La
Metallurgia Italiana
International Journal of the Italian Association for Metallurgy
n. 04 aprile 2022 Organo ufficiale dell’Associazione Italiana di Metallurgia. Rivista fondata nel 1909
La Metallurgia Italiana International Journal of the Italian Association for Metallurgy Organo ufficiale dell’Associazione Italiana di Metallurgia. House organ of AIM Italian Association for Metallurgy. Rivista fondata nel 1909
Direttore responsabile/Chief editor: Mario Cusolito Direttore vicario/Deputy director: Gianangelo Camona Comitato scientifico/Editorial panel: Christian Bernhard, Massimiliano Bestetti, Wolfgang Bleck, Franco Bonollo, Bruno Buchmayr, Enrique Mariano Castrodeza, Emanuela Cerri, Lorella Ceschini, Vladislav Deev, Bernd Kleimt, Carlo Mapelli, Jean Denis Mithieux, Marco Ormellese, Massimo Pellizzari, Barbara Previtali, Evgeny S. Prusov, Dieter Senk, Du Sichen, Karl-Hermann Tacke Segreteria di redazione/Editorial secretary: Marta Verderi Comitato di redazione/Editorial committee: Federica Bassani, Gianangelo Camona, Mario Cusolito, Carlo Mapelli, Federico Mazzolari, Marta Verderi Direzione e redazione/Editorial and executive office: AIM - Via F. Turati 8 - 20121 Milano tel. 02 76 02 11 32 - fax 02 76 02 05 51 met@aimnet.it - www.aimnet.it Immagine in copertina: Shutterstock
Gestione editoriale e pubblicità Publisher and marketing office: siderweb spa Via Don Milani, 5 - 25020 Flero (BS) tel. 030 25 400 06 - fax 030 25 400 41 commerciale@siderweb.com - www.siderweb.com La riproduzione degli articoli e delle illustrazioni è permessa solo citando la fonte e previa autorizzazione della Direzione della rivista. Reproduction in whole or in part of articles and images is permitted only upon receipt of required permission and provided that the source is cited. Reg. Trib. Milano n. 499 del 18/9/1948. Sped. in abb. Post. - D.L.353/2003 (conv. L. 27/02/2004 n. 46) art. 1, comma 1, DCB UD siderweb spa è iscritta al Roc con il num. 26116
La
Metallurgia Italiana
International Journal of the Italian Association for Metallurgy
n. 04 aprile 2022 Organo ufficiale dell’Associazione Italiana di Metallurgia. Rivista fondata nel 1909
Editoriale / Editorial Editoriale
A cura di Christian Bernhard............................................................................................................... pag.04
Memorie scientifiche / Scientific papers Colata Continua / Continuous Casting
Technology and controls improvements on the continuous casting of Acciaierie di Calvisano for improvements on solidification process and new steel grades L. Angelini, P. Frittella, G. Tsymokh, C. Di Cecca, B. Cinquegrana, A. Milan, F.Fredi, M. Bersani, C. Senes, F. Guerra, G. Miglietta, S. Conte, V. Duro, A. Zurru, R. Muhlemann, C. Mapelli, S. Barella, L. Calligarich,
G. Galeazzi, S. Maurina, S. De Monte, C. Persi, S. Spagnul, M. Saba, G. Flor, B. Palm ............................ pag.07
n.04 aprile 2022
Anno 113 - ISSN 0026-0843
Effects of cooling methods and cooling conditions on behavior of thermal distortion and stress generation of steel blooms cast continuously on reverse transformation treatment
K. Isobe................................................................................................................................................ pag.18
Quantification of critical parameters for prediction of surface crack formation in continuous casting R. Krobath, C. Bernhard........................................................................................................................................ pag.29
Reduction of transverse corner cracks in Tata Steel’s Direct Sheet Plant in Ijmuiden "R. Kalter, J.A. Kromhout, M.B. Santillana, A.J.C. Burghardt, J. Link, C. Toeniges, E. Gillebaart,
indice
L. Koomen, G.J.C.H. Goessens, J. van ‘t Hul, S. Meijer"......................................................................... pag.39
Investigations on primary cooling in CC mould through the use of modelling approach
J-F. Domgin, S. Gauthier...................................................................................................................... pag.48
Utilizzo di un sistema di ispezione ottica automatica atto al rilevamento dei difetti di colata continua basato su algoritmi di Machine Learning per l’analisi dell’incidenza delle marche di oscillazione su bramme di acciaio inossidabile austenitico AISI 316L e 316LI
A. F. Ciuffini, F. Di Giovanni, D. Mombelli............................................................................................. pag.59
Attualità industriale / Industry news The intelligent ultra-wide caster for high-quality slabs at Rizhao Shandong
edited by : L. Fischer, Q. Zheng, D. Zhao, J. Yuan, P. Heidemann, I. Olgemöller, J. Wans, R. Wilmes....... pag.71
Powder dosing with mould temperature feedback control in continuous casting of stainless steel for high quality billet surfaces edited by: C. Scarabelli, D. Olivero, F. Bego ......................................................................................... pag.82
Automated mold flux feeders for Industry 4.0 application
edited by: M. Zinni................................................................................................................................ pag.91
Dynamic SuperHeat determination in a continuous casting machine – process, practice and benefits
edited by: P. Hughes-Narborough, P. White, G. Humphrey.......................................................................... pag.96
The mold temperature mapping with Ultrasonic Contactless Technology is the key for the real time initial solidification process control tools
edited by: I.Mazza, S. Miani, G. Schiavon, S. Spagnul................................................................................... pag.107
L’AIM va in trasferta a Bari per un seminario presso “Magna”
Intervista al CEO di Magna ing. Aldo Cirilli..................................................................................................... pag.117
Atti e notizie / AIM news Eventi AIM / AIM events......................................................................................... pag.122 Comitati tecnici / Study groups ........................................................................... pag.124 Normativa / Standards.......................................................................................... pag.126
editoriale - editorial
“
"AIM professional organization
“L'organizzazione profes-
demic caused security measu-
tamente bilanciato le severe
perfectly balanced strict pan-
res with the need for personal
exchange between the participants."
sionale dell'AIM ha perfet-
misure di sicurezza causate
dalla pandemia con la necessità di uno scambio personale tra i partecipanti”.
Assoc.Prof. Dr. Christian Bernhard University of Leoben, Leoben, Austria, EU
A LOOK BACK ON THE 10TH EUROPEAN CONTINUOUS CASTING CONFERENCE
UNO SGUARDO ALLA 10a EDIZIONE DEL CONVEGNO SULLA COLATA CONTINUA
After two pandemic postponements, the 10th European
Dopo due rinvii dovuti alla pandemia, la decima Conferen-
Continuous Casting Conference (ECCC) took place in Bari
za Europea di Colata Continua (ECCC) si è svolta a Bari dal
last October 20 to 22. The successful history of the ECCC
20 al 22 ottobre scorso. La storia di successo della serie di
conference series started in 1991 in Florence. In 2021, AIM,
conferenze ECCC è iniziata nel 1991 a Firenze. Nel 2021,
the Italian Association for Metallurgy, had the honour to
l'AIM, l'Associazione Italiana di Metallurgia, ha avuto l'o-
organize the celebration of the 30th anniversary of this pre-
nore di organizzare nuovamente in Italia la celebrazione
stigious conference in Italy again. More than 250 on-site
del 30° anniversario di questa prestigiosa conferenza. Più
participants and 23 on-site exhibitors, mostly from Euro-
di 250 partecipanti in loco e 23 espositori in loco, per lo
pean countries, provided the worthy scene for 150 spea-
più provenienti da paesi europei, hanno fornito il degno
kers in this hybrid event despite travelling restrictions. AIM
scenario ai 150 relatori in questo evento ibrido, nonostan-
professional organization perfectly balanced strict pande-
te le restrizioni di viaggio. L'organizzazione professionale
mic caused security measures with the need for personal
dell'AIM ha perfettamente bilanciato le severe misure di
exchange between the participants. The conference dinner
sicurezza causate dalla pandemia con la necessità di uno
close to the sea conveyed the feeling of recurred normali-
scambio personale tra i partecipanti. La cena della confe-
ty.
renza vicino al mare ha trasmesso la sensazione di normalità ritrovata.
La Metallurgia Italiana - Aprile 2022
pagina 4
editoriale - editorial
All'interno del programma della conferenza, sei famosi
Within the conference program, six famous plenary spe-
relatori in seduta plenaria hanno affrontato panoramiche
akers addressed overviews on product quality issues and
sui problemi di qualità del prodotto e sullo stato dell'arte
state-of-the-art of numerical simulation in the continuo-
della simulazione numerica nel processo di colata conti-
us casting process. Professor Piotr Scheller reflected the
nua. Il professor Piotr Scheller ha riflettuto sulla storia della
history of continuous casting based on the content of the
colata continua sulla base del contenuto delle conferenze
ECC conferences held between 1991 and 2017. The tech-
ECC tenutesi tra il 1991 e il 2017. Le presentazioni tecniche
nical presentations covered a wide range of topics: Surfa-
hanno coperto una vasta gamma di argomenti, focalizzati
ce quality-related issues, mold fluxes, new technological
in particolare sui problemi legati alla qualità delle superfici,
concepts, and all aspects of the digital transformation re-
i flussi in lingottiera, i nuovi concetti tecnologici e tutti gli
presented the main focus.
aspetti della trasformazione digitale. Il comitato scientifico del 10° ECCC ha selezionato dieci
The scientific committee of the 10th ECCC selected ten
presentazioni eccellenti da pubblicare in questo numero
outstanding presentations for publication in this issue of La
de La Metallugia Italiana. Gli articoli provenienti dal mondo
Metallugia Italiana. The articles from academia and industry
accademico e dall'industria forniscono uno spaccato rap-
provide a representative cross-section of the latest achie-
presentativo degli ultimi risultati nello sviluppo tecnologi-
vements in the technological development of the conti-
co della colata continua dell'acciaio.
nuous casting of steel.
Infine, vorrei cogliere l'occasione per ricordare il prossimo
Finally, I would like to take the opportunity to remember
"2nd International Workshop Surface Quality of Continuo-
the upcoming “2nd International Workshop Surface Qua-
usly Cast Products", organizzato congiuntamente da AIM
lity of Continuously Cast Products”, jointly organized by
e Austrias Metallurgical Society ASMET a Bergamo, dal 1
AIM and Austrias Metallurgical Society ASMET in Berga-
al 2 dicembre. Possiamo aspettarci un altro evento di alta
mo, December 1 to 2. We can expect a further high-quality
qualità con numerosi relatori da tutto il mondo!
event with numerous speakers from all over the world!
La Metallurgia Italiana - April 2022
pagina 5
2
convegno nazionale
trattamenti termici
26-27 MAGGIO 2022 • GENOVA
Organizzato da
MAGAZZINI DEL COTONE • PORTO ANTICO
con il patrocinio di
sponsorizzato da
FEDERATA
COMITATO ITALIANO DEI COSTRUTTORI DI FORNI INDUSTRIALI
UN SA
Group
CMYK
RGB
PANTONE
B/N
100 30 0 10
0 120 185
300 U
nero 100%
0 0 0 70
112 113 115
424 C
bianco 100%
100 72 0 18
0 68 137
280 CV
Chi è invitato a partecipare
Tutta la filiera coinvolta nel settore dei trattamenti termici dei metalli e delle leghe metalliche, ed in particolare: • gli imprenditori, i tecnici e gli operatori delle aziende di trattamento termico e di trattamento superficiale • gli accademici ed i ricercatori del settore metallurgico • i metallurgisti dei laboratori in conto terzi e dei laboratori aziendali • i produttori, i trasformatori ed i fornitori di metalli, leghe metalliche e semilavorati • gli utilizzatori (officine meccaniche, costruttori di stampi e di utensili, produttori di bulloneria, organi di trasmissione, molle, raccordi e, in genere, di tutti i componenti meccanici che necessitino di trattamenti termici).
www.aimnet.it/tt.htm Timetable GIOVEDÌ, 26 MAGGIO 2022 Registrazione dei partecipanti Cerimonia di apertura Apertura dell’area espositiva Sessioni scientifiche e tecnico-applicative Buffet di benvenuto Tavola rotonda Sessioni scientifiche e tecnico-applicative Tavola rotonda
Quote di iscrizione, programma completo, catalogo degli espositori e tutte le info relative al Convegno sono disponilbi online sul sito www.aimnet.it/tt.htm
Termine della prima giornata Cena sociale presso l’Acquario di Genova VENERDÌ, 27 MAGGIO 2022
Segreteria organizzativa
Sessioni scientifiche e tecnico-applicative Pranzo Sessioni scientifiche e tecnico-applicative
Tel. +39 0276021132 . +39 0276397770 info@aimnet.it . www.aimnet.it
Termine del Convegno
Elenco degli espositori STAND ESPOSITORE 1
2 3 4 5
6 7 8
9 10 11
AIR LIQUIDE ITALIA SERVICE GEARCHEM SRL TAV VACUUM FURNACES SPA NICRO SPA SOLIVERI SRL VACUUM SPA GF ELTI SRL TOYO TANSO GNR ECM TECHNOLOGIES GEFRAN
STAND ESPOSITORE 12
13 14 15 16
17 18 19
20 21 22
FUCHS LUBRIFICANTI SPA CIEFFE THERMAL SYSTEMS AICHELIN HOLDING GMBH TREVISAN SRL STRUERS SAS CISAM-ERNST SRL SAET SPA FLUKE PROCESS INSTRUMENTS ATTAS SRL NIPPON GASES INDUSTRIAL SRL MIT.D.
STAND ESPOSITORE 23
24 25 26 27
28 29 30 31 32 33
QUAKER HOUGHTON ITALIA SPA DANIELI CENTRO COMBUSTION SPA EASYDUR SRL MEAPFORNI SRL ICMI FORNI INDUSTRIALI HI-TECH ENGINEERING SRL STREAMLINE SRL VERDER SCIENTIFIC SRL MICROCONSULT SRL GADDA INDUSTRIE SRL
Scientific papers - Continuous Casting
Technology and controls improvements on the continuous casting of Acciaierie di Calvisano for improvements on solidification process and new steel grades L. Angelini, P. Frittella, G. Tsymokh,C. Di Cecca,B.Cinquegrana, A.Milan,F.Fredi, M. Bersani, C. Senes, F. Guerra, G. Miglietta, S. Conte, V. Duro, A. Zurru, R. Muhlemann, C. Mapelli, S. Barella, L. Calligarich, G. Galeazzi, S. Maurina, S. De Monte, C. Persi, S. Spagnul, M. Saba, G. Flor, B. Palm
In last years Feralpi Group is following a route in the direction of increase production of special steel grades for application also in automotive and for mechanical components.
The main actions adopted have been the completion of production chain of special steels coupling the steelmaking facilities
of Acciaierie di Calvisano and the rolling mill for wire rods of Caleotto (100 % acquired in 2020) included in the new Business Unit Specialties realized to promote the production of new products.
For this reason Acciaierie di Calvisano is devoted mainly to improve the product quality and the capabilities of a reliable processes realization in whole steelmaking area.
The solidification process in continuous casting of billets is the core step in this route and for this reason several technolo-
gical developments have been adopted regarding new systems to improve capabilities and devices for casting realization, for process monitoring and control to reach assessed quality aims.
To reach these targets Acciaierie di Calvisano promotes R&D projects and collaborations with technology ad research par-
tner to be supported with an high level of technical contributions in a continuous route of development starting by technological devices availability, process management and simulation/controls.
KEYWORDS: CONTINUOUS CASTING, SIMULATION, PROCESS CONTROL, SOLIDIFICATION MODEL; INTRODUCTION For the continuous casting of billets following aspects can be the origin for process or quality limitations: - breakout events and process instability (steel level in mold, temperature, steel flows….) - internal and surface cracks - porosities - Segregations - Not correct shape - Inclusions content - Scale formation
L. Angelini, P. Frittella, G. Tsymokh, C. Di Cecca, B.Cinquegrana, A.Milan, F.Fredi Feralpi Siderurgica, Italy
M. Bersani, C. Senes, F. Guerra, G. Miglietta, S. Conte, V. Duro, A. Zurru, R. Muhlemann Acciaierie di Calvisano, Italy
C. Mapelli, S. Barella
Politecnico di Milano, Dipartimento di Meccanica, Italy
L. Calligarich, G. Galeazzi, S. Maurina Visiorobotis, Italy
S. De Monte, C. Persi, S. Spagnul Ergolines Lab, Italy
For this reason, efforts on capabilities improvement should be adopted for following aspects: - To control properly the steel feeding from tundish, the liquid steel level and the casting speed depen-
M. Saba
Automazioni Industriali Capitanio, Italy
G. Flor, B. Palm
VDEh-Betriebsforschungsinstitut GmbH, Germany
ding by steel temperature in arrival it is necessary to
La Metallurgia Italiana - April 2022
pagina 7
Memorie scientifiche - Colata Continua maintain the correct metallurgical length, uniform and
and to have a feedback from the product results on-line
stable solidification conditions reducing possible local
enable the possibility to correct in real time eventual
defect formations
abnormal conditions as shapes deformations or possi-
- To correctly control the lubrication in mold with ap-
ble reducing quality conditions.
propriated powder and with correct feeding is fundamental to enable a correct solidification since the be-
CALVISANO STEELMAKING PLANT AND VIEW OF
ginning having enough shell thickness at the exit of
CONTINUOUS CASTING DEVELOPMENTS
primary cooling and to avoid steel sticking on the mold
Acciaierie di Calvisano was founded in 1972 and is located
reducing possible surface defects.
nearby Brescia as part of Feralpi Group.
- To define and control the proper parameters of mold
The plant produces quality and special steel billets, the
oscillation is important to maintain correct solidifica-
intermediate product used for making rebar and wire rod.
tion front formation and to avoid steel sticking on the
The process route at Calvisano steel plant includes an AC
mold
EAF, LF and CCM.
- To control heat flux in mold by primary cooling it is
Continuous Casting Machine of Calvisano is characteri-
necessary to avoid breakout conditions enabling a
zed by 4 casting strands that produce High, Medium and
good surface shell thickness and a proper internal soli-
Low-Carbon Steels billets with different section dimen-
dification formation.
sions, typically from 120x120mm to 160x160mm. Conti-
- To control the secondary cooling depending by ste-
nuous Casting Machine is characterized by presence of:
el temperature in arrival it is necessary to maintain the
• ladle, tundish, mold, SEN (for special steel grades),
uniform and stable solidification conditions to avoid lo-
• primary water cooling system, secondary nozzle spray
cal high reheating (cause of internal cracks) to enable a
cooling system, air cooling bed,
correct liquid pool closure avoiding porosities.
• M-EMS, S-EMS, F-EMS,
- To control and manage the internal liquid pool throu-
• withdrawal rollers and cutting system.
gh Electromagnetic stirring or soft plastic deformations
New installations and developments have been perfor-
it is necessary to obtain homogenous internal solidifi-
med mainly within R&D projects as: SteelPro4.0, Sup-
cation reducing eventual porosities and segregations.
port Cast and Quality Integration (Figure 1).
- To have the correct view of whole process parameters
Fig.1 - R&D project supporting roadmap continuous casting improvements in Acciaierie di Calvisano.
La Metallurgia Italiana - Aprile 2022
pagina 8
Scientific papers - Continuous Casting With the upcoming of new measurement systems tech-
nitoring
nologies and their actual combination, a new approach is
- Application of new systems for process management
under development in Calvisano to improve the process
and control
quality in Continuous Casting.
- Decision level (alert and guidelines for Decision Sup-
The developments have also been supported by a preli-
port System).
minary phase of solidification process off-line simulation
This approach with the main scopes to implement sy-
taken by Politecnico di Milano in different configurations
stems ford:
in order to assess in optimal way the casting configu-ra-
- process and quality improvements; as application of
tion for the new devices.
final stirring for internal segregations reduction and to
Acciaierie di Calvisano is following development of the
guarantee internal and shape quality
global route in figure 2 including as main points:
- process control and results stability also through pro-
- Technological improvements on the devices for li-
cess and product monitoring;
quid pool stirring and mold management
- process data analysis and decision support to enable a
- Application of new technologies for casting billet mo-
route of continuous improvement.
Fig.2 - View of the control system including on-line solidification model coupled with relevant detections.
ACTIONS FOR PROCESS AND QUALITY IMPROVE-
Ultrasonic Level Detection system for process improve-
MENTS
ment
The main actions reported for process and quality impro-
In order to support the on-line process monitoring and
vements are:
control the continuous casting machine has been pro-vi-
- Application of Ultrasonic Level Detection system for
ded with a new Ultrasonic Level Detection (ULD) system
process improvement through automatic powder fee-
developed by Ergolines (Figure 3) to obtain:
ding in mold
- On-line monitoring of the covering powder thickness
- Application of Final Stirrer to enable reduction of in-
on the meniscus
ternal segregation
- Automatic powder feeding - Mold thermal mapping
La Metallurgia Italiana - April 2022
pagina 9
Memorie scientifiche - Colata Continua The quality of casting billets is influenced by the fluid flow
absorption of non-metallic oxide inclusions.
of powder in the mold, particularly at the meniscus. In
The new ULD system it is able to measure the powder
order to minimize inclusion entrapment, it is especially
thickness in mold based on temperature measurement
important to keep at the setting values the liquid steel le-
by ultrasonic signals, by this measurement and having
vel in the mold and the powder feeding rate. As the solid
set a reference value the automatic control of powder is
powder is added on top of the mold, the powder melts
enabled to maintain the stable value of powder thickness
generating a liquid slag pool first and a thin, vertical film
fixed as target so the powder film can be maintained also
of liquid slag then. The final effect consists of a protection
when changes occur on actual process parameters.
the steel meniscus from oxidation, providing thermal in-
The system is applied for the 4 casting strands and fun-
sulation, preventing solidification of the steel surface and
ctionalities are continuously on improvement.
Fig.3 - Scheme of the application of the Ultrasonic Level Detection system.
Application of final Stirrer to enable reduction of internal segregation
The installation of Final EMS Stirring equipment required
the best configuration determination first, then a single line trial installation till prosecution with the completing of the whole 4-line casting machine tooling.
To take into account physical, chemical, electrical and magnetic phenomena, a conspicuous and deep studying phase has started with PoliMi and Ergolines. From Calvisa-
no side, the improvement of the internal quality through a mitigation of carbide and sulfide segregation phenomena between dendrites is one of projects ac-tivities focuses.
The optimization of quality corrective action consists simultaneously both of:
La Metallurgia Italiana - Aprile 2022
- identification of the optimal electromagnetic stirrer
frequency and current to be applied to different steel grades and billet sections;
- identification of the optimal positioning on Continuo-
us Casting machine, considering the present plant constraints.
Steel temperature, liquid diameter estimation and effect of final EMS on liquid steel velocity has been calcu-lated by
PoliMi and Ergolines (Figures 4, 5). For the simulation of Electric and Magnetic module and Turbulent flow a model
with liquid pool and coils of electromagnetic stirrer has been designed. Induction current gener-ated acts with
the magnetic field of the inductor to induce electromagnetic Lorentz force in the molten steel. Along the center
pagina 10
Scientific papers - Continuous Casting line of the liquid pool a Lorentz force up to 240 N/m3 is
of final EMS on liquid steel velocity at billets core at con-
of liquid steel and generates a torque effect that gives to
current density has been simulated as shown in Figure 5.
produced. The same force works on every unit of volume the liquid steel a global rotational movement. The effect
dition of induced magnetic field at 4.8×10^7 A/m surface
Fig.4 - Steel temperature and liquid diameter estimation by predictive solidification modelling.
Fig.5 - Simulation of Lorentz force and effect of final EMS on liquid steel velocity at billets core.
Main benefits obtained with application of final EMS have been (Figure 6) :
- reduction of inner porosity
- reduction of inner carbide and sulfide segregation - improved homogeneity and hardenability
- possibility of increase of casting speed and productivity.
Fig.6 - View of reduction of porosity in the center area of billets with final EMS.
La Metallurgia Italiana - April 2022
pagina 11
Memorie scientifiche - Colata Continua
ACTIONS FOR PROCESS AND PRODUCT MONITORING
Thermal mapping of the billets
dback from product detection to support both:
of defect detection has been developed by BFI coupling
For the control of the process, it is important a direct fee- Deeper understanding of phenomena and relation between results and process management
- Have faster alerts in case of abnormal behaviors and final results on the product
In this way the main billets detection on development are:
- Thermal mapping of the billets for surface defects detection during casting realized with BFI
- Surface defect detection based on laser scanning and deep learning realized with Visiorobotics.
- Shape measurement based on double laser scanning realized by Ergolines/AIC
- Solidified Shell Thickness measurement realized by
In the frame of the RFCS project SupportCast a system
thermal detection through Thermal Camera and a software for on-line data analysis realized in order make evident
eventual surface crack presence in case of abnormal temperatures variability (Figure 7) .
Experimental trial campaign and surface temperature
values have been registered and postprocessed. The system is under testing to improve performances of defect detection and in particular, a defect due to steel level variation in mold has been detected as reduction of local
average temperature. Moreover, an external longitudinal crack (Figure 8) has been detected on the cooled casted billet at room temperature.
Ergolines/AIC
Fig.7 - View page of surface defect detection and thermal mapping realized by BFI.
Fig.8 - Evidence of defect detection obtained through thermal mapping of BFI.
La Metallurgia Italiana - Aprile 2022
pagina 12
Scientific papers - Continuous Casting
Surface defect detection based on laser scanning and deep learning
A system for surface defect detection has been developed
by Visiorobotics within the SteelPro4.0 project using a laser scanning and signal postprocessing for surface status representation (Figure 9).
As off-line application the system has been used to eva-
luate the accuracy in detection of relevant defects. In particular, the laser scanning is able to measure the surface
profile through line scanning with a resolution around
0,2 mm than signals acquired are coupled to show a representation of the billets surface through combination of subsequent profiles. Billets with relevant defects have been used for trials, and capability to detect some defects
In on-line application following steps are available (Figure 10):
- Processing of the data are realized online in real time during the hot billets casting process.
- Surface conditions are shown during the casting in order to support the operators when defects detected
By this approach it has been shown the capability of the systems to detect surface defects while subsequent trials will be necessary to improve the evaluation of smaller defects (till 0,2 mm), selecting areas of scale presence,
through a training system in order to alert in case of defect detected.
has been shown when surface aren’t covered by scale.
Fig.9 - Offline detection with laser scanning and surface representation through Deep Learning.
Laser
Surface
Fig.10 - On-line laser scanning and billets surface reconstruction through Deep Learning.
La Metallurgia Italiana - April 2022
pagina 13
Memorie scientifiche - Colata Continua
Shape measurement by laser scanning and solidified shell thickness measurement
Solid skin thickness measuring system, laser geometry measurement system and Final EMS Stirrers were also designed and tested with the partners Ergolines and AIC.
rement of solid steel phase thickness,
- improve the control of the solidification process, the microstructure of billets and consequentially the quality of the billets product.
A solid skin thickness measuring system, positioned as
The system for laser geometry measurement has been
EMS Stirrer, has been tested in Calvisano plant (Figure 2).
Support System (DSS) in order to:
first trial configuration in correspondence of the Strand The aim of the new system is to send feedback to the solidification simulator in order to:
- calibrate the model parameters based on the measu-
designed (Figure 11) to send feedback to the Decision - alert and recognize cases of abnormal casting conditions
- discard abnormal billets for subsequent billets welding process.
Fig.11 - Geometry measurement system designed output.
SOLIDIFICATION SIMULATIONS AND CC PROCESS
cooling and to liquid/solid transformation (latent heat).
As by the global scheme in figure 2 an off-line solidifica-
section of the billet along casting line
DATA ANALYSIS SYSTEMS
tion model has been implemented by Visiorobotics based on mathematical modelling provided by Polimi (in the fra-
me of the project SteelPro4.0) in order to collect data with
the objective to give indication about the status of the
solidification evolution for the different casting strands in real time.
The mathematical approach realizes a thermodynamic si-
mulation of the solidification evolution during casting taking into account the process parameters as:
- Conditions of steel from tundish (Temperature, Steel grade, liquidus and solidus temperatures) - Casting speed
- Water flow rate and temperature in primary cooling
- Distribution of water flow rate in secondary cooling (spay nozzels)
- Temperature evolution and thermal mapping of each - Liquid, solid and mushy phases mapping.
Relevant parameters taken into account for results validation and as alerts functions are (Figure 12):
- the metallurgical length of the complete solidification of the caster strand.
- the solidified thickness and liquid diameter in relevant sections of the casting line
- the whole thermal mapping of the billets in all sections
The system is improved with continuity and next steps of development will be:
- Adaptation of the simulator to different steel grades of special steels;
- Representation of the effects with different spray nozzles in secondary cooling.
Based on these data are given as output of the simulation of:
- heat fluxes between the hot steel and the cooling sy-
stem resulting in enthalpy decrease due to the water
La Metallurgia Italiana - Aprile 2022
pagina 14
Scientific papers - Continuous Casting
Fig.12 - Scheme of calculation for Solidification off-line model.
The online application follow the general scheme in figure 2 including following steps (Figure 13):
• On-line casting parameters acquisition
• Relevant parameters representation
• Postprocessing and process guidelines and alerts.
• Thermal mapping measurement
The on-line application of solidification model has been
• Comparison of solidification estimation with real
the on-line detection and alert systems are under deve-
• On-line defects detection on-line detections
realized and the whole configuration and coupling with lopment as shown in Figures 13, 14.
Fig.13 - Simulator interface for the single strand. The main functions required to Decision Support System are:
• To Give an overview of billets solidification conditions along the casting line for each strand (Figure 13)
• Autocalibration of solidification modelling thanks to detection of temperature, shell thickness
• Obtain alerts in case of abnormal casting conditions
La Metallurgia Italiana - April 2022
respect (Figure 14):
- Acceptability ranges of casting parameters previously defined
- Casting length, shell thickness, Temperature simulated in reference positions
• Obtain alerts of defects detection thanks to the measurement systems in terms of (Figure 14):
pagina 15
Memorie scientifiche - Colata Continua
- Defects detected by thermal mapping - Shell thickness detected
- Laser surface defect detection
- Laser geometry defects detection
• To suggest guidelines for actions in terms of billets
inspection, level quality assignment or casting parameters modification.
support to on-line process monitoring
- Availability of on-line systems for data collection analysis and solidification process simulation that can
be used for advanced process monitoring and as Decision Support to solidification correction
- Increase of internal competences on solidification, mathematical simulation and process control
The main results aimed by the developments shown in Acciaierie di Calvisano are:
- Reduction of internal defect on billets for special steel in terms of porosity and segregation
- Availability of detection systems that can be used as
Fig.14 - Simulator interface with heat data, simulation output, measurement systems and alerts.
CONCLUSIONS
Acciaierie di Calvisano and Feralpi Group are following a route of strong increase of production of special steel grades with necessity to increase product quality and capabilities of a reliable process.
In this way the application of advanced strategies Industry 4.0 are strongly followed including:
- application of innovative on-line monitoring systems and defect detection
- Application of digital twin strategies with on-line solidification process simulation
- Integrate on-line detection systems with simulation approach to reach advanced capabilities of solidifi-
ca-tion understanding as support to availability of ad-
La Metallurgia Italiana - Aprile 2022
vanced alerts and decision support
These applications and strategies are also the conditions to improve product quality as support to further
development of production of special steel as in industrial strategy of Acciaierie di Calvisano.
For this reason, improvements and adaptations on con-
tinuous casting machine are on-going to implement
further plant sections and devices able to realize and control the solidification process in adapted way as necessary for each steel grade.
To reach these targets Acciaierie di Calvisano follows
R&D projects and collaborations with technology ad
research partner to be supported with a high level of technical contributions in a continuous route of deve-
pagina 16
Scientific papers - Continuous Casting
lop-ment starting by technological devices availability, process management and simulation/controls.
In this way, also the collaborations with research partners as Politecnico di Milano and Visiorobotics has been useful support to realize the mathematical model for solidification simulation used to define optimal pro-
cess configurations for internal quality improvement while the collaboration with BFI is running for surface defect monitoring.
Furthermore, the partnership with a technology deve-
loper as Ergolines, Visiorobotics, and Automazioni Industriali Capitanio enabled developments of innovative systems for defect detection on billets during casting.
The mentioned developments have been realized in
the R&D project “Support Cast” supported by the EU
funding scheme Research Fund for Coal and Steel and in the R&D project “SteelPro4.0” supported by the funding scheme “Accordi per l’Innovatività” promoted by Regione Lombardia.
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
Salandin, F. Chitarin G. (2010-2011). Sistemi per il rimescolamento elettromagnetico di compositi me-tallici allo stato fuso. Università degli Studi di Padova. Mazza, I.; Spagnul, S.; Mantovani, F.; 2015, Review of the Mold Powder Control Technology, Pro-ceedings of METEC 2015, Düsseldorf, Germany Mazza, I.; Spagnul, S.; 2015, A Novel Ultrasonic Sensor for Mold Powder Thickness Control, Proceed-ings of METEC 2015, Düsseldorf, Germany. Olivo, L.; Spagnul, S.; Mazza, I.; 2016, “A New Optical System for Mold Powder Thickness Control by Laser Scanning and MultiSpectral Imaging”, Iron & Steel Technology, Vol. 13, No. 12, pp. 62-69. Mazza, I.; Spagnul, S.; Olivo, L F. Milani, Review of Technologies and Methods for Mold Powder Thickness Control, La Metallurgia Italiana - n. 4 2017 Ma, L., Spagnul, S., & Soleimani, M. (2017). Metal Solidification Imaging Process by Magnetic Induc-tion Tomography. Scientific Reports. Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2018, Contactless Mold Thermal Mapping at Meniscus through an Innovative Ultrasonic Sensor, Proceedings of ICS 2018, Italy, 13-15 June 2018 Li, F., Spagnul, S., Odedo, V., & Soleimani, M. (2019). Monitoring Surface Defects Deformations and Displacements in Hot Steel Using Magnetic Induction Tomography. Sensors, 19(13), [3005] Spagnul S., Mazza I., Miani S., Schiavon G.. METEC Conference 2019 Contactless Mold Thermal Mapping: A New Tool for Metallurgists, Quality Control and Productivity Improvement, January 2020, Conference: AISTech 2020 Authors: Mazza I., Miani S., Spagnul S., Schiavon Bergmann P., Fauser M., Sattlegger D., Steger C., MVTec AD - A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection, MVTec Software GmbH Bergmann P., Fauser M., Sattlegger D., Steger C., Uninformed Students: Student - Teacher Anomaly Detection with Discriminative Latent Embeddings, MVTec Software GmbH Bergmann P., Lowe S., Fauser M., Sattlegger D., Steger C., Improving Unsupervised Defect Segmen-tation by Applying Structural Similarity To Autoencoders
La Metallurgia Italiana - April 2022
pagina 17
Memorie scientifiche - Colata Continua
Effects of cooling methods and cooling conditions on behavior of thermal distortion and stress generation of steel blooms cast continuosly on reverse transformation treatment K. Isobe
In recent years, the treatment of reverse transformation has been adopted in the HCR process of CC-Blooming for steel production to prevent surface cracking on blooming. However, the quenching of bloom in this treatment occasionally causes some troubles, such as occurrence of thermal distortion of bloom and quenching cracks. On the behaviour of thermal distortion of bloom and the generation of stress in cross section of bloom, the effects of kinds of cooling methods (immersion, spray, mist-spray cooling) and cooling conditions for reverse transformation treatment, were analysed in order to prevent these troubles by a model of metallo-thermo-mechanics in this study. The following results were obtained from these analyses. In the several cooling methods and cooling conditions, the time required for the reverse trans- formation treatment under various cooling conditions were clarified. The effects of the three types of cooling methods and various cooling conditions on the cross-sectional shape of the cast bloom after cooling and the stress generated during cooling were analysed from the viewpoint of the several kinds of transformation behaviour and thermal shrinkage. The optimum cooling methods and necessary cooling conditions for shortening the required cooling time and preventing the occurrence of quenching cracks were revealed from the analyses.
KEYWORDS: REVERSE TRANSFORMATION TERATMENT, CONTINUOUS CASTING, HCR PROCESS, METALLO, THERMO MECHANICS, THERMAL DISTORTION, STRESS GENERATION; INTRODUCTION In recent years, the treatment of reverse transformation has been adopted in the HCR process of CC-Blooming for steel production to prevent surface cracking on blooming. This treatment is very effective for toughening of the bloom surface of steel and prevention of occurrence of cracks on hot rolling by refinement of austenite grains (1)-(4). However, the quenching of bloom in this treatment occasionally causes some troubles, such as occurrence of thermal distortion of bloom and quenching cracks. The causes and mechanisms of the occurrence of thermal distortion and quenching
Kohichi Isobe
National Institute of Technology (KOSEN), Gifu College, Japan
cracks haven’t been revealed sufficiently. The behavior of thermal distortion of bloom and the generation of stress in cross section of bloom causing quenching cracks by third cooling for treatment of reverse transformation are affected by kinds of third cooling methods (immersion, spray, mistspray cooling) and cooling conditions for reverse transformation treatment. La Metallurgia Italiana - Aprile 2022
pagina 18
Scientific papers - Continuous Casting
In this study, the causes and mechanisms of thermal distor-
tion analysis with heat transfer analysis and analysis of phase
tion and quenching crack and the effects of kinds of cooling
transformation behavior are necessary to examine thermal
methods and cooling conditions of reverse transformation
distortion and stress generation considering these mu-
treatment were analyzed in order to prevent these troubles
tual effects. Therefore, the behavior of thermal distortion
by the model of metallo-thermo-mechanics (5), (6). The co-
and stress generation were analyzed by the model of me-
oling methods considered in these analyses were immer-
tallo-thermo-mechanics shown in Fig.1(5), (6): COSMAP
sion cooling, spray cooling and mist-spray cooling.
