Tag Archives: Processes

Coilbox Process

The first time I encountered any mention of Stelco’s coilbox process was when prof. Dilewijns asked me to investigate the literature for a process of that name. Now this was way before the days of Google and even the Internet, and the best you could to was to go through the annual summaries of the iron & steel magazines we had in our library. I must admit that I was unsuccessful, but prof. Dilewijns informed me at a later stage that he had managed to unearth some info on the topic.

At the time I think there were only three coilboxes installed in the whole world (from what I read the demand has since picked up, mainly in Western Europe) : one at Stelco, one at Port Talbot, and I can’t remember where the third one was (it may have been a second Stelco plant). That’s as far as the topic went – that is, until I returned to the UK and joined Tinplate R&D in Port Talbot of all places.

At British Steel Tinplate we were only indirectly exposed to the coilbox process, since it made it easier to control scale in the finishing train of the hot strip mill. That’s why at times Llanwern was excluded as a supplier of hot rolled coil to Ebbw Vale, since their Type-D scale was linked with the occurrence of pinholes in the final tinplate material, a problem we never had with feedstock from Port Talbot.

There was again a gap of several years until I started extending use of the traffic light system to Port Talbot’s hot mill. Although I didn’t personally work on the coilbox, I became aware of some of the background to its installation in Port Talbot. In short it was lack of space – I don’t know the full story behind it, but presumably it all started with the charging of longer slabs when the reheat furnaces were last rebuilt. This led to the problem that when the slab had been rolled to intermediate bar in the roughing mill, its length was greater than the distance between the roughing mill and the finishing mill. To counter this problem, the roughing mill was converted from a multi-stand to a single stand reversing mill, and a coilbox was built to receive the semi-finished bar prior to final rolling.

Whether quality entered the picture when making a decision I don’t know, but because the coiled bar showed a more uniform temperature throughout, and there was less of a heat loss when the coiled bar resided in the coilbox, the rolling forces in the finishing train were reduced. On top of that, control of scale build-up in the coil shape was superior to that of a straight bar waiting to enter the first finishing stand, so even if this was not part of the initial reason to install the coilbox, it was a nice to have.

Still, if the process is that good, why isn’t it more common ? After all, if it produced a noticeably superior product, you’d have thought that every mill would be converting to have a coilbox. All I know is what I heard on the grapevine, which is that if the coilbox behaves like a dog, the whole mill behaves like one too.

Whatever the case may be, during the recent rebuild of the hot strip mill there were no plans to get rid of the coilbox. Meaning that whatever the reasons were for installing it in the first place must still hold true today.

For some background reading on the subject, have a look at this : Coilbox technology improves steel quality



At one point during my time at Allied Steel & Wire I was asked to spend some time at the reverse twist process. This is a process which takes lengths of hot rolled rod and twist them several times one way and then the other way, thereby raising the strength of the hot rolled product to levels required by the BS spec (min. 460 MPa) or the various continental specs (min. 500 MPa) through cold deformation.

Then, one day I was told to stop working on the reverse twist process, because it was going to be replaced by the Contistretch process. Also, when I came back from my summer holidays, there had been a rearrangement of responsibilities in Tony Franks’ department, and my area became the new Contistretch department.

I’ve tried to find some background on the process by searching the internet, but found precious little for my efforts. Even Celsa, who must have inherited the process from its previous owner, hardly give it a mention on their website except for when they compare it unfavourably with another product of theirs called Celsamax.

So, failing any available diagram, I’ll try to describe the process in words instead. The machinery consisted of two big wheels, slightly different in diameter, over which the rod coil was pulled in a figure of eight configuration, the difference in diameter imparting an extension to the rod of about 8%. It was a very noisy process, and the wearing of ear muffs was understandably compulsory.

While successful in increasing the strength of the rod to the levels required by the various specifications, the process also flattened the ribs to levels that caused concern for certain continental specifications – the ribs being there to improve adherence between the rebar and the surrounding concrete. At some point there was also concern about safety on occasions when the rod snapped under tension, and the process was modified by imparting the stretch in a way that did not require use of the two big wheels.

However, the rib flattening remained an issue throughout, especially when we started to investigate the use of a 4-rib rather than the standard 2-rib profile. The former already started with a lower rib height and after deformation the remaining height was marginal at best.

