strain limit = 0.00126 for the flare header with a D/t ratio of 160. This translates to a stress of 35.3 ksi (1 ksi = 1,000 psi), which is lower than a yield stress of 38 ksi. Imperfections and residual stress at the circumferential weld lower the strain limit, thus explaining how all three failures occurred at the circumferential welds.

Hoop tensile stress opposes inward buckling of the pipe.

The flare header operating pressure does not exceed 10 psig and this low value added negligible hoop tensile stress to counter buckling.

Select 173 at www.HydrocarbonProcessing.com/RS

107

Safety/Loss Prevention • A 300-ft section encompassing the failure was replaced with a thicker pipe • The D/t ratio for the replacement section was 96 • The saddle supports for the replaced section were redesigned with an angle of contact of 120° • The vertical restraint was modified to a sliding support • Skin thermocouples were installed at the top and bottom of the pipe • The new welds were post-weld-heat-treated to reduce residual stress and to safeguard against environmental cracking.

Metallurgical examination. A close inspection of the flare

FIG. 6. Failed section of 60-in. flare header.

header section removed for metallurgical inspection is shown in FIG. 6. Metallurgical observations included: • Metallurgy of the pipe and weld were verified and found to match with the original design specification • No weld defects were detected; two of the failures were on shop welds and one was on a field weld • No sign of fatigue or environmental cracking was found on the specimens examined • The bottom portion of the pipe, at the weld, buckled inward and the cracks originated at the buckled area • The cracking was due to ductile overload • The crack originated at the toe of the weld from the OD surface of the pipe at the buckled region.

Skin temperature readings taken from the top and bottom of the pipe after replacement revealed a circumferential ther-

GAS DESULPHURISATION WITH THIOPAQ O&G: STABLE & COST EFFECTIVE Award winning technology for sulphur removal and sulphur recovery from gas streams. Take the quick scan at www.paqell.com/scan and receive instant feedback about how THIOPAQ O&G technology matches with your desulphurisation requirements!

www.paqell.com 015-0047 Adv Hydrocabon 180x125mmn 13112012 v4 def.indd 108 MAY 2013 | HydrocarbonProcessing.com

Select 174 at www.HydrocarbonProcessing.com/RS 1

13-11-12 16:48

Safety/Loss Prevention mal gradient of 150°F. In general, the pipe is hotter at the bottom; however, in some instances, the temperature gradient is reversed, with the top of the pipe being hotter.

Discussions. Vertical movement of the pipe causing lift-off

from support will add gravity loads to the pipe. Thermal bowing will negate the free draining capability of the line, leading to localized liquid pooling and associated issues. Thermal bowing is not common with process plant piping, and when it occurs, it is a difficult problem to correct. Most failures due to thermal bowing are fatigue cracking at the circumferential welds. Buckling due to thermal bowing is extremely rare due to installing pipes with favorable D/t ratio. The 60-in. flare header had both a very large D/t ratio and a loading/displacement condition that increased vulnerability to buckling.

When large lateral displacements are imposed on piping, failure generally manifests as localized buckling. Buckling is a failure due to instability and it causes process of achieving equilibrium between external loads, internal resistance and boundary restraint. Strain-based design is typically adopted for displacement controlled designs. Examples are subsea piping or buried lines with large ground movements.

In a piping flexibility analysis, the displacement stress range is compared with the allowable stress range. This is essentially a check against potential fatigue failure due to cyclic tensile stress. Compressive stress or strain limit checks are not part of a piping flexibility analysis. Piping stress analysis softwares

treat pipe as a beam and cannot predict local buckling of the shell elements.

Large D/t ratio also increases susceptibility to failures due to acoustic induced vibration (AIV). AIV is caused by high sound pressure levels inside flare headers during significant relief scenarios. AIV failures typically develop at small bore tieins to the flare header.

Findings. If the D/t ratio of the pipe exceeds 100, these pre-

cautions apply: • When the pipeline is subjected to large bending moments, external pressure or axial compression, strain-based design/buckling assessment using finite element analysis (FEA) should be performed • Equations for stress intensity factors given in ASME B31.3 are valid only for D/t ≤ 100 • During flexibility analysis, corrected stress intensity factors estimated through FEA should be used • Flare headers should be designed with a dead load that includes one quarter full of liquid • Saddle supports for piping with a large D/t ratio require design considerations like pressure vessel saddles (Zick analysis) • The D/t ratio for flare headers should be less than 120.