(7) in this study. Thermo-elastic-plastic analysis, analysis of heat transfer and analysis about transformation behavior
ANALYSIS METHOD
are combined in this model considering of thermal shrinka-
ANALYSIS MODEL
ge and expansion, shrinkage and expansion accompanying
Since the distribution of temperature and stress/strain and
transformation. The effects of cooling methods and cooling
their transition and behavior of phase transformation affect
conditions on the behavior of thermal distortion and stress
mutually during third cooling, analyses combined distor-
generation were studied by this model.
Fig.1 – Schematic view of the model of metallo-thermo-mechanics. (5)
ANALYSIS CONDITIONS
el SCr420(JIS), and the transformation behavior of this steel
The analyses were performed using a finite element method
type was estimated in consideration of the CCT diagram me-
two-dimensional model for total cross section of a bloom
asured in a coarse austenite structure similar to that of blo-
having a square cross section of 200 mm (thickness) x 200
om as cast continuously (8).
mm (width). The analyzed steel type was case hardening ste-
In these calculations, the solidification calculation was not
La Metallurgia Italiana - April 2022
pagina 19
Memorie scientifiche - Colata Continua
performed, but the calculations were carried out under the
measured for the upper surface or lower surface of steel
following conditions. The entire bloom was once heated to
bloom of rectangular cross section, used in the analyses of
1123K, and then was cooled under various cooling methods
the case of mist-spray.
and cooling conditions, until the austenitic phase in the sur-
It’s found that the H during immersion cooling is higher as
face layer of the bloom in the range of 10 mm was transfor-
the stirring is stronger, and H of water jet stirring is higher
med into a low-temperature phase for the refinement of au-
than that of spray cooling or mist cooling in a wide tempe-
stenite structure by reverse transformation. Therefore, the
rature range. H is increasing as increasing of water density in
calculation was performed under the condition of cooling
the case of spray and mist-spray cooling. At the same water
until the volume fraction of austenite in the above range be-
density, H of mist-spray cooling at the lower surface of blo-
came less than 0.1. In addition, calculations were performed
om is a little smaller than H of spray cooling at high tempe-
assuming uniform and non-uniform cooling conditions,
rature range. In the case of mist-spray cooling, H of upper
and the effects of non-uniform cooling conditions on the
surface of steel bloom is greater by the static water on the
cross-sectional shape of the bloom and the stress genera-
surface than H of lower surface in some temperature ranges.
tion behavior during the cooling were discussed. The kinds of third cooling method analyzed in this study, were immersion cooling, in the case of static water, air stirring, and water jet stirring, spray cooling which is performed at a water density of 20, 50, 100 (l/(m2• min)) and mist-spray cooling which is performed at a water density of 20, 50(l/(m2• min)). The bloom surface is treated as a boundary of heat transfer and the heat transfer coefficients (H) in each cooling method and each cooling condition were estimated by the relationship between the surface temperature and H shown in Fig.2 (9)-(11). The heat transfer coefficients, which were
Fig.2 –Relations between surface temperature and heat transfer coefficients of immersion cooling, water spray cooling and mist-spray cooling (on upper surface and under surface).
La Metallurgia Italiana - Aprile 2022
pagina 20
Scientific papers - Continuous Casting
Tab.1 - Analysis conditions in each case.
UNIFORM/NON-UNIFORM
COOLING METHOD
COOLING CONDITIONS (HEAT TRANSFER COEFFICIENTS) Static water
Immersion
Air stiiring 500~1150(l/min) Water jet stirring W=1200 (l/(m2•min))
Uniform Cooling
W=20 (l/(m2•min)) W=50 (l/(m2•min))
Spray
W=100 (l/(m2•min)) W=20 (l/(m2•min)), Lower
Mist-spray
UNIFORM/NON-UNIFORM
NON-Uniform Cooling
W=50 (l/(m2•min)), Lower
COOLING METHOD
UPPER SURFACE
OTHER SURFACES
Immersion
Static water
Water jet stirring
Spray
W=20 (l/(m2•min))
W=100 (l/(m2•min))
W=20(l/(m2•min)), Upper
W=20 (l/(m2•min)), Lower
W=50(l/(m2•min)), Upper
W=50 (l/(m2•min)), Lower
Mist-spray
RESULTS AND DISCUSSION
time decreased with the increase of the cooling strength
te grains in the surface layer of the bloom within a range
cooling, and the required cooling time of the spray coo-
The cooling time required for the refinement of austeniof 10 mm by the reverse transformation treatment for va-
rious cooling methods and cooling conditions was clari-
fied by the calculations. Fig. 3 is a bar graph showing the results of estimated the cooling time required for refining
austenite grains. The required cooling time was determi-
ned as a time at which the volume fraction of austenite was 0.1 or less within a range of 10 mm from the surface of the bloom. As shown in this figure, the required cooling
together with the uniform cooling and the non-uniform ling was shorter than that of the mist-spray cooling, and
the required cooling time of the immersion cooling was
shorter than the spray cooling. In spray cooling or mist
cooling, the required cooling time was shortened as the water density increased, and in immersion cooling, the
heat transfer coefficient could be increased as the stirring intensity was increased, so that the required cooling time was shortened.
Fig.3 – Required cooling time for refinement of austenite grains of bloom surface region by reverse transformation treatment.
La Metallurgia Italiana - April 2022
pagina 21
Memorie scientifiche - Colata Continua
The effects of cooling methods and cooling conditions
on the bloom cross-sectional shape at the end of cooling were studied using the model of metallo-thermo-me-
chanics. As examples, bloom cross-sectional shapes and
contour diagrams of volume fraction of martensite and
bainite and pearlite at end of cooling by uniform immersion cooling with water jet stirring and uniform mist-spray cooling with water volume density of 50 (l/(m2•min)) are
shown in Fig.4, Fig.5. In the following figures of cross section of bloom, the total displacement is shown to be 40 times the actual value so that the effect of deformation
by cooling can be more easily understood. Also, higher colors in the color bar correspond to higher values.
In the case of uniform immersion cooling, the heat tran-
sfer coefficient is larger in a wider temperature range than
tensite generation rate in the case of immersion cooling relatively increased at the corners in the bloom cross section as shown in Fig. 4(a), so the shape of the bloom
corner portion after cooling was changed to the shape protruding outward and the central portion of the surface
was depressed. The bainitic and perlitic transformation didn’t occur in this case(Fig.4(b)).On the other hand, in the case of mist-spray cooling (Fig.5), the amount of marten-
sitic transformation was also larger at the corner side than
at the center of the cross section, but the bainitic and perlitic transformation proceeded at the inside of the cross section during cooling, so that the expansion amount accompanying the transformation is large at the region, so the central part of surface of bloom had a swelling shape.
spray cooling and mist cooling. For the reason, the mar-
(a)Martensite volume fraction
(b) Bainite and pearlite volume fraction
Fig.4 – Cross-sectional shape of bloom and distribution of volume fraction of martensite, bainite and pearlite at the end of uniform immersion cooling. (Water jet stirring)
Fig.5 – Cross-sectional shape of bloom and distribution of volume fraction of martensite, bainite and pearlite at the end of uniform mist-spray cooling. (W=50 l/(m2• min)), lower))
La Metallurgia Italiana - Aprile 2022
pagina 22
Scientific papers - Continuous Casting
The result of comparing the height difference of the surface
and that the height difference of surface became larger as
ling were discussed. From the result shown in Fig.6, it was
stirring in the case of immersion cooling, and that the lower
of bloom at the end of cooling in the case of uniform coofound that the height difference of the upper surface in the
several cases of uniform cooling is about 0.06 to 0.2 mm,
the cooling intensity increased with increasing power of
the water density of spray and mist-spray cooling, the slightly higher the height difference of bloom surface.
Fig.6 – Difference in height of surface of bloom at the end of uniform cooling by several kinds of cooling methods. Fig. 7 and Fig.8 show the cross-sectional shapes of the blo-
of water jet stirring was given to the other three surfaces. In
bainite and pearlite at the end of non-uniform cooling. Fig.
transfer coefficient of the upper surface at a water density of
om and contours of the volume fraction of martensite or 7 shows the results in the case of non-uniform immersion
addition, In the case of mist-spray cooling (Fig.8), the heat
20 (l /( m2•min)) was given only to the upper surface side,
cooling, and Fig. 8 shows the results in the case of non-uni-
and the heat transfer coefficient of the lower surface at a wa-
(Fig.7), the heat transfer coefficient of static water was given
surfaces.
form mist-spray cooling. In the case of immersion cooling only to the upper surface, and the heat transfer coefficient
Fig.7 – Cross-sectional shape of bloom at the end of non-uniform immersion cooling.
(Upper: Static water, Others: Water jet stirring) La Metallurgia Italiana - April 2022
ter density of 20 (l /( m2•min)) was given to the other three
Fig.8 – Cross-sectional shape of bloom at the end
of non-uniform mist-spray cooling. (Upper: W=20(l/ (m2•min), Others: W=20 (l/ (m2•min)), lower)
pagina 23
Memorie scientifiche - Colata Continua
In both the immersion cooling and the mist-spray coo-
surface side was estimated to be due to the increased the
ling, it was found that the bloom width was reduced on
amount of thermal shrinkage caused by the higher coo-
the upper surface side as compared with the lower surfa-
ling strength on the upper surface side.
ce side. In the case of immersion cooling, the martensi-
Fig. 9 is a bar graph showing the height difference on up-
tic transformation amount and expansion amount of the
per, lower and right surface of the bloom at the end of
transformation until the end of cooling do not show a lar-
non-uniform cooling. Since the sides are deformed sym-
ge difference between the upper side and lower surface
metrically, only the values on the right side are shown. In
side, but the cooling strength on the upper surface side
each case, the cooling intensity was changed on the upper
is lower and the temperature on the side of the bloom is
surface side compared with the other surfaces, but in such
higher than that of the lower surface side. Therefore, since
cases, the height difference on the side surface was larger
the temperature transition was high and easily deformed,
than the difference on other surface. The maximum value
it was estimated that the bloom width was reduced on the
of the height difference of each surface was the maximum
upper surface side due to an increase in the amount of de-
on the side where the water density was 20 (l/(m2•min))
formation due to thermal shrinkage of the upper surface
by mist cooling within the cases examined, and the maxi-
during cooling.
mum amount was about 0.4 mm at most. It was presumed
On the other hand, in the case of non-uniform mist-
that such a height difference of bloom surface is not signi-
spray cooling with a water density of 20 (l / (m • min)),
ficant in generating corner defects in blooming.
2
the amount of martensitic transformation is small on the both upper and lower surface side, and the amount of bainitic and pearlitic transformation and expansion amount of the transformation on the upper surface side are larger than the amounts on lower surface side. In spite of these quantitative relations, the width of bloom on upper surface region was reduced in comparison with the lower surface side, the decrease of the bloom width on the upper
Fig.9 – Difference in height of surface of bloom at the end of cooling of several kinds of non-uniform cooling methods.
La Metallurgia Italiana - Aprile 2022
pagina 24
Scientific papers - Continuous Casting
Fig.10 – Distribution of Sxx and volume fraction of
Fig.11 – Distribution of Sxx and volume fraction of
uniform immersion cooling. (Water jet stirring)
uniform immersion cooling. (Air stirring)
martensite when Sxx is maximum during
martensite when Sxx is maximum during
From the viewpoint of the prevention of the occurrence
stress Sxx became the maximum just below the center of
of quenching cracks, the behavior of stress generation in
the surface, because the austenitic phase region just be-
the cooling process by each cooling method was analy-
low the surface at the width center was pulled in the blo-
zed. Fig.10(a) and Fig.11(a) show the distributions of nor-
om width direction by the expansion due to martensitic
mal stress Sxx in the cross section of bloom in the case
transformation at the surrounding surface. On the other
of uniform immersion cooling with water jet stirring and
hand, in the case of immersion cooling with air stirring
air stirring, when the stress Sxx during the cooling is ma-
(Fig.11), the surface layer at the off-corner was pulled in
ximum. Fig.10(b) and Fig.11(b) show the distributions of
the bloom width direction due to the progress of marten-
martensite volume fraction in the cross section at that
sitic transformation inside the cross section of the corner.
time. The Sxx became maximum in the position indicated
It is considered that the maximum stress Sxx occurred at
by the circle. In these cases of uniform immersion coo-
the position, because of the martensitic transformation,
ling, the amount generated bainite and pearlite phase is
the effect of thermal shrinkage and the large deformation
very small, and it is considered that the transformation
resistance at the region caused by the low temperature.
expansion due to martensitic transformation near the surface of bloom and thermal stress cause the generation of maximum Sxx in these cases. In the case of immersion cooling with water jet stirring (Fig.10), the Sxx became maximum in the austenitic phase region just below the surface at the center of the surface. It was seemed that the
La Metallurgia Italiana - April 2022
pagina 25
Memorie scientifiche - Colata Continua
Fig.12 – Distribution of Sxx and volume fraction of bainite and pearlite when Sxx is maximum during uniform mist-spray cooling. (H: W=20 (l/(m2• min)), lower)
Fig.13 – Distribution of Sxx and volume fraction of
Fig.14 – Distribution of Sxx and volume fraction of
uniform spray cooling. (W=100 (l/(m • min)))
non-uniform spray cooling. (Upper: W=20
martensite when Sxx is maximum during 2
martensite when Sxx is maximum during
(l/(m2• min), Lower, Side: W=100 (l/(m2• min))
Fig. 12 shows the distribution of Sxx when the maximum
seen that in this case, martensitic transformation does not
Sxx occurs and the distribution of bainite and pearlite vo-
undergo during the cooling and bainitic and pearlitic tran-
lume fraction in the case of uniform mist-spray cooling
sformation occurs over a large area of the cross section.
with the water density of 20 (l/(m • min)), the weakest co-
As a result, it was found that the expansion due to these
oling intensity. From the analysis for this case, it can be
transformations expanded the upper and lower surface
2
La Metallurgia Italiana - Aprile 2022
pagina 26
Scientific papers - Continuous Casting
layers and increased Sxx at the surface.
ximum Sxx occurs differ greatly, and the maximum Sxx
When the water density on the bloom upper surface side
greatly increased in the case of non-uniform cooling with
is 20 (l/(m • min)), and the other three surfaces are cooled
strong cooling intensity.
by spray at 100 (l/(m • min)), distribution of Sxx and mar-
The maximum Sxx during cooling in various cooling
tensite volume fraction in the cross section of bloom are
methods and conditions is shown in the bar graph of
shown in Fig.14 in comparison with the case of cooling
Fig.15. In the uniform immersion cooling, the maximum
uniformly on all four sides with water density of 100 (l/(m •
Sxx was almost constant regardless of the stirring power
min))(Fig.13). In the case of uniform spray cooling with 100
of the immersion bath. In the uniform cooling with spray
(l/(m • min) shown in Fig.13, the maximum Sxx occurred
and mist-spray, Sxx decreased as the water density de-
on the off-corner surface due to expansion accompanying
creased. In this study, in spray cooling and mist cooling,
martensitic transformation at inside near the corner. In
there are some cases where non-uniform cooling signi-
the non-uniform spray cooling (Fig.14), the upper surfa-
ficantly increased the maximum Sxx in comparison with
ce layer whose temperature is higher than the other three
uniform cooling.
surface layers is pulled due to martensitic transformation
From the results shown in this figure, it was found that
inside the cross section and the thermal shrinkage caused
spray cooling or mist cooling at a low water density of
by constraint from the side surface of which temperature
about 20 (l/(m2• min)) was preferable to reduce the stress
is lower. For the above reasons, it was found that the lar-
of Sxx etc. It was estimated that it was important for de-
gest Sxx occurred at the center of the width of the upper
crease of stress and prevention of quenching crack occur-
surface in this case.
rence to reduce the variation in cooling strength between
Compared with uniform cooling, in non-uniform cooling,
surfaces of bloom.
2
2
2
2
the position, timing, and mechanism at which the ma-
Fig.15 – Max Sxx in the case of each cooling method and each cooling condition.
SAMMARY AND CONCLUSION
lowing results were obtained from these analyses.
In this study, the causes of thermal distortion and quen-
1) In several kinds of cooling methods (immersion coo-
ching crack and the effects of kinds of cooling methods
ling, spray cooling and mist-spray cooling) and various
and cooling conditions in the reverse transformation tre-
cooling conditions, the cooling time required for the
atment were analyzed in order to prevent these troubles
reverse transformation treatment were clarified.
by the model of metallo-thermo-mechanics. The cooling
2) The effects of the cooling methods and various coo-
methods considered in these analyses were immersion
ling conditions on the cross-sectional shape of the cast
cooling, spray cooling and mist-spray cooling. The fol-
bloom after cooling and the stress generated during co-
La Metallurgia Italiana - April 2022
pagina 27
Memorie scientifiche - Colata Continua
oling were analyzed from the viewpoint of the behavior
5) From the viewpoint of preventing the occurrence of
of martensitic, bainitic and pearlitic transformation and
quenching cracks, it was revealed that it was necessary
the expansion accompanied with these transformation
to reduce the variation in cooling strength on the outer
and thermal shrinkage behavior.
surface and to use spray or mist-spray cooling with a si-
In addition, the following findings were obtained regar-
gnificantly reduced water density as a cooling method.
ding the cooling method and cooling conditions. 3) From the viewpoint of shortening the required coo-
Acknowledgement
ling time, immersion cooling is advantageous.
I would like to thank I. Watanabe (NIT Akita college, now
4) The difference in the effect of the cooling methods
DMG MORI COMPANY LIMITED) and K. Fujita (NIT Akita
and the cooling conditions on the cross-sectional sha-
college, now JFE STEEL CORPO.) for their contributions
pe of the bloom after cooling is small.
in the numerical calculations in this study.
REFERENCES [1] [2] [3] [4]
[5] [6] [7] [8] [9] [10] [11]
Tamura I:Tetsu-to-Hagané;1988 74,p.1426-1429. Isobe K: CAMP-ISIJ;2007.20,p.96. Isobe K, Shimakage H, Nakayama A: CAMP-ISIJ;2007.20,p.852. Isobe K: Effect of several kinds of metallurgical factors and thermal hysteresis on the formation of surface crack of high nitrogen content special steels for bar and wire rods. 6th European Conference on Continuous Casting 2008.2008 June 3-6; Riccione, Italy. Milan: AIM;2008.STAINLESS & SPECIAL STEELS,CD. Inoue T, Tanaka K, Nagaki S: Analysis of solid mechanics and phase transformation, Ohkawa Publishing,TOKYO,1995,p.56-107. Isobe K:Tetsu-to-Hagané;2018.104,p.128-137. COSMAP with GiD, Idea MAP. Ohba Y, Kitade S, Shimoguchi H, Takasu I:Tetsu-to-Hagané;2007;93,p.271-280. Mitsuzuka M, Fukuda K:Tetsu-to-Hagané;1978.64 (1978),p.70-77. Mitsuzuka M, Fukuda K:Tetsu-to-Hagané;1983.69,p.268-274. Mitsuzuka M, Fukuda K:Tetsu-to-Hagané;1979.65,p.608-616.
La Metallurgia Italiana - Aprile 2022
pagina 28
Scientific papers - Continuous Casting
Quantification of Critical Parameters for Prediction of Surface Crack Formation in Continuous Casting R. Krobath, C. Bernhard
One focus of the digitalization of the CC process is product quality prediction. The training of quality prediction modules can be achieved by experimental data. A requirement is the performance of experiments with samples simulating conditions close to the process. A challenging task is the forecast of surface cracks. At the Montanuniversitaet Leoben, a unique testing setup was developed, called the In-Situ Material Characterization by Bending (IMC-B) test. This test represents an experimental method, combining solidification, controlled cooling, and deformation in one sequence. It allows the prediction of surface crack formation with respect to the testing conditions. Varying testing parameters enable the determination of the impact of different factors that lead to crack formation during deformation of a casted sample in a 3-point bending test. As output quantification parameters for damage evaluation are provided by defining a critical strain for crack formation. This study shows the susceptibility to transverse crack formation of a low alloyed construction steel in the temperature range of the second ductility trough.
KEYWORDS: SURFACE CRACK FORMATION, IN-SITU BENDING TEST, TRANSVERSE CRACKS, ELASTO-VISCOPLASTIC MATERIAL MODEL, STRAIN ANALYSES, CRITICAL STRAIN; NTRODUCTION Virtual “copies” of the continuous casting machines, called digital twins, are constructs of dynamic and complex unions of specific calculation and monitoring modules [1]. Accurate data, generated by sophisticated experiments, are used in thermophysical models to calculate solidification and strand temperatures [2]. This also serves as the basis for product quality prediction, which can be done by specific implemented quality prediction modules. The goal of digital quality prediction is the forecast of a certain risk for problems under the predominating casting conditions. This is a very challenging task, especially for the surface quality. When focused on the surface
Roman Krobath
Montanuniversität Leoben, Austria
Christian Bernhard
Montanuniversität Leoben, Austria
crack formation, the process factors and parameters that influence the formation and finally lead to surface cracks are widely spread. This results in several possible types of surface cracks and defects, which are summarized in earlier publications [3-5]. Data for the adjustment and further development of surface quality prediction can be generated in the plant by product inspection and quality tracking. This approach has the advantage of a self-learning process. Still, it can
La Metallurgia Italiana - April 2022
pagina 29
Memorie scientifiche - Colata Continua
especially be challenging in the CC process, as the surfa-
gs in the hot ductility of the strand shell [6-13].
ce quality influencing factors are widely spread, and the
Nevertheless, Crowther [14] concluded that due to the
inspection of the semis is impeded due to environmental
high strains to fracture in the tensile experiments, a bend
factors, such as high temperatures and scale on the pro-
test for the simulation of transverse crack formation
duct surface. When used as a single approach, there is the
should give an even more reliable indicator. Such experi-
risk of overlooked influencing factors and misinterpreta-
ments were already developed for reheated samples [15-
tion of results. An additional method for the further de-
17] and in-situ casted samples, improving the approach to
velopment of quality prediction modules is to create data
process conditions [18, 19]. These experiments give pro-
from experimental investigations with respect to the pro-
mising results regarding the indication of surface crack
cess. Therefore, requirements for the plausibility of pro-
formation.
vided data are experimental procedures under near-pro-
This paper explains a bend test for investigating the su-
cess conditions. If this is fulfilled, concerning selected
sceptibility to surface crack formation, which is operated
topics, such experiments can be called “experimental
at the Montanuniversitaet Leoben, called the In-situ Ma-
twins”. For the surface crack formation, an “experimental
terial Characterization by Bending (IMC-B) test. It repre-
twin” should be able to simulate certain process parame-
sents an experimental method combining solidification,
ters and material behavior close to process conditions.
controlled cooling, and defined deformation of samples
Especially when combined with the approach of product
in one testing sequence [20-22].
inspection, models can be tuned and validated. Typical experiments characterizing the strand shell ducti-
EXPERIMENTAL PART
lity are conventional hot ductility tests, usually performed
Methodology
as uniaxial hot tensile tests or compression and torsion
In the IMC-B test, every sample is cast separately in a steel
tests. In consideration of the heat treatment and defor-
mold where solidification and mold cooling occurs. The
mation parameters, reduction of area values can descri-
residual time in the mold is adjusted to the aim of the si-
be certain factors that influence the ductility, e.g., grain
mulated process conditions. Fig. 1, left upper picture, de-
size, precipitations, and the austenite to ferrite transition.
picts the workflow of the method, which is divided into 3
Process-near simulations with, e.g., coarse grain annea-
steps. The split construction of the mold allows fast and
led microstructure or in-situ melted sample material and
easy removal of the sample finishing step I of the test.
fluctuating cooling sequences bring sophisticated findin-
Fig.1 - Workflow of the IMC-B test and schematic cooling curves within the definition of significant points [22].
La Metallurgia Italiana - Aprile 2022
pagina 30
Scientific papers - Continuous Casting Continuing in step II, the sample is cooled down according
IMC-B test. The sequence is defined by significant points
to a specific thermal sequence reaching a defined target
within the three main steps listed and explained on the left-
value, which represents the later on bending temperatu-
hand side. When displacement controlled tests are perfor-
re. This is managed by a stepwise apposition of chamber
med, the deformation characteristics are given by stamp
furnaces with varying temperatures. During this thermal
velocity and displacement. The velocity is constant during
cycle, the sample surface temperature is continuously me-
the loading and unloading stages. The material behavior
asured with an optical pyrometer. For bending, the sample
during bending is simulated with the Finite Element (FE)
is positioned in bending equipment on the bearing stamps,
software package Abaqus. The material model parameters
where the temperature field is homogenized at bending
are adjusted to the material behavior of the in-situ casted
temperature. At a defined starting point, isothermal ben-
samples, which can be significantly different from rehea-
ding of the sample takes place, realized with the movement
ted samples. The constitutive material model combines
of the bending stamp. The well-controllable facility allows
elastic, plastic, and viscoplastic effects, allows accura-
accurate bending parameters, either force or displacement
te analyses of the sample’s stress and strain fields, and is
controlled. This deformation of the sample should simu-
explained by Krobath et al. in a recently published work
late local strand shell deformation, e.g., during bending
[23]. Fig. 2 shows a draft of the FE model with the mesh and
and straightening. Finally, the sample is cooled to room
the significant control and evaluation points. A quarter of
temperature by smooth cooling to minimize the thermal
the sample with symmetry planes in x- and z-direction is
and transition-induced stresses and strains or quenched in
used. The translational and rotational values of the bearing
order to freeze the microstructure. Every sample is desca-
are fixed (control point “support”). Evaluation of the force
led, and the surface is investigated in detail with a digital
and displacement takes place at the control point “punch”.
microscope. The defects and cracks are documented with
Two-dimensional evaluations of stresses and strains on the
respect to their position on the sample surface.
sample surface in the x-direction depending on the distance to the bending axis are displayed as mean values over
Fig. 1 shows schematic temperature-time curves for the
the sample width.
Fig.2 - FE model for simulating material behavior in the 3-point bending test [23]. A main characteristic of the IMC-B test is the usage of
trolled heat withdrawal, the resulting structure is similar
in-situ cast samples. The coatings enable controlled heat
to the structure of a strand shell surface; see Presslinger
flux in the mold resulting in directional dendrite growth
et al. [24] and Fig. 3 a). Metallographic investigations reve-
and associated coarsening of the dendrite and columnar
al the coarse columnar austenitic grain structure of IMC-B
grain structure, as shown in Fig. 3 b). Due to the well-con-
samples, depicted in Fig. 3 d) for a 0.17 wt.% C steel. The
La Metallurgia Italiana - April 2022
pagina 31
Memorie scientifiche - Colata Continua
mean grain size depends mainly on the heat withdrawal in
the surface but also, depending on the steel composition
the mold and can therefore also be adjusted to the grain
and here namely the content of certain elements like Si,
size at the slab surface, as shown by Reiter et al. [25]. The
Al and Mn, in intergranular and inner oxidation. A recent
steel composition also plays an important role in the ap-
study by Krobath et al. [22] shows the harmful impact of
pearance of the austenite grain structure in directional so-
a certain stronger intergranular oxidation on the forma-
lidification. Ohno et al. [26] show these correlations with
tion of surface cracks in the bending experiments. It poin-
samples investigated with rapid directional solidification
ts at a possible strong influence of oxidation reactions of
equipment. A typical coarse columnar austenite grain
the sample surface with respect to tested conditions and
structure for plain carbon steels in the range of 0.1 to 0.2
composition. It makes the results of the bending experi-
wt.% C is shown in Fig. 3 c). IMC-B-tests may be perfor-
ment dependent on the cooling conditions. If this effect
med either after controlled cooling in air or protection by
should be suppressed, the sample may also be shielded
inert gas, such as Ar. Usually, the bending tests are perfor-
by inert gases during cooling and surface oxidation is mo-
med after cooling in air. This results in strong oxidation of
stly prevented.
Fig.3 - Comparison of microstructures: a) Solidification structure strand shell; b) Solidification structure IMC-B sample; c) Coarse columnar austenite grains (scheme) [26]; d) Typical coarse columnar austenite grains in IMC-B sample - 0.17 wt.% C steel. Testing Parameters
The basic composition for the following results is listed in Tab. 1. It represents an Al deoxidized 0.17 wt.% C construction steel. The significant points for the temperature sequence are shown in Tab. 2.
Tab.1 - Basic steel composition - all values in wt.%. STEEL COMPOSITION C
Si
Mn
P
S
Al
N
0.17
0.4
1.55
0.01
<0.004
0.03
<0.008
Tab.2 - Significant points in the temperature sequence. TEMPERATURE SEQUENCE
La Metallurgia Italiana - Aprile 2022
tm [s]
Tm [°C]
Th [°C]
th-e [s]
tb-s [s]
TbX [°C]
45
~1180
1050
160
700
700-1100
pagina 32
Scientific papers - Continuous Casting
The liquid steel is poured into the mold with a starting tem-
mum strain and the strain distribution on the sample. Con-
a surface temperature of ~1180°C afterwards.
splacements up to 5 mm. An advantage of the three-point
perature of ~1550°C. The time of 45 s in the mold results in The holding temperature Th is 1050°C (except for samples with a bending temperature Tb of 1100°C) and the total time
at the start of bending is 700 s. It is related to the start of the straightening process of a slab caster with a casting speed
of 1.2 m/min and a slab thickness of 225 mm; see Krajewski et al. [27].
Tb ranges between 700°C and 1100°C in 50°C steps. The
testing sequence is done in air atmosphere. After the ben-
ventional IMC-B tests are performed with maximum dibending test is an induced strain distribution on the sample surface. This enables the investigation of the impact of strains from 0 to a maximum strain on the formation of
cracks with respect to the distance from the bending axis in a certain range of strain rates. Fig. 4 a) depicts the total
strain distributions in the x-direction (LE11) for two cases of stamp displacement and resulting maximum strains at
Tb = 900°C, displacement = 3 mm - max. strain = 3.5% and
ding test, each sample is cooled to room temperature
displacement = 5 mm - max. strain = 5.9%. It illustrates that
depending on the value of Tb. The bending stamp velocity
but it also covers the strain ranges of the sample with a
with mild cooling rates between ~5 °C/min to ~1 °C/min, for the loading and unloading step is 0.04 mm/s. Stamp di-
splacements of 3 mm and 5 mm are applied for the whole temperature range.
Tests get reproduced at Tb = 850°C and Tb = 1050°C. Stamp
displacements of 2 mm, 3 mm, 4 mm and 5 mm are tested at the bending temperature of 900°C. For the validity of reproduction, in this case every test is performed twice.
the higher deflection results in a higher maximum strain, lower deflection. A total displacement of 5 mm leads to
maximum strains in the x-direction of between 5.9% (Tb = 900°C) and 6.4% (Tb = 1000°C) for the investigated tem-
perature range. Differences in the strain distributions with
same displacements are according to changes in the material behavior between the bending temperatures.
RESULTS
Simulation
The displacement of the stamp triggers the induced maxi-
Fig.4 - a) Strain distributions at Tb = 900°C; max. strain in bending axis 3.5% (displacement = 3 mm) and max. strain in bending axis 5.9% (displacement = 5 mm); b) Mean strain values in x-direction and the corresponding strain rates against the distance to the bending axis corresponding to the bending cases in a).
La Metallurgia Italiana - April 2022
pagina 33
Memorie scientifiche - Colata Continua
The detailed strain evaluations of the two bending cases
obtain information about the temperature zone where
are visualized in Fig. 4 a) are shown in the diagram in Fig.
they formed. Cracks with paths through the bulk of au-
4 b). It contains the mean strain values in x-direction and
stenite grains and fully dendritic morphology indicate
the corresponding strain rates against the distance to the
defects formed in the first ductility trough during solidi-
bending axis. The strain rates coincide nearly fully within
fication. The triggers are remaining liquid phases at the
s . The values correlate
liquid-solid interface, enriched with segregated elements
to surface strain rates during straightening in continuous
and a lower melting point. Initialized strains are concen-
casting, which are calculated to be in the range of max.
trated at these phases, resulting in the formation of pores
values of 3 ∙ 10
s
-1
to 4.7 ∙ 10
-1
growing to cracks with increasing strain. The temperature
parameters [28]. In contrast, the gap between the strains
range reaches from solidus to the “zero ductility tempera-
gets higher with decreasing distance to the bending axis.
ture” ~50°C below solidus [30, 31]. To evaluate the ducti-
At an interval of 40 mm, the strains of both cases are still
lity in the area of 700°C to 1100°C, these defects cannot
below 0.5%. The test with a higher deflection reaches 2%
be considered. Nonetheless, they provide information
surface strain at ~25 mm distance. For the test with lower
concerning the behavior during solidification in the mold.
deflection, it is obtained at 17.5 mm. This strain is known
Cracks formed due to the bending of the sample are inter-
as the guide value for maximum mechanically induced
granular (IG) at the austenite grain boundaries and range
strains during straightening [3,14]. As the value can shift,
from ~50 μm to ~3 mm in length. Fig. 5 shows different
e.g. due to thermal strains and higher strains in the case
crack arrangements. Fig. 5 a) depicts a typical singular
of notch effects in oscillation marks or surface pores [29],
crack and b) crack clusters. Singular IG cracks are the
the higher strains are also considered for interpretation.
most frequently observed crack type. The crack clusters
The diagram expresses the advantage that positions of
are areas with many orientated cracks in which a separa-
documented cracks can be directly related to a certain in-
tion of the singular cracks is already difficult. They often
duced strain.
form when a very critical state is simulated in samples with
-1
to max. ~6 ∙ 10
-4
s , with respect to machine
~4 ∙ 10
-4
s
-5
-5
-1
a substantial number of cracks. Network cracks, shown in Surface Crack Investigation
Fig. 5 c), are unoriented cracks that can form on samples
Every sample surface is analyzed in detail. The defects
where pre-defects such as intergranular high temperature
must be carefully investigated, as the morphology can
oxidation takes place [23].
Fig.5 - a) Singular orientated IG cracks; b) Crack clusters; c) Network cracks. Quantification Parameters
results for the described testing parameters and the two
the total number of cracks. It indicates the surface damage
the max. strain mean value for the investigated tempera-
The first quantification parameter of the crack formation is
due to the bending test, taking into account all testing pa-
rameters, including the bending case. Fig. 6 a) shows the
La Metallurgia Italiana - Aprile 2022
bending cases in Fig. 4. The labeled explanation shows
ture range of 700°C to 1100°C. At 3.6% the four bending temperatures between 850° and 1000°C show a number
pagina 34
Scientific papers - Continuous Casting
of IG cracks between 18 and 28. All other remaining sam-
ples revealed no IG cracks at the sample surface (except one crack at 800°C). The test at 850°C was successfully reproduced with values of 18 and 22. The temperatures
where no cracks formed can be described as not critical,
but because of the only minor differences in the number
of cracks at the other temperatures, no clear statement as to which temperature is the most critical is possible for this deflection. In comparison, the higher strains lead to significantly higher gaps. The samples in the lower temperature range of 700°C to 750°C still show no cracks. This
leads to the first conclusion that this steel reveals no susceptibility to surface cracking at these temperatures with
the current testing sequence. 37 cracks are documented for a bending temperature of 800°C. The high temperatures 1050°C and 1100°C show 59 and 11 cracks, respecti-
vely. A second sample at 1050°C with 64 cracks shows a similar result. The most critical temperatures are also
found to be between 850°C and 1000°C. But for the higher stains, a considerable deterioration from 113 (1000°C) to
sponding crack distributions at the surfaces. The mesh indicates the crack positions and the crack density 40 mm
left and right of the bending axis in the temperature range
of 800°C to 1000°C. Categories for segments visualize the crack distributions. White means no cracks, yellow means
1 to 5 cracks and red means more than 5 cracks. The highest crack density is clearly visible for the sample bent at 850°C with the higher maximum strain.
Within the total number of cracks, noncritical temperatu-
res and temperatures with significantly more cracks can be explained. To reach a quantified appraisal of the crack formation regarding continuous casting, a critical strain
for first crack formation is established. According to the crack positions and the development of the strains with
respect to the maximum stamp displacement and bending temperature, the strain for first crack formation is
determined. As this is a significant value, the critical strain ε2 defines the lowest strain value where two cracks have
already formed on the sample surface with an accuracy of 0.01%. This ensures the validity of this parameter.
229 (850°C) cracks is identified. Fig. 7 shows the corre-
Fig.6 - a) Number of IG cracks for the whole temperature range and the two bending cases; b) Corresponding critical strain ε2 for all samples.
Fig.7 - Crack distributions Tb = 800°C, 850°C, 900°C, 950°C, 1000°C La Metallurgia Italiana - April 2022
pagina 35
Memorie scientifiche - Colata Continua
The diagram in Fig. 6 b) presents the ε 2 values according
(max. strain 6.3%) is again very accurate. Fig. 8 a) illustra-
to the samples in a). The noncritical temperatures 700°C
tes the number of cracks for samples with continuously
and 750°C show no cracks, which means the value of ε 2
increasing maximum strain at Tb = 900°C. For the validity
exceeds 6.3% and is therefore not relevant for conven-
of reproduction, every test is performed twice. The two
tional continuous casting. At 800°C, 1050°C and 1100°C
samples with maximum strains below the critical strain li-
the values of ε2 range between 4.25% (800°C) and 3.88%
mit don’t reveal significant cracks on the surface. When
(1050°C). This correlates to the samples with maximum
that “barrier” is exceeded, the number of cracks increa-
strains of 3.6%, which show no cracks, as the critical
ses with a nearly exponential progression. But the criti-
strains exceed the maximum induced strain. The sam-
cal strain values in diagram b) confirm that the first crack
ples with the higher number of cracks at 850°C to 1000°C
formation doesn’t depend significantly on the maximum
also reveal the lowest values of ε 2. At 900°C and 950°C all
induced strain. The mean value of critical strain is 2.95%,
four values are nearly superimposable at 3%. At 850°C the
which is marked as the critical strain level. Fig. 9 shows
most critical behavior, with ε 2 of 2.49%, is observed. The
crack distributions corresponding to diagram Fig. 8 b).
reproducibility at 850°C (max. strain 3.6%) and 1050°C
The crack density increases with higher strains.
