Wednesday 7 October 2015

Thermite welding at Laverton - a special report

A four unit team from Haigh Rail Ltd came today in 4 vans to weld up the first ten pairs of rails towards Broadway. Each team busied itself with a separate set of the 18 joints in question.

This exothermic welding process was rather spectacular, and you will excuse the larger number of pictures in this posting, which will describe how it was done. Your blogger, although well acquainted with Pway work, had never seen this, and no doubt many of our readers will find it interesting as well.

On a cold and blustery day, the 4 vans lined up along the rail head. Each van held the materials and gas bottles that support a crew to make a weld.

Pending the welding, the rails were not permanently fixed to the sleepers, being clipped only on every other sleeper, and without pads. Now that the welds have been effected, we can properly clip up this stretch.

The first step is to create a gap between the rails to be joined that is exactly the right size. This gap is made with a special cutting torch that slices off the end of the rail of the desired thickness.

With the proper gap, the rails are then checked for a perfect alignment, and corrected where necessary with wedges, and sometimes, with brute force.

Using a special frame clamped to the rail, a pair of moulds is attached around the joint to be welded.

The joints around these moulds are then sealed using a special sand with a resin in it, which makes it sticky so that it can be tamped into place.

The joint is then heated using a special gas burner which blasts a flame into the mould, and which erupts back out on the top, with a rather spectacular effect of a V Twin flame that roars out. A bit like my motorbike then.

Mick explains how the heat can be adjusted to the right degree.

A single use crucible containing the Aluminium and iron oxide powder is then place on top of the mould, and ignited with a special taper.

To the left and right of the moulds are pots which catch the molten Aluminium which is released as part of the process. This is lighter than the iron which sinks to the bottom and into the joint, so the Aluminium floats to the top and can be collected at the sides.

Shortly after the reaction starts, the plug at the bottom of the crucible melts, the iron flows into the joint below, and the aluminium into the sides, with a roar of flames and heat.

It doesn't take long for the reaction to conclude, after which you simply lift off the single use crucible, leaving two pots of molten Aluminium at the sides.

These pots are then upturned into the empty crucible for disposal.

Once the apparatus has been removed, the remains of the moulds are left, as well as a stub, which is too big to be ground off. With the use of this hydraulic cutter, the stub is then cut off.

The cutter just needs a lot of pumping, to force its jaws together through the stub, which is still hot.

This is the stub that is left at the end of the thermite welding process.

A portable motorised rail grinder is then used to grind off the remains of the stub, and to achieve a completely flat rail surface.

This picture shows the untreated stub in the foreground, and rail grinding in progress at the rear.

Finally the inside of the web is cleaned up using a Hilti with a needle gun attachment.

Surprisingly, there were different forms of crucible in use. Each supplier of the materials has his own, slightly different system. The next couple of images show you the types used by one of the other teams, just as curious as the first.

A few joints away, a second team is seen setting up their pair of rails, which were not where they should be. Although great care was taken to spread the ballast out evenly using a laser level on the JCB, it can still happen that there is a height difference between rails to be welded, such as is the case on the right. Rails with concrete sleepers are notoriously heavy, so you can see the effort that was needed to shift them, even if only a few millimeters.

This team was using a French single use crucible, which had to be filled with the powder kept in special tubs. This is what it looks like, the silver colour being accounted for by the Aluminium.

The process generates heat up to 3000 degrees C, which is considerable. According to a recent television documentary, it is thought that the extreme heat of the molten Aluminium from the 9/11 planes burning accounted for the unexpected weakening of the metal framework inside the towers, not simply the crash.

This French system has trays facing the other way, and we can see here how the powder is about to be lit from one of the pre-heating torches. A lid goes on top.

Mick supervises the process, and gives advice based on the many years he's been doing this.

The lid goes on. In the foreground is a tube of sealant, which is used in this process instead of the resinous sand used in the first.

The burn takes place, and after a short while the aluminium starts to flow into the two receptacles on either side. You can imagine the heat, even from several feet away.

In fact, there is so much heat released that it lingers, and frozen GWSR volunteers can avail themselves of it and warm their hands.

A cup of coffee and a chat in Andy's car was also a rather reviving pleasure, out of the wind on top of this exposed embankment.

The rail grinder then came along here as well, and showed us how it can do sides as well as the top. A neatly designed machine.

Then, finally, a third type of crucible was in use at the Laverton end. This is the classic type of crucible, which is multi use (ie not throwaway.)

This steel crucible is mounted on a frame, which allows it to be swung into place once it is loaded, and the moulds have been attached.

Here is a view into the top of the mould. The gas burner blows flames into the centre opening, and the flames, once round the joint, come out again through the two holes on the side, so making the 'Vee' effect.

The pots collect the molten Aluminium.

Here a special plug is inserted into the bottom of the multi use crucible. This melts at a predetermined temperature, and allows the molten iron at the bottom to flow into the joint.

The Aluminium and iron oxide powder is then poured into the top, and the whole thing lit with the taper.

As this takes place, the gas burner is heating the up the joint (flames at the front), and when it is at the right temperature, gauged through the use of stop watches, the crucible is swung over the mould and ignited.

There isn't a picture of this one going off, but there are two videos.

The first, longer one, shows the burn using the single use metal container, as well as the removal of the stub with the hydraulic jaws:

The second film, rather shorter, shows the multi use crucible ignited, which seemed a bit more spectacular still:

Finally, a quick check at the Broadway goods shed showed that the 45T excavator had finished the job for which it was hired. Our engineer and our contractors now have a good understanding of the structure of the embankment there, and we await their options and recommendations.