My input in all this ? Although officially I was the development metallurgist, in practice I took on the role of QA manager. In this capacity I was in charge of the testhouse, the practices of which I compiled into a comprehensive QA manual. I was also responsible for dealing with all the auditors and ensuring that we passed the annual audits. The problems with even minor non-compliances is that at times dealing with them forces you into a corner, and the rib profile of the 4-rib rebar was one of those instances.

I never saw the end of where the 4-rib story went, because I got my redundancy before it came to a head. But I’m in no doubt that it would come to a head. As I may have mentioned elsewhere, the main reason why I was deemed surplus to requirements was that I hardly had done any work as a development metallurgist. There was only one attempt at seeing how a vanadium micro-alloyed would perform under Contistretch conditions (the resulting increase in strength was hardly worth it, something that I could have predicted beforehand), and for the remainder I was more in support of department objectives to improve formability, increasing the size range to 16mm rebar and improving the rib height after deformation.

I don’t know whether the Contistretch process is still there. If it isn’t then it would require a higher alloyed steel composition to achieve the strengths specified in the standards, a cost that would have to be compared with the transportation of rod coil from Castle to Tremorfa Works, and the upkeep of an additional process step.

The process appeared to fill a gap between hot rolled products and wire drawn products, but I have no idea whether this gap was economically viable.

Electro-Slag Refining

During my last three years at university, there was only one other person doing metallurgy in my year, and that was a person called Charles Gheenen. For his thesis he decided to do some work at Professor Dilewijns’ laboratory, which involved working on the new Electro-Slag Refining (ESR) equipment they had acquired in their off-site premises in Zwijnaarde.

I see that Wikipedia calls it “Electro-slag Remelting”, but essentially it involves letting a current go through a molten slag between a bar of steel (the one to be refined) and the baseplate, where the bar of steel gradually melts and deposits itself onto the baseplate, leaving most of its impurities behind in the slag. I seem to remember that it was being used on an industrial scale in Russia, but I’m not aware that it’s much in vogue n the current steel industry, where making clean steel is produced through processes such as vacuum degassing stations.

Anyhow, Charles did a great job on hoovering up most of the literature on the subject, and since throughout his thesis year they had very little luck in producing any electro-refined steel, the literature survey took up the lion’s share of his thesis. Charles was also employed for a while by Prof. Dilewijns’ laboratory as a researcher, starting a little bit earlier than I did, and seeing as he had done the initial work on ESR was the natural person to head the sequel thesis done by a student whose name escapes me now.

This student was now in the unenviable situation that there was very little literature left for him to the first part of his thesis that hadn’t already been covered in the thesis that Charles had written. Although in the end he was more successful in producing actual steel from the equipment, there wasn’t much of it and surely not enough to come to many conclusion of how it affected the steel properties.

On top of that, Charles left his position at the laboratory for Union Minières near Antwerp (I think that’s where his father used to work) after only a few months, and I had to take over as leader for the project. I remember going to Zwijnaarde on a number of occasions, and trying out different configurations to sharpen the tip of the steel bar so that we could get the localised current high enough to start the melting process.

At some point there was also a minor panic when the student mentioned the possibility of noxious gases emanating from the molten slag, so we had to wear gas masks with microphones in them so that we could communicate between some of the workers on the top platform and those observing the process from below. I also remember that the first time we were successful we came back in triumph, declaring “we have a little one !” (referring to the fact that although successful, the resulting refined steel was less than a foot in length).

All in all, the student in question just scraped through for his thesis with 50% of the points, which confirmed in my mind that it’s not a good thing to do a follow-up thesis, since yours will always be compared, and often unfavourably, with that of your predecessor (Charles’ thesis received a rating of 70%).

I’m not sure what happened to the Electro-Slag Refining equipment. At no time in my subsequent time at Prof. Dilewijns’ laboratory did I hear any further mention of it. Had the equipment been on loan, and had been returned from where it came from ? Whatever the case, as I’ve mentioned earlier, I think subsequent developments in the steel industry made ESR a superfluous process, unless you wanted to go for small quantities of steel in a super-refined state.