1

LITERATURE CITED Gresnigt, A. M. and R. J. Van Foeken, “Local buckling of UOE and seamless steel pipes,” 11th International Offshore and Polar Engineering Conference, Stavanger, Norway, June 2011.

JULY 30–31, 2013

Norris Conference Centers – CityCentre Houston, Texas

GulfPub.com/GTL

Hydrocarbon Processing Introduces the Inaugural Gas-to-Liquids Technology Forum Gulf Publishing Company’s inaugural Gas-to-Liquids (GTL) Technology Forum will investigate the technology and trends at work as GTL usage and projects become increasingly popular. The two-day technology conference will feature industry keynotes and presentations from GTL technology experts.

GTL is poised to become an increasingly important part of the North American energy industry. As the natural gas boom in North America continues and new technologies emerge to reduce costs, company interest is increasing—and so is investment. Project announcements and planning from industry innovators like Sasol, Shell and BP are beginning to ramp up in this energy niche.

GTL Technology Forum is your chance to connect with top operators and technology leaders from across the global hydrocarbon processing industry. Sponsorship at this event will benefit the following types of companies active in the GTL sector: GTL technology providers, GTL engineering companies, major and independent operators, catalyst companies, specialized equipment manufacturers, turbine specialists and others.

For more information, contact your local Hydrocarbon Processing sales representative or Bret Ronk, publisher, at +1 (713) 520-4421 or Bret.Ronk@GulfPub.com.

Hydrocarbon Processing | MAY 2013 109

MARKETPLACE / Gerry.Mayer@GulfPub.com / +1 (972) 816-3534

&MJNJOBUF 7BMWF $BWJUBUJPO s 0LACE ONE OR MORE DIFFUSERS DOWNSTREAM OF A VALVE TO ELIMINATE CAVITATION s %LIMINATE NOISE s %LIMINATE PIPE VIBRATION s 2EDUCE VALVE lRST COSTS s 2EDUCE VALVE MAINTENANCE

s Specialty Engineering â&#x20AC;&#x201C;Static Equipment â&#x20AC;&#x201C;Rotating Equipment s Metallurgical and Materials Lab s Field Service

1,000 GPM AMINE PLANT 1250 PSI AVAILABLE NOW

Specialists in design, failure analysis, and troubleshooting of static and rotating equipment www.knighthawk.com

Houston, 4exas 4el: s s Fax: s s

More info: www.gas-corp.com/amineplant Gas CorporaĆ&#x;on of America 800-762-6015 * gascorp@wf.net * www.gas-corp.com

Select 202 at www.HydrocarbonProcessing.com/RS

Select 203 at www.HydrocarbonProcessing.com/RS

PROCESS TOOLS SOFTWARE t VIDEO t BOOKS

PEPAC 1 P R Reservoir Engineering Programs a and Economic Evaluation SSpreadsheets

#5 3ERVICES ,,#

0ARKVIEW #IR %LK 'ROVE 6LG ), 0HONE s RCRONFEL CUSERVICES NET WWW CUSERVICES NET Select 201 at www.HydrocarbonProcessing.com/RS

$POTJTUT PG SFTFSWPJS FOHJOFFSJOH programs written in Basic and three FDPOPNJD FWBMVBUJPO TQSFBETIFFUT 5IF reservoir engineering programs are linked UISPVHI POF NBTUFS NFOV BOE ĂśWF NBJO menus. However, each program is selfcontained and can be used independently.

5IF QSPHSBNT JO UIJT QBDLBHF BSF TFMFDUFE for their usefulness in day-to-day operations that fall into six main categories: t 175 QSPQFSUJFT GPS PJM XBUFS BOE HBT t 3FTFSWPJS FOHJOFSJOH t /BUVSBM HBT FOHJOFFSJOH t &DPOPNJD FWBMVBUJPO t 5SBOTJFOU XFMM UFTU BOBMZTJT t 8BUFSøPPE EFTJHO DBMDVMBUJPOT 4PGUXBSF JT BDDPNQBOJFE XJUI B comprehensive user manual.