Fig.8 - a) Number of IG cracks for continuously rising maximum strains at Tb = 900°C; b) Corresponding critical strain ε2 at Tb = 900°C.
Fig.9 - Crack distributions Tb = 900°C, max. strain 3.5%; 4.7% and 5.9%.
La Metallurgia Italiana - Aprile 2022
pagina 36
Scientific papers - Continuous Casting
DISCUSSION
trough with the IMC-B test is exemplified. The most es-
Samples with differing maximum strains show depen-
sential characteristics of the IMC-B test can be summed
dency in the maximum strain and the number of cracks,
up as follows:
but similarity in the critical strain ε 2. This points at certain
• Similar microstructure compared to the strand shell
strain-induced underlying mechanisms. More precisely, a
- directional dendrite growth and columnar austenite
high amount of deformation-induced AlN precipitates for
grains.
the highest strains is plausible. Further on, these results
• Surface phenomena due to contact with atmosphere -
indicate that for a steel composition with a certain set of
no inert atmosphere in conventional IMC-B testing.
testing parameters, the critical strain for first crack forma-
• Spectrum of strains and strain rates on the sample sur-
tion doesn’t depend on the maximum induced strain. In
face - information about the first crack formation.
fact, ε2 represents a factor that shows the surface ductility
The results of the evaluation of a 0.17 wt.% C steel are
of a current state dependent on the whole set of testing
expressed with the number of cracks and the critical
parameters. In addition, the highest maximum strain va-
strain for crack formation. The most important results
lues enable the most detailed grading of results and pro-
are:
vide more information of the material. The results permit
• The most critical temperature range is revealed at
an evaluation regarding surface cracking in the second
850°C to 1000°C.
ductility trough. Basically, the critical strains are higher
• Noncritical behavior is observed at 700°C and 750°C.
than the mechanically induced strains in a conventional
• The lowest critical strain value of 2.49% shows the
straightener (marked areas Fig. 6 b) and Fig 8 b)). In the
sample with a bending temperature of 850°C.
case of the investigated steel grade and testing parameters, the risk of surface cracking is moderate, although the
ACKNOWLEDGEMENTS
most critical temperature is 850°C. It should be kept in
The authors gratefully acknowledge the funding support
mind that the total strains on the strand surface can incre-
of K1-MET GmbH, metallurgical competence center. The
ase with additional thermal strain contributions or deep
research program of the K1-MET competence center is
oscillation marks. The results are valid for the current
supported by COMET (Competence Center for Excel-
testing parameters and steel composition. Changes in
lent Technologies), the Austrian program for competen-
the parameters can, however, have a detrimental impact
ce centers. COMET is funded by the Federal Ministry for
on the surface ductility; for example, less cooling in the
Climate Action, Environment, Energy, Mobility, Innova-
mold, which leads to high temperatures of >1200°C for a
tion and Technology, the Federal Ministry for Digital and
certain time after the mold, can trigger pre-defects on the
Economic Affairs, the Federal States of Upper Austria,
sample surface and bring about poor ductility in the criti-
Tyrol and Styria as well as the Styrian Business Promotion
cal temperature range; see Krobath et al. [23]. Therefore,
Agency (SFG). Besides the public funding from COMET,
it should be kept in mind that the testing conditions have
this research project is partially financed by scientific par-
to be as near as possible to the real casting process whi-
tners and the industrial partners voestalpine Stahl and Pri-
ch is simulated, to reach the best goals in optimizing the
metals Technologies Austria.
process. Using the data of the IMC-B results for surface crack prediction models and calculations proves highly beneficial since a vast number of scenarios can be simulated to gain information for different mechanisms leading to surface cracks. CONCLUSION In the present study, the determination of the susceptibility to surface cracking in the range of the second ductility
La Metallurgia Italiana - April 2022
pagina 37
Memorie scientifiche - Colata Continua REFERENCES [1]
[2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]
Leitner R, Fuchshuber D, Brugger C, Pennerstorfer P. Digital Twin for continuous casters – the playground for metallurgists and process engineers. 4th European Steel Technology and Application Days. 2019 Jun 24-28; Dusseldorf, Germany. Steel Institute VDEh; 2019. Preuler L, Louhenkilpi S, Bernhard C, Ilie S, Taferner M. Utilization of Experimental Data as Boundary Conditions for the Solidification Model Tempsimu-3D. BHM 165; 2020; 5: 237-242. Brimacombe JK, Sorimachi K. Crack formation in the continuous casting of steel. Metall. Mater. Trans. B 8; 1977; 3: 489-505. Szekeres ES. A Review of Strand Castings Factors Affecting Transverse Cracks, 6th International Conference on Clean Steel. 2002 Jun 10-12; Balatonfüred, Hungary. OMBKE; 2002. p. 324-338. Xia G, Burgstaller K, Fürst C, Harrer B, Milosch E, Samoilov A. Influence of Steel Composition on Crack Formation on the Continuous Casting Steel Slab. 8th European Continuous Casting Conference. 2014 Jun 23-26; Graz, Austria. Leoben: ASMET; 2014, p. 729-739. Schwerdtfeger K. Crack susceptibility of steels in continuous casting and hot forming. Verlag Stahleisen GmbH: Dusseldorf, Germany, 1994. Crowther DN, Mintz B. Influence of Carbon on Hot Ductility of Steels. Mater. Sci. Tech. 2; 1986; 7: 671–676. Suzuki HG. Study of workability of continuous cast strand steels. PhD Thesis; Tohuku University; 1981. Abushosha R, Vipond R, Mintz B. Influence of sulphur and niobium on hot ductility of as cast steels. Mater. Sci. Tech. 7; 1991; 12: 1101–1107. Mintz B, Banerjee JR. Influence of C and Mn on hot ductility behavior of steel and its relationship to transverse cracking in continuous casting. Mater. Sci. Tech. 26; 2010; 5: 547–551. Mintz B, Crowther D.N. Hot ductility of steels and its relationship to the problem of transverse cracking in continuous casting. Int. Mater. Rev. 55; 2010; 3: 168–196. Revaux T, Deprez P, Bricout JP, Oudin J. In situ solidified hot tensile test and hot ductility of some plain carbon steels and microalloyed steels. ISIJ Int. 34; 1994; 6: 528-535. Schwerdtfeger K, Spitzer KH. Application of reduction of area-temperature diagrams to the prediction of surface crack formation in continuous casting of steel. ISIJ Int. 49; 2009; 4: 512-520. Crowther DN. The Effects of Microalloying Elements on Cracking During Continuous Casting. The Use of Vanadium in Steel - Vanitec Symposium. 2001; Bejing, China. Schmidt F, Gross O. The Continuous Casting of Small Sections. Steel Int. 3; 1967; 1: 11-23. Lankford WT. Some Considerations of Strength and Ductility in the Continuous Casting Process. Metall. Trans. 3; 1972; 6: 1331-1357. Burden MH, Funnel GD, Whitaker AG, Young JM. Origins of defects in continuously cast blooms produced on a curved mould machine. Solidification and Casting of Metals. 1979; London; UK; p. 279–286. Yasumoto K, Maehara Y, Nagamichi T, Tomono H. Effect of Thermo-Mechanical History on Surface Cracking of As-Cast Low Carbon Low Alloy Steel Slabs. Tetsu to Hagane 73; 1989; 10: 1381–1388. Crowther DN, Green MJW, Mitchell PS. The influence of composition on the hot cracking susceptibility during casting of microalloyed steels processed to simulate thin slab casting conditions. Mater. Sci. Forum 284-286; 1998; 469-476. Krobath R, Bernhard C, Ilie S, Six J, Hahn S, Pennerstorfer P. A new method for the experimental simulation of surface crack formation in continuous casting. La Metallurgia Italiana 5; 2017; 4: 5–13. Krobath R, Bernhard C. A New Experimental Setup for the Simulation of Surface Crack Formation in the Continuous Casting Process. AISTech. 2017 May 8-11; Nashville, TN, USA. Association for Iron & Steel Technology; p. 1827–1837. Krobath R, Bernhard C, Ilie S, Six J, Hahn S, Pennerstorfer P. The role of grain boundary oxidation on surface crack formation under continuous casting conditions. BHM 164; 2019; 11: 461-465. Krobath M, Krobath R, Bernhard C, Ecker W. Elasto-Viscoplastic Material Model of a Directly-Cast Low-Carbon Steel at High Temperatures. Materials 13; 2020; 2281. Presslinger H, Mayr M, Tragl E, Bernhard C. Assessment of the Primary Structure of Slabs and the Influence on Hot- and Cold-Rolled Strip Structure. Steel Res. Int. 77; 2006; 2: 107-115. Reiter J, Bernhard C, Presslinger H. Austenite grain size in the continuous casting process: Metallographic methods and evaluation. Mater. Charact. 59; 2008; 737-746. Ohno M, Tsuchiya S, K. Matsuura. Microstructural Features and Formation Processes of As-cast Austenite Grain Structures in Hypoperitectic Carbon Steels. ISIJ Int. 55; 2015; 11: 2374–2382. Krajewski P, Krobath R, Bernhard C, Miettinen J, Louhenkilpi S, Ilie S, Schaden T. A Novel Approach for the Simulation of Surface Crack Formation in Continuous Casting. BHM 160; 2015; 3: 109–116. Zhang X, Zhu H, Huang W, Shi B, Tieu AK. Investigation on slab continuous straightening technology and deformation analysis. Int. J. Mater. Prod. Tec. 47; 2013; 126-137. Suzuki M, Hayashi H, Shibala H, Toshihiko E, In-Jae L. Simulation of transverse crack formation on continuously cast peritectic medium carbon steel slabs. Steel Res. 70; 1999; 10: 412-419. Bernhard C, Pierer R, Tubikanec A, Chimani C. Experimental characterization of crack sensitivity under continuous casting conditions. Continuous Casting and Hot Rolling Conference. 2004; Linz; Austria. Ilie S, Reiter J, Presslinger H, Fluch J, Bernhard C. Characterization of hot tear segregations in continuous casting of slabs. 6th European Continuous Casting Conference. 2008 Jun 3-6; Riccione, Italy.
La Metallurgia Italiana - Aprile 2022
pagina 38
Scientific papers - Continuous Casting
Reduction of transverse corner cracks in Tata Steel’s Direct Sheet Plant in IJmuiden R. Kalter, J.A. Kromhout, M.B. Santillana, A.J.C. Burghardt, J. Link, C. Toeniges, E. Gillebaart, L. Koomen, G.J.C.H. Goessens, K. van Eijk, J.P. van ‘t Hul, S. Meijer
In the thin slab casting and rolling plant in Tata Steel in IJmuiden, breakouts and quality issues occurred due to the occurrence of transverse corner cracks. Transverse corner cracks are thought to originate during the casting process and become apparent in the slab and in the hot rolled coil. In order to tackle the problem a multi-fold approach was used.
Analysis of breakout shells indicated an insufficient shell lubrication. Trials with adapted mould oscillation practice were
conducted to control crack formation. The different mould oscillation practices did not result in improved corner crack performance. Mould measurements showed that the liquid mould slag layer was sometimes very thin. An alternative
mould powder with reduced free carbon content was tested, leading to improved slag infiltration and a reduction in breakouts due to corner cracks.
In the product mix, both low carbon and high strength low alloyed (HSLA) steel grades show a large percentage of transverse
corner cracks, making the solution direction difficult. A thermomechanical study was done to verify in what temperature
range the onset of nitride and carbide precipitation occurred, which was linked to the crack sensitive temperature region in secondary cooling. With a FEM model, the slab temperatures in secondary cooling were calculated. It was shown that
the cooling of the slab corners was too high, leading to cold corners in the bending points. Adaptations were made to the upper secondary cooling segment to reduce cooling of the corners, which resulted in hotter slab corners at caster exit.
Additional data analysis showed that next to casting parameters, also rolling mill parameters play a significant role in the corner crack performance. Especially the edger draft proved to be an important parameter. Successful trials were started with increased edger draft, which lead to improved corner crack performance.
KEYWORDS: THIN SLAB CASTING, TRANSVERSE CORNER CRACKS, BREAKOUTS, MOULD POWDER, MOULD OSCILLATOR, SECONDARY COOLING; INTRODUCTION The Direct Sheet Plant The thin slab casting and rolling facility of Tata Steel in IJmuiden, the Direct Sheet Plant (DSP), is used for the production of high strength low alloyed (HSLA), low carbon and electrical steel grades. Slab widths range between 1000 and 1560 mm and the operational casting speed is up to 6 m/min. The caster is a vertical and liquid bending type machine with a 1.1 m mould, followed by a seven-strand secondary cooling area. In the second segment, liquid core reduction is applied, squeezing the slab from 90 to 72 mm thickness [1].
Rudi Kalter
Tata Steel, PO Box 10000, 1970 CA IJmuiden
J.A. Kromhout, M.B. Santillana, A.J.C. Burghardt, J. Link, C. Toeniges, E. Gillebaart, L. Koomen, G.J.C.H. Goessens, K. van Eijk, J.P. van ‘t Hul, S. Meijer Tata Steel, IJmuiden
The DSP suffers from the occurrence of transverse corner cracks. A main issue related to transverse corner cracks is the occurrence of breakouts. The cracks are formed in the mould, causing a weak spot. Below the mould
La Metallurgia Italiana - April 2022
pagina 39
Memorie scientifiche - Colata Continua the as-formed shell of the weak spot ruptures, causing
steels are cast more often, the problems are in the same
the breakout. An example of a transverse corner crack as
order of magnitude. An example of a transverse corner
found in a breakout shell is shown in Fig. 1 (right). It should
crack is shown in Fig. 1 (left). Edge cracks can clearly be
be noted that this is a corner crack found on the cold slab,
detected by the Parsytec system of the rolling mill, and
so it is not certain that this crack was also present during
by combining Parsytec data in combination with operator
casting.
inspection, the customer is shielded from receiving
Another issue is the occurrence of edge cracks which
defective material. However, edge cracks cause a large
manifest in hot rolled coils for both low carbon and HSLA
financial burden due to downgrading and scrapping of
steel grades. The percentage of edge cracks on hot rolled
material.
HSLA steels is obviously larger, but since low carbon
Fig.1 - Left: Edge crack as captured by the parsytec system in the hot strip mill in the DSP. Right: Transverse corner crack as found on a breakout shell. Origin of transverse corner cracks
that the cracks occur due to micro segregation of MnS
Several mechanisms causing transverse corner cracks are
in the oscillation mark. The degree of segregation was
proposed in literature. Brimacombe [2] comprehensively
strongly affected by the depth of the oscillation mark. The
reviewed the stresses and strains in continuous casting
MnS has a low melting point, making the grain structure
that can give rise to cracks in the solidifying shell.
in the oscillation mark weaker. It was advised to use slow
Mechanical properties are affected by several factors,
cooling of the slab, a higher frequency and smaller stroke
i.e. temperature, chemistry, microstructure, and thermal
of the mould oscillation and avoid a Mn/S below 80.
history.
In a continuous caster the steel is subjected to a
One of the proposed mechanisms is that the cracks are
combination of thermal and mechanical deformation.
formed in the mould during initial solidification of the
Mechanically induced stresses can be caused by mould
steel shell [3]. In general, local process disturbances
friction, ferrostatic pressure, machine condition i.e. roll
related to mould heat transfer and slag infiltration as well
pressure and misalignment, bending and unbending, and
as the depth of the oscillation marks may result in the
bulging. Additionally, thermal stresses, like temperature
formation of weak spots and transverse corner cracks
gradients in the shell and shrinkage associated with
[4]. This could be attributed to the chemistry of the
cooling, may lead to mechanical deformation of the
steel, where manganese-sulfides can be formed. Harada
shell. Transverse corner cracks can also originate when
et al [5] present a thorough analysis of the formation of
the strand is subjected to mechanical deformation in
transverse cracks in conventional cast slabs in relation
the secondary cooling area in the temperature range
with manganese and sulfur content. The conclusion is
of the ductility trough [6], especially when the slab is
La Metallurgia Italiana - Aprile 2022
pagina 40
Scientific papers - Continuous Casting in the bending or unbending area. The precipitation
another part about the edger force in the rolling mill.
of carbonitrides in steels alloyed with Al, Nb, Ti, and V, influences the low ductility of steels in this temperature
BREAKOUTS AND STRAND LUBRICATION
range. The carbides and nitrides precipitate at the grain
As mentioned in the introduction, several breakouts
boundaries, making the steel brittle and crack sensitive.
occurred due to the presence of transverse corner cracks.
As long as the corners of the slab remain outside the
Inspection of breakout slabs also revealed the presence
ductility trough, the occurrence of transverse corner
of small bleeders at the narrow faces of slabs. It is well
cracks can be reduced. In literature, several authors have
known that bleeders can be caused by a (local) lack of slag
reported on the use of chamfered moulds (e.g. Rivaux et
infiltration during casting.
al [7]) in order to reduce cooling of the slab corners in the
Measurements of the slag pool thickness (or liquid pool
mould. In the DSP, a slightly curved narrow face plate is
depth) in the mould of the DSP showed a thin slag layer
used, as described by Hibbeler et al [8]. During this study,
with values between 2 and 5 mm. In particular the very low
trials were performed with straight narrow face copper
values around 2 mm or lower close to the narrow faces
plates, but they did not show an improvement in corner
were thought to be insufficient for proper slag infiltration
crack performance.
during casting. The measurements were done at casting
Adaptations in secondary cooling can be made. Modern
speeds of 5.0 and 5.8 m/min.
casting machines are equipped with width-dependent
It was concluded that the breakouts due to transverse
secondary cooling [9]. When this is not installed, width
corner cracks were related to processes in the mould and
dependent cooling can be achieved by changing the
to a lesser extent to precipitation. Predominantly a lack of
nozzles at the sides of the slab [10], [11]. This leads to
strand lubrication at the narrow faces of the mould, plays an
hotter slab corners and thus a reduced occurrence of
important role in the occurrence of breakouts. Improving
corner cracks. Isaev et al [12] indicate a clear dependence
strand lubrication could be realized by increasing the
of the occurrence of transverse corner cracks on the slab
liquid slag layer thickness. In principle this would also be
width. They relate this to the over cooling of the slab
helpful in controlling the occurrence of transverse corner
edges when the width is too small, so they proposed a
cracks in the hot rolled product [5],[13].
careful width dependent cooling. It should be noted that the weak spots in the shell as
Mould powder
formed in the mould could be made visible as cracks
The presence of free carbon in mould powder (C free) is
in the secondary cooling area since due the uneven
the dominant factor influencing powder melting. Free
solidification, thermal stresses may arise upon further
carbon is added to the mould powder in order to control
cooling. This underlines the need for stable casting
the melting rate during casting. Lowering the free carbon
operations in the mould, especially at the meniscus.
content in mould powder will result in increasing the melting rate and finally in the formation of a thicker
Outline
liquid slag layer during casting [14]. It is reported that a
In this paper we present steps taken to reduce the
thicker slag layer in the mould can be a cause of more rim
number of transverse corner cracks and breakouts. Not all
formation during casting - in particular at the corner areas
approaches were successful, but for completeness they
and in combination with meniscus fluctuations [13],[14].
will be mentioned.
Besides, a thicker slag layer can result in some increase of
The first part is about the reduction of corner cracks
slag entrapment during casting [15].
formed in the mould due to lack of lubrication. This is
It was found that the free carbon content always shows
done by investigating effects of mould oscillation and
values of 3.8 or 3.9 wt.%. The supplier was asked to
mould powder.
decrease the free carbon content to 3.2 wt.% using the
The second part is split in two, a part related to the
same free carbon sources and maintaining the same
chemistry and secondary cooling area in the caster and
mould powder components and hence the same powder
La Metallurgia Italiana - April 2022
pagina 41
Memorie scientifiche - Colata Continua and slag properties.
steels suffer from transverse corner cracks at high casting
Plant trials with the adapted version of the standard
speeds, while for some of the HSLA’s, the casting speed
mould powder were done at all casting speeds and
seemed to have no effect.
casting widths with special attention for the process and
Chemistry
product performance. Based on mould measurements,
Thermodynamic calculations have been performed with
the average slag layer thickness increased with 18%.
the Thermo-Calc package, using the TCFE9 database. The
After the introduction of the adapted version of the
grades considered are niobium and vanadium containing
mould powder, no breakouts due to transverse corner
HSLA’s and plain low carbon steels.
cracks occurred. Furthermore, coil inspection showed a
While cooling down, the low carbon steel starts fully
significant decrease of edge cracks. No effects were found
liquid and then forms a fully ferritic phase. Around 1400oC,
on other casting process parameters like strand friction,
the material has a transformation to the austenitic phase,
powder consumption and bulging of the strand. The
which remains until the steel is cooled down to 900 oC.
only parameter changing slightly is the presence of non-
At 900oC, the transformation to the ferritic phase follows
metallic inclusions in the strip due to slag entrapment.
and below 700 oC, a cementite phase starts to appear
Occasionally, some more rim formation was observed,
(seen as pearlite in combination with the ferrite phase
which can be explained by the fact that the liquid pool is
in the microstructure). This pearlite phase is brittle and
thicker, and more slag may quench on the mould copper
possibly crack sensitive, but as it appears at much lower
plates.
temperatures than the ones considered during continuous
Currently, the adapted mould powder with a reduced free
casting, no problems should be expected.
carbon content is used as standard mould powder for low
The HSLA steels cast at the DSP contain niobium and
carbon and HSLA steel grades at all casting speeds and
vanadium. Upon cooling, these grades also follow
casting widths. Special attention is given to the control
a full transformation to ferrite and then to austenite
of rim formation during casting. This can be done by
around 1450oC. Different from the low carbon steel is
controlling the mould level fluctuations.
that precipitation of vanadium carbonitrides V(C,N) and niobium carbonitrides Nb(C,N) takes place, making the
Mould oscillation
steel less ductile and more brittle. The fact that a multiple
Parallel to the work on strand lubrication and mould
precipitation appears in this steel, is consistent with the
powder, plant trials were done on changing the setting
theory that some micro-alloyed steels have two ductility
of the mould oscillator. Some authors [5] report that the
troughs [17]. According to the steel grade specifications,
presence of transverse corner cracks can be reduced
the vanadium can be in a range with a minimum, maximum
with increasing oscillation frequency and with decreasing
and aim concentration. Thermo-Calc calculations were
stroke of the oscillator. It should be noted, however, that
performed to verify the influence of vanadium and nitrogen
this is reported for conventional slab casting with mould
concentration on the onset temperature for precipitation.
oscillation frequencies in the order of 100 cpm and with
Fig. 2 shows the influence of the vanadium content on
strokes above 5 mm.
the temperature of onset of precipitation of vanadiumoscillation
nitrides. It is clear that for higher vanadium contents, the
frequencies between 350 min and 370 min and the stroke
onset of precipitation is at a higher temperature. This
was between 3 and 4.5 mm. No significant differences in
temperature is reached higher in the secondary cooling
corner crack performance were observed.
area, making the steel more prone to cracking.
EDGE CRACKS REDUCTION BY IMPROVING THE
Secondary cooling
SECONDARY COOLING STRATEGY
Data analysis showed more transverse corner cracks
It was observed that the HSLA steel grades suffer from a
for smaller width slabs. This is thought to be due to the
higher amount of transverse corner cracks than the low
cooling intensity at the corner areas, as also stated by
carbon steels. Also, it was observed that the low carbon
Isaev [12]. In the cooling segments, flat spray nozzles are
Trials
were
performed
with
maximum
-1
La Metallurgia Italiana - Aprile 2022
-1
pagina 42
Scientific papers - Continuous Casting
used for slab cooling. When a smaller width slab is cast,
replaced by nozzles with a 35% lower flow rate than the
the spray nozzles have an impact on the narrow face,
original nozzles. The adapted spray nozzles are indicated
overcooling the corners (Fig. 3). Modern thin slab casting
by the green spray nozzles in Fig. 3. This approach is
machines are usually equipped with a width dependent
comparable to the one described in reference [10].
secondary cooling system, but in the design of the DSP,
Fig. 3 shows the configuration of the spray nozzles in the
this is not incorporated.
secondary cooling area in the caster. It is seen that when
Trials were done with adaptation of the outer spraying
the slab width is smaller than 1350 mm, the spray nozzle
nozzles in nine spray bars in the upper segment in the
angle is such that it impacts the narrow face of the slab,
secondary cooling area. The outer two nozzles were
leading to over cooling of the corners of the slab.
Fig.2 - Onset of precipitation temperatures for different vanadium contents in the HSLA steel grade.
Fig.3 - Spray nozzle configuration for narrow and broad slabs in the secondary cooling area. The outer two (in green) spray nozzles were replaced in the plant trials for spray nozzles with a 35% lower flow rate than the original nozzles. Model calculations
slab corners would go through the bending areas in the
Model calculations, using the inhouse developed 3D finite
caster during the ductility trough. Fig. 4 shows the corner
element model Slab3D [16], were performed to verify
temperature of the slab for cast widths of 1.1, 1.3 and 1.5
the impact of the adaptation in secondary cooling on the
m. The red line shows the temperatures for the original
slab temperature. This was done to verify whether the
spray nozzle configuration and the green line shows
La Metallurgia Italiana - April 2022
pagina 43
Memorie scientifiche - Colata Continua
the results for the adapted spray nozzle configuration.
average slab surface temperature as measured by the line
It is seen that when the slab width is 1.1 m, there is no
scan for a trial with casting speed of 5.7 m/min and cast
significant difference in slab temperatures, as expected,
width of 1450 mm. As predicted by the model calculations,
since the adapted spray nozzles are not impacting the slab
temperature differences between the original and trial
surface. For 1.3 and 1.5 m slab width, a significant increase
situations are small, but a clearly higher slab corner
in slab corner temperature is seen with the adapted spray
temperature is measured during the trials. The transverse
nozzles configuration, especially between 1.5 and 5 m
corner crack performance with the adapted secondary
from meniscus. Further down the secondary cooling
cooling segment is significantly improved for low carbon
zone, the temperatures equalize. For the adapted spray
steels.
nozzle configuration the corner temperatures are about
A positive side effect of the adapted spray cooling is
10 – 20 C higher at the end of the casting section.
that also bulging of the slab was reduced. This can be
o
explained since the flow to the slab is redistributed such Plant trials
that the cooling in the middle of the slab is increased
Trials were performed with the adapted top segment in
while the cooling at the edges is decreased. This makes
the secondary cooling area. At the end of the secondary
the shell in the middle of the slab stronger and more
cooling area, 16 meters from the meniscus, a line scan
bulging resistant.
temperature measurement is available. Fig. 5 shows the
Fig.4 - Model calculations of the temperature at the slab corners for the original spray nozzle configuration (red) and for the configuration where the outer three spray nozzles at the first secondary cooling segment were replaced by nozzles with a 35% lower water flow rate.
Fig.5 - Slab temperature as measured at 16 m from meniscus. Result for low carbon steel, cast at 1450 mm slab width and at 5.7 m/min. The red line indicates the result for the trial with adapted spray nozzles and the blue line indicates the result for the original spray nozzle configuration. The dotted lines indicate the temperature at plus and minus one standard deviation. La Metallurgia Italiana - Aprile 2022
pagina 44
Scientific papers - Continuous Casting
EDGE CRACKS REDUCTION BY OPTIMIZING THE
which exerts a force on the narrow sides of the slab. It
Data analysis
significant correlation with the occurrence of edge cracks:
EDGER FORCE IN THE ROLLING MILL
A random forest data analysis was performed to detect the main parameters influencing the occurrence of edge
cracks in the hot rolled material. Since the plant consists 1
of both a continuous thin slab caster and a rolling mill, also the rolling mill parameters were considered. The
rolling area consists of two roughing and five finishing stands. In front of the roughing stands an edger is installed
was observed that the edger force of the rolling mill had a
With a lower edger force more edge cracks occurred (Fig.
6). Trials were done with increasing the edger draft in order to exert a higher edger force and the trial results show that
indeed a causality can be found between edger force and edge crack performance. In this chapter, the effect of the
edger force and a possible mechanism for reducing the edge cracks will be elaborated on.
Fig.6 - Distribution of transverse corner cracks with varying edger force. Data analysis on coils in the period from January 2017 to June 2019 clearly shows the dependence of edge crack formation on edger force.
Fig.7 -Top row: From left to right, the edger pushes on the narrow face with a draft of 0, 10 or 20 mm. Note that the narrow sides of the slab always bulge outward by 10 mm due to the liquid core reduction of the caster. Due to the inward pressure of
the edger force, the broad face close to the slab edge bulges, forming a dog bone. Bottom row: From left to right, the action
of the first roughing stand is depicted. First the dog bone is pushed back, causing the re-widening of the slab until it almost reaches the original width. Then further rolling action reduces the slab thickness to roughly 30 mm.
Note the change in terminology: In the milling area we refer to edge cracks whereas in the casting area we refer to transverse corner cracks. 1
La Metallurgia Italiana - April 2022
pagina 45
Memorie scientifiche - Colata Continua
Proposed mechanism of edge crack reduction
takes place where the broad face thickness near the slab
The edger is set to a certain draft, which is the value with
corner increases. It is hypothesised that with the edger
which the slab width is decreased. When the edger draft is
pushing on the narrow face, weak spots in the form of
larger, subsequently the edger force is larger. However,
micro cracks are decreased in size. These smaller micro-
since the slab width to thickness ratio is large (14 < W/T
cracks lead to reduced local stress build up, making the
<22) the effect of the edger on the final width of the coil is
material less prone to cracking. In the first roughing stand,
minimal. This is in contrast to the working of the edger for
the dog bone is pushed outward, but also deformation in
conventional slabs with 4 < W/T < 10, in which the edger is
the length is established, reducing stresses in the strip
used for width control. In practice at the DSP it is observed
edges and hence reducing the sensitivity to edge cracks,
that an edger draft of 10 mm results in a decrease of width
this is a fully three dimensional process, schematically
at the end of the rolling mill of 2 – 3 mm.
indicated as a 2D image in Fig. 7. After the first roughing
Fig. 7 shows the stages the slab narrow face goes through
stand, the dog bone has vanished, bringing back the width
when the edger draft is applied (top row) and when the
of the strip to its original width. The rougher reduces the
roughing action is applied (bottom row). When the edger
thickness to ~30 mm.
pushes on the narrow face, a dog-bone deformation CONCLUSIONS
Fig.8 - Schematics of the integrated approach for reduction of transverse corner cracks in Tata Steel’s Direct Sheet Plant. In order to reduce transverse corner cracks in Tata Steel’s
a sufficiently thick liquid slag layer thickness in the
Direct Sheet Plant in IJmuiden, an integral approach was
mould is needed in order to provide enough liquid slag
followed to determine the most significant parameters
for infiltration between the mould and the solidifying
influencing the transverse corner crack performance. This
shell. The liquid slag layer thickness was increased by
is visually illustrated in Fig. 8. Significant parameters were
reducing the free carbon content in the mould powder.
found in the casting as well as in the rolling area of the
Note that the rest of the mould powder chemistry was
machine, making it necessary to incorporate both aspects
kept the same. Besides, a reduction in the occurrence
in the analysis. Below, the most important findings are
of edge cracks was obtained by increasing the slag layer
summarized:
thickness. The mould oscillator setting was also tested,
- To reduce breakouts due to transverse corner cracks,
La Metallurgia Italiana - Aprile 2022
but did not have a dominant effect on the edge crack
pagina 46
Scientific papers - Continuous Casting
performance.
found between the vanadium content in the steel and
Process parameters that significantly affect the occurrence
the occurrence of transverse corner cracks. Also, the
of edge cracks are
amount of sulfur in the steel was positively correlated
- Slab width: More transverse corner cracks for small
to corner cracks.
width slabs. A reduced corner cooling in the secondary
- Edger force of the rolling mill: A lower edger force
cooling area is shown to help reduce the occurrence of
correlates with more transverse corner cracks. A
transverse corner cracks.
mechanism for reduction of transverse corner cracks
- Chemistry: More transverse corner cracks were
due to the edger action was proposed.
observed for HSLA steel grades. A correlation was
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
Li, H. X., Zhu, W. Z., Ruvalcaba, D., Mortensen, D., van der Plas, D. W., Fjaer, G. G. and Zhuang, L. Z. Influence of Liquid Core Reduction on Stress-strain Distribution and Strand Deformation in a Thin Slab Caster. ISIJ International, (2016, Vol 56, N.9, pp1616-1624) Brimacombe, J. K. and Sorimachi, K. Crack formation in the continuous casting of steel. Metall. Trans. B, (1977, vol. 8B, pp. 489–505) Ramirez Lopez, P. E., Mills, K. C., Lee, P. D. and Santillana, M. B. A unified mechanism for the formation of oscillation marks. 7th ECCC, Dusseldorf, Germany, 2011. Santillana, M.B., Botman, G. Dekker, E. and Thomas, B.G. 3D thickness measurement technique for continuous casting breakout shells. 7th ECCC, Dusseldorf, Germany, 2011. Harada, S., Tanaka, S., Miumi, H., Mizoguchi, S. and Horiguchi, H. A formation mechanism of transverse cracks on CC slab surface. ISIJ International (1990, Vol. 30, N.4, pp 310-316) Mintz, B. The Influence of Composition on the Hot Ductility of Steels and to the Problem of Transverse Cracking, ISIJ International, (1999, Vol 39, N.9, pp833-855) Rivaux, B. Yao, B, Domgin, J-F. and Moravec, R. Chamfer mold technology for corner crack reduction: understanding and design optimization through numerical simulation. 9th ECCC, Vienna, Austria, 2017 Hibbeler, L.C., Thomas, B.G., Schimmel, R.C. and Abbel, G. The thermal distortion of a funnel mold. Metallurgical and materials transactions B (2012, 43 (5), p. 1156) Krasilnikov, A., Lieftucht, D. Reifferscheid, M. Slab corner temperature with the “HD spray” secondary cooling concept. 8th ECCC, Graz, Austria, 2014 Lu, Y. L., Wang, O. Li, Y.G., He, S. P., Pan, S.S., Zhang, J. G. and Hu, B. Prevention of transverse corner cracks in continuously csat steel slabs using asymmetric secondary cooling. Ironmaking & Steelmaking (2011 Vol 38, N.8, pp 561-565) Lee, E-K., Lim, S-H., Park, T-H., Cho, W-J. Prevention of transverse corner cracks in continuous casting steel slabs by structural modification of secondary cooling nozzles. ESTAD/METEC 2019, Dusseldorf, Germany Isaev, O.B., Kislitsa, V.V., and Fedosov, A.V. Studies of the conditions of formation of transverse corner cracks on the surface of continuous-cast slabs. Metallurgist (2012, Vol. 55, N. 9-10) Riboud, P.V. in ’Metallurgie des Stranggießens, Gießen und Erstarren von Stahl’, (Herausgeber K. Schwerdtfeger), Eigenschaften und Aufgaben von Stranggießschlacken, Verlag Stahleisen mbH, Düsseldorf, Germany, 1992. Kromhout, J.A. Mould powders for high speed continuous casting of steel, Doctoral Thesis, Delft University of Technology, The Netherlands, 2011. Feldbauer, S., Jimbo, I. Sharan, A. Shimizu, K. King, W. Stepanek, J. Harman, J. and Cramb, A.W. Physical properties of mold slags that are relevant to clean steel manufacture. 78th Steelmaking Con-ference,Nashville, USA, 1995 Burghardt, A.J.C., van der Plas, D.W., Ruvalcaba, D., and Santillana, M.B. Modelling of strand deformation to assess the castability of high alloyed steels. International Symposium on Liquid metal pro-cessing and casting, Leoben, Austria, 2015 Ludlow, V., Bain, K.G., Riaz, S. Muller, K., Wans, J. Alvarez de Toledo, G., Zajac, S. and Senk, D. Precipitation of nitrides and carbides during solidification and cooling in continuous casting. (2006, La Rev. Metall., Vol 103(1), pp. 17–24)
La Metallurgia Italiana - April 2022
pagina 47
Memorie scientifiche - Colata Continua
Investigations on primary cooling in CC mould through the use of modeling approach J.F.Domgin, S. Gauthier
Primary cooling and heat flux extraction in mould are key parameters in the continuous casting process: They directly control the solidification in mould and then will affect the quality of products. Solidification behaviour is at the origin of several defects or issues in the final products (Cracks, slivers, breakage/tearing…). A modeling tool, based on the use of the CFD Fluent software, was developed and applied to evaluate the effect of several parameters on solidification behavior generated in the mould: Casting speed, water cooling characteristics, copper plate material and design, slag layer thickness... The numerical results clearly show the importance of some parameters (Slag layer thickness…) on solidification behaviour while other are less or of second order (Temperature of water cooling…). This numerical tool is very useful to better understand the effect of these parameters and how to optimize them for a better control of the solidification and a better quality of products. This tool was also applied to study the origin of Ni coating cracking on copper plates observed in some ArcelorMittal plants and to propose some recommendations for solving this issue.
KEYWORDS: CC MOULD, PRIMARY COOLING, HEAT FLUX EXTRACTION, MODELLING, CFD, NUMERICAL SIMULATION, NI COATING DAMAGE; INTRODUCTION In continuous casting, the mold where the liquid steel starts to solidify is a critical component because it controls the initial solidification and then the quality of the products. Surface condition of the mold can also affect premature failure, then low wear life will affect the productivity of the continuous casting plant. The mold for slab products is constituted with 4 water-cooled plates (Broad and narrow faces). It is made of copper alloy which facilitates an optimal combination of thermal and mechanical properties. The inner surface of the mold is generally coated with nickel or ceramics to protect the substrate. When the casting plant is operating, an important
Jean-François Domgin, Ségolène Gauthier
ArcelorMittal Maizières R&D, FRANCE
thermal flux is transferred from the molten steel, which is in contact with the inner mold surface, to the water-cooled outer side of the mold. The mold is exposed to a high time-varying temperature combined with a high thermal gradient across the mold wall. This cyclic thermal loading promotes thermal fatigue cracks in the meniscus area, which is the most thermally stressed and strained region of the mold.