QST Atlas

When I joined Allied Steel & Wire, my first involvement was with the Tempcore process, and the qualification trials of the 40mm rebar. Over time I got to investigate the whole size range from 16mm to the upper limit of 40mm, as well as material from trials on higher strength materials, and from time to time, competitor’s material.

Wherever possible, I started to get in the habit of summarising the steel chemistry, quenching parameters, mechanical properties, the hardness profile across the bar diameter, and the corresponding microstructures, which formed part of their individual reports. And then it occurred to me that this set of summarised data could easily be compiled into what became the QST (Quench-and-Self Tempered) Atlas. Two typical examples are shown at the bottom of this blog.

The result of all these efforts ? Preciously little. It didn’t fit in any management-led scheme, it was just a metallurgist collating information of possible use for the future. The first problem was that this was in the days that all information was reported in paper copies, and the hope was that these paper copies were being held in a place where they would be consulted – optimistic ? Probably. It certainly didn’t help that this was not the days of servers, where information could be collated in a suitable format for easy analysis and wide distribution.

On top of that, the report came out when the days of the bar mill were numbered, and as such the Tempcore process was on its way out, together with any information that pertained to it.

Still, I can’t help but think that if AS&W had made better use of its metallurgists, and they were allowed to work closer with the production managers to mark the direction in which developments would lead, then this atlas might have been something that people wanted to use and peruse, with a view of making a better informed use of the Tempcore process. Maybe trials to see if the process could be sped up, or the quench flow rate modified, or the deformation pattern for each stand better understood.

In short, if all those factors had been in place, this could have represented a highlight of my career. As it was, it was all too little, too little – with as little impact as a drop in the ocean.

The Tempcore Process

For the first few years at Allied Steel & Wire, I was the development metallurgist for the Tremorfa Bar Mill, which used an interesting type of process to produce high strength reinforcement bar with the addition of alloying elements such as molybdenum or vanadium. We called it the Quench-and-Self-Tempered (QST) process, although officially its name was the Tempcore process, a process originally developed by the Centre de Recherche Métallurgique (CRM) in Liège.

The way the process works is that after a billet has been hot rolled to the correct diameter rebar it then enters a quenching chamber when still white hot, and high pressure water jets quench the bar from all directions until the outside reaches temperatures below 200°C. When the bar leaves the quenching chamber there is still enough heat left in the centre of the bar for the outside to be reheated to temperature in the order of 400°C, after which ii is allowed to cool down to room temperature on the cooling bed.

What this does to structure of the steel is that the outside turns to martensite, a hard and brittle phase of the iron-carbon diagram, during the quench, and the subsequent reheat or temper phase softens the martensite and makes it more ductile. The resulting microstructure displays various types of bainite near the surface, which gradually grades into a ferrite-pearlite structure in the centre.

My first job was to collect the data for qualification trials on the newly developed 40mm QST bar, something that went without too much of a problem as far as the mechanical properties were concerned. It did, however, run into some issues with end splitting in what had started off as the north end of the billet, where most of non-metallic inclusions congregated. Since the defect confined itself to the very ends of the bar, this turned out not to be of much importance, since an extra crop got rid of that portion of the bar.

I started investigating different types of rebar, of various dimensions and chemistry, which ultimately led to trials on a high strength type of bar which aimed to emulate Macalloy tie bars without the need for expensive alloys (I intend to have this as the topic of another blog). Unfortunately that work only came to fruition after I had left for the Contistretch process, so was not involved in subsequent approval trials.

And then came the moment when the directors of the steel plant and of the rod mill ambushed the third director, and carved up his domain amongst themselves, the bar & section mill going to the steel plant, and Tremorfa Bar Mill to the rod mill. The last I heard of it was that the bar mill was to be closed and its product range produced on the rod mill, which was to be converted into a rod-and-bar mill. Not sure what the reasoning behind it was, and how the modification of the rod mill to handle bars was effected, but presumably something must have been done to achieve the required strength without re-introducing expensive alloys in the chemistry.

All I know is that when you go on the Celsa website, there is no mention of the Tremorfa Bar Mill, and hence the Tempcore process is no longer in use in the UK. Presumably the economics must have dictated this, and presumably if the properties are acceptable, economics dictate which way the production goes.