"VUIPS . " .JBO Price: $250

Visit our website to see all that Gulf Publishing Company offers Select 204 at www.HydrocarbonProcessing.com/RS

www.GulfPub.com

SURPLUS GAS PROCESSING/REFINING EQUIPMENT

Phone: +1 713-520-4426 Email: Software@GulfPub.com

25 MMCFD x 1100 PSIG PROPAK REFRIGERATION PLANT 28 TPD SELECTOX SULFUR RECOVERY UNIT 1100 BPD LPG CONTACTOR x 7.5 GPM CAUSTIC REGEN NGL/LPG PLANTS: 10â&#x20AC;&#x201C;600 MMCFD AMINE PLANTS: 60â&#x20AC;&#x201C;3300 GPM SULFUR PLANTS: 10â&#x20AC;&#x201C;180 TPD FRACTIONATION: 1000â&#x20AC;&#x201C;25,000 BPD HELIUM RECOVERY: 75 & 80 MMCFD NITROGEN REJECTION: 25â&#x20AC;&#x201C;100 MMCFD MANY OTHER REFINING/GAS PROCESSING UNITS We offer engineered surplus equipment solutions.

Bexar Energy Holdings, Inc. Phone 210-342-7106 s Fax 210-223-0018 www.bexarenergy.com s Email: info@bexarenergy.com

Select 205 at www.HydrocarbonProcessing.com/RS

110 MAY 2013â&#x20AC;&#x201A;|â&#x20AC;&#x201A;HydrocarbonProcessing.com

"QQMJDBCMF UBY TIJQQJOH BOE IBOEMJOH BQQMZ

Simulations and Design Calculations for the Petrochemical Processing Industry

New Version! Now includes Plate Fin Exchangers. For the designer or operator in the petroleum and chemical industry.

Using over 40 years of design and operating experience by the author, the program calculates units that enable the inexperienced technical user to determine accurate design or evaluate operations.

For the design and evaluation of process equipment: t "JS XBUFS TUFBN BOE BNJOF QSPQFSUJFT t &RVBUJPO EFSJWBUJPO t &NJTTJPOT DBMDVMBUJPOT GSPN TUPSBHF BOE PS WFOUT t 'MBSF TZTUFN EFTJHO PS FWBMVBUJPO t -JOF TJ[JOH t 4FQBSBUPS 4DSVCCFS EFTJHO t 'MBTI DBMDVMBUJPOT t 5SFBUJOH 1MBOU TJNVMBUJPO .&" %&" .%&" %("

t 5&( EFIZESBUJPO TJNVMBUJPO t .PMF TJFWF EFIZESBUJPO TJNVMBUJPO t "JS DPPMFE FYDIBOHFS EFTJHO t 4IFMM BOE UVCF FYDIBOHFS EFTJHO t $FOUSJGVHBM DPNQSFTTPS EFTJHO BOE SBUJOH t 3FDJQSPDBUJOH DPNQSFTTPS EFTJHO BOE SBUJOH t $FOUSJGVHBM QVNQ EFTJHO BOE SBUJOH t 3FDJQSPDBUJOH QVNQ EFTJHO BOE SBUJOH t *OTUSVNFOUBUJPO %FTJHO $POUSPM WBMWFT 3FMJFG WBMWFT 0SJÜDF øPX BOE NFUFS SVO DBMDVMBUJPOT

t 1MBUF 'JO &YDIBOHFS "VUIPS %PO -PWF 1I % 1 & Price: $3,295

Visit HydrocarbonProcessing.com

MARKETPLACE / Gerry.Mayer@GulfPub.com / +1 (972) 816-3534

Select 206 at www.HydrocarbonProcessing.com/RS

Flexware®

CUSTOM REPRINTS

®

Turbomachinery Engineers

Turbomachinery Training Compressors Steam Turbines Gas Turbines Performance Analysis, Evaluation, Troubleshooting Problem Resolution, Case Studies, Maintenance, Reliability

Take advantage of your editorial exposure.

Train with the Best

REPRINTS ARE IDEAL FOR: Q Product announcements Q Sales aid for your field force Q PR materials and media kits Q Direct mail enclosures Q Trade shows Q Conferences

Give yourself a competitive advantage with reprints.