La Metallurgia Italiana - Aprile 2022
pagina 48
Scientific papers - Continuous Casting
Several numerical studies both in 2D [1,2] or in 3D [3] have
nical degradation that reduces the mechanical properties
been developed in the last 10 years to investigate the in-
of the deposit [10]. The inner surface of the mold under-
fluence of different casting parameters on fluid flows and
goes high fluctuating thermal stresses, which can lead to
on the initial solidification in the mold. Some of them
thermal fatigue.
[1,2] are able to directly simulate the slag behavior in the
In order to better understand the effect of different pa-
mould with its infiltration between solidified shell and
rameters on solidification in mould and on temperature
the mould walls. Some other are focused on the thermal
distribution on copper walls, a parametric study was laun-
behaviour of the copper plates [4,5] taking into account
ched through the use of Computational Fluid Dynam-ics.
the precise design of the copper plates with the water-co-
The 3D numerical tool was also applied to an industrial
oled channels and the coating characteristics. Several
issue to better define the origin of mould damages occur-
metallurgical studies [6,7,8] have been performed to de-
ring in some ArcelorMittal plants.
termine the complex damage mechanism that occurs in the mold during its service life. A study [9] observed crack
NUMERICAL MODEL
propagation through the copper substrate with a depth
The new CFD model is used to calculate heat transfers
ranging from 0.6 mm to 5 mm. All cracks were found wi-
occurring in the mold between the cooling water and the
thin the area approximately 100 to 140 mm from the top of
liquid steel. A 3D geometry with detailed description of
the mold (i.e. 0 to 40 mm below the meniscus position).
water cooling channels within the copper plates is consi-
The mold in the meniscus zone is damaged because of a
dered. The hot face of the copper plates is coated with a
detachment of the coating material and mechanical de-
nickel layer. In between the copper plates and liquid steel,
gradation assisted by the chem-ical attack of low melting
a slag layer is simulated. This layer is considered with a
point elements on the Cu substrate which thus becomes
constant thickness layer. Fig 1. shows a top view of the
embrittled. The coatings are affected by a thermo-mecha-
geometry.
Fig.1 - Geometry used in the CFD model (top view). Flow of liquid steel and water in the cooling channels is calculated. Heat transfers between the different regions are also calculated and solidification of liquid steel is taken into account. The Fluent software is used to solve the following equations: Continuity equation: (1)
La Metallurgia Italiana - April 2022
pagina 49
Memorie scientifiche - Colata Continua
Momentum equation (Navier-Stokes): (2)
Where ui and u j are the velocity vectors (m/s), t is the time (s), x is the coordinate corresponding to the direction i or j (i, j = x, y or z) (m), ρ is the fluid density (kg/m 3), P is the pressure (Pa), g is the gravitational acceleration (m/s²), Sm is a
momentum source term related to the influence of mushy zone on fluid flow velocities (N/m3) and μeff is the effective viscosity (Pa.s):
(3) Where μ0 is the molecular viscosity (Pa.s), μt is the turbulent viscosity (Pa.s), Cμ is the specific heat of fluid (J/kg.K), κ is the turbulent kinetic energy (m ²/s²) and εεis the turbulent dissipation rate (m ²/s3). The realizable (κ-ε) turbulence model
is used to simulate turbulence. Energy equation:
(4)
Where H is the enthalpy (J/kg), k eff is the effective thermal conductivity (W/m.K), T is the temperature (°K) and S is source term due to solidification (J/(m3.s)). We use the enthalpy-porous model for the solidification heat transfer process of liquid steel. The liquid-solid mushy zone is treated as a porous zone with a porosity equal to the liquid fraction. The momentum source term (Sm) is given as: (5)
Where β is the liquid volume fraction, ε is a small num-
As we have considered that the liquid steel is introduced
ber (0.001) to prevent division by zero, Amush is the mushy
at the top of the mold with a homogeneous vertical velo-
zone constant equal to 1.108, and up is the solid velocity
city, the domain corresponds to ¼ of the mold. The mesh
due to the pulling of solidified material out of the domain.
is made of hexahedral structured cells with very thin cells
Water is fed in the channels by the bottom side of the ge-
in the region of solidified shell formation geometry. The
ometry with a constant temperature whereas liquid steel
mesh size is approximately 9.5 million cells.
enters in the domain by the top face with a constant temperature and a homogeneous vertical velocity. Pressure
PARTAMETRIC STUDY BY NUMERICAL APPROACH
outlet conditions are considered at outlets and no slip
Table 1 summarizes the main parameters considered in
conditions are considered at wall surfaces. Heat transfers
the numerical simulations. Table 2 gives some details
between the different regions is calculated by considering
about these variable parameters with the associated case
coupled conditions at interfaces. Water temperature in-
reference. The objective of these simulations was to de-
creases as is goes through the channels toward the top
termine the impact of different parameters on primary co-
of the mold because of the energy extracted by the liquid
oling in terms of heat flux extracted, temperature distri-
steel. Zero heat flux is considered at the external surfaces
bution on plate and solidified shell thickness at the exit of
of the domain.
the mold.
La Metallurgia Italiana - Aprile 2022
pagina 50
Scientific papers - Continuous Casting
Tab.1 - Main parameters for numerical simulation. Mould dimensions
1750x229 mm²
Casting speed
Variable
Superheat
25°C
Mold characteristics
Height: 0.8 m Copper plate thickness: Variable Conductivity: 377 W/m.°K
Cooling water characteristics
Temperature inlet: Variable Water flow rate in broad face: Variable Water flow rate in narrow face: 35 m3/h
Slag layer
Thickness: Variable Conductivity: 1.2 W/m.°K
Ni layer
Thickness: 3 mm Conductivity: 80 W/m.°K
Tliquidus and Tsolidus
1535°C, 1522°C
In these numerical simulations, slag layer thickness, ca-
reliable and robust data is available. Then no simulation is
sting speed, inlet temperature of water cooling, cooling
proposed here. Nevertheless, the air layer plays the role
water flow rate in broad face and thickness of the copper
of strong insulator (=Thermal resistance) and it drastically
plate were investigated. Air gap between infiltrated slag
reduces the extracted heat flux in the mould.
and solidified product could have been simulated but no Tab.2 - Cases simulated with values for parameters tested by numerical simulation.
Table 3 delivers the detailed results about heat flux ex-
and broad faces and solidified shell thickness at the exit of
tracted and temperature variation (Outlet-Inlet) in narrow
the mold. All of these results will be de-tailed later.
La Metallurgia Italiana - April 2022
pagina 51
Memorie scientifiche - Colata Continua
Tab.3 - Cases simulated with values for parameters tested by numerical simulation.
Effect of mould powder/slag properties
ckness infiltrated between the solidified product and the
(Reference versus cases 1 and 2 from Table 2)
mould walls.
Mold powder characteristics can directly influence heat
As expected, the slag layer plays the role of insulator
flux extracted in mould: Higher viscosity mold powders
(=Thermal resistance) between hot liquid steel and cold
are being promoted for the purpose of lowering heat tran-
cooling water by:
sfer. The result is a hotter slab within the mold that can
• Decreasing heat flux in mould and shell thickness at
affect slab-shrinkage factors and impacts lower part of the
mould exit with an increase of slag layer thickness, al-
mold. Powders with lower viscosity break down faster and
most linear (See Figure 2)
can inhibit proper lubrication. Too much heat transfer wi-
• Decreasing mold surface temperature with an increase
thin the mold can result in heavy oscillation marks on the
of slag layer thickness
slab, which could be at the origin of surface defects like
When no slag is considered in the simulation, very high
cracks. Then depending on cast grades, different mold
copper surface temperature (>900°C) is reached which is
powder properties are used.
very detrimental for copper plates properties.
In this study we focus on the effect of the slag layer thi-
Fig.2 - Effect of infiltrated slag thickness on heat flux extraction (Left) and on solidified shell thickness (Right). La Metallurgia Italiana - Aprile 2022
pagina 52
Scientific papers - Continuous Casting
Effect of casting speed
rements of the heat flux extracted in the mould for the
(Reference versus case 3 from Table 2)
same casting conditions. In these casting conditions, the
For slab CC machine, casting speed can vary between 0
slag layer thickness was reduced from 1 mm to 0.8 mm
(When stoppage occurs) and 2m/min (Max value in some
when casting speed was increased from 1 to 1.5 m/min.
ArcelorMittal plants). Then heat transfer in mould could
This value is quite in good agreement with values found
be greatly affected by this parameter.
in literature.
Moreover, it exists some correlations between slag layer
Then according to the results of this new simulation, an
infiltrated and casting speed [11]. In our simulations, the
increase of the casting speed induces and increase of the
slag layer thickness is fixed to a constant value. In these
heat flux extracted in the mould but also a decrease of the
new simulations about casting speed effect the thickness
solidified shell thickness at the exit of the mould (See Fi-
was adjusted in order to agree with the industrial measu-
gure 3).
Fig.3 - Effect of casting speed on extracted heat flux (Left) and on solidified shell thickness (Right) along the mould height on broad face.
Effect of cooling water characteristics
from 250 m3/h to 180 m3/h leads to:
In [1] the 2D numerical results show no effect of the wa-
• Higher water temperature increase (From 7.6°C to
ter flow rate on the solidified shell thickness at mold exit,
10.5°C),
only a weak influence is observed close to the meniscus.
• Slightly lower heat flux extracted (From 1.257 to 1.248
In [12], the experimental results show that a decrease of
MW/m²) with variation almost linear,
the water flow rate in broad face induces smaller hook
• A similar shape of heat flux profile along mold height,
depth.
• A direct impact on mold surface temperature level but
In our simulations, we teste 2 different parameters con-
no impact on temperature distribution,
cerning primary cooling water: The flow rate through the
• No effect on solidified shell thickness at mold exit.
broad face and the inlet temperature. Then, due to change in temperature level, especially at Water flow rate
meniscus location, we can imagine that this parameter can
(Reference versus cases 4 and 5 from Table 2)
affect solidified structures and then impact surface quality
Figures 4 and 5 show that a decrease of water flow rate
of the products (Cracks…).
La Metallurgia Italiana - April 2022
pagina 53
Memorie scientifiche - Colata Continua
Fig.4 - Effect of cooling water flow rate on extracted heat flux and on water temperature increase.
Fig.5 - Effect of cooling water flow rate on broad face temperature distribution (Up) and temperature profile along the mould width (Down). Water inlet temperature
• No real effect on heat flux extracted in mould,
(Reference versus case 6 from Table 2)
• No effect on solidified shell thickness at mould exit,
In that configuration, by decreasing the inlet temperature
• No effect on water temperature at outlet,
of the water cooling, the numerical results illustrated on
• An effect with a decrease of the temperature distribu-
Figure 6 and in Table 3 show that there is:
tion on mold surface especially at meniscus location.
La Metallurgia Italiana - Aprile 2022
pagina 54
Scientific papers - Continuous Casting
Then, due to change in temperature distribution, espe-
some plants mention seasonal effects on products quali-
cially at meniscus location, we can imagine that this para-
ty. Could the seasons have an impact on inlet temperatu-
meter can affect solidified structures and then impact sur-
re of water cooling and then affecting thermal behavior in
face quality of the products (Cracks). From time to time,
the mold?
Fig.6 -Effect of inlet cooling water temperature on broad face temperature distribution (Left) and profile along the mould width (Right).
Effect of copper plate thickness
Then, clearly it appears with these results that the heat
(Reference versus case 7 from Table 2)
transfers in the mould will be affected by this parameter
Decreasing copper plate thickness from 40 mm to 30 mm
throughout its lifetime. Wear of the plates occurs sequen-
changes temperature distribution and heat trans-fers in
ce after sequence and plates are remachined regu-larly.
the mould. According to Figure 7 and Table 3, with thinner
Therefore, without any specific correction, the heat tran-
plates, we observe:
sfers in the mould will change continuously and should
• Lower mold face temperature: Higher efficiency of wa-
affect quality of products. Special attention has to be paid
ter cooling
to this parameter especially when critical limits are rea-
• Higher extracted heat flux but with limited effect,
ched.
• No impact on solidified shell thickness at mould exit.
Fig.7 - Effect of copper broad face thickness on broad face temperature distribution (Left) and along the mould width (Right).
La Metallurgia Italiana - April 2022
pagina 55
Memorie scientifiche - Colata Continua
Other parameters affecting heat transfers in mould
Based on this parametric study, a lot of parameters affect
Copper materials have a relatively low hardness and thus
heat transfers in mould and then the initial solidifi-cation.
low resistance to abrasive wear. For this reason, a high
Such a numerical tool is very useful to evaluate and adjust
degree of wear can occur in the lower part of the mould
process parameters for casting good prod-ucts quality
where the strand shell causes increased levels of stress.
and reduce incidents during casting which could cause
To improve the service life of moulds, dedicated coatin-
some damages to the mould.
gs (Ni, Cr, ceramics…) can be applied. When it concerns coatings for mould plates, a distinction has to be drawn
INDUSTRIAL APPLICATION: Copper plates damage in
between coatings for metallurgical protec-tion to im-
CC mould
prove the surface quality of the products (Prevention of
From time to time, some ArcelorMittal plants face some
cracks), and anti-wear coatings to improve resistance to
issues with copper plates damage due to cracking of co-
abrasion. In this parametric study, this parameter was not
ating.
considered to change and was fixed to a constant value. However, this parameter plays the role of another thermal
Industrial issue
resistance and it will directly affect the heat transfers in the
High temperatures directly influence the base copper
mould.
and coating performance. The mechanical properties of
Finally, the copper material itself and its nature (CuAg
copper, nickel-base products, can be damaged during
versus CuCrZr for example) can affect the heat transfers
high-temperature casting. Cracking of a coating in the
in the mould. Due to properties modification, especially
meniscus region caused by mechanical breakdown will
the thermal conductivity, the cooling efficiency will be
subsequently deteriorate the copper-base material. An il-
affected and then the mould temperature itself will be
lustration of such a phenomenon is presented on Figure
modified. In this parametric study, this property was not
8.
considered to change and was fixed to a constant value.
Fig.6 -Ni coating damage on broad face. Then, in order to better understand the situation in the
behavior (Before and after casting)
plant some investigations were carried out: (1) Numerical simulation to determine the possible
Investigations by numerical simulation by R&D and in
origin of the damage in terms of heat transfer in the
the plant
mould
The ArcelorMittal plant concerned by this issue uses 2 dif-
(2) Industrial checking of water quality and Ni coating
ferent mold designs for casting (Different designs of co-
La Metallurgia Italiana - Aprile 2022
pagina 56
Scientific papers - Continuous Casting
oling channels) and they observed that one mold is more
damage is observed, with higher temperature level (At
affected by this issue than the other. This issue occurred
least 30°C more).
in summer time with high weather temperature and just
These results are obtained in standard casting conditions
after maintenance operation at cooling towers.
(=Normal conditions). We can suppose that in degraded
The numerical results applied for these 2 configurations
conditions (With higher inlet temperature, with reduced
(D1 and D2) are illustrated in Figure 9. They clearly highli-
water flow rate…) the temperature heterogeneity and the
ght that mold D1 presents more heterogeneous tempe-
temperature level could be exaggerated.
ratures especially close to meniscus location, where the
Fig.9 - Effect of plate design on broad face temperature (Left) and on temperature profile along width (Right) depending on plate design. These numerical results are interesting, but they are not
water cooling efficiency. It directly causes the hot face
enough to explain the issue met with the mould coating
temperature of the Cu mold to rise to an unacceptable le-
damage.
vel. This high temperature situation accelerates the failure of Ni coating and the erosion of Cu by presence of zinc in
Then, at the same time, the plant started some internal in-
the steel. For operations where scrap materials are used
vestigations to better understand the situation.
in the melt, the result can lead to zinc concentrations that
• During a maintenance operation of the cooling boxes
penetrate into the grain and further expand the cracks
of the mould, the operators discovered important cor-
through the coating and copper substrate. The mold desi-
rosion inside the boxes with some channels blocked by
gn has amplified this phenomenon but is not at the origin
corrosion. It means that the cooling efficiency by water
of the issue.
is weaker than in normal conditions.
Corrective actions were proposed, and everything comes
• During machining of plates, the operators measured
back in order…
the presence of Zn at high level in plates in Ni coating area. It means that Zn penetrates the grain and further
CONCLUSION
expand the cracks through the coating and copper sub-
Continuous casting is a very complex process depending
strate.
on many parameters. This numerical investigation based on the use of a 3D CFD numerical tool shows that several
Explanations and recommendations
parameters directly affect and control heat transfers in the
Extremely high temperature on the hot face of the Cu
mould and initial solidification which is a key element for
mold at meniscus area is found the cause of the mold he-
products quality. Numerical tool as well as online sensors
avy erosion problem. The corrosion of cooling boxes has
will be more and more useful to better control and opti-
affected the water flow rate and has significantly reduced
mize the CC process in the future.
La Metallurgia Italiana - April 2022
pagina 57
Memorie scientifiche - Colata Continua
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
X. Zhang, L. Zhang, H. Wang, W. Yang, Y. Ren, Mathematical modeling on the influence of casting parameters on initial solidification at the meniscus of slab continuous casting, Metallurgical and Materials Transactions B, Published online: April 2019 Pavel E. Ramirez-Lopez, P.D. Lee and K.C. Mills, Explicit Modelling of Slag Infiltration and Shell Formation during Mould Oscillation in Continuous Casting, ISIJ International, Vol. 50, pp. 425-434, 2010 C. Tscheuschner, M. Schäperkötter, P. Müller, Analysis and Optimisation of mould cooling conditions by application of CFD modelling and vericiation in industrial trials, Vienna, ECCC2017 A. Gupta, R.K. Singh, A. Paul, S. Kumar, Effect of mould geometry, coating and plate thickness on the thermal profile of continuous casting mould, The Journal of Southern African Institute of Mining and Metallurgy, Vol. 118, May 2018 L.C. Hibbeler et al., A reduced order model of mould heat transfer in the continuous casting of steel, Applied Mathematical Modelling, 2016 S. Barella, A. Gruttaduria, C. Mapelli, D. Mombelli, Investigation of failure and damage on a continuous casting copper mould, Eng. Fail. Anal. 36, 2014 C. Pandey, M. Raj, R. Mishra, V.K. Tripathy, N. Bandyopadhuau, Failure of nickel coating on a copper mold of a slab caster, J. Fail. Anal. Prev. 8, pp. 3–11, 2008 J. Srnec Novaka, A. Lanzuttia, D. Benasciuttib, F. De Bonaa, L. Moroa, A. De Luca, On the damage mechanisms in a continuous casting mold: Afterservice, material characterization and finite element simulation, Engineering Failure Analysis 94, pp. 480–492, 2018 F. Faries, J. Rawson, A. Rose, M. Bugdol, Mould Coatings for Continuously Cast Billet Production, Technical steel research – Final Report, 1996 S.M.M. Hadavi, A.A. Zadaeh, M.S. Jamshidi, The effect of thermal fatigue on hardness of hard chromium electroplating, J. Mater. Process. Technol. 147, pp. 385–388, 2004 P.E. Ramirez Lopez, P. Lee, K.C. Mills, B. Santillana, A new approach for modelling slag infiltration and solidification in a continuous casting mould, ISIJ International, Vol. 50, pp. 1797-1804, 2010 X. Zhang, L. Zhang, H. Wang, W. Yang, Y. Ren, Oscillation marks and subsurface hooks in ultra low carbon steel of continuous casting slabs, Vienna, ECCC2017
La Metallurgia Italiana - Aprile 2022
pagina 58
Scientific papers - Continuous Casting
Utilizzo di un sistema di ispezione ottica automatica atto al rilevamento dei difetti di colata continua basato su algoritmi di Machine Learning per l’analisi dell’incidenza delle marche di oscillazione su bramme di acciaio inossidabile austenitico AISI 316L e 316LI A. F. Ciuffini, F. Di Giovanni, D. Mombelli
La colata continua è attualmente il processo principale per la solidificazione dell’acciaio e si è affermata grazie all’e-
levata produttività, alla qualità metallurgica del prodotto e all’elevato grado di automazione che la contraddistingue.
Tuttavia, come tutti i processi industriali, soffre della genesi di alcuni difetti tipici che possono compromettere l’integrità strutturale dei semilavorati oltre che inficiare la qualità del prodotto. Per questo motivo sono stati messi a punto sistemi di ispezione automatica delle bramme capaci di monitorare la superficie dei semilavorati di colata continua al fine di limitare la deformazione plastica e le successive lavorazioni di prodotti con difettosità critiche e/o pericolose.
Sfruttando le potenzialità del sistema di ispezione installato su una delle linee di colata dell’acciaieria di Acciai Speciali Terni S.p.A., è stata analizzata l’incidenza delle marche di oscillazione presenti su bramme di acciaio inossidabile austenitico AISI 316L e 316LI al fine di correlarne distribuzione e profondità ai parametri di processo. Dall’analisi è
emerso che la velocità di colata è un parametro critico che influenza proporzionalmente la distribuzione delle marche lungo la bramma. L’analisi ha messo in luce una minore lubrificazione degli spigoli aumentando frequenza e profondità delle marche di oscillazione in questa posizione, soprattutto sull’estradosso a causa del peso del semilavorato stesso. L’occorrenza preferenziale di tale difettosità su un lato delle bramme è associata a peculiarità intrinseche degli impianti
complessi (ad esempio leggeri disallineamenti). La concentrazione di azoto, diversa per i due acciai analizzati, concor-
re a modificare il comportamento sugli spigoli. I risultati hanno evidenziato che nelle bramme di transizione (avvio e chiusura colata) il 316L presenta una incidenza di marche di spigolo maggiore del 316LI, mentre la tendenza si inverte
per le bramme centrali della colata, per le quali, la maggior disponibilità di azoto del 316L aumenta la viscosità cinematica della polvere di colata, migliorando la lubrificazione.
PAROLE CHIAVE: COLATA CONTINUA, ACCIAI INOSSIDABILI, SISTEMI OTTICI PER RILEVAMENTO AUTOMATICO DIFETTI, MARCHE DI OSCILLAZIONE; INTRODUZIONE Il processo di solidificazione in colata continua copre attualmente il 90-95% della produzione industriale di acciaio e la totalità dei semilavorati [1]. L’affermazione di tale processo a discapito della colata in lingotto deriva
Andrea Francesco Ciuffini
Acciai Speciali Terni, Viale Benedetto Brin 218, 05100 Terni
principalmente dalla possibilità di ridurre il numero de-
Federica Di Giovanni, Davide Mombelli
gli stadi del ciclo di produzione e lavorazione dell’acciaio
20156 Milano
e soprattutto di ridurre il consumo energetico legato al
Politecnico di Milano, Dipartimento di Meccanica, Via La Masa 1,
riscaldo dei semilavorati. Questo si accompagna ad un
La Metallurgia Italiana - April 2022
pagina 59
Memorie scientifiche - Colata Continua
aumento della produttività dell’impianto e alla possibilità
lore dal cuore liquido e durante la discesa il battente
di colare in sequenza il contenuto di più siviere, con una
ferrostatico cresce progressivamente, portando la pel-
conseguente riduzione dei costi di esercizio.
le solida a deformarsi fino a ritornare in contatto con le
La colata continua permette di ottenere semilavorati con
pareti della lingottiera. Il comportamento della barra in
una lunghezza teorica infinita, con sezione e forma ap-
solidificazione, insieme alla contrazione termica, gene-
propriate per la diretta utilizzazione da parte dei treni di
ra dei vuoti all’interfaccia tra acciaio e lingottiera, che
laminazione, senza dover procedere alle operazioni preli-
vengono riempiti dal lubrificante e dai gas liberati dalla
minari di sbozzatura [2].
sua decomposizione [4]. Il movimento oscillatorio del-
L’impianto inoltre è altamente automatizzato, ciò com-
la lingottiera (tipicamente con andamento sinusoidale),
porta una riduzione dell’incidenza della manodopera
adottato per evitare incollamenti e garantire uno scor-
essendo richiesti soltanto tecnici nelle sale di controllo.
rimento uniforme della barra, ha la peculiarità di avere
Il processo in continuo permette inoltre di ottenere una
una velocità di discesa maggiore rispetto alla velocità
struttura a grano più fine ed un dendritismo colonnare
di colaggio (negative strip time – NST). Quest’ultimo
meno pronunciato a causa del rapido raffreddamento a
parametro ha un’influenza diretta sulla qualità del se-
cui è sottoposto, con conseguente miglioramento delle
milavorato: in generale, tempi più lunghi portano alla
proprietà meccaniche del prodotto.
formazione di marche di oscillazione più profonde ma
Il cuore dell’impianto della macchina di colata continua-
migliorano la lubrificazione, efficace per prevenire rot-
mente è indubbiamente la lingottiera, responsabile della
ture [5].
formazione della prima pelle solida dell’acciaio in transito
Le marche di oscillazione sono particolari difetti, deri-
e del conferimento della geometria della sezione del se-
vanti dal processo di colata continua, legati al movimen-
milavorato. La corretta gestione del flusso termico in lin-
to oscillatorio della lingottiera. Esternamente si presen-
gottiera è fondamentale per garantire una qualità ottimale
tano come ondulazioni sulla superficie del semilavorato
delle bramme e può essere regolato attraverso la polvere
mentre internamente la microstruttura riporta una tipica
di colata. Il processo di solidificazione e il flusso di metal-
conformazione a uncino, associata al movimento oscil-
lo liquido nella lingottiera hanno un impatto decisivo sulla
latorio verticale della lingottiera, che altera localmente in
formazione di difetti e cricche, sia all’interno che sulla su-
modo significativo il trasferimento del calore e il flusso
perficie della bramma stessa [3].
del metallo liquido nella regione del menisco, favorendo
I fattori che influenzano lo scambio termico tra l’acciaio e
la formazione di cricche trasversali. Tra le cause che por-
la lingottiera sono:
tano alla formazione di questo difetto, oltre al movimento
• la temperatura di colata: se elevata porta alla formazio-
oscillatorio della lingottiera, vi sono: la distribuzione ter-
ne di una pelle troppo sottile;
mica all’interno della lingottiera stessa, le proprietà termi-
• la composizione chimica dell’acciaio: Al, Mn, P, S favo-
che e meccaniche del lubrificante, il contenuto di carbo-
riscono lo scambio termico; C<0.20% lo peggiora;
nio (al diminuire della percentuale di carbonio aumenta la
• l’intensità del raffreddamento primario: portate d’ac-
profondità del difetto), la pressione dinamica generata dal
qua eccessive causano un brusco raffreddamento
lubrificante e dai movimenti oscillatori, e la solidificazione
dell’acciaio. Ne consegue che la velocità di raffredda-
del menisco, che genera la presenza di pareti solide pro-
mento genera un rapido ritiro della barra e un distacco
tese verso il centro della barra. Osservando la forma del
prematuro dalla lingottiera e quindi una diminuzione
menisco di solidificazione dell’acciaio si evince la genesi
dell’efficienza degli scambi termici;
delle marche di oscillazione (Fig. 1) [6].
• la lubrificazione della lingottiera: se effettuata con olio favorisce gli scambi termici per mezzo di fenomeni di “cracking” dell’olio stesso. Il guscio di acciaio solidificato continua a ricevere ca-
La Metallurgia Italiana - Aprile 2022
pagina 60
Scientific papers - Continuous Casting
Fig.1 - Genesi delle marche di oscillazione (a) [7] e fenomeni di flusso nella zona del menisco [6] (b). Poiché la formazione delle marche di oscillazione è inevi-
la marca, con possibili inglobamenti di scoria (Fig. 2a);
tabile, risulta fondamentale ottimizzare i parametri di pro-
2. eccessivo ripiegamento del menisco con conse-
cesso affinché si possa ottenere il corretto bilanciamento
guente rottura del guscio solidificato e sanguinamen-
tra lubrificazione e profondità delle marche. Se non per-
to da fondo gola. Una rottura del menisco solidificato
fettamente sotto controllo, le marche di oscillazione pos-
comporta una fuoriuscita di acciaio liquido segregato e,
sono generare due tipici difetti:
in alcuni casi, può inglobare scoria presente all’interno
1. overflow di acciaio liquido durante la formazione del-
della marca di oscillazione (Fig. 2b) [8].
Fig.2 - Schematizzazione del fenomeno di overflow di acciaio liquido (a); fenomeno di rottura del menisco solidificato (b) [8]. Sistema di classificazione automatica dei difetti
mercati internazionali. Inoltre, per l’acciaio inossidabi-
Negli ultimi anni il mercato degli acciai inossidabili è
le, un’adeguata qualità superficiale non è solamente un
diventato sempre più esigente in termini di qualità, so-
vezzo estetico, ma influisce in modo significativo sulle
prattutto per quanto concerne l’aspetto superficiale
proprietà meccaniche e di resistenza alla corrosione. I
dei prodotti finiti. È fondamentale garantire un elevato
difetti derivanti dal processo di colata continua, se non
standard qualitativo, in linea con le richieste e le aspet-
rilevati e opportunamente rimossi, possono compro-
tative del cliente, al fine di incrementare le vendite sui
mettere anche in modo grave e irreparabile la qualità del
La Metallurgia Italiana - April 2022
pagina 61
Memorie scientifiche - Colata Continua
prodotto laminato, con un conseguente danno econo-
re tramite un processo manuale di apprendimento che
mico non trascurabile per l’azienda. Per questo motivo
fornisca criteri di identificazione dei difetti e regole ap-
sono stati messi a punto sistemi di ispezione automatica
positamente strutturate. Infatti, dato che il sistema si
per l’analisi dei difetti superficiali e sub-superficiali in
basa sull’associazione di forma, colore, dimensione e
linea di prodotti da colata e da laminazione. L’obietti-
ciclicità, è necessario introdurre dei criteri che possano
vo principale è quello di ispezionare i prodotti al fine di
aiutare a distinguere tra loro tutte le tipologie di difetti,
identificare immediatamente la presenza dei difetti più
sebbene presentino caratteristiche simili.
gravi. Ciò permette di intervenire sui parametri di pro-
L’obiettivo di questo lavoro è proprio quello di utiliz-
cesso per evitarne la genesi. Inoltre, con l’utilizzo di
zare il sistema automatico di rilevazione difetti per l’a-
questo strumento si riducono notevolmente i tempi di
nalisi della posizione e dell’intensità delle marche di
ispezione ma soprattutto si riesce ad evitare che un di-
oscillazione al fine di identificare possibili correlazioni
fetto grave di colata, se non rilevato prima delle opera-
con i parametri di processo (velocità di colata, polveri
zioni di laminazione successive, possa influire in modo
di colata e contenuto di azoto dell’acciaio). La scelta è
irreparabile sulla qualità del prodotto finito.
ricaduta sulle marche di oscillazione perché sono il di-
L’ispezione ottica si basa su un sistema combinato di
fetto rilevato con minor probabilità di errore dal sistema
sensori 2D e 3D, che consentono di identificare in modo
di riconoscimento e perché presenti su tutti i prodotti di
accurato i difetti, indicandone estensione e profondità.
colata continua, essendo intrinseche del processo.
Per l’ispezione della superficie della bramma vengono utilizzate in totale dieci telecamere: quattro per la su-
Procedura sperimentale
perficie superiore (intradosso), quattro per la superfi-
L’attività sperimentale si è concentrata su due marche di
cie inferiore (estradosso) e due per le superfici laterali,
acciaio inossidabile austenitico, appartenenti alla stessa
collocate dopo la via a rulli, nella zona di movimenta-
famiglia, il 316L e il 316LI, che differiscono tra loro per il
zione e recupero delle bramme. Affinché un difetto sia
diverso contenuto di azoto in lega (Tab. 1).
rilevato correttamente, è necessario istruire il softwaTab.1 - Composizione degli acciai inossidabili 316L e 316LI [9]. C
Si
Mn
Ni
P
S
Cr
Mo
N
316L
≤0,03
≤1
≤2
10÷13
≤0,045
≤0,015
16,5÷18,5
2÷2,5
≤0,11
316LI
≤0,03
≤1
≤2
10÷13
≤0,045
≤0,015
16,5÷18,5
2÷2,5
≤0,03
Ogni sequenza analizzata è costituita da più colate, iden-
tificate a mezzo della seguente nomenclatura: la prima e l’ultima bramma della sequenza sono, rispettivamente, la
51R e la 83R; tutte le altre bramme della sequenza sono numerate da 51 a 5X, mentre l’ultima di ogni colata è la
83. Per semplificare la visualizzazione dei risultati ottenuti, la profondità delle marche è stata definita mediante
una classe di profondità senza utilizzare il valore reale. Le diverse classi di profondità corrispondono a: • classe 1: profondità < 1 mm
• classe 2: 1 < profondità < 1.5 mm
• classe 3: 1.5 < profondità < 3 mm • classe 4: 3 < profondità 4.5 mm • classe 5: profondità > 4.5 mm
La Metallurgia Italiana - Aprile 2022
Inoltre, la posizione della marca sulla bramma è stata normalizzata rispetto alla larghezza e alla lunghezza della
bramma stessa, in quanto non tutti i semilavorati presi in esame hanno la medesima larghezza.
I parametri presi in considerazione sono stati la velocità
di colata, la capacità di lubrificazione della polvere di colata e il contenuto di azoto in lega dell’acciaio. Risultati e discussione
Effetto della velocità di colata
Inizialmente si è cercata una possibile correlazione tra
la presenza, la profondità delle marche di oscillazione,
la loro disposizione sulla superficie dei semilavorati e la velocità di colata.
pagina 62
Scientific papers - Continuous Casting
In letteratura è riportato che, in generale, un aumento
presenza e la profondità delle marche di oscillazione nel
negative strip time (NST), a cui corrisponde una riduzio-
peso paniera, posizione asta tampone. Questi ulteriori
della velocità di colata comporta una diminuzione del ne della profondità delle marche di oscillazione. I dati sperimentali raccolti sono concordi [10].
Nel grafico in Fig. 3 è riportata la variazione dei princi-
pali parametri di processo che possono condizionare la
corso di una intera sequenza tipo: velocità interessata, due parametri permettono di testimoniare la bontà degli
scambi siviera e la fluidità, e quindi la qualità, dell’acciaio colato al procedere della sequenza.
Fig.3 - Variazione dei parametri di processo (velocità di colata, livello asta tampone, peso paniera) durante la colata. Fig. 3. Di conseguenza la prima parte del semilavorato è
colata arriva a regime, presenta invece un minore quanti-
si traduce in una maggiore concentrazione delle marche
è pressoché simile in entrambe le marche di acciaio (Fig.
creata a velocità inferiori rispetto a quelle di regime. Ciò di oscillazione al centro della larghezza della bramma. La
porzione restante di bramma, una volta che la velocità di
tativo di marche di oscillazione centrali. La distribuzione 4), le differenze più marcate verranno considerate in seguito.
Fig.4 - Distribuzione delle marche di oscillazione sulla prima bramma della sequenza (codice 51R), acciaio 316L (a) e acciaio 316LI (b).
La prima bramma di ogni colata, poiché viene prodotta
individuarli nei prodotti.
di transizione e può presentare delle oscillazioni centrali
cato in arancione in
durante il cambio siviera viene considerata una bramma molto più marcate. Ciò è dovuto al fatto
che i cambi siviera costituiscono un punto di discontinuità nella colata di una sequenza, nasce quindi la necessità di
La Metallurgia Italiana - April 2022
Nel caso di un cambio siviera particolarmente lento (indiFig. 3), è necessario diminuire la velocità di colata al fine di impedire un abbassamento eccessivo del livello dell’acciaio liquido in paniera, per non comprometterne la fluidodi-
pagina 63
Memorie scientifiche - Colata Continua
namica. Per questo è più probabile il rischio di intrappolamento di inclusioni esogene all’interno dell’acciaio. Nel caso di cambi siviera lenti, la velocità di colata cala ancor
più significativamente che rispetto al transitorio di inizio sequenza. Questo comporta l’insorgenza di marche più
profonde in prossimità degli spigoli e di una maggior con-
centrazione di marche di minor profondità al centro della
bramma (Fig. 5). Come nel caso precedente, si osserva la presenza di un maggior numero marche di oscillazione sul 316L (Fig. 5a) rispetto al 316LI (Fig. 5b), soprattutto nella parte centrale.
Fig.5 -Distribuzione delle marche di oscillazione su una bramma prima di colata (codice 51) in corrispondenza di uno scambio siviera lento, acciaio 316L (a) e acciaio 316LI (b). Talvolta si rende necessario aprire lo scaricatore della
tà lungo gli spigoli della bramma (Fig. 6). In questo caso
siviera durante un cambio mediante lancia ad ossigeno
l’origine delle marche di oscillazione è dovuta all’intrap-
(zona gialla di
polamento degli ossidi generati dal getto di ossigeno al
Fig. 3). Questa operazione, nonostante non comporti ral-
contatto con l’acciaio residente in siviera. Tali ossidi non
lentamenti così drastici nella velocità di colata come nel
riescono a flottare in paniera e si ritrovano finemente di-
caso di cambio siviera lento, può comunque provocare
spersi nell’acciaio in colata e solidificazione.
l’insorgenza di un numero di marche di media intensi-
Fig.6 - Distribuzione delle marche di oscillazione su una bramma prima di colata (codice 51) corrispondente all’apertura con ossigeno della siviera, acciaio 316L. In generale anche la seconda bramma di ogni colata pre-
dice 51 di ciascuna colata. Questo è dovuto al protrarsi del-
centrali, seppur in misura minore rispetto alle bramme co-
le marche di oscillazione centrali presenti nei primi metri
senta un’incidenza maggiore di marche di oscillazione
La Metallurgia Italiana - Aprile 2022
le due possibili cause analizzate precedentemente. Infatti,
pagina 64
Scientific papers - Continuous Casting
della bramma codice 52 potrebbero essere generate da una
con ossigeno della siviera (Fig. 7). Tuttavia, risulta evidente
bi siviera problematici che avevano interessato la bramma
bramma della sequenza (codice 52) sia inferiore rispetto al
velocità di colata non ancora a regime, causata dagli scam-
precedente (codice 51). Analogamente, potrebbe verificar-
si un residuale trascinamento di ossidi dovuti all’apertura
come la numerosità delle marche riscontrate nella seconda semilavorato che la precedeva (codice 51).
