GWR 813 Failure

Could the casting be welded? There are specialist cast iron welders who might be able to advise.

 

I think it would be not worth while having the casting welded. it is over 100 years old. The Fund is going ahead with a new casting in steel.

 

Given the work required to get at it, would you trust a repaired item and hope it doesn't break again?

 

Wondering that myself. I was also wondering about the advisability (necessity?) of replacing 'handed' components in pairs.

 

If I'm understanding the earlier posts and pictures correctly, it's a single cast stretcher housing a transverse spring, so is both hands in one, so to speak.

As for the relative merits of casting and fabrication, while fabrication has its virtues, I'll note that the Beachy Head team recently decided to replace an existing fabricated bogie stretcher with a casting. They needed to change it anyway, but it's clearly not as straightforward as: "Fabrications: cheap, good. Castings: expensive, bad." I think the lack of accessibility of the welds for regular inspection was cited, as well as authenticity.

 

Quite a bit to chew on there. Cheers. Can't help but think of 72010's manifold 'spider' casting ....

 

I think one of the things that, ten - fifteen years ago, counted against castings and led various groups to investigate fabrication was the cost of pattern making, particularly when an expensive pattern might be amortised over just a single one-off casting. (And, related, the scarcity of traditional pattern-making skills).

I think that reason is diminishing. I note in post 20 above that the broken casting has gone for 3D scanning. Take the image produced and it is a relatively small job (for someone with suitable skills!) to turn it into a drawing from which a poly pattern can be machined, all done in ones and zeros. That leads to a marked decrease in cost of casting for one-off components.

Obviously there are still issues with casting quality, but with a fabricated part you have similar issues with weld quality. So it feels to me that the balance between casting and fabrication for one-off components is possibly swinging back towards casting, relative to where it was twenty years or so ago.

Tom

 

Attached is a shot of the chassis of the ex-Slough Hudswell Clarke taken when it was on the Mid Hants. You can make out the transverse spring under the cab floor, and the cover plate / inspection hatch / whatever that also showed in pics of 813.
Pat

 

For the most part I agree with Tom but I take issue with the suggestion that weld quality might be a problem. That should only be the case if it is done by an iffy welder (like me) and there is enough technology out there to check the welds on completion.

 

Does the liquid metal-pattern interface display signs of instability in the lost foam process? It is a concern when casting aluminium alloys by means of this method.

 

Presumably this means the loco is out of action for the remainder of the summer at least?

 

Reading through just a page or 2 of this thread., makes me think back to the early/mid 1970s'. IIRC the owners of 6201 Princess Elizabeth made quite a splash in the railway media when they retubed the boiler..... Yet today that sort of thing is looked on as almost routine, just like the driver oiling round all the bearings etc. Just look at what folk have got upto in the last fifty [yes nearly fifty] years when it come to replacing even building complete locomotives [i.e. Tornado] or making really quite complicated replacement parts..... without thinking sort of thing.

 

For what it's worth, I've commissioned a few 3D scans, mostly for fabrications where we need the as-welded form. I've never put out a single casting for manufacture.

The process of scanning and interpreting is increasingly routine; the main trick is in getting a 'watertight' surface that the software can actually recognise as a closed solid. Even then, relatively run of the mill CAD software has got quite good at stitching up surfaces containing gaps where the part was sitting on a support.

Taking the example of scanning an existing part as a pattern, once you've got that watertight solid, things like scaling up for a shrinkage allowance and adding a machining allowance on the appropriate surfaces are basically trivial with current software.

Given that 813 had a casting already, it seems to me that in the interests of authenticity, if it's cost (and performance) competitive to produce a new casting, that is more desirable than to replace with a fabrication. Using a stronger steel (or maybe SG iron) casting in place of the original grey iron seems like an acceptable allowance for the original having broken. No, you probably wouldn't do it that way in modern industry, but there's not much modern about 813 either.

 

26 May 2022

 

There was a time not so long ago when a piece of three-quarter inch plate across the top and screwed and bolted to the existing casting bits would have put it back in service quite quickly and probably seen it to the end of its working life.

 

A jury-rig solution is one thing, but surely preservation is something a bit more ambitious?
Pat

 

Two opinions, both 'Worthy' in their own way. One of My engines finished It's working life with a pot mender in the fire box crown, have a guess whether I retained It after restoration?

 

We did similar on nuclear power stations

Sent from my SM-A105FN using Tapatalk