Call us today!

EXPERT INSTRUCTORS WITH YEARS OF GLOBAL EXPERIENCE “I feel your course is very bene¿cial because of the technical content. It is one step more advanced than a typical OEM course ...” Yuthana Preechalai, PTT Aromatics and Re¿ning, PLC, Rayong, Thailand

• • • •

FREE PERFORMANCE ANALYSIS SOFTWARE Maximize Plant Production & Reliability Optimize Your Condition Based Equipment Reliability Program Con¿rm OEM Performance Guarantee Optimize Equipment Utilization

Denver For additional information, please contact Foster Printing Service, the official reprint provider for Hydrocarbon Processing.

Call 866-879-9144 or sales@fosterprinting.com

www.Àexwareinc.com sales@Àexwareinc.com 1-724-527-3911 Select 207 at www.HydrocarbonProcessing.com/RS Hydrocarbon Processing | MAY 2013 111

MARKETPLACE / Gerry.Mayer@GulfPub.com / +1 (972) 816-3534 Providing Value through Process Expertise

Kinetics Process Improvements

SPECIALIZE IN BASIC DESIGN/REVAMPING* â&#x20AC;˘ â&#x20AC;˘ â&#x20AC;˘ â&#x20AC;˘ â&#x20AC;˘

Ammonia Plants Methanol/H2-CO Plants Re-rating Primary Reformers Propane Dehydrogenation Plants Acrylic Acid & Oxo-Alcohol Plants

WABASH SELLS & RENTS BOILERS & DIESEL GENERATORS FAST EMERGENCY SERVICE

48,&. ($6, 3UHFDVW &RQFUHWH

800-704-2002

www.wabashpower.com

* Including Process Design, Technology Evaluation, Simulation Modeling, Troubleshooting, Revamp studies, Project Cost Estimates & Customized Process Training

FAX: 847-541-1279 847-541-5600

KPI, Inc., Houston, Texas-USA; Mail: info@kpieng.com

Phone: 281-773-1629; Fax: 832-565-9360; Web: www.kpieng.com

Select 208 at www.HydrocarbonProcessing.com/RS

Select 209 at www.HydrocarbonProcessing.com/RS

0VUQFSGPSNT .FUBM 8PPE PS .BTPOSZ

Why Should You Filter Your Water?

6TFE GPS &MFDUSJDBM t .FDIBOJDBM t #MBTU (FOFSBUPST t 4VCTUBUJPOT t 6UJMJUJFT 1VNQT t $PNNVOJDBUJPOT t )B[NBU .BJOUFOBODF t 3FTUSPPNT t 4IFMUFST 'JFME 4UBUJPOT t -BCPSBUPSJFT t 4UPSBHF t &"4*Ź41"/ 3PPGT DMFBS TQBOT VQ UP Y t 4J[FT GSPN Y Y UP Y Y t *OTUBMMT JO KVTU IPVST NJOJNBM TJUF QSFQBSBUJPO t 1SF FOHJOFFSFE BTTFNCMFE BOE PVUÜUUFE t 3FTJTUBOU UP SPU XFBS øPPE CMBTU EJTBTUFST

ZZZ (DVL6HW%XLOGLQJV FRP

HP113

Select 210 at www.HydrocarbonProcessing.com/RS

L\Ze^ _hkfZmbhg k^]n\^l ma^ a^Zm mkZgl_^k kZm^ Zg] bg\k^Zl^l ma^ pZm^k ik^llnk^ ]khi makhn`a ma^ a^Zm ^q\aZg`^k Zg] ibi^l' Bg _Z\m% hg^ lmn]r aZl lahpgmaZm '))+ _hnebg` pbee bg\k^Zl^ infibg` g^^]l [r +) '

The Best Engineered Water Filteration Solution Always Costs Less

SOFTWARE t VIDEO t BOOKS The Operatorâ&#x20AC;&#x2122;s Role in Achieving Equipment Reliability Series

+/0+ L EZ <b^g^`Z ;eo]% Ehl :g`^e^l% <: 2)),- NL: !1))" ++/&*2-+ !,*)" 1,2&+1+1 ?Zq !,*)" 1,2&/101 ppp'm^de^^g'\hf bg_h9m^de^^g'\hf Select 211 at www.HydrocarbonProcessing.com/RS