Fig.7 - Distribuzione delle marche di oscillazione su una bramma intermedia seconda di colata (codice 52), acciaio 316L (a) e 316LI (b). Una sostanziale differenza rispetto ai casi analizzati fino ad
timizzato (Fig. 8). Tuttavia, la presenza di marche di oscil-
corrispondenti a colaggi con velocità costanti e di regime,
indice di un problema di lubrificazione degli spigoli stessi.
ora si può notare nelle bramme centrali di ogni sequenza, per cui l’impianto lavora alle condizioni per cui è stato ot-
lazione, anche profonde, sugli spigoli, potrebbe essere
Fig.8 - Distribuzione delle marche di oscillazione sulle bramme intermedie, acciaio 316L (a) e 316 LI (b).
Effetto della capacità di lubrificazione della polvere di
zioni ottimali si infiltra tra la lingottiera e la pelle solida
colata
della bramma in formazione in maniera omogenea sia sul-
La composizione, le proprietà e le caratteristiche delle
le facce larghe, sia sulle facce strette, sia agli spigoli del
polveri di colata determinano in modo significativo la sta-
semilavorato
bilità del processo di colata continua e la qualità dei semi-
Tuttavia, dai dati riportati in Fig. 8, è emerso un compor-
lavorati prodotti.
tamento non ottimale della polvere di colata in corrispon-
Durante il colaggio la polvere fonde sulla superficie
denza degli spigoli, che comporta inevitabilmente un
dell’acciaio, formando uno strato liquido, che in condi-
peggioramento complessivo della qualità delle bramme
La Metallurgia Italiana - April 2022
pagina 65
Memorie scientifiche - Colata Continua
prodotte.
sull’asse delle ordinate in Fig.9, con l’intento di quantifi-
Gli spigoli, per ovvi motivi geometrici, costituiscono una
carne la presenza effettiva in determinate posizioni. Inol-
zona critica in cui la pelle solida della bramma in forma-
tre, i difetti sono stati riclassificati anche in base alla loro
zione subisce un’estrazione termica estremamente inten-
presenza sulle superfici superiore (intradosso) ed inferio-
sa poiché proveniente da due direzioni perpendicolari.
re (estradosso) per individuarne un eventuale contributo
La pelle solida risulta in questa zona più spessa e fredda
alla loro genesi. Dai grafici emerge un chiaro sbilancia-
e di conseguenza meno incline a deformarsi e quindi ad
mento della distribuzione delle marche di oscillazione di
assecondare i movimenti della lingottiera. Ne consegue,
spigolo, che tendono a concentrarsi maggiormente da un
sia per questo, sia per i confinamenti geometrici imposti,
lato. La spiegazione risiede in problematiche di carattere
la genesi di maggiori sforzi d’attrito tra la lingottiera e la
impiantistico: è ragionevole pensare che ci possa essere
pelle solida della bramma. Tuttavia, a questo incremen-
un leggero disallineamento nella struttura della macchina
to degli sforzi d’attrito non corrisponde un aumento della
di colata continua. Ciò comporta che la bramma in forma-
lubrificazione, né un aumento locale della viscosità cine-
zione tenda a scivolare e ad appoggiarsi maggiormente su
matica della polvere di colata liquefatta, che ne permette-
uno dei due lati della lingottiera. Ne deriva un maggiore
rebbe una migliore infiltrazione.
flusso termico, e soprattutto maggiori sforzi di attrito do-
Quindi, la polvere di colata utilizzata dovrebbe essere
vuti ai movimenti relativi tra lingottiera e pelle solida della
ottimizzata aumentandone leggermente la viscosità, per
bramma, sul lato destro, con la conseguente formazione
permettere una migliore infiltrazione tra lingottiera e pel-
di più marche di oscillazione su questo lato.
le solida della bramma, specie alle più basse temperature
In aggiunta, la presenza maggiore di marche di oscilla-
raggiunte nella regione degli spigoli. Questo senza che
zione di spigolo sull’estradosso è un’ulteriore conferma
venga compromesso il buon rendimento a regime otte-
dell’ipotesi precedente. Ciò è dovuto all’azione della for-
nuto sulle facce larghe dei semilavorati.
za di gravità che, facendo appoggiare la bramma sull’e-
I dati riguardanti la numerosità dei difetti sono stati norma-
stradosso, genera un maggiore attrito tra lingottiera e pel-
lizzati rispetto al loro totale per bramma, come riportato
le della bramma in formazione.
Fig.9 - Distribuzione delle marche di oscillazione sul totale: 316L intradosso (a) e estradosso (b); 316LI intradosso (c) e estradosso (d).
La Metallurgia Italiana - Aprile 2022
pagina 66
Scientific papers - Continuous Casting
Effetto del contenuto di azoto
profondità, e conseguentemente la gravità, delle marche di
sa famiglia e presentano tenori di elementi in lega identici,
Tuttavia, è possibile evidenziare che per le bramme centrali
Le marche di acciaio 316L e 316LI appartengono alla stes-
con la sola differenza del contenuto di azoto, come riportato in Tab. 1. Questo permette la valutazione dell’effetto
della variazione del contenuto di azoto e una sua possibile
interazione con la polvere di colata rispetto alla formazione
delle marche di oscillazione, poiché tutti gli altri parametri di processo sono invariati. Per raffinare ulteriormente il
confronto tra le due marche acciaio ed evidenziare anche eventuali piccole variazioni sulla numerosità delle marche di oscillazione, sono state prese in considerazione soltanto le marche di oscillazione di spigolo, che, come visto pre-
cedentemente sono le più critiche, e sono state normalizzate rispetto alla lunghezza totale della bramma prodotta.
I dati così ottenuti sono riportati in Fig. 10 e mostrano che per le bramme di avvio, transizione e chiusura di colata (51,
51R, 83, 83R), soggette a decrementi della velocità di colata, l’acciaio 316L, avendo un tenore di azoto più elevato,
è più suscettibile alla genesi di marche di oscillazione di spigolo. Questo fenomeno è dovuto alla minore deforma-
bilità a caldo [11], conseguenza anche di piccoli incrementi
del tenore si azoto in lega, che limitano la formabilità della
pelle solida della bramma in formazione e accentuano la
oscillazione.
della sequenza l’acciaio 316L, con un tenore di azoto più elevato, è al contrario meno suscettibile alla genesi di marche di oscillazione di spigolo rispetto al 316LI. Come visto
precedentemente, una volta raggiunte le condizioni di la-
voro a regime ottimali, le marche di oscillazione presenti
sulle bramme di questa tipologia d’acciaio sono prevalentemente di spigolo e generate più per una locale scarsa lubrificazione dovuta a vincoli geometrici. In questo caso,
il più alto contenuto di azoto risulta avere un effetto positivo sul sistema, contrastando l’effetto dei maggiori sforzi
interni dovuti all’incremento del gradiente termico. Infatti,
il maggior quantitativo di azoto dell’acciaio 316L, ricombinandosi, dà luogo alla formazione di microbolle gassose
di N2 che, flottando sulla superficie dell’acciaio liquido, interagiscono con la polvere di colata liquefatta, diminuendone la densità e, conseguentemente, aumentandone la
viscosità cinematica e quindi la capacità d’infiltrazione [12].
Una maggiore capacità d’infiltrazione permette una migliore lubrificazione degli spigoli tra lingottiera e pelle della
bramma in formazione riducendo così la profondità e la gravità delle marche di oscillazione.
Fig.10 - Rappresentazione del valore medio del numero di marche di oscillazione di spigolo per unità di lunghezza.
Conclusioni
del processo quali velocità di colata, lubrificazione e con-
ne visiva dei prodotti di colata continua, sono state messe
316L e AISI 316LI.
Attraverso l’utilizzo di un sistema automatico di ispezio-
in relazione le caratteristiche delle marche di oscillazione (profondità, posizione e numerosità) con alcuni parametri
La Metallurgia Italiana - April 2022
tenuto di azoto per due marche di acciai austenitici, AISI L’interpretazione dei risultati ha permesso di stabilire che:
1. la velocità di colata influisce significativamente sulla pro-
pagina 67
Memorie scientifiche - Colata Continua
fondità delle marche di oscillazione e sulla loro disposizio-
ne sulla superficie del semilavorato. In particolare, velocità inferiori a quella di regime favoriscono la genesi di marche di oscillazione centrali;
2. il comportamento lubrificante della polvere non è omogeneo sull’intera superficie del semilavorato, a svantaggio
degli spigoli, zona particolarmente critica, caratterizzata da una maggiore incidenza delle marche di oscillazione, e
3. il tenore di azoto condiziona il comportamento della polvere di colata, andandone a modificare fisicamente la
viscosità cinematica. A velocità di colata di regime, l’acciaio 316L trae vantaggio dal maggior tenore di azoto in lega,
beneficiando di una migliore lubrificazione sugli spigoli, da
cui consegue una minore gravità delle marche di oscillazione.
dell’estradosso;
BIBLIOGRAFIA [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
(2008) Casting, in ASM Handbook, ASM International, Materials Park (OH), pp. 41–55. Irving, W.R. (1993) Continuous casting of steel, CRC Press, Leeds. Sguigna, D. (2019) Dispense del corso di siderurgia. Lukin, S. V., Shestakov, N.I., and Strashko, T.I. (2007) Heat transfer of slab with mold wall in continuous-casting machine. Steel Transl., 37 (3), 198–201. Cibulka, J., Krzok, R., Hermann, R., Bocek, D., Cupek, J., and Michalek, K. (2016) Impact of oscillation parameters on surface quality of cast billets. Arch. Metall. Mater., 61 (1), 283–288. Sengupta, B.G., and Ojeda, J. (2006) Mechanism Of Hook And Oscillation Mark Formation In Ultra-Low Carbon Steel. Carbon N. Y., 1, 112–117. Lei, Z.S., Zhang, B.W., Deng, R., and Ren, Z.M. (2002) Study on meniscus temperature fluctuation during mold oscillation in continuous casting by modeling experiments. J. Shanghai Univ., 6 (3), 236–239. Camisani-Calzolari, F.R., Craig, I.K., and Pistorius, P.C. (2003) A Review on Causes of Surface Defects in Continuous Casting. IFAC Proc. Vol., 36 (24), 113–121. Steelnumber.com (2016) European Steel and Alloy Grades / Numbers. 41–43. Takeuchi, E., and Brimacombe, J.K. (1984) The formation of oscillation marks in the continuous casting of steel slabs. Metall. Trans. B, 15 (3), 493–509. Barella, S., and Gruttadauria, A. (2017) Metallurgia e Materiali Non Metallici, Esculapio. Svyazhin, A.G., Kaputkina, L.M., Bazhenov, V.E., Skuza, Z., Siwka, E., and Kindop, V.E. (2015) Phases and defects upon the solidification of nitrogen-alloyed stainless steels. Phys. Met. Metallogr., 116 (6), 552–561.
Lavoro vincitore del Premio Aldo Daccò 2021 ex aequo con il lavoro “Evaluation of the use of foundry sand cores on
solidification and on the characteristics of structural castings: a comparison between organic and inorganic cores and validation with simulation” già pubblicato sul numero di marzo 2022”.
La Metallurgia Italiana - Aprile 2022
pagina 68
Scientific papers - Continuous Casting
Use of an automatic optical inspection system for the detection of defects arising from continuous casting processes, based on Machine Learning algorithms, for the analysis of incidence of oscillation marks on AISI 316L and 316LI austenitic stainless-steel slabs Nowadays, continuous casting is the technology for the solidification of steel that is prevailing throughout the world.
It has established itself thanks to the high productivity, the metallurgical high-quality of the products and the elevated degree of automation that distinguishes it from other processes. However, defects arising from continuous casting can
compromise the structural integrity of the semi-finished products as well as affect the quality of the final product. For this reason, automatic slab inspection systems have been developed in recent years to monitor the surface of the semi-finished casting products, in order to limit further processing of products with critical and dangerous defects.
By exploiting the potential of the inspection system installed on one of the casting lines of Acciai Speciali Terni S.p.A., it was possible to evaluate the impact of oscillation marks on AISI 316L and 316LI austenitic stainless-steel slabs. The aim of this research is to correlate marks distribution and their depth to the main process parameters.
The analysis showed that the casting speed is a critical parameter that proportionally affects the oscillation marks distribution along the slab. In addition, the analysis revealed a lower lubrication of the edges, as oscillation marks increase in
terms of frequency and depth in this position, especially on the bottom side, because of the weight of the semi-finished
product itself. The preferential occurrence of this defect on one side of the slabs is associated with a slight misalignment of the casting machine. The different nitrogen content of the two steel grades analysed contributes to modify the casting powder lubricating behaviour on such edges.
The results show that the AISI 316L stainless-steel transition slabs (i.e., those at the beginning and at the end of the sequence) are far more affected by the presence of edge oscillation marks than AISI 316LI. The trend is reversed for the cen-
tral slabs of the sequence, for which the greater availability of nitrogen in AISI 316L leads to an increase of the kinematic viscosity of casting powder, resulting in improved lubrication.
KEYWORDS: CONTINUOUS CASTING, STAINLESS STEELS, OPTICAL SYSTEMS FOR AUTOMATIC DEFECT DETECTION, OSCILLATION MARKS;
La Metallurgia Italiana - April 2022
pagina 69
Premio
Felice De Carli 2022
L’Associazione Italiana di Metallurgia, per onorare la memoria del prof. Felice De Carli, ex Presidente dell’AIM, istituì nel 1968 un premio da assegnare ad un giovane ricercatore di cittadinanza italiana, che non avesse superato il 32° anno di età al momento della data di presentazione della domanda di concorso al premio e che avesse dimostrato di possedere un’adeguata maturità nel settore della ricerca metallurgica fondamentale e applicata. L’Associazione ha deciso di bandire nuovamente il concorso per l’assegnazione del premio, consistente in una somma dell’importo di 1500 Euro e in una medaglia di ricordo recante l’effige del prof. Felice De Carli. Per concorrere al premio occorre presentare domanda, anche a mezzo e-mail, con il testo di uno o più articoli originali del concorrente oltre che il curriculum vitae e l’elenco dei lavori già pubblicati o in corso di stampa. La domanda va inviata, entro il 15 luglio 2022, alla Segreteria AIM, e-mail: info@aimnet.it. La Commissione Giudicatrice, nominata dal Consiglio Direttivo AIM, a suo insindacabile giudizio, sceglierà l’Autore meritevole del premio, sia in base all’esame del lavoro inedito che dei titoli presentati. La consegna del premio avverrà nel corso del 39° Convegno Nazionale AIM, che si terrà a Padova dal 21 al 23 settembre 2022. Milano, 5 aprile 2022
Per informazioni e candidature:
Associazione Italiana di Metallurgia
Via Filippo Turati 8 - Milano t. +39 02 76021132 - t. +39 02 76397770 www.aimnet.it - info@aimnet.it
Industry news - Attualità industriale
The intelligent ultra-wide caster for high-quality slabs at Rizhao Shandong
edited by: Z. Quiang, Z. Dengbao, J. Yuan, P. Heidemann, I. Olgemöller, J. Wans, R. Wilmes, L. Fischer The single-strand caster is designed for an annual production of up to 1.5 million tons of steel slabs with widths of up to 3,250 mm and a thickness of 150 mm. This means the plant is able to cast the widest slabs in the world. Structural steels as well as micro- and low-alloy steel grades are produced on the caster, and peritectic grades make up approx. 45 percent of the overall production output. The slabs are then processed to sheet metal plate and hot strip in the Steckel Mill using the hot charging technique. The jumbo caster is equipped with smart and proven technologies, such as HD moldTC/FO, X-Pact® Level Control, X-Pact® Width Control, X-Pact® Solid Control, which are implemented assuring high-quality slab production. In the maintenance shop the digital aligning assistant HD LASr (High Definition Laser Aligning System remote) gets segment and mold alignment performed easily and perfectly. The PQA® (Product Quality Analyzer) system documents, monitors and secures the entire production process in the continuous casting plant. The plant is also equipped with Industry 4.0 technologies, they provide the smart processing of plant data and automation of processes. Based on the steel grades to be produced, the X-Pact® Tech Assist automatically selects the optimal technological parameter settings for the metallurgical process.
KEYWORDS: SLAB CASTING, ULTRA-WIDE CASTER, HIGH-QUALITY STEEL GRADES, DIGITALIZATION, INDUSTRY 4.0 TECHNOLOGIES; INTRODUCTION Shandong Iron & Steel Group set up a new green field production base producing prime quality steels in Rizhao, Shandong Province in China. The steelmaking plant consists of 4 x 210 t BOF, 2 x 210 t ladle furnace, 3 x 210 t RH, 2 x 2 strand slab caster, 1 x 1 strand ultra-wide slab caster (#3), 2 other relocated slab caster. #3 is an ultra-wide medium thickness slab caster. The format range of slabs is 150 mm thickness and 2,000 – 3,250 mm width. Steel grades being produced are mainly structural steels, micro- and low-alloy steels. The slabs are hot charged to the SMS group supplied Steckel Mill, which rolls them to sheet plate and hot strip. The single-strand caster is designed for an annual production of max. 1.5 million tons
Zheng Qiang, Zhao Dengbao
Shandong Iron & Steel Group Rizhao Corp. China
Jihua Yuan
SMS group China
Peter Heidemann, Ingo Olgermöller, Dr. Jochen Wans, Ronald Wilmes, Lothar Fischer SMS group China
of steel slabs. After intensive technical negotiation and market investigation SDIS Rizhao Base finally chose SMS group as the supplier of the jumbo caster. Basing on its successful jumbo caster projects and rich experience, advanced mechanical design, technology La Metallurgia Italiana - April 2022
pagina 71
Attualità industriale - Industry news know-how and latest electrical and automation X-Pact®
logy for its high-quality steel grades and high output pro-
packages, SMS group supplied SDIS Rizhao Base one
duction.
jumbo continuous caster including Industry 4.0 techno-
Fig.1 - Production line at continuous casting plant, in particular: the ultra-wide caster, furnace for Steckel mill, Steckel plate mill in SDIS Rizhao Base. The caster project scope comprised the design, manu-
ded SMS group the final acceptance certificate,
facture and supply of core mechanical parts, the electrics
following the successful commissioning of the new conti-
and automation, and the supervision of erection and com-
nuous caster for ultra-wide slabs 6 months after the
missioning.
first cast.
SDIS Rizhao Base, Shandong province in China, has awar-
Fig.2 - Jumbo caster in production.
La Metallurgia Italiana - Aprile 2022
pagina 72
Industry news - Attualità industriale
Fig.3 - JThe widest slabs cast by caster # 3 at SDIS.
PLANT DESCRIPTION AND MAIN DATA
cro-alloyed, alloyed steel grades, which are used in lar-
The single-strand ultra-wide caster is designed for the
ge buildings, bridges, ships, pipelines, ocean platforms,
production of qualified steel slabs in flexible formats,
boiler and pressure vessels and machinery, provided in
which match the 3500 mm Steckel mill production mix in
thicknesses of 150 millimeters for Steckel mill. During ca-
SDIS Rizhao Base.
sting, the width can be steplessly set from 2,000 to 3,250
The range of grades produced comprises structure, mi-
millimeters.
Tab.1 - Main technical data.
La Metallurgia Italiana - April 2022
pagina 73
Attualità industriale - Industry news
The caster features a vertical length of approx. 2.5 m thus
for bulging or cracking. The strand bending as well as the
tering segregation bands in the slab. The strand guide is
several points enabling minimum bending / unbending
accounting for uniform strand solidification and for cen-
composed of 9 segments with a smart roller configuration for optimum strand support minimizing tendencies
straightening within one segment are performed each over strains for best internal and surface quality.
Fig.4 - 3D caster design. SMS group’s intelligent slab casting modules are linking design, process, technology and automation.
cares for exact taper and slab width adjustment.
The X-Pact® Width Control is equipped amongst others with:
INTELLIGENT MOLD
The mold, being of compact type and featuring a length of
900 mm enables the casting of 150 or 180 mm thick slabs as a future extension. The narrow sides can be online set and
controlled according to individual tapers by X-Pact® Width Control. A soft clamping function during the width change results in reduced adjustment forces and less wear of the
copper plates. The electromechanical narrow side adjust-
ment featuring upper and lower high-precision spindles
Delta Speed Adjustment serves to adjust bigger width changes within shorter transition length. It allows lower
compression (width decrease) or reduced gap (width increase) during adjustment.
The width and taper adaption function ensures an optimi-
zed width and taper setup of the mold. Thus, it optimizes
the slab width and the heat flux on the narrow sides, which sustains the casting process by uniform melting of casting powder in this sensitive area.
Fig.5 - Delta speed adjustment with considerably reduced transition lengths. La Metallurgia Italiana - Aprile 2022
pagina 74
Industry news - Attualità industriale With X-Pact® Level Control, using an eddy-current mea-
ckers thus predicting and minimizing breakouts.
suring system, a stabile mold level is achieved. A hydrau-
Further, the heat flow over the mold width and height is
lic actuator system consisting of a rapidly moving hydrau-
measured and the temperature distribution in the mold
lic cylinder with integrated position transducer actuated
is displayed as a map.
by a control-valve moves the tundish stopper rod to keep
Summing up HD moldTC/FO visualizes the processes in the
constant mold levels. The stopper mechanism featu-
mold, helps for optimizing the mold taper settings, cares
res high thermal stability and tight tolerances in order to
for minimizing sticker breakouts and permits the opti-
translate the rapid hydraulic movement onto the desired
mization of casting powder for an improved slab surface
stopper rod movement.
quality. One set of HD mold FO narrow side is supplied in the
The intelligent mold comprises also the HD moldTC/FO (TC
project. By a much higher density of measuring points
= Thermocouples or FO = Fiber Optics) enabling an un-
compared to thermocouples, it can prevent gutter forma-
disturbed casting process for increased plant availabili-
tion for best slab quality.
ty. Based on a reliable thermocouple or the fiber optical installation in the mold, the system allows detecting sti-
Fig.6 - example of broad face HD moldFO. Copper plate coating plays an important role in the quality
decreased and the heat flow through shell and mold pow-
of highly sophisticated steels sensitive to cracking at high
der into the mold is kept more homogeneous.
temperatures. Without coating or following the wrong
Copper plate life time can be increased using the UNI-
coating philosophy pinches can form close to the
GUARDTM concept with 1mm full face coating by factor
meniscus zone and grow up to longitudinal cracks in the
2.5 for broad faces and 3.5 for narrow faces [2].
edge zones of the solidifying and shrinking slabs. UNIGUARDTM metal-ceramic coating from SMS group
RIGID MACHINE HEAD DESIGN
leads to higher copper plate surface temperature at sta-
By leaving the mold’s foot rollers the strand is guided
bile casting speed compared to e.g. the formerly coating
smoothly to the first roller of the well-aligned segment 0
concept with NiCo. Contraction of the strand shell is thus
and accordingly to segment 1.
La Metallurgia Italiana - April 2022
pagina 75
Attualità industriale - Industry news
Fig.7 - Compact machine head. During insertion, segment 0 and segment 1 are easily in-
to assure an optimal strand support, slab geometry and
stalled as they are self-positioning and self-connecting to
quality of ultra-wide, medium thick slabs.
the water supply lines by water-connecting plates. Fur-
First, a smaller bow radius is used to lower the machine
thermore for quick maintenance, mold, segments 0 and 1
height and reduce the ferrostatic pressure, which leads to
can be removed as a package in one step.
very slim segment frame design for easy installation and
The leaf-spring guided oscillation system is actuated by 2
removal. Secondly the segment frames are strengthe-
hydraulic cylinders. With online adjustment of stroke and
ned with heavy and big size plates to bear much higher
frequency, the hydraulic oscillation offers complete flexi-
ferrostatic pressure coming from ultra-wide slabs. The
bility. Sinusoidal and asymmetric sinusoidal wave forms
segments built with rigid plate design have very long seg-
are possible. X-Pact® Mold Oscillation controls and mo-
ments life time as well.
nitors reliably the synchronized movement of the cylin-
Thirdly the roller bodies of the segments are split into
ders on the left and the right side. The system cares for
four parts to satisfy the strict requirements imposed on
reduced and very even oscillation marks and a good di-
the strand guiding system of ultra wide slabs for minimum
stribution of casting powder, important for a slab surface
roller deflection. All these minimizes segment and roller
quality that enables hot charging.
deflection, which significantly increases with increasing
The smart and well proven leaf-spring guided system
casting width.
reduces the total oscillation weight, is mechanically we-
In addition, the segment body construction was engine-
arfree and almost maintenance-free. It reaches the best
ered with the help of the finite-element method (FEM) to
oscillation shape, which is favorable for strand shell and
achieve proper segment stiffness.
copper plate interaction.
The cold/hot strand is driven on both sides at the upper
Different oscillation curves are selected depending on
and lower center roller to ensure exact guiding of all slab
the used steel group, suggested by the X-Pact® Tech As-
sizes and an optimal force transmission on the ultra-wide
sist. It offers real-time displays of measured values of the
strand. The drive control X-Pact® Strand Drives provides
oscillator like positions, pressures and neg. strip, moni-
a high degree of synchronicity between the two electric
tors strand friction forces helping to indicate bad or in-
drives per segment and the interaction of all segments
sufficient mold powder and hydraulic or mechanical pro-
during casting. It ensures a predefined distribution of the
blems.
total load among the segment motors which are pressed on the cold/hot strand.
OPTIMUM STRAND SUPPORT AND HANDLING
Precise aligning of the strand guide is one of the important
The strand support and segments are specially designed
keys for producing high-quality slabs.
La Metallurgia Italiana - Aprile 2022
pagina 76
Industry news - Attualità industriale
For caster #3 at SDIS Rizhao Base the reliable, easy, pre-
group software undertaking 3D measurement and au-
cise and digital measuring system HD LASr (High Defini-
tomatic storage of data. It perfectly aligns the mold and
tion Laser Aligning System remote) was provided by SMS
strand guide and digitalizes the measuring data to increa-
group. The digital aligning assistant combines the best
se quality and decrease maintenance costs.
available hardware laser tracker and the intuitive SMS
Fig.8 - HD LASr [segment] used for segment alignment. Perfect edge temperature avoiding corner cracks is reali-
Dynamic soft reduction® and safe casting.
zed with width-dependent secondary cooling. Up to five control loops each on loose and fixed side ensure flexible
The X-Pact® Gap Control is designed for segments with
adaptation of cooling media distribution over the ultra-wi-
position control. Its functions include
de slab width in the critical segments. The fix distance of
•
Dynamic soft reduction® to improve the internal slab
nozzle to slab maintains constant impact and constant
quality, ensures a well-adjusted taper for each casting
heat transfer. An increasing product range is realized by
situation.
air mist cooling.
•
Intelligent Taper Tracking to avoid negative effects on
X-Pact® Solid Control (replaces the previous DSC®) and
mold level, prevents segment overload and avoids
includes the well-proven model for temperature control
high withdrawal forces by smooth changing the ther-
of the secondary cooling. Its extended functionality en-
mal taper in all casting situations.
sures even and controlled strand shell growth, offers the
•
Segment Gauge Control to ensure defined slab geo-
option of selecting "dry casting" in the horizontal strand
metry, prevents internal cracks caused by imperfect
guide, improves the edge quality through controlled
segment transition. It compensates the segment ben-
edge cooling for all slab widths, calculates and precisely
ding and cylinder elongation under load conditions.
regulates the final solidification point and ensures with a homogeneous solidification front for besterformance of
La Metallurgia Italiana - April 2022
•
Withdrawal Reduction to prevent high segment forces after strand stop and roll slippage for strand start
pagina 77
Attualità industriale - Industry news
•
by controlling the transmitted torque.
(Product Quality Analyzer). Based on data acquisition and
Slant Control to prevent mechanical damages of the
administration of the above systems the X-Pact® Cast Op-
rolls/segments, decreases the failure rate and impro-
timizer tracks the complete material from arriving at the
ves slab geometry.
ladle turret, transport through the caster until leaving the run out area. During this, the system optimizes the com-
X-Pact® Cast Optimizer is centrally embedded in the ove-
plete casting process using several functions beyond the
rall automation infrastructure. It features interfaces to the
tracking as e.g. casting speed optimization, determination
Level 1 and other Level 2 systems of the caster and is con-
of heat transition, supervision of current steel analysis or
nected to other Level 2 systems, the Level 3 and the PQA®
cut-length optimization.
Fig.9 - Control pulpit of #3caster. PQA® AND INDUSTRIE 4.0 The steel grades produced by the steelworks encompass structural steels as well as micro and low-alloy steel grades used in high buildings, bridges, ship and marine, boiler and pressure vessels, machinery, pipelines as per the below table. Tab.2 - SDIS Rizhao Base steel grade groups.
La Metallurgia Italiana - Aprile 2022
pagina 78
Industry news - Attualità industriale
Fig.10 - SMS group smart plant concept. SDIS Rizhao Base also ordered latest quality enhancing digital solutions for the casting plant from SMS group.
PQA® (Product Quality Assessment & Assurance, also known as QES® Quality Ensuring System) is an integrated product quality system. It covers the essential quality aspects by the acquisition of relevant process data and
events from the steel melt shop, the caster up to the slab yard or plate / coil yard in case of direct hot slab charging.
The quality-determining variables and process parameters are automatically evaluated at regular intervals. The re-
sults are used to improve process control and to create a
reproducible setting of the desired quality characteristics. Further benefits of the PQA® are the improvement of qua-
lity management and the full transparency of the process and product quality. PQA® helps to assure a superior slab
quality and to compare the results for hot or cold-charged
slabs. It stabilizes operational performance, enhances con-
fidence in quality decisions and enables the integration of continuous improvement by flexible rule adaption.
Additionally, the plant is equipped with Industry 4.0 te-
chnologies developed by SMS group for smart plant data
processing and process automation. Based on the steel
grades to be processed, the X-Pact® Tech Assist automatically selects the optimal technological parameter settings for the metallurgical process.
X-Pact® Process Guidance provides automatically - whenever needed - all relevant process information and
prompts operator screens. All plant control and maintenance functions can be conveniently executed from the
operator's station. X-Pact® Business Intelligence combi-
nes the data from different systems, enabling interaction with dynamic applications and dashboards. PRODUCTION & PRODUCTS
After the start of the hot commissioning the caster has shown a stable production curve. As an example: In the last two months of 2019, 904 heats, which means 210,707.00 t liquid steel were cast.
The production of different steel grade groups in these two months has been portioned as per the following graph.
Fig.11 - Production of steel grade groups in Nov. & Dec. 2019. La Metallurgia Italiana - April 2022
pagina 79
Attualità industriale - Industry news
The slab widths cover nearly the whole format spectrum; however emphasis is on the medium width range of 2,250 – 2,750 mm.
Fig.12 - Produced slab widths in Nov. & Dec. 2019. The slab surface quality records for Nov. 2019 are only 9 slabs with surface defects (cracks). There are no slabs with surface defects in Dec. 2019.
Fig.13 - Slab piles in slab yard. The superior internal quality is highly-demanded for slabs to be rolled for plate products in Steckel mill. The macro solidification analysis show that class C internal quality is achieved in the casting process.
Fig.14 - Macro solidification analysis: Class C 0.5 (central segregation, cracks, porosity, holes, ..) according to Chinese Slab Quality Standard YB-T4003-2016.
La Metallurgia Italiana - Aprile 2022
pagina 80
Industry news - Attualità industriale
CONCLUSIONS
vertical bending caster supplied by SMS group, commis-
producing prime quality steel grades in Rizhao, Shandong
modules of SMS group. The latest smart and well-proven
SDIS Rizhao Base is a new green field production base Province in China, set up by Shandong Iron & Steel Group (SDIS).
SDIS Rizhao Base is successfully producing high quality ul-
tra-wide slabs comprising of demanding steel grades. The
sioned in 2019 is equipped with the Industry 4.0 slab casting
technologies such as HD modules, various X-Pact® packages and the PQA® system assure the ultra-wide slab quality achieving more than 99% defect-free slabs.
REFERENCES [1] [2]
Newsletter SMS group, no 4 2019 Innovative mold coating technology positively influences strand surfaces, Stahl und Eisen “135 (2015), Vol. 12, Pp 87 – 95
La Metallurgia Italiana - April 2022
pagina 81
Attualità industriale - Industry news
Powder dosing with mould temperature feedback control in continuous casting of stainless steel for high quality billet surfaces edited by: C. Scarabelli, D. Olivero, F. Bego, I. Mazza, S. Spagnul, G. Schiavon
Ergolines has a consolidated experience in the implementation of closed-loop mould powder feeding with inductive or optical sensors. In the recent years, Ergolines has also implemented closed-loop mold powder thickness control based on the ultrasonic sensor (ULD). The present installations of UT-MAP are working to measure and control mould powder thickness, providing a feedback signal to the automatic powder feeders (MPF). The temperature trends were cross-correlated with the results of steel quality analysis on tracked billets: billets cast under manual powder feeding featured several occurrences of bleedings and powder entrapment events caused by the saw-tooth meniscus perturbation. ULD+MPF was installed in Cogne Acciai Speciali CAS, Aosta Plant (Italy), in April 2019 and the application was developed by CAS Technicians. Evidences are that thermal data obtained by ultrasonic sensor are strictly related to chemical composition of the steel and to casting process parameters. A fine tuning quality control, based on thermal data (ULD) controlling automatic powder feeding (MPF), was developed in Cogne Acciai Speciali. The optimization of the powder quantity according to CAS specific know how was targeted to minimize powder entrapment phenomena, providing a more stable steel meniscus temperature. Defects occurrence on billet surfaces were reduced by 75%. On Final products, bright bars 100% eddy current controlled, the scrape rate due to cracks related to entrapment phenomena has been reduced by 50%.
KEYWORDS: MOLD THERMAL MAPPING, ULTRASONIC SENSOR, MOLD POWDER THICKNESS CONTROL, STAINLESS STEEL, HIGH QUALITY, TECHNOLOGY, INNOVATION; INTRODUCTION The crucial impact of the initial solidification on steel quality is well-established1-2. Furthermore, the occurrence of defects in the cast products is directly related with the dynamics of the fluid flow in the mold and with meniscus stability 3,4. The temperature distribution along the copper tube in the meniscus region provides key data on heat transfer, skin formation and meniscus stability. However, traditional techniques to measure the copper temperature involve invasive probes such as thermocouples (TC) or optical fibre cables (OFC) 5, which require expensive ma-
C. Scarabelli, D. Olivero, F. Bego Cogne Acciai Speciali, Italy
I. Mazza, S. Spagnul, G. Schiavon Ergolines Lab, Italy
chining of each single copper tube and complex management of multiple cable outlets in a critical area. Due to these drawbacks, TC- or OFC-based mold thermal monitoring is typically limited to slab casting and plate molds, being not feasible in the everyday casting practice of small
La Metallurgia Italiana - Aprile 2022
pagina 82
Industry news - Attualità industriale
sections. In fact, due to installation criticalities and costs,
information on the skin formation and heat transfer in the
TC and OFC application to billets and blooms is typically
mold, potentially enabling to fine-tune key casting para-
restricted to research projects and validation of simula-
meters such as casting speed, primary cooling and mold
tion models for scientific purposes. In order to overco-
lubrication to minimize the breakout risk and enhance
me these limitations, Ergolines developed an ultrasonic
productivity. The ULD sensor therefore represents a key
sensor (ULD) able to measure the copper temperature in
tool to Metallurgists, Quality Control experts and Pro-
a fully contactless way, without the need to machine the
ductivity Managers6-10.
copper tube. In fact, the ULD sensor is installed within the water jacket and features one single cable outlet.
ULD INSTALLATION
The ULD measures the copper temperature at several lo-
In contrast to TC or OFC, which require expensive machi-
cations along the mold wall, providing real-time data on
ning of the copper tube and feature several cable outlets,
the meniscus thermal profile and copper temperature
Ergolines’ ULD is installed within the water jacket, requi-
trends over time.
ring no machining of the copper tube at all. Furthermore,
Historic ULD data collected on the steel plant database
the ULD has one single cable outlet, enabling a fast con-
can be correlated with the quality control data acquired
nection and disconnection of the signal cables (Fig. 1).
on tracked billets. These data can in turn be analyzed and
The front side of the sensor is aligned with the inner wall
exploited by the CCM specialists to improve the casting
of the water jacket, enabling the primary cooling water to
practice, leading to significantly improved steel quality.
flow in front of the sensor.
Furthermore, the temperature trends provide important
Fig.1 - Ergolines’ ultrasonic sensor (ULD): Non-invasive installation in the water jacket.
Fig.2 - Concept of Meniscus Thermal Profile. Colors are used to associate thermal profile points with copper volumes.
La Metallurgia Italiana - April 2022
pagina 83
Attualità industriale - Industry news
Fig.3 - Temperature trends measured by a 4-points ULD sensor. Color code is the same as the one used in Fig. 2 (T4 corresponds to the volume which is the closest to the top of copper).