The 15 trenches were back filled today, and the slope of the embankment regraded to give a gentler slope, starting at the row of yellow tipped posts. Unfortunately, this will no longer give room for the siding that was there, but there will be the double track main line, on the right. The photograph was taken from the position of the former siding.


  1. Slightly off topic, but you did mention it. Molten steel was found at wtc 7 rubble too, and that wasn't hit by a plane. Infact, it fell at free fall speed for 2.5 seconds, or 8 floors, symmetrically into it's own footprint.

  2. When the welds are in place jo, will the track be slewed and ballasted ? Will the section connecting the laverton running line railhead also be welded? Thanks for an informative blog. Do you have many parts for the station at broadway? Would like to know what cool railwayana you've managed to find!

    1. As I've never seen CWR being laid, I'm not sure of the exact order of things.
      We do have to re-attach the rail to the sleepers (currently only loosely attached, every other sleeper) and there will be a lot more welded rail, all the way to Broadway, which will see the only expansion gap. The rail also has to be stretched, at a certain temperature.
      There will be a tamper coming in too, in due course.

      At Broadway a small group of us are pursuing various items of railwayana which will be vital to an authentic GWR station. It only comes up occasionally, so you can't just go out and order it, you need to strike while the iron is hot. In this way we picked up the two GWR lamp posts, and the V board 'BOOKING OFFICE'. We need more of both.
      We have spotted three or four items which are of interest at a forthcoming auction, so fingers crossed.

      We would be interested in the following: (all GWR type)

      Wooden benches (all wood, for under the canopy)
      Ticket racks
      Lamp posts
      V boards
      Platform seats of the earlier 'script' type (not the later 'shirt button' design)
      Fire buckets and racks
      A water tower
      Diamond pavers
      'Passengers must cross by the Footbridge' cast iron notice

  3. You mentioned your surprise at the two methods. Was there any explanation about this? I mean, I thought it was one firm that did the job. Sorry if I am a bit thick about it -or do the chaps just like variety in the job? .

    1. Howard. I'd presume it was a 'Training Exercise' (for the differeing methods). I've been told that when the CWR was welded through Bishop's Cleeve, THAT was done in the pursuance of 'Training' and was actually laid to 100mph standard.

      Whether true or not? I'll stand to be corrected.

      We seem to be fortunate in that we have certain persons who have 'contacts' that they utilise in order to provide Training facilities to Contractors at a nil or reduced cost to the Railway.

  4. Great piece Jo, thanks for that.

    Why do you not use Fish Plates to connect the rails? This would maintain that old mesmerising clackerty clack, that was so evocative of rail travel in the forties and fifties.

    1. Using fishplates means lots of maintenance, because they lead to dipped joints and bent rail ends. Bump-bump-bump, rather than clickety-click.
      If you saw Chris Tarrant in Burma recently, you'd see what happens! We only have a small team of volunteers and up to 15 miles of joints to maintain, so the idea is to reduce the load on them, out of sight of the passenger.

  5. On the subject of Diamond pavers, if you are referring to the blue type used on the platform at Gotherington, then there used to be some at a reclamation yard in Worcester.

    Worcester Reclamation Ltd

    Railway Goods Yard, Woodbury Lane, Norton
    WR5 2PT Norton

    01905 351615

    Could be worth a try? After all, as (I think) their site falls within the newly planned Rail interchange, they, themselves, will no doubt be 'reclaimed' in the not too distant future.

    1. Thanks, Ken. The issue is cost really, not availability. You can get them new, but at a price. What would be nice is if someone said, hey, I've got a yard full of these I don't want. We need about 50yds worth for the pavement.

  6. Jo,
    There used to be a separate fund raiser site for the BAG, to enable donations to be channelled directly to the Group. It was also used to raise funds for the signals and the signal box window-frames. That facility seem to have disappeared and the only route for donations is through the GWSR's general site, with no guarantee of use for a particular cause within the overall Company.
    Is there some way we can make donations online to help purchase these smaller items of furnishings ?

    1. Hi Mike,
      It is true that the new Broadway blog does not have a donation facility.
      The best way to help the railway would be to buy shares in the next share issue, to fund the Broadway extension itself. We will need a really big effort from our supporters for this one. It will be another EIS issue, so potentially you could claim a rebate from HMRC. The share issue should launch in late spring 2016.
      The best way to help the BAG in particular would be to go in person, make a donation, or peruse the bric a brac stall.
      You could also send them a cheque, addressed to the BAG treasurer via Toddington. BAG do buy heritage items ( we are going to have a go this weekend in fact) so help with this would certainly go a long way.

    2. Thanks Jo.
      I am overseas and no longer have a UK bank account. Bank drafts (cheques) are expensive if for small amounts. I prefer to make small donations (20 to 50 pounds) on an irregular basis and the old system, which accepted credit/debit card payments was ideal for this. I guess I'll just have to save up and make a larger payment in a year or so's time.

  7. When I worked on SVR, they had their own trespass notices cast in Aluminium to a GWR style. Maybe a liase with them could help with cast replica "cross the line" boards. Also GWR used "WHISTLE" boards, metal letters on wooden boards, within station limits as opposed to the "SW" cast ones; which could be fabricated quite cheaply and authenticly. Hope this helps.

    1. Thank you, Paul, a useful suggestion. I do have a line into the SVR, and they have already helped us.
      I heard on Monday that there are two 'Cross the line' boards at Winchcombe, but we need to ascertain the owner first, and ask politely whether we might have them.