Presented by Heinz Bloch Here is a three-part lecture series featuring Heinz Bloch, respected author and lecturer on process equipment reliability. Each program introduces important and proven recommendations that the operator can use to improve equipment reliability in their facility. t 1BSU Introduction and Overview, and why there is no reliability without operator involvement t 1BSU Common Misunderstandings with Equipment Reliability Impact t 1BSU Avoiding Machinery Failures

Price: $595*

Detailed and up-to-date information for active construction projects in the refining, gas processing, and petrochemical industries across the globeâ&#x20AC;&#x201A;|â&#x20AC;&#x201A;ConstructionBoxscore.com

www.GulfPub.com 1IPOF

&NBJM 4PGUXBSF!(VMG1VC DPN *Applicable tax, shipping and handling apply

112 MAY 2013â&#x20AC;&#x201A;|â&#x20AC;&#x201A;HydrocarbonProcessing.com

ADVERTISERS INDEX / HydrocarbonProcessing.com The first number after the company name is the page on which an advertisement appears. The second number, appearing in parentheses, after the company name, is the Reader Service Number. There are two ways readers can obtain product and service information: 1. Go to www.HydrocarbonProcessing.com/RS. Follow the instructions on the screen, and your request will be forwarded for immediate action.

2. Go online to the advertiser's Website listed below.

Company

Website

Page

ae Solutions.............................................................46–47

RS#

Company

Website

Page

RS#

Company

Page

Website

RS#

(165)

GE Measurement & Control, Inspection Technology ...........26

(155)

PARCOL SpA ....................................................................15

(152)

Aggreko ........................................................................ 41

(162)

GE Oil & Gas, Surface Pumping Systems ...........................34

(86)

Pentair Porous Media .....................................................60

(64)

Ametek Process Instruments ...........................................45

(164)

ARCA Regler GmbH......................................................... 72

(169)

Avondale .......................................................................90

(77)

Gulf Publishing Company Construction Boxscore .............................................T-98 Events—GTL............................................................. 109 Events—IRPC ...........................................................6–7 Marketplace........................................................110–112 Hermetic Pumpen GmbH ................................................ 32

(158)

Hoerbiger ......................................................................30

(156)

HTRI .............................................................................. 22

(154)

www.info.hotims.com/45679-165

www.info.hotims.com/45679-162

www.info.hotims.com/45679-164 www.info.hotims.com/45679-169 www.info.hotims.com/45679-77

Axens ........................................................................... 116

(51)

Babbitt Steam Specialty Co. ............................................40

(160)

BETE Fog Nozzle .............................................................84

(58)

Borsig GmbH ................................................................. 19

(153)

www.info.hotims.com/45679-51

www.info.hotims.com/45679-160 www.info.hotims.com/45679-58

www.info.hotims.com/45679-153

Cameron ....................................................................... 52

(55)

Carboline Company ........................................................88

(171)

CB&I ...........................................................................T-96

(54)

Cudd Energy Services ......................................................71

(168)

Curtiss Wright Flow Control Company, DeltaValve ................... 68

(67)

Curtiss Wright Flow Control Company, Farris Engineering.........73

(56)

Dresser-Rand.................................................................67

(62)

Elliott Group ................................................................. 115

(52)

www.info.hotims.com/45679-55

www.info.hotims.com/45679-171 www.info.hotims.com/45679-54

www.info.hotims.com/45679-168 www.info.hotims.com/45679-67

www.info.hotims.com/45679-56 www.info.hotims.com/45679-62 www.info.hotims.com/45679-52

www.info.hotims.com/45679-155 www.info.hotims.com/45679-86

www.info.hotims.com/45679-158

www.info.hotims.com/45679-156 www.info.hotims.com/45679-154

(83) (54)

Idrojet ........................................................................... 77

(170)

ILTA ............................................................................. 104

(59)

InduMar Products ........................................................T-95

(172)

Inpro/Seal, A Waukesha Bearings Business .....................103

(74)

John Zink Company ....................................................... 27

(80)

Johnson Screens ............................................................ 55

(91)

Kobe Steel Ltd................................................................48

(82)