ULD WORKING PRINCIPLE
Furthermore, each copper temperature can be plotted
The copper temperature across the meniscus is characte-
against time, providing the copper temperature trends.
rized by a marked gradient, corresponding to the meni-
An example of copper temperature trends measured over
scus thermal profile (Fig. 2). The profile typically displays
time is displayed in Fig. 3. The four temperature points are
a maximum slightly below the meniscus, while skin deta-
labelled T1, T2, T3 and T4, where T4 conventionally cor-
chment generally corresponds to a temperature decrease
responds to the volume which is the closest to the top of
in the area immediately below the profile peak. The ULD
the copper tube. Casting start and casting end are clearly
generates an ultrasonic beam which crosses the water gap
seen by a sharp increase and decrease in the temperature
and propagates along the copper wall of the mold. The
trends, respectively, while temperature variations during
ultrasound is then collected by an array of receivers and
casting reflect the dynamics of the steel meniscus and ca-
the signals are processed by a dedicated inversion algo-
sting powder.
rithm developed by Ergolines. Since ultrasound velocity
The copper temperature data measured by the ULD can
in copper depends on temperature, copper temperatures
be stored in the steel plant database and correlated with
can be determined by signal processing based on mathe-
the results of Quality Control analysis on tracked billets,
matical inversion. The algorithm determines the copper
providing crucial information to Metallurgists and CCM
temperatures at several locations along the mold wall in
experts. In fact, by analyzing the ULD data, the casting
the meniscus region. The scheme in Fig. 2 explains this
specialists can fine-tune key casting parameters including
concept in the case of four temperature points. Due to
mold powder thickness, casting speed and primary water
the physical dimension of the ultrasonic beam, each tem-
cooling, leading to an optimized casting practice and im-
perature is averaged over a copper volume of typically 2
proved steel quality.
cm (colored copper regions in Fig 2). The meniscus ther3
mal profile is obtained by plotting the copper tempera-
Application of ULD to Automatic Mold Powder Feeding
tures measured by the ULD against the vertical coordina-
Ergolines has a well-established experience on the design
te of the corresponding copper volume. This concept is
of automated powder feeders and in the implementation
explained in Fig. 2 by assigning to each profile point the
of closed-loop powder thickness control based on de-
same color as the copper volume over which the tempe-
dicated sensors (electromagnetic, optical and ultrasonic
rature is averaged.
sensors)11-19. Besides providing real-time mold thermal
La Metallurgia Italiana - Aprile 2022
pagina 84
Industry news - Attualità industriale
mapping, the ULD signal can be used to close the control
gy developed by Ergolines16. In fact, it is well-known that
loop of an automated powder feeding machine, enabling
the ability to keep the powder thickness constant during
accurate and real-time control of the mold powder thick-
casting is crucial to obtain high quality steel based on a
ness (PTC-Powder Thickness Control). Closed-loop PTC
reproducible process12,15. By combining ultrasonic mold
based on the ULD can be implemented either on pre-in-
thermal mapping with ULD-based automatic powder fee-
stalled powder feeding machines or on Ergolines’ Mold
ding, the metallurgists can determine the optimal value of
Powder Feeders (MPF, Fig. 4). The MPF option enables to
powder thickness for each steel grade and keep the pow-
fine-tune the powder thickness in real-time with high ac-
der thickness constant with Ergolines’ MPF technology,
curacy, based on an innovative powder dosing technolo-
leading to enhanced and reproducible steel quality.
Fig.4 - An example of Ergolines’ Mold Powder Feeder (MPF). INDUSTRIAL OPERATION
Typical casting speeds range from 1,2 m/min to 1,6 m/min
The closed-loop powder thickness control, ULD+MPF,
with a superheat of about +40°C.
was implemented in Cogne Acciai Speciali, Aosta Plant (Italy), targeting the scrap reduction on final products
Product Quality Improvement
through the improvement of billets surface quality.
As discussed before, the goal of the project started in Co-
The Continuous Casting machine in which the system was
gne Acciai Speciali was the improvement of the quality
tested has features listed hereafter:
in the final product, with particular reference to defects
• 80 Tons ladle; 15 Tons tundish;
arising in peeled bars and coming from the steelmaking
• Curved machine 10 m radius; 4 strands;
process.
• Sliding gate + radioactive Metacom-Berthold for level
On peeled bars, typical anomalies due to casting and put
control;
into evidence at final eddy current control are the “crack
• Hydraulic oscillation;
with powder entrapment”. Although the only analysis of
• M-EMS + F-EMS;
the surface morphology of the defect, fig.5, and even the
• Level2 automation.
eddy current itself cannot clearly identify the nature of the defect, the SEM exam, fig.6, reveals the root cause: conti-
CAS casts only Stainless Steel grades, in three different
nuous casting powder entrapped.
billet sizes: • 160 mm x 160 mm; • 220 mm x 270 mm; • 280 mm x 340 mm.
La Metallurgia Italiana - April 2022
pagina 85
Attualità industriale - Industry news
Fig.5 - Cracks on peeled bar.
Fig.6 - Powder entrapment (micro etching).
The powder entrapment that generates the defect on pe-
countermeasure could be to remove it by grinding.
eled bar, evidently comes from the continuous casted
But, even not considering the loose of productivity due
billet. In fact in some occasions anomalies are already
to grinding, in many cases powder entrapments are sub-
evident there: in fig.5 a billet with a surface defect is pre-
cortical and their presence is subtle and not ascertainable
sented. This type of defect is visually detected on billet
with a visual check. That is why Cogne Acciai Speciali de-
surface and it is generated by lack of powder lubrication in
cided to investigate the phenomenon and to counteract it
the mould during casting.
at root: in mould.
The defect presented in fig.5 is extremely evident, and a
Fig.7 - Cracks with powder entrapment. PROCESS KEY-PARAMETERS
6. Oscillation control (stroke, frequency);
To prevent powder entrapment during continuous ca-
7. Primary cooling rate;
sting operations, there are many key-parameters to take
8. SEN geometry and immersion depth;
in account:
9. Mould geometry (taper, corner radius).
1. Steel level control;
Moreover all the parameters must be set in agreement
2. Powder type (viscosity, basicity, melting temperatu-
with steel chemical composition.
re, %C free); 3. Powder feeding control;
Cogne Acciai Speciali, CAS, worked in order to stabilize
4. Casting speed;
the “Powder Feeding Control” parameter by joining
5. Casting temperature;
the ULD system with the MPF system, both systems in-
La Metallurgia Italiana - Aprile 2022
pagina 86
Industry news - Attualità industriale
stalled in Cogne Continuous Casting Machine.
SYSTEM APPLICATION IN INDUSTRIAL OPERATION
The main idea was to set the Powder Dosing on the MPF
From April 2019 in Cogne Acciai Speciali, Aosta Plant
(automatic powder feeding) according to the temperatu-
(Italy), the UT-MAP system (ULD) is working to measure
re measured by the ULD sensor The approach suggested
and control mould powder thickness, providing a fee-
by CAS was implemented by Ergolines trough a software
dback signal to the automatic powder feeders (MPF).
update, so that tests were industrially operated in the Cogne Plant.
Fig.8 - Ultrasonic sensor installation and working principle.
ULD Sensor is set in each mould water jacket on lateral
the instantaneous mold Powder Dose (%), serving as fe-
side.
edback to drive MPF (automatic mold powder feeder) in
There are four point of temperature detection (T1, T2, T3,
closed-loop mode.
T4) on the mould for Cogne ULD (see Fig.8). T3 is the temperature at the meniscus point.
Data collected in casting clearly show that T3 (meniscus temperature) changes with %Powder Dose (see Fig.9). In
ULD system measures the temperature profile and calcu-
particular the trend of Fig.9 put into evidence how:
lates a % of Powder Dose. Each casting strand (4 in Co-
Low Powder Dose% -> Low lubricating layer thickness ->
gne CC machine) has a MPF powder feeder. We set a %
Steel attached to the mould -> High T3 temperature.
of Powder Dose on the ULD+MPF HMI. The ULD measure
Fig.9 - T3 (°C) vs %Powder Dose. La Metallurgia Italiana - April 2022
pagina 87
Attualità industriale - Industry news
This approach makes it possible, setting the correct
quantity obtained with the use of the ULD+MPF system in
%Powder Dose, the optimization of the lubrication insi-
conjunction with the correct %Powder Dose guarantee a
de the mould. The optimal %Powder Dose parameter is
more stable steel meniscus temperature (see Fig.10) with
function of the behavior in solidification of different steel
respect to the manual powder adduction. In CAS expe-
grades, so finding the way of predicting its correct value
rience the only MPF system, automatic powder feeding,
(analytically or by experience) is the key to improve the
without feed-back from ULD, not guarantee the same sta-
billet quality surface.
bility of steel meniscus temperature.
During Cogne analysis, from May to December 2019 in
Moreover, the implementation of the ULD+MPF makes it
more than 700 heats, all the temperature data were col-
possible, managing the data, to put the casting process
lected and managed in LEVEL2 automation with a dedica-
under an additional control, also using control chart to
ted report.
control powder feeding stability (Fig.11), so that
Collected data show that the optimization of the powder
minimizing powder entrapment phenomena.
Fig.10 - ULD Plant Data.
Fig.11 - Control Chart – T3 standard deviation vs Casting month.
La Metallurgia Italiana - Aprile 2022
pagina 88
Industry news - Attualità industriale
INDUSTRIAL DATA KNOW-HOW
(see Fig.13). The interpretation makes reference to the
With ULD+MPF we are able to optimize powder feeding
following relations:
according to steel because we have a measure very sensi-
Low Viscosity -> Low lubricating layer thickness -> Steel
ble to process parameters variation.
attached to the mould -> High T3 temperature.
As discussed before in general terms for all the grades,
Also, it is noticed that meniscus temperature is correla-
for austenitic stainless steel the meniscus temperature is
ted to oscillation parameters which control powder con-
strictly correlated to steel grade (chemical composition,
sumption during lubrication in the mould (see Fig.14), so
ferrite content), because the chemistry control the shrin-
that High frequency -> High powder consumption -> Low
kage of solid steel shell in the mould (see Fig.12).
lubricating layer thickness -> Steel attached to the mould
In this case the phenomenon could be interpreted as fol-
-> High T3 temperature.
lows:
By analyzing a real-time thermal map of the copper
Low Ferrite content -> Low shrinkage -> Steel attached to
mould, we have the possibility to understand deeply the
the mould -> High T3 temperature.
most critical topics about the initial solidification and to
Another important fact that has to be taken into account
propose effective solutions, including fine-tuning of the
is that meniscus temperature is also correlated to Pow-
casting key-parameters, improvement of the operative
der Viscosity, because the flux reology control the lubri-
casting practice and optimized billet surface.
cating layer thickness between solid shell and the mould
Fig.12 - T3 (°C) vs Ferrite content.
Fig.13 -T3 (°C) vs Powder viscosity (dPa.s).
La Metallurgia Italiana - April 2022
pagina 89
Attualità industriale - Industry news
DEFECT OCCURRENCE
quantity really minimize powder entrapment phenomena,
2019, there was a reduction of 75% of powder entrapment
We have confidence that ULD data, collected in CAS plant,
Using ULD+MPF system, from May 2019 to December defects on billet surface and by 50% the scrape rate due to cracks on final peeled bars. CONCLUSIONS
ULD+MPF (Ergolines system) was installed in Cogne Acciai Speciali CAS, Aosta Plant (Italy), in April 2019. A fine tuning quality monitoring, based on thermal data (ULD) control-
ling automatic powder feeding (MPF), was developed according to CAS specific know how. CAS plant data, more than 700 heats, show that the optimization of the powder
providing a more stable steel meniscus temperature.
can be useful for a future development of this system. We want to set alarm on Continuous Casting LEVEL2 automa-
tion to separate billets when the meniscus temperature
(T3 Ergolines) is out of optimal range, defined by very fine tuning of several plant data (chemical analysis, continuous
casting parameters, billet surface defects, scrape rate on final peeled bars).
In this way we plan to reduce even more the scrape on final product.
REFERENCES [1] [2] [3] [4] [5] [6]
[7]
[8]
[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
Lait, J.E.; Brimacombe, J.K., Solidification During Continuous Casting of Steel, Continuous Casting, Vol. 2, 1984, p. 171. Emi, T., Surface Defects on Continuously Cast Strands, The AISE Steel Foundation, Pittsburgh, PA, Ch. 21, 2003, p. 1. Thomas, B.G.; Modelling of Continuous-Casting Defects Related to Mold Fluid Flow, 3rd Internat. Con-gress on Science & Technology of Steelmaking, Charlotte, NC, AIST, Warrendale, PA, 2005. Thomas, B. G.: On-line detection of quality problems in continuous casting of steel. Modelling, Control and Optimization in Ferrous and Nonferrous Industry, Chicago, IL, TMS, Warrendale, PA, 2003. Ecke, W.; Applications of Fibre Bragg Grating Sensors, IPHT Jena, The 19th International Conference on Optical Fibre Sensors, Perth, 2008. Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2019: Fully Contactless Mold Thermal Mapping and Breakout Prevention for Billets and Blooms: An Innovative System to Improve Metallurgy, Quality and Productivity, Proceedings of AISTech 2019, 6-8 May 2019, Pittsburgh, Pa., USA Spagnul, S.; Mazza, I.; Miani, S.; Schiavon, G.; 2019; Breakout Prevention for Billets and Blooms through Contactless Mould Thermal Mapping: A New Tool for Metallurgists, Quality Control and Produc-tivity Improvement, Proceedings of METEC 2019, 25-29 June 2019, Düsseldorf, Germany Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2019: Real-time and Contactless Mould Thermal Moni-toring: Improving Metallurgy, Quality and Productivity of Billets and Blooms, BHM Berg- und Hüt-tenmännische Monatshefte, Austrian Society for Metallurgy of Metals (ASMET) and Bergmännischer Verband Österreich (BVÖ) (Proceedings of ESSC and Duplex 2019, Sept. 30 – Oct. 02 2019, Vienna, Austria) Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2018, Contactless Mold Thermal Mapping at Meniscus through an Innovative Ultrasonic Sensor, Proceedings of ICS 2018, Venice, Italy, 13-15 June 2018. Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2020: The mold temperature mapping with Ultrasonic Contactless Technology is the key for the real-time initial solidification process control tools, Proceedings of ECCC 2020, 17-19 June 2020, Bari, Italy (pre-print). Spagnul, S.; Olivo, L.; Schiavon, G.; Mazza, I.; 2017; A Compact Mold Powder Diffuser with Built-In Opti-cal Powder Thickness Measurement, Iron and Steel Technology, Dec. 2017, Vol. 14, No. 12, pp. 64-70. Mazza, I.; Spagnul, S.; Olivo, L.; Milani, F.; 2017, Review of Technologies and Methods for Mold Pow-der Thickness Control, La Metallurgia Italiana , No. 4, pp. 30-37. Spagnul, S.; Olivo, L.; Schiavon, G.; Mazza, I.; 2017, A New Instrumented Mold Powder Diffuser with Built-in Optical Sensor for Powder Thickness Control, Proceedings of ECCC 2017, Vienna, Austria, 26-29 June 2017. Olivo, L.; Spagnul, S.; Mazza, I.; 2016, A New Optical System for Mold Powder Thickness Control by La-ser Scanning and Multi-Spectral Imaging, Iron and Steel Technology, Dec. 2016, Vol. 13, No. 12, pp. 62-69. Mazza, I.; Spagnul, S.; Mantovani, F; 2015, Review of the Mold Powder Control Technology, Proceed-ings of METEC 2015, Düsseldorf, Germany. Spagnul, S.; Padovan, M.; Bianco, A.; Mantovani, F.; 2014, Latest Enhancements in Mold Powder Thick-ness Control as a Result of a New Propulsion System Implemented in Flux Feeding, Proceedings of ECCC 2014, Graz, Austria. Spagnul, S.; Mantovani, F.; 2011, A Reliable Powder Control based on an Automatic Closed Loop Sys-tem including Measurement, Powder Feeding and Powder Thickness Control, Proceedings of METEC 2011, Düsseldorf, Germany. Mazza, I.; Spagnul, S.; 2015, A Novel Ultrasonic Sensor for Mold Powder Thickness Control, Proceed-ings of METEC 2015, Düsseldorf, Germany. Spagnul, S.; Mazza, I.; Schiavon, G.; Miani, S.; 2019: An Innovative Ultrasonic System for Steel Level Control, Proceedings of 7EFRS 2019, International Iron and Steel Symposium, 26-27 Sept. 2019, Izmir, Turkey.
La Metallurgia Italiana - Aprile 2022
pagina 90
Industry news - Attualità industriale
Automated mold flux feeders for Industry 4.0 application M. Zinni
New sensor technologies have been applied to the mold flux feeder, automating the feeding processes and enabling data collection of process information previously unavailable. Analysis of the information to determine operational, maintenance and quality relationships critical to sustainability and operational efficiency is now possible. The current automation and technology of the Imerys feeding systems, and examples of observations from the mold flux feeder of mold level variability and bias flow will be presented. The objectives using edge computing, ERP systems, augmented reality and cloud analytics of previously isolated data will be discussed with the outcome of integration with operational, maintenance and inventory management solutions.
KEYWORDSS: MOLD FLUX FEEDERS, SENSORS, AUTOMATION, MOLD FLUX, CONTINUOUS CASTING;
INTRODUCTION: MOLD FLUX FEEDERS Applying mold flux with feeding equipment has been a standard practice since the introduction of granular mold flux for continuous casting of steel. Imerys equipment designed specifically for mold flux application followed during the late 1990’s [1] as the conversion from powder to granular mold flux occurred. The equipment provides even, constant flux addition, replacing a manual addition practice where the operator would add the flux as a batch application. The negative impact of the batch addition practice has been measured and well documented over the years for the effect on mold level and meniscus stability [2], and ultimately quality and operational performan-
Mike Zinni
Imerys Steelcasting, USA
ce. Despite the improvement over the manual additions practice, the early feeding equipment still had limitations and lacked the ability to measure, control and collect information from the process. The advancement of flux feeding automation has been developed with the addition of sensors to automate the flux feeding and enable collection of process information.
La Metallurgia Italiana - April 2022
pagina 91
Attualità industriale - Industry news
1. Load cells for consumption monitoring
the next step is to use the information for self diagnostics,
The information from load cells allows consumption to
monitoring and analysis.
be monitored and collected. Programs for independent mold feeding by application rate are now possible for
EXAMPLE OF LASER MOLD LEVEL MONITORING
billet and bloom casters. Alarms for feeding issues and
Mold level control has a major impact on the steel quali-
machine empty status can be observed.
ty, and much work has been done to improve the stability and is well documented [3]. There is a heavy reliance on
2. Lasers for feeding control
the level control system, to be precise in response to the
Lasers for slab casting allow measuring of the flux thick-
uncertainties and disturbances of the casting process. The
ness with calibration to the mold level system. The rate
necessity for tuning and filtering of the mold level sen-
of flux application is controlled to maintain a steady flux
sor for process control can lose sight of what is actually
thickness.
occurring in the mold, and consequently can give a false impression of stability.
3. Thermal flux thickness control The temperature of the flux surface provides an indica-
For slab casters, the mold flux feeder uses lasers to mea-
tion of the flux thickness from the insulation of the layer
sure the distance to the top surface of the flux. The sensor
thickness.
positions are calibrated to the mold level as a reference point through the feeder PLC program to read in mm (or
4. Silos with automatic flux selection
inches) of flux thickness. The outcome is a controlled fe-
Silos with valving and level measurement allow auto-
eding to keep the flux thickness constant at the desired
matic flux changes from the HMI and/or customer si-
thickness.
gnal. Silos refilling is also indicated. The use of high precision lasers also provides additional 5. Robotic arm for distribution
information based on the absolute distance change seen at the surface of the flux layer, and gives a second view of
A programmable linear actuator moves the flux feeding
the mold fluctuations and actual level control. During ap-
outlet across the mold surface for optimal flux distribu-
plication, variations can be observed that may go unnoti-
tion at all points in the mold. Control of the speed and wi-
ced with the caster steel mold level system and collected
dth of coverage are realized.
data in the caster system. Laser mold level monitoring is based on the principles below:
The control automation and data acquisition bring additional benefits to the feeding process at the mold. Main-
- The lasers measure independent of each other, on
taining a stable, even flux layer on the steel provides con-
each side of the mold
stant thermal insulation at the meniscus and removes the
- Using the unfiltered signal for quick response, fluctua-
unknown variability of an operator controlled flux layer
tions occur together in magnitude and frequency indi-
practice. Feeding under controlled conditions displays
cating true mold level variability
information that can be critical to the quality and operational stability of the process regarding mold level and bias
The following figures demonstrate the measurements:
flow. Moving forward, it is not enough to simply control and collect, but to take advantage of the information with predictive modelling and monitoring. As the evolution of the feeding equipment continues into 4.0 integration,
La Metallurgia Italiana - Aprile 2022
pagina 92
Industry news - Attualità industriale
Fig.1 - The lasers (red and light blue) and feeding signals (purple and dark blue) are displayed on the chart. The setpoint distance for each laser (orange and green) are followed for control of the mold flux thickness. The trend is monitored over ½ hr, using a laser scale of +/- 1.25 inches. The agreement of the magnitude and behavior of the laser signals demonstrates the presence of mold level fluctuations.
Fig.2 - During an unstable mold event capture for 300 seconds, the lasers (red and green) were recording the actual mold level variation. The customer mold level signal (purple) is on the same scaling and followed the same trend, but with a filtered response not representing the true mold level fluctuations.
Fig.3 - The graph shows good agreement in signal behavior and amplitude fluctuation. between the laser signal (blue) and the mold level (black) on the same scale range, giving confidence to the mold level control performance. La Metallurgia Italiana - April 2022
pagina 93
Attualità industriale - Industry news
EXAMPLE OF BIAS FLOW MONITORING
using the lasers and load cells, the feeding rates can me-
Bias flow can be detrimental to steel quality in the con-
asure the bias effects at the steady state flux thickness. A
tinuous casting mold [4]. The additional flow to one side
phenomenon is observed when bias flow is present that
of the mold results in excessive turbulence causing mold
the feeding of mold flux becomes unbalanced to maintain
level fluctuations, flux entrapment and can also inhibit the
a constant flux thickness. This is indicative of an uneven
steel shell formation. Determination of bias flow is diffi-
thermal profile to melt the flux into slag. Figure 3 displays
cult and often not apparent as it occurs below the mold
a bias flow event captured prior to a tube change.
flux layer. With the automatic feeding of the mold flux
Fig.4 - At steady state speed (grey) and flux thickness (yellow and dark blue), a bias flow condition occurs prior to a tube change. During the bias flow event, a 4:1 difference of mold flux consumption was measured.
Bias flow is a phenomenon that can be noticed with the
also a tool used [5], but measured information identifies
automated feeding of the flux feeder. This information
actual events in real time that have an impact on quality
properly analyzed can potentially assist with understan-
and productivity.
ding the flow relationship with cast speed and width,
Future objectives are applying decision making and pro-
the presence of plugging, and help evaluate tube design
cess steering occurring in real time. This will first require
changes in operation.
identification of correlations from the data. To assist with the collection and combination of data from the devices
DATA ANALYTICS
and process on a centralized platform, edge computing
The examples of mold level and bias flow monitoring are
devices now exist. In addition, to provide the initial analy-
just 2 observations from the sensor information presen-
sis, cloud database platforms can be accessed to perform
ted by the mold flux feeder. There are likely more key
remote evaluation of the edge data, to enable continuous
observations that can be made through evaluating the in-
improvement.
formation and finding correlations. The traditional use of
The outcome from the data analysis will bring benefits
data is to gather information and store it in a database for
by continuously developing software to assist in the fol-
analysis after an event. Modeling of the casting process is
lowing ways:
La Metallurgia Italiana - Aprile 2022
pagina 94
Industry news - Attualità industriale
•
Comparing signals to verify optimal operation of de-
allowing the expert to freeze views, annotate areas, share
vices
their computer screen onto the glass lenses, and translate
Determining events linked to quality and operational
language in real time. This tool is currently being brought
stability
into the Imerys organization and will benefit the customer
Evaluating mold flux performance parameters and
by increasing availability, and lowering intervention time
optimization
and cost.
Evaluating caster equipment (ex. new tube design,
Industrial remote access devices are also providing re-
mold level control adjustments)
mote support for PLC and HMI maintenance allowing a
•
Feeder maintenance and performance management
secure connection to the feeder. The access point can li-
•
Automated reports of events and overview of pro-
mit authorization only to specific devices, such as the PLC
ductivity
and HM of the feeder, restricting access to the industrial
Inventory management with ERP connectivity to cu-
network for cybersecurity purposes. Connection to the
stomer and vendor
system can be accomplished through the local network,
• • •
•
wifi or a dedicated 4G network. Programs can be updated MAINTENANCE OBJECTIVES
as new features become available, or more automation is
The flux feeder is designed for low maintenance, but oc-
requested remotely.
casionally it is necessary to troubleshoot an issue. Having a workforce close to the customer is a benefit for reaction
CONCLUSION
time, but does not always bring the necessary knowledge
Mold flux feeding equipment is a known and accepted te-
and experience required. Augmented reality technology
chnology in steelmaking. The feeder has been further de-
to connect with an equipment expert located remotely is
veloped with sensors and automation that are presenting
now available to bring an expert onsite. Smart glasses al-
additional opportunities beyond consistent flux feeding.
low a specialist to work together with an onsite resource.
The data from the sensors manifest events in real time that
Wearing a set of glasses and microphone on a hard hat,
can impact steel quality and operational performance of
the local resource can make a connection through a mo-
the continuous caster. The Industry 4.0 objective for the
bile phone. The expert connects via laptop virtually co-
feeder is to bring the information from an isolated envi-
ming onsite to see what the local resource sees. The on-
ronment for increased analysis, automation, diagnosis
site resource has a field of vision that is also collaborative,
and monitoring.
REFERENCES [1] [2] [3] [4] [5]
Schoner F. Zinni M. Apparatus for introducing granular mold flux onto the top of a slab being cast within a continuous casting mold. US Pat. 6,474,398 B1, 2002 Nov. 5 Lieftucht D. Reifferscheid M. Schramm T. Krasilnikov A. Kirsch D. HD mold a new fiber optical based mold monitoring system. Iron and Steel Technology. 2013 Dec. Suzuki D. Formulation of mold level control model by molten steel flow analysis method. Nippon Steel Technical Report No. 89. 2004 January; 46-49 Zhang L.Cai K. Li J. Wan X. Thomas B. Investigation of fluid flow and steel cleanliness in the continuous casting strand. Metallurgical and Material Transaction B. 2005 June Thomas, B. Mathematical models of continuous casting of steel slabs. Annual Report to Continuous Casting Consortium UIUC. 1993 August 18.
La Metallurgia Italiana - April 2022
pagina 95
Attualità industriale - Industry news
Dynamic superheat determination in a continuous casting machine - process, practice and benefits P. Hughes-Narborough, P. White, G. Humphrey
There is an increasing need to accurately control the parameters associated with casting liquid steel in a continuous casting machine. One of the main parameters is superheat, which is defined as the elevated temperature of the liquid steel above its liquidus. Liquidus is defined as being the first point of freezing. Accurate measurement of this elevated temperature is required to reliably control the casting speed of the machine. and in order to reflect the conditions in the Continuous Casting mould, which is where the liquid steel actually freezes at the start of the solid product forming process. A sensor mounted close to this and continuously monitoring the temperature is the superior solution. In order to complete the superheat picture, an accurate assessment of the liquidus must be provided in conjunction with continuous temperature. The uncertainty in the liquid steel temperature and the uncertainty in the liquidus calculation has led to a situation where liquid steel is sent to be cast with excessive superheat with a resultant over use of energy and materials. Heraeus Electro-Nite (HEN) have developed CasTemp Superheat package as a means for enhancing the visualisation of dynamic Superheat during casting, and ultimately helping to improve the control of casting through optimal use of the features included in the package. An overview and early adoption of the system is given highlighting the potential benefits for controlling superheat thereby reducing energy requirements and improving the associated process control parameters.
KEYWORDS: SUPERHEAT, LIQUIDUS, PRODUCTIVITY, ENERGY SAVING, PRODUCT QUALITY, OPTIMISATION, PROCESS CONTROL; INTRODUCTION There are significant pressures on steel makers and casters to ensure that their plants are operating to achieve several objectives. Amongst many factors these are some of the key: safety, productivity, low cost, environmentally friendly – low emissions, high yield, excellent product quality, minimal scrap. Once the steel has been processed through primary and secondary refining then the casting machine must ensure that the steel is cast with all aforementioned factors in mind and accurately control the parameters associated with casting liquid steel in a
Philip Hughes-Narborough, Peter White, George Humphrey
Heraeus Electro - Nite (UK) Ltd
continuous casting machine. One of the main parameters is superheat, which is defined as the elevated temperature of the liquid steel above its liquidus. Liquidus is defined as being the first point of freezing and solidus as the last point. These points and the freezing range are defined by the steel chemistry with the carbon content being the most significant factor.
La Metallurgia Italiana - Aprile 2022
pagina 96
Industry news - Attualità industriale
Accurate measurement of this elevated temperature can
analysis, validity of method model and timeliness. A new
be achieved by several methods. In order to reflect the
CasTip sensor has been developed to precisely measure
conditions in the Continuous Casting mould, which is
the liquidus in the CC tundish which is instantaneous and
where the liquid steel actually freezes at the start of the
accurate to +/- 0.5oC. The system includes integration
solid product forming process, a sensor mounted close
of the CasTip direct reading liquidus sensor to measure
to this and continuously monitoring the temperature is
tundish liquidus and hence offer a current dynamic
the superior solution. In order to complete the picture, an
superheat and a forward prediction towards end of cast.
accurate assessment of the liquidus must be provided in
The system is available through software control and
conjunction with Continuous temperature measurement
integration of inputted plant data. There is an excellent
to provide a reliable, accurate and dynamic superheat1.
correlation between CasTip and liquidus calculated from equivalent reference samples across a range of carbon and
As a result of some investigations into how much effort
some low alloy steels. Ultimately, there should be one
was put into regular understanding of superheat by
liquidus value for a given heat of steel, and HEN believe
different steel makers, it became apparent that it is not
that the CasTemp Superheat system (1) focusses closer
an area of regular examination. Indeed, several plants
to this value than previous approaches, particularly in
used do not review their liquidus practice and the basis
higher carbon steels. This allows for a dynamic superheat
of understanding was often not evident. Many plants are
to be available which gives the operator the potential to
using methodologies for calculating liquidus based on
optimise the casting process control.
equations and formulae derived in the middle of the 20th century, perhaps with some empirical adjustment over
BACKGROUND
the intervening decades. The current situation is that
Why is Superheat necessary?2
there appears to be no common standard and variation
In ideal conditions, steel would be cast with the minimum
in the liquidus calculation has led to significant process
amount of superheat for that grade to achieve low energy
issues, particularly when plants are making high carbon
cost, best product quality and productivity recognising
and / or more highly alloyed steels. The future steel grade
the superheat is a balance between metallurgy and
markets are continually evolving, and, as an example,
productivity within the machine design constraints. In
electrical steels are becoming more in demand to service
ideal circumstances maximum casting speed would be
the electric car market. These are high in silicon (~3%)
maintained at all times, which would mean minimum
and outside the usual product range for a standard carbon
steel residence time during casting if the superheat is
steel producer.
correctly positioned to allow for safe draining of the ladle and continuance of the sequence. However, there
In terms of practical solutions, it seems therefore that a
is uncertainty of the measurements in a traditional setting
system that delivers superheat rather than relying an ever-
and there is the constant operational battle to keep plant
changing model (for calculating liquidus) would be of
and equipment online and available to produce steel to
significant benefit to steel makers and casters. Delivery of
meet the casting times. These uncertainties refer not only
improvements to process control via productivity, energy
to temperature measurement but also to weighing of
savings (and emission reductions), and product quality
ladle and tundish contents and accounting for unknown
can simultaneously be achieved. The traditional method
variables such as slag. A small deviation from the expected
of calculating the liquidus is through an analysis of a
can lead to ladles emptying earlier than expected or
chemical sample and by use of a mathematical formula,
overrunning and this can cause major disruption back
equation or model that has limitations due to accuracy of
through secondary refining and onto to the primary
La Metallurgia Italiana - April 2022
pagina 97
Attualità industriale - Industry news
refining furnace.
potentially producing steel with lower than expected quality at slower rates. Excess superheat; uncertainly in
In fact, the caster design is usually predicated on optimum
measurement and lack of consistency and standardisation
product quality at maximum casting speed so that full
lead to sub-optimal performance. Most of the casting is
benefits of product quality and productivity are achieved
generally broken down into a simple relationship between
simultaneously. The imperative should therefore be
superheat and casting speed. The higher the former, the
to cast at maximum speed for that grade. However,
slower the latter. The principle is robust to ensure that
because a ladle’s previous thermal history cannot always
the strand product emerges from the mould with enough
be accounted for, then there is an empirical adjustment
shell thickness to retain the liquid and solidifying core. As
based usually on a discretionary subjective analysis of this
the mould and machine can extract only a finite amount of
thermal load which accounts for a part of the uncertainty.
heat at any speed then excessive heat has to be removed by exposing the strand to a lengthier cooling period.
How does superheat influence the casting parameters?
The liquidus temperature is grade dependent, varying
Good metallurgical knowledge – theoretical and practical
in commercial carbon steels from about 1450°C to
to ensure that robust casting techniques are applied
1530°C dependent upon steel chemistry. It is normally
at minimal cost and a desire to utilise the machine
modelled via linear formula from the steel chemistry,
effectively. Low superheat is a desirable aim to cast steel
each plant favouring its own derivation of formula.
because of the inherent better as cast product quality
Superheat will vary in the range (+00C) to +700C. Low
derived from equiaxed zones in the as cast product. There
superheat indicates an increased risk of freezing, and high
are the practical constraints of ensuring the maximise
superheat indicates an increased risk of breakout. There
yield from each heat and specifically ensuring enough
are several areas of improvement for the casting operator
superheat to allow a controlled changeover to the next
to address through improved control of superheat. These
heat without jeopardising the sequence continuity. In
include productivity, product quality, energy reduction,
fact, the observations seen in a majority of steel plants
operational control.
is that steel is presented to the caster with excessive superheat. Generally, this amount of superheat is far in
Productivity
excess of a sensible and controlled quantity for a known
The casting machine should be run at the designed
and predictable period of casting:
maximum speed for the relevant grade in order to generate tonnes and minimise casting time and if not,
Variability in the reliability of steel assets over many years
then why not? There is always a scheduling constraint
has led many steel makers and management to regard
between primary and secondary refining, where heats are
the continuation of a sequence and the buffer of excess
batch processed in fixed quantities (dependent on ladle
superheat in case of local difficulties to override the sum
size) and continuous casting. The latter is unconstrained
of constraints identified such as energy cost, productivity
by heat size but has its own constraints in terms of design:
and indeed a potential impact on product quality.
throughput / casting speed being chemistry grade related
The desire for liquid tonnes can seem to be the only
and the metallurgical length will define what is the
imperative that matters and as long as the caster is casting
practical casting speed (particularly in a slab caster) and
there is a knowledge that production is occurring. Indeed,
the dimensional requirements of the customer output.
metal in mould is often a key benchmark. Efficiency,
Logistical constraints can also impact but in principle the
cost management and quality are often deemed lower
caster should be driven at the maximum possible for the
priorities leading to sub optimal use of the machine and
grade to maximise productivity but also benefit from the
La Metallurgia Italiana - Aprile 2022
pagina 98
Industry news - Attualità industriale
design characteristics of the machine in terms of product
quality. A further reduction in superheat is then available
quality. As an example of a small increase in casting speed,
to be taken advantage of as the casting time is reduced as
by casting at 1.05 m/min rather than 1.0m/min, this 5%
the ladle does not take so long to be emptied.
uplift in speed could generate 50,000 tonnes of output per annum (in a million tonnes output caster). Another way to
As this is a critical area of control it is often considered
view this is to consider a 5% speed increase would reduce
just another piece of data and Fig.1 – Casting Parameters,
the casting time and empty the same ladle weight in 95.2%
shows how variation in liquid steel temperature can be
of the original time. For a 120tonne ladle casting in one
displayed on a typical steel plant amongst other casting
hour (at 2t/min) this would take 57 minutes (at 2.1t/min)
parameters. This type of display, although full of useful
generating a 3 minutes saving. For a 20-ladle sequence
information, masks the critical temperature infections that
approximately one hour is saved, enabling one additional
occur and does not aid the operator in making important
cast to be made in the same overall time period. This
and timely decisions.
improves output, refractory utilisation, yield and product
Fig.1 - Casting Parameters.
Product quality
which is capital and operationally expensive to install
As product quality is related to superheat, a lot of effort has
and maintain and the effects are more beneficial at lower
gone into caster machine design to ameliorate the impact
superheat. As the sump is determined by metallurgical
of superheat control to some extent. Soft reduction
length then the designed location is positioned when
has allowed casters to significantly improve internal
stable casting at maximum speed occurs.
segregation and Electromagnetic stirring (EMS) can have a similar effect. However, to address the superheat control
Energy
where action can be taken is a considered way forward.
Energy costs are a significant part of the steelmakers’
Product quality is linked to caster design and casting
cost base. The actual makeup of the energy will vary from
speed. Segregation is more likely to occur in steel casting
site to site and the cost will indeed vary according to the
with higher superheat. This is a result of the formation
season, day and time. However, the fundamental factor
of a smaller area of equiaxed zone and a larger dendritic
of raising the one tonne of liquid steel by 1 degree is not
area potentially leading to intercolumnar quality issues.
insignificant. A simplified calculation attempts to outline
Although EMS has mitigated these effects to some extents,
the main cost factors (see Figure 2 Cost factors in Steel
not all casting machines are fitted with this technology
Plant Energy).
La Metallurgia Italiana - April 2022
pagina 99
Attualità industriale - Industry news
Fig.2 - Cost factors in Steel Plant Energy.
Emissions are directly related to energy consumption and
METHODOLOGY
as current technologies are carbon based, a reduction in
What is the CasTemp Superheat system?
energy consumption will lead to a reduction in emissions.
There are 3 main components to achieve an accurate reliable and dynamic superheat:
Operational Control
• CasTemp – a sensor to reliably and accurately measure
Modern caster control requires sophisticated knowledge
the liquid steel temperature
of the casting process. To maximise casting speed to
• CasTip - to reliably and accurately measure the liquidus
design levels then a good understanding of the process
of that heat of steel
and the metallurgy is required. Of course, there may be
• Delivery platform – to marshal the information via an
significant factors that constrain the caster: metal supply
easy to use, digital and accurate instrument system for
(timeliness), previous practice influencing how the change
operator control
process is implemented along with other regulations
This equipment is shown in Fig.3 – Equipment for
and ways of working. However, by avoiding plant critical
CasTemp Superheat
events, such as breakouts and freeze offs, improvement is possible 3.
Fig.3 - Equipment for CasTemp Superheat. La Metallurgia Italiana - Aprile 2022
pagina 100
Industry news - Attualità industriale
The traditional model for relying on a potentially outdated
profile. This sensor can measure from before the start
model in order to calculate the liquidus and to measure
of preheating until the end of casting without any
the liquid steel temperature with a single isolated dip is
further interaction from the casting operator crew which
not sufficient to ensure that the modern-day caster can
significantly improves the safety of continuous caster area
perform to its optimum capability. The dip measurement
by removing the need for personnel to take dips and for
is replaced by continuous temperature measurement and
dipping equipment to be readily available nearby to the
a sensor mounted close to the tundish outlet provides
tundish, causing a hazard in itself, see Figure 5 to show
real time and accurate information to the operator, see
the close proximity of the installed sensor to the tundish
Fig.4 – Cross section of Tundish showing temperature
outlet.
Fig.4 - Cross section of Tundish showing temperature profile.
Fig.5 - CasTemp Sensor Installed in a Tundish.
It is clear from the installation that the thermocouple is
of that heat and into the start of the next heat as the inter
measuring the liquid steel temperature without any undue
heat mixing is completed, usually around 10 minutes. In
influence from other refractory bodies within the tundish
addition, once a CasTip is taken, a customer defined and
as a minimum clearance is calculated. Once the safe,
configurable, critical action limit, is displayed. The time
reliable and accurate CasTemp continuous temperature
that this will be reached as well as a predicted end of heat
measurement system is installed and measuring then
superheat value will also be advised if the algorithm for
the liquidus is taken via the CasTip sensor. This liquidus
the prediction is activated. To achieve all this, the system
measurement is necessary every heat as there can be a
must communicate bi-directionally with the plant level 2
change from heat to heat and sometimes a difference
via an established network link and display effectively to
can be noted. This can be due to a variety of factors in
the operator via
the steel making process from incorrect alloy additions
Fig.6 – CasTemp Superheat Operator Display.
to insufficient mixing or some problem with the alloying system. It is therefore more likely that higher alloyed steels are more likely to differ in liquidus values on a heat by heat basis. Once the CasTip is taken it will be displayed on the instrument screen alongside the ongoing CasTemp measurement and be readily observable for the remainder
La Metallurgia Italiana - April 2022
pagina 101
Attualità industriale - Industry news
Fig.5 - CasTemp Superheat Operator Display. The following features are available to the operator
temperature operations every 10 to 20 minutes throughout
• Liquidus evaluation via CasTip measurement
the casting periods of many hours and every day.
• Dynamic Superheat determination
Once CasTemp is proven the customer will then refer to
• Rate of change determination
their traditional practices and try to draw comparisons
• Visual trending
even though the measurement systems are in completely
• Visual forward prediction of superheat to end of cast
different zones of the tundish. As CasTemp measures
• Critical limits indication
as close as possible to the exit of the tundish then it is a
• Displays available in the control room and on the
very close approximation to the steel temperature within
casting floor
the mould. The mould is where solidification begins and
This then gives the operator at the caster (and elsewhere
the constraints within the mould such as consistent flow
in the steel shop) to be given guidance as to the likely
patterns and zones of temperature variation do not make
course of the Superheat during the casting of the heat. As
this a sensible position for temperature measurement. The
the process is dynamic then any changes in parameters,
liquid steel flow through the tundish and the introduction
such as the casting speed, are reflected through to the
of furniture can significantly affect what temperature is
prediction and new advice can acted upon.
actually measured because of the residence time in the tundish of the steel can be affected. Fast flow and the
RESULTS
absence of flow control can lead to short circuiting of hot
In the initial phase of trial work the customer needs to
steel straight into the mould.
have the system proven. This is understandable, the CasTemp continuous temperature measurement system
The next phase of trial work is for the CasTemp superheat
is a new way of working and is often felt to be counter
to be evaluated by the customer. Again, the first
intuitive to best practice. However, the system is proven
comparison will be to consider any existing liquidus
to operate extremely well and improves the overall safety
calculation and do a direct comparison with the CasTip
of the caster operations by removing the need for manual
measurement. In general, for most low carbon, lightly
La Metallurgia Italiana - Aprile 2022
pagina 102
Industry news - Attualità industriale
alloyed steels this gives good agreement, usually with 2-4
difference between the CasTemp and CasTip.
degrees C. However, for some steels there is a significant difference. E.g. electrical and high carbon steels. This is
Fig.7 – Historical data Trends shows the detail of the first
fundamentally because the liquidus models are outdated
few heats of a sequence when considered historically. It is
and require significant effort to be maintained with the
clear that there is considerable amount of superheat and
developing steel grades. However, the most important
some scope for reduction, which then makes the need
consideration is not the absolute value of the liquidus
for end of casting prediction far more relevant, see Fig.8
as determined by CasTip but the Superheat, which is
– Superheat.
Fig.7 - Historical data Trends.
Fig.8 - Superheat.
Fig.9 – CasTip difference from evaluated liquidus, shows
requirement to then reduce the Superheat can be a lengthy
how by measuring the superheat and understanding
management process and despite identifying key factors
how an incorrect calculation from traditional techniques,
as identified earlier, such as productivity, energy saving
particularly on highly alloyed steels (in this case carbon)
and difficulties with casting some grade, e.g. high carbon,
leads to an excess of superheat during the critical start up
customers seem reluctant to make changes to practice
period of the first ladle and in this case resulted in a strand
on a small scale to identify routes to improvement. There
breakout on the billet caster.
is a preference to remain with existing practice despite
In many cases Superheat can be seen to be in the range 30
knowing it could be improved.
to 50 degrees, with many examples exceeding this. The
Fig.9 - CasTip difference from evaluated liquidus.
La Metallurgia Italiana - April 2022
pagina 103
Attualità industriale - Industry news
Figures 10 and 11 are real examples of steel grades split
adjusted to aim for lower values. The impact on product
into a general high carbon group and non-high carbon
quality has not been assessed but it is clear that there
group where the CasTemp Superheat system has been
are demonstrable energy savings (with comparative cost
routinely employed. Table 1 shows both groups exhibit
savings) derived from avoiding attaining high superheat
high maximum but also high end of ladle superheat. The
and then having to remove it during casting.
preparation of the steel during steel making needs to be
Fig.10 - High Carbon Grades Superheat
Fig.11 - Non-High Carbon Grades Superheat
Maximum and EOL.
Maximum and EOL
Tab.1 - CSummary of maximum Superheat and EOL for High Carbon and Non-High Carbon steel grades. High Carbon grades
Non-High Carbon grades
Number of Ladles
77
197
Max SH average
37.8
40.8
End of Ladle SH average (EOL)
19.8
28.1
Average drop in SH
18.0
12.6
% casts below 10ºC SH EOL
7.8% (6)
2.5% (5)
DISCUSSION
ladle furnace and to primary steelmaking
An analysis therefore must take account of a wide body of
• Prediction of events should lead to improved planning
information, these include:
and control
• Maximum superheat during casting and at the end of
• Margin for unexpected misjudgements are reduced
ladle
• Virtuous cycle of casting shorter time requires a lower
• Duration of casting time
superheat.
To incorporate the temperature management system to take account of the thermal history of the ladle would
Better control leads to improved safety reduction in break
be of benefit to the casting machine but even without
out events and as importantly freeze off events as it is
this knowledge a significantly improved process control
clear what the direction of travel is and how long it will
package to take account of delivered liquid steel is
take to get there.
available to dynamically deliver the superheat., thus:
It can help to define what is the correct endpoint
• Evolution – improved control at caster to back feed to
La Metallurgia Italiana - Aprile 2022
superheat of a ladle. Each will have its own characteristic
pagina 104
Industry news - Attualità industriale
dependent upon the thermal history; refractory content
is required to make. High silicon content can alter the
and type and planned / actual duration of casting. The steel
liquidus value by a large amount. Using existing models
grade, residence time and the processes that have been
and testing methodology can lead the steel maker to
undertaken and for how long: such as stirring, heating and
develop casting practices that are not compatible with
alloying. The slag content and heat enthalpy will also be a
current practice. This might mean slower casting reducing
variable effect as will lidding a ladle and the point at which
productivity and affecting the economics of the plant and
this is done, for example either at end of processing or
the grade being produced.
only during the casting process.
Unexpected consequences can also occur such as unexpected freezing using a model that as shown can
An isolated Dip is unverifiable and by itself gives
provide a significant variation from reality 4. Other grades
no indication until the Dip measurement is taken. A
now moving into the commercial arena are TWIP,
continuous trend can start to indicate trends within a
TRIP, Corrosion resistant rebar and high manganese.
short period of time as the datapoints are produced
Most of these newer grades are significantly different
typically on a 15 second interval. Therefore, over a
in metallurgical chemistry and will therefore require
casting sequence of 24 hours then 5000datapoints can be
refinement. This places a challenge for the caster to keep
generated compared with approximately 4 or 5 per hour
pace with developments and maintain optimal casting
with a Dip measurement. Each continuous datapoint is
productivity.
then a fraction the cost of a single Dip type measurement. Product quality What will the future hold for the evolution of steel
It is well known that superheat affects the solidification
grades?
structure and there is a compromise with ideal and
From earlier it is clear that changing market conditions
practical casting conditions. Some of this can be mitigated
– e.g. light weighting; EVs; packaging requirements that
by utilising, for example EMS in the mould or strand but
there will continue to be a need to innovate in the steel
these have their own extra costs associated with capital
grade but the basic tenets of steel casting will remain
investment, operation and maintenance.
fundamentally the same but the technological and
In the caster of the future there will be more demand
operational imperative to ensure safe and efficient casting
from close control of the superheat to match the product
will progress. CasTemp Superheat is designed to assist
quality. Indeed, heating in the tundish may well become
the casting operator to achieve those aims.
more prevalent.
The use of CasTemp for reliability to ±1 degree for liquid steel measurement and a high degree of accuracy in
CONCLUSIONS
liquidus too is required to provide accurate and reliable
Dynamic superheat control of a ladle casting is a valuable
superheat.
and useful tool for the operator to accurately control the process parameters. There are significant benefits in terms
There are several scenarios where superheat control
of productivity, product quality, energy, and operational
can be significantly improved:
control.
New grades of steel - Technological evolution and market changes has driven the need for new grades of steel,
Some investigations have shown that only a few steel
greater light weighting in vehicular traffic and packaging
companies regularly review their liquidus and superheat
leads metallurgists to develop different grades of steel.
practices although it is considered to be essential to
Grades for electric car batteries are significantly different
have robust practices for successful casting. Trials at
in chemistry from existing grades that a steel maker
several plants have shown that superheat is higher than
La Metallurgia Italiana - April 2022
pagina 105
Attualità industriale - Industry news
is optimum leading to sub optimal practices of casting slower than desired for throughput and product quality reasons. A considerable amount of effort is required throughout the whole steel making and casting process to ensure the delivery of ladles of steel that can be cast in the desired superheat range. The CasTemp superheat system has the ability to forward predict the superheat towards the end of the ladle with a high degree of precision. This allows the operator to cast at a lower superheat with the confidence that even with a small error in delivery temperatures, the process can be managed through the casting cycle to maintain sequence casting
effectively
and
efficiently.
The
uncertainty
surrounding previous systems is effectively eliminated.
REFERENCES [1] [2] [3] [4]
Pagden, Hale, Whitaker, Hughes-Narborough et al. Dynamic Superheat Determination: A Fresh Approach. 2018 AISTech Conference Proceedings Liang, Cheek, Mustoe Low Superheat Casting Technology Utilized Through Control of the Steel Temperature in the Tundish. 1997 AISE Annual Convention and Exposition John Pischak, Alex England, Steve Walker Process Improvements at a Continuous Caster Using CasTemp SuperHeat 2019 AISTech Conference Proceedings Addes, Sabol Development and Implementation of the Process Model for Controlling Casting Superheat Temperature/ 1996 Steelmaking Proceedings
La Metallurgia Italiana - Aprile 2022
pagina 106
Industry news - Attualità industriale
The mold temperature mapping with Ultrasonic Contactless Technology is the key for the real-time initial solidification process control tools I. Mazza, S. Miani, G. Schiavon, S. Spagnul
The Ultrasonic Thermal MAPping (UT-MAP) technology is the new Ergolines’s approach for real-time temperature de-
tection of the copper mold, enabling to obtain key information on the first solidification based on the heat flux extracted in
the meniscus area. The intrinsic features of the system enable the implementation of several process control tools, such as mold powder thickness control, liquid steel level control or thermal mapping of the copper mold. Potential developments include breakout prevention and quality tracking.
The ultrasonic technology is different with respect to thermocouples or optical fibres because it is contactless, requiring
no machining of the copper tube at all: In fact, since the ultrasonic sensor is installed on the water jacket, the copper tube remains unaltered and can be replaced easily.
Ergolines’ ultrasonic sensor measures the copper temperature at several locations along the copper mold. Each temperature is averaged over a “copper volume” of about 2 cm3. The number of measuring points depends on sensor model and number of sensors installed. Each sensor can measure up to sixteen temperature points. The sensor provides two different datasets: the copper temperature trends over time and the meniscus thermal profile, namely the vertical temperature distribution in copper in the meniscus region.
One interesting application of Ergolines ultrasonic sensor is the mold level control in open stream casting (ULD - Ultrasonic
Level Detector). A single array of detection points (typically four to sixteen) can precisely estimate the liquid steel level, as shown by comparing the ultrasonic steel level measurement with the radiometric signal.
When steel level is controlled by a pre-installed radiometric sensor, the ultrasonic sensor can be used to keep the powder
thickness constant by associating it with a powder feeding machine operating in closed-loop mode with the feedback from the ultrasonic sensor.
Mold thermal mapping provides powerful information also for breakout prevention. By analysing the thermal map, an algorithm calculates any deviation from standard conditions generating warning and alarms based on custom thresholds which can be set by the steel plant metallurgists. Historic data form the ultrasonic sensor can be collected in the steel plant
database, correlated with quality control reports on tracked billets/blooms and used to improve the casting parameters and
operative practice. Ergolines’ ultrasonic sensor therefore represents a key tool for Metallurgists, Quality Control experts and Productivity managers.
KEYWORDSS: ULTRASONIC SENSOR, CONTACTLESS, MOLD THERMAL MAPPING, MOLD LEVEL CONTROL, MOLD POWDER FEEDING CONTROL, STEEL LEVEL CONTROL, BREAKOUT DETECTION; INTRODUCTION: INITIAL SOLIDIFICATION AND COPPER MOLD TEMPERATURE PROFILE The initial solidification which takes place within the copper mold tube has always been a main focus for steelmakers, engineers and academics, because the performance of the Continuous Casting Machine and the quality of the
Isabella Mazza, Stefano Miani, Giovanni Schiavon, Stefano Spagnul ERGOLINES LAB s.r.l.; Area Science Park, Bldg. R3;
Padriciano, 99 34149 Trieste, Italy; infosteel@ergolines.it
cast products are highly influenced by this crucial part of the solidification process1-4.
La Metallurgia Italiana - April 2022
pagina 107
Attualità industriale - Industry news During the last decades, technical efforts have been made
the meniscus area, which progressively decreases with a
in order to develop solidification models, validated
hyperbolic law if the solidification conditions are in a ste-
with dedicated instrumented molds. Industrial solutions
ady state situation. The heat flux peak located in the me-
based on thermocouples (TC) or optical fibers cables
niscus area is typically of the order of 4-6 MW/m2, and it
(OFC)5 were developed to detect the copper mold tem-
progressively decreases to less than 1 MW/m 2 at the mold
perature, providing mold thermal monitoring mainly ad-
exit. The steel meniscus is located slightly above the peak
dressing breakout detection or qualitative control of mold
of the copper temperature. As a consequence, the posi-
lubrication. However, both TC and OFC require invasive
tion of the liquid steel level can be determined from the
machining of the copper mold, while the ultrasonic sen-
copper thermal profile by applying dedicated algorithms
sor developed by Ergolines provides a fully contactless
based on heat flux modeling.
measurement of the copper temperature (UT-MAP – Ultrasonic Thermal MAPping).
Due to the high thermal conductivity of copper, the heat
The detection of the copper mold temperature provides
flows from the hotter to the colder face of the mold with a
real-time information on the effectiveness of heat tran-
fast step response: If a step temperature of 1000°C is im-
sfer in the mold: The measured temperature profile is in
posed at the inner wall of the mold in order to simulate an
fact related with the heat flux flowing from the liquid steel
increase of the steel level (or a bleeding), it takes approx.
through the mold walls. Beside many other parameters,
0.6 s for the temperature to cross a 10-mm-thick copper
the heat flux is directly linked to the thickness of the soli-
wall reaching the 67% of the steady state temperature
dified steel skin and the lubrication effectiveness.
(Fig. 1).
The heat flux in the mold is characterized by a peak in
Fig.1 - Simulation of heat flux in the mold and temperature step response inside a copper tube.
Both TC and OFC probes are installed into channels ma-
this reason, the thermal answer of these probes is not su-
chined within the copper, requiring expensive and invasi-
itable for a fast-reacting control since the heat flow needs
ve machining. The probes are cemented into the channels
too much time to cross the copper and the resin layers
by means of thermo-conductive resins (and/or springs).
(even considering no air in between).
However, these resins are much less conductive than the copper, thus creating a small thermal resistance which
Furthermore, the drawbacks of this kind of installation af-
slows down the step response of the TC/OFC probes. For
fect the cost of spare parts since every single mold must
La Metallurgia Italiana - Aprile 2022
pagina 108
Industry news - Attualità industriale
be machined. In addition, the installation is complex due
CONTACTLESS ULTRASONIC SENSOR
to the presence of multiple cable outlets, because each
Ergolines solved these problems by using a completely
TC/OFC probe comes with its own cable cemented into
different approach: Ergolines developed a non-intrusive,
the copper, which must be disconnected at each mold
real-time and fully contactless ultrasonic system for mold
change.
thermal mapping, requiring no machining of the copper
This technology is generally not used for mold level con-
tube at all, providing a cost-effective system, with a long
trol but mainly to prevent the breakouts, and today it is
life and virtually no maintenance 6-11.
limited to the CCM for slabs. This is the case not only be-
Ergolines’ UT-MAP (Ultrasonic Temperature MAPping sy-
cause of the importance of slab CCMs, but also because
stem) has very compact dimensions. As a main key-point,
the slab mold is made of flat plates, which are much easier
the installation of the UT-MAP ultrasonic sensor is inside
to machine with respect to copper tubes. For billet CCMs,
the water jacket (Fig. 2). This is a considerable advantage
which employ curved copper tubes, any CNC machining
because it requires minimal machining of the water jacket
of the copper becomes complex and much more expen-
only (long life component) while the copper tube remains
sive.
unaltered and can be replaced easily.
Fig.2 - Installation of Ergolines’ ultrasonic sensor in the water jacket for Ultrasonic Thermal MAPing (UT-MAP).
The Meniscus Thermal Profile is obtained by plotting the
lowest position in the mold. A schematic representation
four temperature values of each copper volume in front of
of the UT-MAP concept is provided in Fig. 3 and 4.
the sensor versus their respective vertical positions along the mold side. In the case of a 4-points thermal profile, the temperatures are labelled T1, T2, T3 and T4, where T4 is the closest to the top of the copper tube and T1 is in the
La Metallurgia Italiana - April 2022
pagina 109
Attualità industriale - Industry news
Fig.3 - Concept of ultrasonic temperature measurement: 1) Ultrasound propagates through the water gap and along the mold wall; 2) signal processing converts the ultrasonic signals into copper temperatures (colored beams); 3) the meniscus thermal profile is reconstructed (red line).
Fig.4 - Schematic representation of a typical UT-MAP measuring area: Copper temperature is mapped vertically in the meniscus region
The mechanical engineering of the sensor has been con-
ding on the extension of the mold area to be thermally
ceived to be modular. Presently there are different confi-
mapped.
gurations, based on either 4, 8 or 16 detection points. The sensor with 4 measuring points is a small unit of approxi-
The ultrasonic sensor provides the copper temperature
mately 40x50x200 mm, extending to 260 mm in length
trends and the meniscus thermal profile in real-time (Fig.
for the 16-point model. Thanks to the sensor modularity,
5).
multiple modules can be installed on the mold, depen-
La Metallurgia Italiana - Aprile 2022
pagina 110
Industry news - Attualità industriale
Fig.5 - Example of copper temperature data measured by the ultrasonic sensor: The location where each temperature is measured is represented in the diagram above the graph. The working principle of the ultrasonic sensor relies on
nic wave crosses the complete copper volume from the
the dependence of ultrasound velocity on copper tem-
inner to the outer tube walls with high frequency. The
perature. A dedicated algorithm developed by Ergoli-
sensor measures the temperature values several times
nes based on mathematical inversion and FEM modeling
per second, providing real-time information about what
enables to determine the copper temperatures by inver-
happens on the hot face (i.e., steel level fluctuation cor-
ting the ultrasonic signals. In the example in Fig. 5, the
responding to instantaneous temperature variations).
data provided by UT-MAP represent the average copper
Beside its non-invasive installation and contactless mea-
temperature at four vertical locations in the meniscus re-
surement, one further advantage of the ultrasonic sensor
gion. Each temperature is averaged over a copper volume
with respect to TC or OFC is its “direct” (contactless) me-
3
of approximately 2 cm in between the hot and the cold
asurement of copper temperature with virtually no delay
face of the copper tube.
due to heat transmission through cementing resins: In fact, before reaching the TC or OFC probe, the heat must
The system is able to detect even small temperature va-
propagate through the resin used to cement the probe
riations of the measured volume because the ultraso-
into the channel, and only after thermalization of the resin
La Metallurgia Italiana - April 2022
pagina 111
Attualità industriale - Industry news
does the probe reach thermal equilibrium with the copper
Contrary to the radiometric, the ultrasonic sensor detects
and provide a correct temperature measurement. On the
the actual steel level and it is not affected by powder ad-
other hand, the ultrasonic sensor provides a fully contact-
dition.
less temperature measurement, which involves no probes inside the copper and no cementing resin. As a con-
When casting in open stream, the control feedback is
sequence, the copper temperature variation is detected
made on the motors of the withdrawal unit. The ULD has
by the ultrasonic sensor instantaneously, with no delay
been tested as a stable and reliable system for steel level
due to the thermalization of the resin. As a consequence,
regulation and automatic casting start in open stream ca-
the ultrasonic sensor response is faster with respect to TC
sting.
or OFC. The radiometric system has a real time detection of the Summarizing, when compared to the TC or the OFC, Er-
steel movement and needs a proper number of samples
golines’ ultrasonic technology is more performing for at
to have a more stable information (i.e. moving average
least two fundamental reasons:
and filtering of the signals). The reaction time of the sy-
• The non-invasive installation in the water jacket, which
stem is fast, and this is the reason why up to today this is
leaves the copper tube unaltered
the winning system on the market.
• A faster response to any temperature variation asso-
The systems based on TC or OFC need more time to react
ciated with a steel level fluctuation due to contactless
due to resin thermalization, as discussed above, and are
measurement
therefore intrinsically less reactive and precise than the ultrasonic system.
MOLD LEVEL CONTROL Ergolines’ ultrasonic sensor can also be applied to mold
ULD INDUSTRIAL APPLICATION TO STEEL LEVEL
level control (ULD-Ultrasonic Level Detector). The fast
CONTROL
response time provides an interesting technical advanta-
Beside its application to mold thermal mapping and clo-
ge to Ergolines’ ULD compared to the TC or OFC when
sed-loop powder thickness control, Ergolines’ ultrasonic
the application needs a reactive process control tool such
sensor can also be used for steel level control12. Fig. 6 be-
as the mold steel le vel control or the breakout detection
low shows comparative field data from a pilot installation
system. The steel level detection by the radiometric sy-
in open casting, where liquid steel level was simultane-
stem is based on the density variation of the mass moving
ously measured by the ultrasonic and radiometric sen-
inside the reading window (steel and mold powder), whi-
sor, with steel level control based on the radiometric fe-
le the ULD reads the actual steel level based on the heat
edback. The blue graph is the radiometric setpoint, while
transfer conditions from the steel to the cooling water
the red one is the steel level measured by the ultrasonic
(meniscus thermal profile).
sensor (data are not filtered). The steel level is expressed
The radiometric mold level reading is an average of liquid
as percentage of the reading range of the two sensors,
steel level and powder thickness based on the respective
while time is expressed is seconds x 10 4. This casting se-
densities of steel and powder. In fact, mold powder has a
quence was particularly interesting since it offered the
density which is typical 1/3 with respect to the density of
opportunity to measure the sensor response to a step
liquid steel. As a consequence, each time powder is added
variation of the radiometric setpoint: The steel level was
manually, the radiometric reads the powder addition as a
kept at 75% of the radiometric range for about one hour
fictitious increase of the steel. Consequently, the system
and then lowered to 50%. It can be seen that the red trend
reacts by decreasing the actual steel level to restore the
of the ultrasonic sensor follows very closely the blue ra-
initial value of the radiometric setpoint. When the powder
diometric level at both casting start and step variation.
is consumed, the steel level rises again for the opposite mechanism of powder consumption.
La Metallurgia Italiana - Aprile 2022
pagina 112
Industry news - Attualità industriale
Fig.6 - Comparison of ultrasonic (red) and radiometric sensor (blue). Steel level was controlled by radiometric and measured with ultrasonic sensor (HMI unfiltered raw data sampled at 1 Hz). Another interesting casting sequence is shown in Fig. 7. Also in this case the steel level was controlled by the radio-
metric sensor (blue graph) and it was simultaneously measured with the ultrasonic sensor (red graph) for comparati-
ve purposes. Since the ultrasonic range is different from the radiometric one, the ultrasonic graph has been scaled in order to be compared with the radiometric trend (for this rea-
son the ultrasonic level before casting start has a non-zero
value). The plot also displays casting speed (green) and the
temperature of primary cooling water (black), which is also
the temperature of the outer case of the ultrasonic sensor. The vertical axis is the steel level expressed as percentage
of the radiometric range. The reason why this sequence is interesting is that three clogging events are clearly visible (at times 1400 s, 1900 s and 2500 s), offering the opportu-
nity to assess the ultrasonic sensor response to a “spike”,
namely an abrupt increase in the steel level immediately followed by an abrupt decrease. Clogging occurs when
steel momentarily freezes at the tundish outlet, causing
the steel flow in the SEN to diminish. When the operator
removes the solidified steel with an oxygen lance, the liquid steel level in the mold rises abruptly. This event can be seen as a spike in the radiometric signal (at times 1400 s,
1900 s and 2500 s). The radiometric feedback reacts to the spike by increasing the speed of the withdrawal unit (see green trend at the same instants of time), and then lowering
it again as the steel level approaches the setpoint again. It
can be clearly seen that the ultrasonic steel level measurement (red trend) follow the radiometric signal (blue) very
closely during the whole sequence. It is interesting to note that the ultrasonic signal clearly detects the clogging even-
ts. Even if the ultrasonic response reflects the thermal delay due to heat propagation through copper, the graphs in Fig. 6-7 demonstrate that the ultrasonic sensor is able to reliably measure the liquid steel level both in stationary and
transient regimes (such as casting start, step variations and clogging).
Fig.7 -Comparison between ULD and radiometric level signals. Steel level was controlled by radiometric and measured with ultrasonic sensor. Open casting with clogging events. La Metallurgia Italiana - April 2022
pagina 113
Attualità industriale - Industry news
APPLICATION TO QUALITY TRACKING AND MOLD
to improve the casting practice.
As previously reported in the literature6, Ergolines’ ultra-
Based on Ergolines experience on powder thickness con-
to quality tracking in billet casters. The results reported
used to implement automatic powder feeding in closed-lo-
POWDER THICKNESS CONTROL
sonic thermal mapping system was successfully applied in Fig. 8 demonstrate the advantage of applying ultrasonic
thermal monitoring to the copper mold. The graph shows the temperature trends measured at different locations in
the meniscus region during a three-ladle casting sequence. The first 120 minutes correspond to manual feeding, clearly recognizable from the typical sawtooth pattern in the trends, which is due to the well-known
13-15
radiometric
feedback reaction to manual powder feeding. After minute
120, an automatic powder feeder operated in open-loop was activated: It is seen that the temperature trends stabilize significantly. At about minute 155 an 8% increase in the casting speed is seen to cause an increase in the T1 and T2
temperatures, which are measured below the meniscus, demonstrating UT-MAP effectiveness in monitoring the
local heat transfer conditions in the mold. The ultrasonic temperature trends were correlated with data from Quality Control on tracked billets: A marked reduction of both sur-
face (bleedings) and internal defects (defects due to pow-
der entrapment events) was found in the billets cast under automated feeding if compared to the ones casted before
minute 120 under manual powder addition. In fact, the sta-
bility of liquid steel level in the mold is known to have a direct impact on steel quality1,2. These results are an exam-
ple of how the ultrasonic sensor can be used in UT-MAP mode to provide key information which can be exploited
trol technologies13-20, the ultrasonic sensor can also be
op mode. In fact, the position of the liquid steel meniscus can be determined from the meniscus thermal profile by
means of a dedicated algorithm developed by Ergolines. If the steel level is controlled by a pre-installed radiome-
tric sensor, then the ultrasonic sensor can be used to keep the powder thickness constant. This is possible since the radiometric sensor response is affected by both steel and
powder, while the ultrasonic sensor detects the actual steel level. As a consequence, when powder is added by the automatic feeding machine, the radiometric feedback
causes the actual steel level to decrease (3 cm of powder attenuate the gamma rays roughly as 1 cm of liquid steel). This decrease in the steel level is detected by the ultrasonic
sensor and used by a PID controller to drive the powder flow rate accordingly, keeping the powder thickness constant, leading to increased meniscus stability and significant quality improvement.
As a further example, the effectiveness of Ergolines’ UT-
MAP technology associated with fine-tuned mold powder
dosing in closed-loop mode has been demonstrated by extensive data collection campaigns on special steels, le-
ading to high quality billet surfaces, as recently reported in the literature21.
Ergolines’ ultrasonic sensor applied to mold powder thickness control is currently installed in several steelplants.
Fig.8 - UT-MAP trends detect difference between manual feeding, leading to defects occurrence on tracked billets, and automatic feeding (open-loop), leading to marked defect reduction6.
La Metallurgia Italiana - Aprile 2022
pagina 114
Industry news - Attualità industriale
OTHER POTENTIAL DEVELOPMENTS
• Insufficient or improper lubrication
Breakout detection
• Non correct position of the SEN
Since the UT-MAP system has been designed to be modular, its extension to different areas of the mold is possi-
The UT-MAP data can be collected in the plant database
ble. The concept of mold thermal mapping, today applied
and used by the metallurgists to associate the segmen-
to the slab casters only, can be applied to the billet or blo-
ts of the cast billets where the temperature fluctuations
om curved mold. In general, the upper area of the mold is
were detected with a qualitative benchmarking, thus pro-
more interesting for thermal mapping applications, reser-
viding a qualitative tracking of the product.
ving the mapping of the last segment of the mold mainly for the tapering optimization.
The problems in general can be resumed into two categories: Surface quality and Geometrical quality. On one
The thermal mapping applied to the first 200-300 mm has
hand, surface quality is mainly influenced by local / spot
been demonstrated to be enough to effectively imple-
deviation from the normal conditions: bleedings, surface
ment the breakout detection systems. It is evident that
or subsurface cracks are an example. On the other hand,
the thermal mapping extended to the complete mold is
geometrical problems are related with a non-homogene-
more “safe and complete” but the major problems related
ous behavior of the heat transfer along the four sides of
to breakout are generated by bleeders and sticking in the
the mold. An example of this case is the rhomboidity, whi-
initial solidification.
ch is generated when uneven heat flux conditions occur
Typical temperatures trends, as deeply documented in li-
in the mold.
terature, between two consecutive vertical points where the sticker occurs show a double temperature peak (up-
CONCLUSION
per and lower detection point), with a propagation speed
Ergolines’ developed and successfully tested an ultraso-
a bit lower than the casting speed. The UT-MAP software
nic sensor for contactless temperature mapping of the
can recognize a deviation from the normal conditions and
copper mold tubes (UT-MAP – Ultrasonic Temperature
provide a diagnostic variable to the strand PLC characte-
MAPping). The sensor is able to monitor the temperatu-
rized by “normal”, “warning” and “alarm” mode, based on
re of the copper mold tube in real-time, providing both
custom thresholds which can be set by the steel plant me-
temperature trends over time and the meniscus thermal
tallurgists.
profile. Contrary to thermocouples or optical fibre cables, which require invasive and expensive machining of each
Quality tracking
copper tube, the ultrasonic sensor is installed in the wa-
A further application of mold thermal mapping is related
ter jacket and therefore does not require any copper ma-
to the billet quality tracking. It has been proven by ULD in-
chining at all. Correlation of ultrasonic sensor data with
stallations on SBQ CCM, that some of the parameters can
quality control reports on track billets demonstrated the
be associated to quality issues. What the UT-MAP provi-
effectiveness of the UT-MAP technology in providing key
des is the copper temperature at several points (4 up to 16
information to the metallurgists, enabling improved steel
points) in the meniscus area. From this data, key informa-
quality. Ergolines’ ultrasonic sensor, currently used in se-
tion on the quality of the heat exchange in the monitored
veral steel plants for powder thickness control, can also
area can be obtained. In fact, the factors related with heat
been applied to steel level control, as shown by compara-
flux variations include:
tive field-data with the radiometric sensor on pilot plants.
• Steel level fluctuation • Abnormal oscillation mark formation • Bleedings • Stickers
La Metallurgia Italiana - April 2022
pagina 115
Attualità industriale - Industry news
REFERENCES [1] [2] [3] [4] [5] [6]
[7]
[8]
[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
Thomas, B.G., Modelling of Continuous-Casting Defects Related to Mold Fluid Flow, 3rd Internat. Congress on Science & Technology of Steelmaking, Charlotte, NC, AIST, Warrendale, PA, 2005. Thomas, B. G., On-line detection of quality problems in continuous casting of steel. Modelling, Control and Optimization in Ferrous and Nonferrous Industry, Chicago, IL, TMS, Warrendale, PA, 2003. Lait, J.E. ; Brimacombe, J.K., Solidification During Continuous Casting of Steel, Continuous Casting, Vol. 2, 1984, p. 171. Emi, T., Surface Defects on Continuously Cast Strands, The AISE Steel Foundation, Pittsburgh, PA, Ch. 21, 2003, p. 1. Ecke, W., Applications of Fibre Bragg Grating Sensors, IPHT Jena, The 19th International Conference on Optical Fibre Sensors, Perth, 2008. Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2019: Fully Contactless Mold Thermal Mapping and Breakout Prevention for Billets and Blooms: An Innovative System to Improve Metallurgy, Quality and Productivity, Proceedings of AISTech 2019, 6-8 May 2019, Pittsburgh, Pa., USA Spagnul, S.; Mazza, I.; Miani, S.; Schiavon, G.; 2019; Breakout Prevention for Billets and Blooms through Contactless Mould Thermal Mapping: A New Tool for Metallurgists, Quality Control and Productivity Improvement, Proceedings of METEC 2019, 25-29 June 2019, Düsseldorf, Germany Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2019: Real-time and Contactless Mould Thermal Monitoring: Improving Metallurgy, Quality and Productivity of Billets and Blooms, BHM Berg- und Hüttenmännische Monatshefte, Austrian Society for Metallurgy of Metals (ASMET) and Bergmännischer Verband Österreich (BVÖ) (Proceedings of ESSC and Duplex 2019, Sept. 30 – Oct. 02 2019, Vienna, Austria) Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2018, Contactless Mold Thermal Mapping at Meniscus through an Innovative Ultrasonic Sensor, Proceedings of ICS 2018, Venice, Italy, 13-15 June 2018. Mazza, I.; Miani, S.; Schiavon, G.; Spagnul, S.; 2020: The mold temperature mapping with Ultrasonic Contactless Technology is the key for the real-time initial solidification process control tools, Proceedings of ECCC 2020, 17-19 June 2020, Bari, Italy (pre-print). Schiavon, G.; Mazza, I.; Spagnul, S.; Miani, S.; 2020, Contactless Mold Thermal Mapping: A New Tool for Metallurgists, Quality Control and Productivity Improvement, Proceedings of AISTech 2020, Cleveland, Ohio, USA (pre-print). Spagnul, S.; Mazza, I.; Schiavon, G.; Miani, S.; 2019: An Innovative Ultrasonic System for Steel Level Control, Proceedings of 7EFRS 2019, International Iron and Steel Symposium, 26-27 Sept. 2019, Izmir, Turkey. Spagnul, S.; Padovan, M.; Bianco, A.; Mantovani, F.; 2014, Latest Enhancements in Mold Powder Thickness Control as a Result of a New Propulsion System Implemented in Flux Feeding, Proceedings of ECCC 2014, Graz, Austria. Mazza, I.; Spagnul, S.; Olivo, L.; Milani, F.; 2017, Review of Technologies and Methods for Mold Powder Thickness Control, La Metallurgia Italiana , No. 4, pp. 30-37. Mazza, I.; Spagnul, S.; Mantovani, F; 2015, Review of the Mold Powder Control Technology, Proceedings of METEC 2015, Düsseldorf, Germany. Spagnul, S.; Olivo, L.; Schiavon, G.; Mazza, I.; 2017; A Compact Mold Powder Diffuser With Built-In Optical Powder Thickness Measurement, Iron and Steel Technology, Dec. 2017, Vol. 14, No. 12, pp. 64-70. Spagnul, S.; Olivo, L.; Schiavon, G.; Mazza, I.; 2017, A New Instrumented Mold Powder Diffuser with Built-in Optical Sensor for Powder Thickness Control, Proceedings of ECCC 2017, Vienna, Austria, 26-29 June 2017. Olivo, L.; Spagnul, S.; Mazza, I.; 2016, A New Optical System for Mold Powder Thickness Control by Laser Scanning and Multi-Spectral Imaging, Iron and Steel Technology, Dec. 2016, Vol. 13, No. 12, pp. 62-69. Mazza, I.; Spagnul, S.; 2015, A Novel Ultrasonic Sensor for Mold Powder Thickness Control, Proceedings of METEC 2015, Düsseldorf, Germany. Spagnul, S.; Mantovani, F.; 2011, A Reliable Powder Control based on an Automatic Closed Loop System including Measurement, Powder Feeding and Powder Thickness Control, Proceedings of METEC 2011, Düsseldorf, Germany. Scarabelli, C.; Olivero, D.; Bego, F.; Spagnul, S.; Mazza, I.; Schiavon, G.; 2020: Powder dosing with mould temperature feedback control in continuous casting of stainless steel for high quality billet surfaces, Proceedings of ECCC 2020, 17-19 June 2020, Bari, Italy (pre-print).