Linde Engineering North America Inc. ..............................42

(73)

www.info.hotims.com/45679-170 www.info.hotims.com/45679-59

www.info.hotims.com/45679-172 www.info.hotims.com/45679-74

www.info.hotims.com/45679-80 www.info.hotims.com/45679-91

www.info.hotims.com/45679-82

(76)

Prosernat ......................................................................45

(163)

Quest Integrity Group LLC.................................................31

(157)

Samson GmbH ...............................................................66

(167)

www.info.hotims.com/45679-163

Hytorc ........................................................................... 37 www.info.hotims.com/45679-54

www.info.hotims.com/45679-64

Petro-Canada Lubricants ..................................................2 www.info.hotims.com/45679-76

Hydro, Inc......................................................................20 www.info.hotims.com/45679-83

www.info.hotims.com/45679-152

www.info.hotims.com/45679-157

www.info.hotims.com/45679-167

Sandvik Materials Technology ......................................... 74

(61)

Servomex Ltd............................................................... 106

(175)

Smith & Burgess LLC .......................................................83

(72)

SO.CA.P. s.r.l. ................................................................. 41

(161)

www.info.hotims.com/45679-61

www.info.hotims.com/45679-175 www.info.hotims.com/45679-72

www.info.hotims.com/45679-161

Spraying Systems Co ...................................................... 10

(66)

Summit Industrial Products, Inc. .....................................36

(159)

T.F. Hudgins, Inc .............................................................65

(166)

Team Industrial Services ................................................. 23

(95)

www.info.hotims.com/45679-66

www.info.hotims.com/45679-159

www.info.hotims.com/45679-166 www.info.hotims.com/45679-95

Tiger Tower Services ....................................................... 18

(75)

Total Safety ...................................................................56

(99)

Trachte USA ..................................................................107

(173) (151)

www.info.hotims.com/45679-75

www.info.hotims.com/45679-99

www.info.hotims.com/45679-173

Emerson Process Management, Fisher ............................. 33 Emirates .........................................................................13

(68)

Linde Process Plants ....................................................... 14

(85)

UOP, A Honeywell Company ............................................24 Vega Americas, Inc. .........................................................12

Exxon Lubricants & Specialities ........................................51

(71)

Paharpur Cooling Towers, Ltd. .........................................28

(102)

Weir Minerals Lewis Pumps ............................................. 16

(94)

Flexitallic LP ....................................................................5

(93)

Paqell ......................................................................... 108

(174)

Zyme-Flow, Decon Technology........................................89

(92)

www.info.hotims.com/45679-68

www.info.hotims.com/45679-73

www.info.hotims.com/45679-71

www.info.hotims.com/45679-93

www.info.hotims.com/45679-85

www.info.hotims.com/45679-102 www.info.hotims.com/45679-174

www.info.hotims.com/45679-151 www.info.hotims.com/45679-94 www.info.hotims.com/45679-92

This Index and procedure for securing additional information is provided as a service to Hydrocarbon Processing advertisers and a convenience to our readers. Gulf Publishing Company is not responsible for omissions or errors.

SALES OFFICES—EUROPE

Bret Ronk, Publisher Phone: +1 (713) 529-4301 Fax: +1 (713) 520-4433 E-mail: Bret.Ronk@GulfPub.com www.HydrocarbonProcessing.com

SALES OFFICES—NORTH AMERICA IL, LA, MO, OK, TX Josh Mayer Phone: +1 (972) 816-6745, Fax: +1 (972) 767-4442 E-mail: Josh.Mayer@GulfPub.com

AK, AL, AR, AZ, CA, CO, FL, GA, HI, IA, ID, IN, KS, KY, MI, MN, MS, MT, ND, NE, NM, NV, OR, SD, TN, TX, UT, WA, WI, WY, WESTERN CANADA Diana Smith Phone/Fax: +1 (713) 520-4449 Mobile: +1 (713) 670-6138 E-mail: Diana.Smith@GulfPub.com

CT, DC, DE, MA, MD, ME, NC, NH, NJ, NY, OH, PA, RI, SC, VA, VT, WV, EASTERN CANADA Merrie Lynch Phone: +1 (617) 357-8190, Fax: +1 (617) 357-8194 Mobile: +1 (617) 594-4943 E-mail: Merrie.Lynch@GulfPub.com