La Metallurgia Italiana - Aprile 2022
pagina 116
L’AIM va in trasferta a Bari per un seminario presso “Magna”: perché? Ce ne parla il CEO di MAGNA ing. Aldo Cirilli.
Dopo la laurea a pieni voti in ingegneria presso il Po-
litecnico di Bari, l’ing. Aldo Cirilli ha avuto una vasta esperienza di lavoro in differenti stabilimenti all’in-
terno del Gruppo Getrag (Germania, Italia e Messico)
con crescenti responsabilità, fino a ricoprire la posi-
zione di Amministratore Delegato dello stabilimento
di Irapuato in Messico, dove è rimasto tre anni. Successivamente, con l’acquisizione di Getrag da parte di Magna, si è trasferito nello stabilimento di Neuenstein in Germania come General Manager, e poi nuo-
vamente a Modugno come Amministratore Delegato.
Ing. Cirilli, perché Magna ospita il seminario AIM
senza negli USA: in 8 ore abbiamo raggiunto più di
Il seminario “Trasmissioni nell’automotive: dall’ac-
comunicazione in presenza è fondamentale.
del 16 e 17 giugno?
ciaieria al processo di pallinatura”, organizzato da AIM all’interno della nostra Azienda il 16 e 17 giugno,
nasce come una proposta “amichevole”, formula-
ta prima della pandemia dall’Arch. Alberto Rosso *, dall’ing. Enrico Morgano * e dall’ing. Ettore Camarda *, che è stata poi favorevolmente accolta e fortemen-
te voluta dai vertici Magna: oggi l’evento è finalmente in fase di realizzazione.
L’evento in Magna è, a mio parere, un’opportunità per tutti gli attori.
La pandemia ha dimostrato la nostra necessità di re-
lazioni sociali. I risultati dei soli rapporti virtuali non
sono del tutto efficaci: questa riflessione si è rafforzata dall’esperienza di un mio recente incontro in pre-
La Metallurgia Italiana - April 2022
quanto avevamo prodotto in sei mesi di video-call. La Siamo persone relazionali!
La personale esperienza lavorativa in Germania mi ha insegnato l’importanza di fare “rete” tra le aziende che insistono sul territorio: le realtà che vivono
le stesse difficoltà hanno molto da offrire e molto da imparare reciprocamente. Lo scambio, culturale e
tecnologico, è assolutamente da promuovere. Natu-
ralmente all’interno dello scambio deve essere preservato e tutelato il know-how condiviso, per evitare che il vantaggio competitivo venga meno.
Magna: quale storia e quale strategia per il futuro?
Il Gruppo Magna è presente in circa 434 sedi differenti
pagina 117
in 28 paesi con circa 160 000 dipendenti e un fatturato di circa 36 miliardi di dollari.
È il 3° fornitore diretto OEM (Original Equipment Manufacturer) a livello mondiale ed il 1° sul mercato americano.
Fornisce sistemi di propulsione e di trasmissione, sistemi di visione, auto guida, elettronica di bordo, sellerie, cruscotti, sistemi di illuminazione e altra com-
ponentistica ad alto contenuto tecnologico e nella
Business Unit Magna Steyr, per conto dei costruttori d’auto, sviluppa e assembla interi veicoli come mo-
delli di nicchia o con equipaggiamenti speciali per
nuovi servizi di guida autonoma, car sharing, traccia-
bilità, comfort ed entertainment a bordo. Il software e la elettronica di bordo saranno predominanti sulla
meccanica, e per questo sono state avviate collaborazioni con diverse software house. La mobilità vista non solo come una necessità ma anche come un’e-
sperienza, da vivere innanzitutto in sicurezza, attraverso sistemi intelligenti. Un concetto chiave ispirato e sostenuto da Swamy Kotagiri, da gennaio 2021 nuovo CEO di Magna International, che ha fatto del precorrere i tempi il suo cavallo di battaglia.
BMW, Audi e Mercedes e prossimamente anche per Fisker, che con il modello Ocean sarà in concorrenza
diretta con Tesla. Stellantis è uno dei clienti principali, soprattutto nel mercato americano attraverso i gruppi Chrysler e Jeep.
Magna accelera nel mercato dell’elettrico: da anni
sviluppa sistemi di propulsione elettrica (full o hybrid), ha allestito diversi stabilimenti per la produzione di questi sistemi e di recente, attraverso la Joint
Venture con LG Electric, che produce motori elettrici, ha accresciuto il suo know-how e la sua capacità produttiva a fronte di una domanda in forte crescita.
Magna è quindi pronta alla sfida dell’elettrico, ma sarà necessario un periodo di transizione.
Gli operatori del settore si stanno lanciando sull’elettrico con grandi risorse, fermando di fatto gli inve-
stimenti sui prodotti intermedi di transizione: Magna sta raccogliendo l’opportunità di questo vuoto con
prodotti ibridi di cui c’è grande richiesta in questo momento.
La recente ed instabile situazione geopolitica avrà un’influenza decisiva in questa transizione: mancano ancora le infrastrutture e la crisi attuale, energetica e geopolitica, rende il percorso ancora più difficile.
La transizione non si limiterà, però, a quella elettri-
ca. Magna ha istituito una nuova Business Unit chiamata “New Mobility” per studiare le potenziali nuove esigenze della mobilità del futuro, oggi ancora ine-
spresse, e sviluppare i prodotti che le soddisfino:
La Metallurgia Italiana - Aprile 2022
Perché Magna di Modugno è un’eccellenza?
Lo stabilimento di Modugno, costruito da prato verde dalla multinazionale tedesca “Getrag”, nota per la produzione di trasmissioni per autoveicoli, è nato nel
1996, ed è stato poi acquisito nel 2016 dalla Business Unit Powertrain del Gruppo Magna, che ha introdotto, secondo una visione imprenditoriale molto forte del suo fondatore Frank Stronach, una forte decen-
tralizzazione della gestione aziendale, concedendo margini di autonomia molto ampi.
Esiste un clima di grande attaccamento del personale dipendente di Modugno: l’80% dei collaboratori è presente sin dalla apertura dello stabilimento. Quasi tutti i dipendenti erano allora alla prima esperienza la-
vorativa e sono stati reclutati attingendo alle migliori risorse delle Scuole superiori e del Politecnico di
Bari. Oggi i dipendenti vedono nell’Azienda qualcosa che loro stessi hanno contribuito a creare e poi rinnovato nel tempo. Il senso di appartenenza e la vo-
glia di primeggiare a livello mondiale sono ancora più forti, da quando l’Amministratore Delegato è italiano. Lo stabilimento di Modugno, ad esempio, è stato no-
minato “Best Employer” nel settore Automotive da Statista, per conto del Corriere della Sera nel 2021 e
nel 2022. È lo stabilimento del gruppo con maggiori iniziative sulla sostenibilità, con importanti investi-
menti in produzione di energia rinnovabile o recupero delle risorse, uno degli stabilimenti produttivi
pagina 118
più fedele allo standard Magna, con frequenti riconoscimenti frutto della passione e della dedizione delle
ni robuste e lungimiranti.
sue persone.
La reputazione dello stabilimento di Modugno a livello di gruppo è alta e permette di guardare con sereni-
tà alla destinazione di finanziamenti, con prospettive incoraggianti per i livelli occupazionali.
Con la forte decentralizzazione che è parte del DNA
di Magna, oggi c’è uno spirito imprenditoriale più marcato con una grande proattività nella ricerca di
commesse e acquisizione di nuovi clienti: ci si avvale di economie di scala e competenze tecniche globali al servizio della realtà locale. Grazie a volontà, caparbietà ed inventiva, di recente è stata vinta una com-
messa per un cliente premium per la produzione di
trasmissioni ibride. La sfida è stata vinta contro sta-
bilimenti che sulla carta potevano offrire soluzioni più economiche o produrre in contesti territoriali più favorevoli.
Nella ricerca dei fornitori, si stanno privilegiando realtà locali, competenti e qualificate, con l’obiettivo di coinvolgerli nella crescita dello stabilimento.
I costi logistici stanno esplodendo, e questo costrin-
ge ad esplorare nuove soluzioni: le merci, ad esempio, viaggiano principalmente su rotaia, forma di trasporto più vantaggiosa e sicura.
Pochi mesi fa lo stabilimento Magna di Modugno ha deciso di continuare ad investire con ECM/Silco con l’acquisto del nostro forno da trattamento n° 10.
La transizione verso l’elettrico richiede iniziative ed
investimenti considerevoli ed urgenti: i grandi grup-
pi stanno mobilitando grandi capitali, ma il sistema industriale tutto deve attivarsi per non perdere per strada le piccole e medie imprese e per incoraggiare
le grandi a preferire il nostro Paese; purtroppo constatiamo una reazione politica lenta e tardiva, in cui
non si manifesta una visione strategica di insieme. Paesi come la Germania o la Francia, ancora una volta,
davanti alle stesse difficoltà, dimostrano di essere più tempestivi nell’affrontarle e nello sviluppare soluzio-
La Metallurgia Italiana - April 2022
Ingegner Cirilli, ci racconti anche un po’ della sua storia.
La mia ventennale esperienza con Getrag prima e ora con Magna, grazie alle differenti opportunità che mi
sono state offerte, mi consente di affrontare le varie situazioni da prospettive differenti e più ampie.
Ho trascorso diversi anni in Messico prima come
Plant Manager e poi come Amministratore Delegato, curando anche il Program Management e la parte Commerciale della Divisione Stati Uniti.
Successivamente ho assunto la responsabilità del più
grande stabilimento tedesco che realizza trasmissioni per i modelli Premium dei grandi marchi, stabilimento che produce anche in esclusiva i cambi per
Ferrari. Sono molto orgoglioso di essere stato Diret-
tore dell’unico stabilimento che produce cambi per “il cavallino” e ne conservo un esemplare in ufficio
che mi è stato donato come “saluto” quando sono rientrato in Italia.
Successivamente ho scelto di rientrare a Modugno per condividere la mia esperienza con i miei connazionali: il mio senso di appartenenza all’Italia è molto
forte. La mia famiglia, siamo in 4, mi ha sempre seguito, mia figlia è nata in Messico. Abbiamo moltissimi amici messicani e tedeschi e moltissimi bei ricordi.
A Modugno sto cercando di introdurre schemi di
organizzazione per certi versi nuovi. Nella realtà te-
desca le collaborazioni con scuole superiori ed università sono fortemente desiderate, in alcuni casi un obbligo imposto alle aziende, con manodopera qua-
lificata e formata negli ambiti di interesse dell’azienda stessa. Ho quindi rafforzato i rapporti con gli ITS lo-
cali favorendo gli stage degli studenti e sottoscrivendo un accordo quadro triennale con il Politecnico su tematiche di comune interesse.
L’arruolamento del personale si sta trasformando in
pagina 119
modo significativo: abbiamo sempre meno bisogno
di manovalanza in senso stretto, mentre aumenta la richiesta di personale qualificato, ad esempio
di softwaristi. Ci dovrebbe essere una risposta del mondo accademico a queste nuove esigenze dell’in-
dustria e la Scuola dovrebbe contestualizzare il suo ruolo nella formazione.
Io consiglio vivamente alle nuove leve che si affac-
ciano al mondo lavorativo di accumulare esperienze all’estero dopo un percorso formativo italiano (come
il Politecnico di Bari ed altre eccellenze nazionali): si nota molto in sede di colloquio chi ha avuto espe-
rienze nel Nord Europa, dove c’è una straordinaria
Ma Bari vecchia non è solo cultura, attrae anche i più giovani, grazie ai tantissimi ristoranti e locali notturni sorti negli ultimi anni.
A pochi passi dal capoluogo, si incontrano vivaci
cittadine, ricche anch’esse di storia e tradizione ga-
stronomica, che non hanno certo bisogno di presentazioni: il panorama mozzafiato di Polignano a mare,
i trulli di Alberobello, Castel del Monte nel nord barese, lo scenografico Ponte acquedotto di Gravina in Puglia…
Insomma, la Puglia è una terra ricca di perle, tutta da scoprire.
capacità organizzativa ed una disciplina, che a noi talvolta manca.
All’italiano è riconosciuta una grande capacità di rea-
zione, di gestione efficace dell’imprevisto in situazioni fuori controllo: l’esperienza Covid ne è l’esempio
più recente: lo stabilimento a Modugno ha gestito l’emergenza prima e meglio del resto del gruppo Ma-
gna ed i protocolli di sicurezza per la gestione dell’emergenza pandemica sono stati presi come riferimento dagli altri stabilimenti. Di contro io noto una
difficoltà tutta italiana a pianificare sul lungo periodo:
confidando troppo nella nostra capacità di recupero
all’ultimo momento, partiamo in ritardo e non sempre centriamo l’obiettivo. Dovremmo investire meno tempo nella dialettica inconcludente e sostenere in-
vece le grandi eccellenze che il nostro Paese offre, su tutti i fronti.
E per finire… ing. Cirilli cosa ci consiglia di fare a Bari nei momenti liberi?
Bari è una città in continua evoluzione, che oggi offre moltissime opportunità per i turisti.
Imperdibile una passeggiata nel centro storico, con il suo dedalo di vicoli e viuzze ricche di profumi e tradizioni, che portano ai luoghi più noti della città: la cattedrale di San Sabino, la basilica di San Nicola ed
il Castello svevo, fortezza del XIII secolo, oggi sede museale.
La Metallurgia Italiana - Aprile 2022
pagina 120
11th European Stainless Steel Conference - Science & Market & 7th European Duplex Stainless Steel Conference & Exhibition
Bardolino · Verona · Italy, 15-17 June 2022 Organized by
In cooperation with
Sponsored by
siderweb THE ITALIAN STEEL COMMUNITY
Patronized by
www.aimnet.it/essc.htm
Atti e notizie - AIM news
Eventi AIM / AIM events CONVEGNI 27° CONVEGNO NAZIONALE TRATTAMENTI TERMICI – Genova, 26-27 maggio 2022 http://www.aimnet.it/tt.htm ESSC & DUPLEX 2021 - 11TH EUROPEAN STAINLESS STEEL CONFERENCE SCIENCE & MARKET & 8TH EUROPEAN DUPLEX STAINLESS STEEL CONF. & EXHIB. – Bardolino (Verona), 15-17 giugno 2022 http://www.aimnet.it/essc.htm 39° CONVEGNO NAZIONALE AIM – Padova, 21-23 settembre 2022 http://www.aimnet.it/nazionaleaim
IWSQ 2022 - 2ND INTERNATIONAL WORKSHOP ON SURFACE QUALITY OF CONTINUOSLY CAST PRODUCTS – Bergamo, 1-2 dicembre 2022 https://www.aimnet.it/iwsq-2.htm CORSI E GIORNATE DI STUDIO Giornata di Studio LE OPPORTUNITA’ REGIONALI PER LO SVILUPPO DI UN ECOSISTEMA CIRCOLARE NEL SETTORE METALLURGICO – Milano c/o Regione Lombardia, 27 aprile Corso DIFETTOLOGIA NEI GETTI PRESSOCOLATI: METALLIZZAZIONI - 3 maggio – webinar FaReTra Corso ADDITIVE METALLURGY – 4-10-17-24 maggio – webinar FaReTra Giornata di Studio SVILUPPO TECNOLOGICO DEGLI IMPIANTI DI PRODUZIONE A CALDO DI VERGELLA DI ALTA QUALITÀ PER USI SPECIALI – Lecco c/o Caleotto, 5 maggio Giornata di Studio ALLUMINIO NELL’AUTOMOBILE: LA STORIA INCONTRA IL FUTURO – Ferrara, 20 maggio Corso Modulare METALLOGRAFIA – ibrido – Milano, 31 maggio, in modalità webinar 7-8-9-14-15-28-29 giugno, 5-6 luglio, settembre Giornate di Studio GETTI PRESSOCOLATI PER APPLICAZIONI STRUTTURALI – Brescia c/o Idra, 14 giugno e c/o Museo Mille Miglia, 24 giugno Seminario TRASMISSIONI NELL’AUTOMOTIVE: DALL’ACCIAIERIA AL PROCESSO DI PALLINATURA – Modugno, (Bari) c/o Magna, 16-17 giugno Corso MICROSCOPIA ELETTRONICA IN SCANSIONE – SEM - III Edizione – Milano, Lecco, 21-22 giugno Corso TRIBOLOGIA INDUSTRIALE – ibrido – in modalità webinar, 22-23 giugno e Modena, 29-30 giugno Corso modulare FONDERIA PER NON FONDITORI – 30 giugno, 1-7-8-12-13-14 luglio – webinar FaReTra Giornata di Studio ANALISI INFORTUNI – giugno SCUOLA DI METALLURGIA DELLE POLVERI – Imola c/o SACMI, 5-6 luglio Metallurgy Summer School SURFACE ENGINEERING OF METALS – Bertinoro (FC), 24-27 luglio Corso MASTER PROGETTAZIONE STAMPI – itinerante, settembre/ottobre/novembre/dicembre Giornata di Studio LEGHE PER ALTA TEMPERATURA PRODOTTE CON TECNOLOGIE ADDITIVE – Firenze (c/o Baker Hughes), 15 settembre Giornata di Studio MICROSCOPIA ELETTRONICA APPLICATA ALLA FAILURE ANALYSIS – Padova, 20 settembre
La Metallurgia Italiana - Aprile 2022
Per ulteriori informazioni rivolgersi alla Segreteria AIM, e-mail: info@aimnet.it, oppure visitare il sito internet www.aimnet.it
ROLLING 2022 - 12th INTERNATIONAL ROLLING CONFERENCE – Trieste, 26-28 ottobre 2022 https://www.aimnet.it/rolling-12/
pagina 122
Atti e notizie - AIM news
Giornata di Studio IL MONDO INDUSTRIAL: ASPETTI METALLURGICI E METODOLOGIE DI CONTROLLO – Torino c/o CNH Industrial Village, settembre Giornate di Studio CEMENTAZIONE VS NITRURAZIONE – Provaglio d’Iseo c/o Gefran, 6 ottobre Giornata di Studio DALLA SCELTA ALLA REALIZZAZIONE DI COMPONENTI TRATTATI TERMICAMENTE: COME EVITARE PROBLEMI TECNICI E CONTRATTUALI? – Brescia, 9 novembre Corso PROVE NON DISTRUTTIVE – ibrido: Miano, 16-17 novembre Giornata di Studio RIVESTIMENTI E ADDITIVE MANUFACTURING – Milano, novembre
www.aimnet.it
Giornata di Studio TECNICHE DI CARATTERIZZAZIONE DEI MATERIALI – Vicenza, novembre Giornata di Studio VERIFICA E MANUTENZIONE DI STRUTTURE METALLICHE STORICHE: ASPETTI STRUTTURALI E METALLURGICI, DIAGNOSTICA E TECNICHE DI INTERVENTO – Milano, novembre Giornata di Studio TRATTAMENTI SUPERFICIALI SU ALLUMINIO, A SCOPO DECORATIVO E PROTETTIVO – autunno/inverno FaReTra (Fair Remote Training) - FORMAZIONE E AGGIORNAMENTO A DISTANZA Modalità Asincrona (registrazioni) Corso METALLURGIA PER NON METALLURGISTI Giornata di Studio PRESSOCOLATA IN ZAMA Giornata di Studio LA SFIDA DELLA NEUTRALITÀ CARBONICA Corso itinerante METALLURGIA SICURA Corso modulare TRATTAMENTI TERMICI Corso modulare I REFRATTARI E LE LORO APPLICAZIONI Corso GLI ACCIAI INOSSIDABILI 11ª edizione Corso FAILURE ANALYSIS 11a edizione Giornata di Studio PERFORMANCE E DEGRADO DEI MATERIALI METALLICI UTILIZZATI IN CAMPO EOLICO: CAPIRE PER PREVENIRE Giornata di Studio DIFETTI NEI GETTI PRESSOCOLATI: POROSITA’ DA GAS Corso PROVE MECCANICHE Corso di base LEGHE DI ALLUMINIO Giornata di Studio GREEN ECONOMY E ASPETTI AMBIENTALI PER L'INDUSTRIA DEI RIVESTIMENTI Giornata di Studio IL CICLO DI FABBRICAZIONE DI UNA VALVOLA. NORMATIVE, PROGETTO, ACCIAIO, FUCINATURA COLLAUDO Corso CORROSIONE PER NON CORROSIONISTI Giornata di Studio TECNOLOGIE DI FORMATURA DELLE ANIME IN SABBIA PER GETTI IN LEGA LEGGERA
L’elenco completo delle iniziative è disponibile sul sito: www.aimnet.it
(*) In caso non sia possibile svolgere la manifestazione in presenza, la stessa verrà erogata a distanza in modalità webinar
La Metallurgia Italiana - April 2022
pagina 123
Atti e notizie - AIM news
Comitati tecnici / Study groups CT CONTROLLO E CARATTERIZZAZIONE PRODOTTI (CCP) (riunione telematica del 5 aprile 2022)
Notizie dal Comitato •
Il presidente Toldo dà il benvenuto all’ing. Enrico Mariani, invitato alla riunione, che dopo essersi presentato viene accolto come membro
del CT CCP. L’ing. Mariani ha lavorato 7 anni presso Fomas e 20 anni presso Rodacciai, sempre in ambito qualità. Ora da due anni è consulente e titolare di Emmeconsulting, e lavora per diverse aziende, tra cui principalmente SMT che rappresenta nel comitato.
Manifestazioni in corso di organizzazione •
Per la GdS sulla “Microscopia elettronica applicata alla failure analysis” Toldo ha preparato una bozza di programma sulla falsariga di quel-
lo dell’edizione precedente, e ha interpellato i possibili relatori: Irene Calliari ha accettato di fare una introduzione alla GdS, poi diversi
relatori presenteranno casi pratici. Restano da definire ancora un paio di presentazioni. La sede sarà a Padova presso Element il 20 set•
tembre 2022, il giorno prima del Convegno Nazionale AIM di Padova. La partecipazione alla giornata sarà mista, in presenza o in remoto.
Corso PdD: oltre a Trentini e Cusolito, entrano a far parte del comitato organizzatore ristretto Bertolini e Giroletti. Trentini ha ottenuto la partecipazione di AIPnD come nella precedente edizione. Visto il successo dell’edizione 2019, si cercherà di ottenere la partecipazione degli stessi docenti. Saranno esaminati i questionari di valutazione compilati dai partecipanti alla precedente edizione per ottenere ulteriori spunti di miglioramento. Si cercherà di introdurre anche il tema della valutazione dello stato di servizio di un impianto tramite CdD. Anche gli sponsor del 2019 saranno nuovamente interpellati. Il corso si svolgerà in modalità mista nei giorni 16 e 17 novembre 2022.
Iniziative future •
Un tema per una futura GdS è quello della vita residua e riattivazione della vita degli impianti, e quindi delle nuove filosofie nell’utilizzo dei
CnD per valutare lo stato di servizio di un impianto. Si forma un comitato organizzatore ristretto che valuterà entro la prossima riunione la fattibilità di questa manifestazione e l’interesse che potrebbe generare nel pubblico.
CT TRATTAMENTI TERMICI E METALLOGRAFIA (TTM) (riunione telematica del 10 febbraio 2022)
Notizie dal Comitato •
Il presidente Petta informa che Enrico Morgano ha lasciato Stellantis, e continuerà a partecipare al CT TTM con la sua nuova affiliazione
•
Petta introduce poi Jacopo Tatti di Stellantis, già collega di Morgano, che viene accettato dai partecipanti e rappresenterà da oggi la sua
•
di SILCO.
azienda in seno al CT. Tatti si presenta raccontando di lavorare nel campo della metallurgia, gruppo bulk, occupandosi di leghe leggere, componenti colati, prodotti lunghi e trattamenti termici.
Petta infine introduce Alessandro Zago di Cogne Acciai Speciali. Zago, laureato al Politecnico di Torino e con dottorato di ricerca, si
occupa di metallurgia e metallografia come responsabile del laboratorio. Zago viene accettato come membro del CT e Petta si dichiara soddisfatto della presenza di un acciaiere tra i membri del CT.
Manifestazioni in corso di organizzazione •
La GdS sullo shot-peening - coordinatore Morgano – è confermata presso Silco a Rivalta (TO) il 28 aprile. Il programma tecnico e l’orga-
•
Il “Convegno Nazionale Trattamenti Termici” (Genova, 26 e 27 maggio 2022) è confermato; la cena sociale si terrà all’Acquario di Geno-
nizzazione sono definiti.
va. Si sono aggiunti 3 ulteriori espositori con lo stand, portando così il numero a 31. Bassani conferma che a livello organizzativo tutto è pronto e anche il numero degli iscritti sta salendo. Gli spazi disponibili sono ampi sia nella zona espositiva che nell’aula delle presenta-
•
zioni scientifiche; quindi, non ci sono problemi di capienza.
Il corso “Metallografia” è stato completamente ristrutturato rispetto alle precedenti edizioni per la fruizione in modalità ibrida. L’inizio è
fissato per il 31 maggio 2022 con una giornata propedeutica svolta in presenza, cui seguirà il Modulo 1 sui materiali ferrosi, composto da
sette mattinate di lezione da remoto e una giornata nei laboratori del Politecnico di Milano (ripetuta per due gruppi diversi, così da poter •
•
dar modo a tutti di seguire le preparazioni metallografiche). Il Modulo 2 sarà relativo ai materiali non ferrosi e si terrà dopo l’estate.
Anche il seminario “Trasmissioni nell’automotive: dall’acciaieria al processo di pallinatura” – coordinatori Morgano e Rosso – è confermato per il 16 e 17 giugno 2022 a Modugno (Bari) presso Magna. I relatori si sono dichiarati tutti disponibili nonostante i continui rinvii dovuti alla situazione sanitaria.
La GdS “Cementazione e nitrurazione” si terrà presso Gefran (Provaglio di Iseo – BS) il 6 ottobre 2022 – coordinatori Desario e Buzzi.
La Metallurgia Italiana - Aprile 2022
pagina 124
Atti e notizie - AIM news
Sono previste 8 presentazioni per parlare di terminologia, caratteristiche ottenibili con cementazione e nitrurazione, controlli di processo e misurazioni e macchinari. Le presentazioni comprenderanno casi di failure analysis. Al termine è prevista una visita allo stabilimen•
to.
La GdS "Dalla scelta alla realizzazione di componenti trattati termicamente: come evitare problemi tecnici e contrattuali" – coordinatori Vicario, La Vecchia, Cusolito, si svolgerà il 9 novembre 2022. E’ stata illustrata ai membri del CT una bozza del programma con nove presentazioni e una tavola rotonda finale.
Iniziative future •
GdS “Il Mondo Industrial - aspetti metallurgici e metodologie di controllo”: IVECO CNH, che dovrebbe ospitare la manifestazione, non
•
Seminario congiunto con CT Metallurgia delle Polveri e Tecnologie Additive: anche in questo caso nessuna conferma da parte delle
ha comunicato novità circa la possibile data. Si rimanda ogni decisione.
aziende ospitanti. L’evento resta in sospeso e comunque sarà per il 2023.
CT METALLI E TECNOLOGIE APPLICATIVE (MTA) (riunione telematica del 07 aprile 2022)
Iniziative future •
•
•
L’organizzazione della GdS sulla sostenibilità ambientale nel campo delle costruzioni civili sta procedendo. Il presidente Loconsolo ha già trovato alcuni relatori e da altri sta aspettando una risposta. Il coordinatore Stella si sta occupando del tema della sicurezza sismica e contro gli incendi. La data possibile è l’autunno 2022, in attesa di fissare un termine più preciso.
La GdS “Utilizzo leghe di nichel in saldatura”, più volte rimandata, viene per il momento sospesa per poter ripensare daccapo a tutto, visti i nuovi, e probabilmente futuri, problemi di approvvigionamento e/o costo di materiali quali il titanio e il nichel. Se ne riparlerà alla prossima riunione.
Per la manifestazione sui metalli nel settore automotive non ci sono novità, se ne riparlerà nelle prossime riunioni.
CT MATERIALI PER L’ENERGIA (ME)
(riunione telematica congiunta del 13 aprile 2022) Notizie dal Comitato •
Merkling espone la nuova attività che sta impegnando il gruppo di lavoro italiano Creep a proposito della caratterizzazione di un pipe di
acciaio grado 93. Alcune analisi hanno portato alla luce un problema strutturale (ferrite d primaria ai bordi dei grani austenitici al momento della tempra). Sono state segnalate anche fasi complesse ricche di W e Nb, non riconducibili per ora a fasi note. In Germania si stanno
•
conducendo valutazioni parallele e i due gruppi si riuniranno in maggio per discutere come impostare le future attività.
Il Presidente Gavelli invita ad affrontare il tema della stesura delle specifiche, da molto tempo trascurato. Viene sottolineata l’importanza di favorire una migliore comunicazione che eviti fraintendimenti sulle specifiche tra i vari attori coinvolti (clienti, produttori, utilizzatori, ecc.) e viene proposto di organizzare un evento che possa trattare l’argomento in modo più generale. Si istituisce un sottogruppo di lavoro all’interno del Comitato che si occuperà dell’organizzazione.
Manifestazioni in corso di organizzazione •
Gariboldi sta lavorando con Paola Bassani del CT Metallurgia Fisica per mettere a punto il programma di massima della manifestazione sul
Thermal Storage, che sarà definito nella prossima riunione. Il titolo potrebbe modificarsi in Thermal Management. Lo svolgimento viene per ora fissato nella prima metà di ottobre 2022.
Iniziative future • •
Manifestazione su materiali per rinnovabili: non ci sono aggiornamenti o sviluppi da segnalare.
Manifestazione su Materiali per Nucleare: il presidente Gavelli invita tutto il CT a pensare alla possibilità di organizzare un evento su questo argomento, in modo da poterne discutere alla prossima riunione.
La Metallurgia Italiana - April 2022
pagina 125
Atti e notizie - AIM news
Normativa / Standards Norme pubblicate e progetti in inchiesta (aggiornamento 31 marzo 2022)
legati e acciai automatici - Parte 3: Acciai
Drilling and production equipment —
da cementazione
Wellhead and tree equipment
UNI EN ISO 19901-5:2016 Industrie del petrolio e del gas naturale Requisiti specifici per le strutture in mare - Parte 5: Controllo del peso durante la
Norme UNSIDER pubblicate da UNI nel mese di marzo 2022
progettazione e la costruzione
e ISO/DIS – aprile 2022
UNI EN ISO 10438-4:2009
prEN – progetti di norma europei
Industrie del petrolio, petrolchimiche e
UNI EN ISO 683-3:2022 Acciai per trattamento termico, acciai legati e acciai automatici - Parte 3: Acciai da cementazione
del gas naturale - Sistemi di lubrificazione,
prEN ISO19901-5
sistemi di tenuta e sistemi di regolazione
Petroleum and natural gas industries
ad olio e loro accessori - Parte 4: Sistemi
- Site-specific assessment of mobile
di regolazione per tenute a gas a secco
offshore units - Part 3: Floating units (ISO 19905-3:2021)
UNI EN ISO 6368:2022 Industrie del petrolio, petrolchimiche e del gas naturale - Sistemi di tenuta a gas secco per compressori assiali, centrifughi e rotativi a vite e per espansori UNI EN ISO 35102:2022 Industrie del petrolio e del gas naturale - Operazioni in climi artici - Fuga, evacuazione e salvataggio da installazioni in mare aperto UNI EN ISO 19901-5:2022 Industrie del petrolio e del gas naturale Requisiti specifici per le strutture in mare aperto - Parte 5: Controllo del peso
petrolifere,
petrolchimiche
e del gas naturale - Protezione contro la corrosione esterna dei tubi di risalita mediante
rivestimento
Norme UNSIDER pubblicate da CEN e ISO nel mese di marzo 2022
prEN ISO 19905-1 Petroleum and natural gas industries - Site-specific assessment of mobile
ISO 23717:2022
offshore units - Part 1: Jack-ups (ISO/DIS
Steel wire and wire products — Hose
19905-1:2022)
reinforcement wire prEN ISO 16961 ISO 23213:2022
Petroleum, petrochemical and natural gas
Carbon steel wire for bedding and seating
industries - Internal coating and lining of
springs
steel storage tanks (ISO/DIS 16961:2022)
ISO 15590-3:2022
prEN ISO 14556
Petroleum and natural gas industries —
Metallic
Factory bends, fittings and flanges for
pendulum impact test - Instrumented test
pipeline transportation systems — Part 3:
method (ISO/DIS 14556:2022)
materials
-
Charpy
V-notch
Flanges
UNI EN ISO 18797-2:2022 Industrie
Progetti UNSIDER in inchiesta prEN
-
Parte
2:
Manutenzione e riparazioni in campo di rivestimenti per tubi di risalita
prEN 10278 ISO/TS 12835:2022
Dimensions and tolerances of bright
Qualification of casing connections for
stainless and other cold drawn steel
thermal wells
products
ISO 11257:2022 Iron
ores
for
shaft
direct-reduction
feedstocks — Determination of the low-
Norme UNSIDER ritirate da UNI nel mese di marzo 2022
UNI EN ISO 683-3:2019 Acciai per trattamento termico, acciai
La Metallurgia Italiana - Aprile 2022
temperature
ISO/DIS
–
internazionali
progetti
di
norma
reduction-disintegration
index and degree of metallization
ISO/DIS 24131-1 Internal protection by polymeric lining for
ISO 10423:2022
ductile iron pipes — Part 1: polyurethane
Petroleum and natural gas industries —
lining
pagina 126
Atti e notizie - AIM news
ISO/DIS 24131-2
ISO/FDIS 4298
Internal protection by polymeric lining for
Manganese ores and concentrates —
ductile iron pipes — Part 2: epoxy lining
Determination of manganese content — Potentiometric method
ISO/DIS 19905-1 Petroleum and natural gas industries
ISO/FDIS 3421
— Site-specific assessment of mobile
Petroleum and natural gas industries —
offshore units — Part 1: Jack-ups
Drilling and production equipment — Offshore conductor design, setting depth and installation
ISO/DIS 16961 Petroleum, petrochemical and natural gas industries — Internal coating and lining of steel storage tanks ISO/DIS 14556 Metallic materials — Charpy V-notch pendulum impact test — Instrumented test method
Progetti UNSIDER al voto FprEN e ISO/FDIS – aprile 2022
FprEN – progetti di norma europei FprEN ISO 10113 Metallic materials - Sheet and strip Determination of plastic strain ratio (ISO/ FDIS 10113:2019)
ISO/FDIS
–
internazionali
progetti
di
norma
ISO/FDIS 19901-2 Petroleum and natural gas industries — Specific requirements for offshore structures — Part 2: Seismic design procedures and criteria ISO/PRF 7992 Iron ores for blast furnace feedstocks — Determination of reduction under load ISO/FDIS 4954 Steels for cold heading and cold extruding
La Metallurgia Italiana - April 2022
pagina 127
2th International ROLLING Conference | Trieste (Italy), 26-28 October 2022 Organised by
in cooperation with
Thanks to a long and successful series, the ROLLING conferences have become the premier international event for rolling practitioners worldwide. The target audience embraces the whole metals rolling community: flat and long products rolling, hot and cold rolling and ferrous and non-ferrous rolling. The forum will cover product and process topics, encompassing rolled product properties, quality and applications, in addition to the design, control and management of mill assets. The Conference will bring together rolling practitioners, steel producers, plant designers and researchers and it will provide a forum for best practices and state-of-the-art technology. It will also take a look at developments in the foreseeable future. CALL FOR PAPERS All contributions on the Conference topics as described below are welcome. Prospective authors are invited to submit a tentative title and an abstract of 400 words (in English) by May 13, 2022 to the Organizing Secretariat. The abstract should provide sufficient information for a fair assessment. There are two ways to submit papers: • to fill in the form online: www.aimnet.it/rolling-12/ • to send the requested information by e-mail to: aim@aimnet.it CONFERENCE SCHEDULE October 26, 2022: Opening session and opening of the exhibition October 27, 2022: Session + Social event – Cocktail October 28, 2022: Sessions + Plant visit LANGUAGE The official language of the Conference will be English.
main sponsor
TOPICS Topics: Ferrous and Non Ferrous • Hot Strip Mills • Steckel Mills • Plate Mills • Rolling Mills for Long Products, including Pipes and Tubes • Thin slab casting and inline rolling • Cold Rolling, • Galvannealing • Coating Processes • Thermomechanical Processes, TMCP • Mathematical Modeling and Simulation • New Rolling Processes • Flat Products, including Coated Products • Long Products, including Pipes and Tubes • Automation, Measurements and Control Technology • Re-heating furnaces – Oxidation and Descaling • Equipment and maintenance • Surface modification and steel deep-working • Clean rolling • Energy and Environmental management • Health & Safety ORGANISING SECRETARIAT
AIM – Associazione Italiana di Metallurgia Via Filippo Turati 8 20121 Milano, Italy Tel. +39 0276021132 or +39 0276397770 http://www.aimnet.it
www.aimnet.it/rolling-12/