CLASSIFIED SALES Gerry Mayer Phone: +1 (972) 816-3534, Fax: +1 (972) 767-4442 E-mail: Gerry.Mayer@GulfPub.com

DATA PRODUCTS Lee Nichols Phone: +1 (713) 525-4626, Fax: +1 (713) 520-4433 E-mail: Lee.Nichols@GulfPub.com

FRANCE, GREECE, NORTH AFRICA, MIDDLE EAST, SPAIN, PORTUGAL, SOUTHERN BELGIUM, LUXEMBOURG, SWITZERLAND, GERMANY, AUSTRIA, TURKEY Catherine Watkins Tél.: +33 (0)1 30 47 92 51 Fax: +33 (0)1 30 47 92 40 E-mail: Watkins@GulfPub.com

ITALY, EASTERN EUROPE Fabio Potestá Mediapoint & Communications SRL Phone: +39 (010) 570-4948 Fax: +39 (010) 553-0088 E-mail: Fabio.Potesta@GulfPub.com

RUSSIA/FSU Lilia Fedotova Anik International & Co. Ltd. Phone: +7 (495) 628-10-333 E-mail: Lilia.Fedotova@GulfPub.com

UNITED KINGDOM/SCANDINAVIA, NORTHERN BELGIUM, THE NETHERLANDS Michael Brown Phone: +44 161 440 0854 Mobile: +44 79866 34646 E-mail: Michael.Brown@GulfPub.com

SALES OFFICES—OTHER AREAS AUSTRALIA—Perth Brian Arnold Phone: +61 (8) 9332-9839 Fax: +61 (8) 9313-6442 E-mail: Australia@GulfPub.com

CHINA—Hong Kong Iris Yuen Phone: +86 13802701367, (China) Phone: +852 69185500, (Hong Kong) E-mail: Iris.Yuen@GulfPub.com

BRAZIL—São Paulo Alfred Bilyk Phone/Fax: 11 23 37 42 40 Mobile: 11 85 86 52 59 E-mail: Brazil@GulfPub.com

INDIA Manav Kanwar Phone: +91-22-2837 7070/71/72 Fax: +91-22-2822 2803 Mobile: +91-98673 67374 E-mail: India@GulfPub.com

INDONESIA, MALAYSIA, SINGAPORE, THAILAND Peggy Thay Publicitas Singapore Pte Ltd Phone: +65 6836-2272 Fax: +65 6634-5231 E-mail: Singapore@GulfPub.com

JAPAN—Tokyo Yoshinori Ikeda Pacific Business Inc. Phone: +81 (3) 3661-6138 Fax: +81 (3) 3661-6139 E-mail: Japan@GulfPub.com

KOREA D. S. Chai Dongmyung Communications, Inc. Phone: +82 (2) 391 4254 Fax: +82 (2) 391 4255 E-mail: Korea@GulfPub.com

PAKISTAN—Karachi S. E. Ahmed Intermedia Communications Phone: +92 (21) 663-4795 Fax: +92 (21) 663-4795

REPRINTS Rhonda Brown, Foster Printing Service Phone: +1 (866) 879-9144 ext. 194 E-mail: RhondaB@FosterPrinting.com

Hydrocarbon Processing | MAY 2013 113

Engineering Case Histories

A. SOFRONAS, CONSULTING ENGINEER http://mechanicalengineeringhelp.com

Case 72: Interaction between disciplines when troubleshooting When troubleshooting, we can become fixated on the discipline that we are most familiar—in the author’s case, mechanical engineering. Investigating the failure of a machine may have a group of mechanical engineering specialists involved in the troubleshooting effort. When one’s specialty is vibration analysis, the first thought for the failure could be a torsional or linear vibration problem. A stress analyst will focus on developing an analytical model to describe this failure. Conversely, the materials engineer will be investigating corrosion issues as the root cause. Likewise, the process and control engineers will be reviewing computer simulations and searching for possible abnormal operations.

It’s human nature to want to analyze problems with the tools that you are most familiar in using. I certainly have done this during my career.

Solution. When the solution is obvious, then the one-discipline approach can work. A specialist may have seen a very similar failure and solved the problem. Indeed, most day-today failures are resolved by part replacement or with field personnel applying a simple fix. Replacing damaged bearings because the oil was contaminated, and correcting the root cause for the contamination, will mitigate future failures.

Silver-bullet solutions. A silver-bullet, single solution does not work well on complex multivariable system problems. Erroneous solutions can result in severe consequences to equipment, personnel and careers with the one-discipline approach.

This can be addressed by using a structured-team problem-solving approach, with all of the necessary disciplines represented and a team leader who has been selected by management. In this way, all of the potential causes can be identified, using the multi-discipline team. There is no purpose in outlining technical problem-solving methods that are well documented.1, 2 Most of us have attended seminars or have been involved in such sessions. We are familiar with the benefits.

Follow through. I have noticed during my career that there are times when the structured-team methods have not been applied correctly. There are many reasons for the lapse in application and include: • A deadline is approaching, and it is imperative that this deadline be met. Result: The uninformed make critical decisions.

• No one knows that such an approach will help, and no one has been trained for it.

114 MAY 2013 | HydrocarbonProcessing.com

• The problem solving is well under way before it is realized that the approach is chaotic, and that a more structured method is needed.

• There is not much urgency in solving the problem; no one wants to spend any money, delegate personnel or be responsible.

• The solution to the problem is started with no action plans and no definite responsibilities.

• Various disciplines that are needed to solve the problem are not available at the site.

Who is responsible? It is management’s job to address all of these listed reasons. The question becomes: When is it necessary to utilize a structured-team approach to solve a problem?

From my viewpoint, when safety, litigation concerns, major production losses or careers may be at risk, then a structured-team approach is needed. Many excellent articles have been written on when just making a repair is inadequate and more should be done.3 A quick computer search on engineering disasters will show many such failures, and, unfortunately, most were avoidable. There are many reasons to consider structured-team problem-solving approaches, and training in root-cause analysis methods, such as the Kepner-Tregoe or TapRooT methods.

Due to the present litigious society, it is very prudent to invest in methods that produce a “paper trail” and show that a sincere and thorough effort was made to address the concern.

Detailed and concise calculations provide a method to document that a sincere effort has been made to understand the concern. Other ways are to include the input of well-known technical experts familiar with the equipment or process involved and have experience with similar failures.

LITERATURE CITED Bloch, H. P. and F. K. Geitner, Machinery Failure Analysis and Troubleshooting, Gulf Publishing Company, Houston, p. 343.

2 Kepner, C. H., and B. B. Tregoe, The Rational Manager: A Systematic Approach to Problem Solving and Decision Making, Kepner-Tregoe, Second Ed., 1976.

3 Bloch, K., “Extreme failure analysis: Never again a repeat failure,” Hydrocarbon Processing, April 2009, pp. 87–97.

1

DR. TONY SOFRONAS, P.E., was worldwide lead mechanical engineer for ExxonMobil Chemicals before retiring. He now owns Engineered Products, which provides consulting and engineering seminars on machinery and pressure vessels. Dr. Sofronas has authored two engineering books and numerous technical articles on analytical methods. Early in his career, he worked for General Electric and Bendix, and has extensive knowledge of design and failure analysis for various types of equipment.

Q

Customer:

Q

Challenge:

Q

Result:

Oil refinery, Russia.

High reliability in extreme operating conditions.

Elliott centrifugal compressors have operated for more than twenty years without overhaul.

They turned to Elliott for reliable compression solutions.

Many of Russia’s largest refineries turn to Elliott compressors and turbines for critical applications such as hydrotreating and hydrocracking, fluid catalytic cracking (FCC), coking and alkylation. Who will you turn to?

EBARA CORPORATION

C O M P R E S S O R S

Q

T U R B I N E S

Q

G L O B A L

S E R V I C E

The world turns to Elliott. www.elliott-turbo.com

Select 52 at www.HydrocarbonProcessing.com/RS

Stimulate the heart of your hydroprocessing unit ImpulseTM, the catalyst technology that combines the stability you recognize with the activity you need 4JOHMF TPVSDF *40 t *40 t 0)4"4 www.axens.net Select 51 at www.HydrocarbonProcessing.com/RS