Current and Proposed New-Builds

Trying to remember the reason for Clan going for wooden rather than poly/sandprint but the pattern might come in handy for a few other standards that are extent

 

OTOH, you can also share the CAD to make a new poly pattern if you are so minded, so I am not sure ability to share with other groups is much of a reason on its own to prefer a physical pattern.

Tom

 

Once you've got the CAD all drawn up, how expensive is it to print a new pattern each time - I would imagine not very?

 

Because the cylinder casting is steel?

 

Talking to one of my S15 colleagues in the past he is against poly patterns because of the residue remaining on the casting. Is it a problem when fettling / machining?
Pat

 

Yep that might be it, ... more chance of bubbles or summink ?

 

I'd be interested to know who is making quality castings from pollypatterns at the moment.

Premier Patterns is only a patternshop, most of his pollypattern castings were being done by Shakespeares before they went, and I've not been very impressed by some of the offerings we've had recently; we had two goes with two different foundries on a set of pollypatterned cylinders for a narrow gauge loco recently, and even the second attempt (although usable unlike the first) left a lot to be desired.

It's worth noting that there are a few "non traditional" routes to producing castings for railway loco parts now.
As well as normal pollypatterns, there is a variation where pollypatterns are used more like a wooden pattern, and broken up to get them out of the sand before the metal is poured (I think this is how "Warwickshire"'s block was done).
Another fairly new technology (more applicable to smaller parts than cylinders) is sand printing, where a conventional sand mould is 3D printed,rather than created with a pattern. For moderate items (say 100kg) this is only marginally more expensive than polly patterns, and the finish is much better.
Also worth considering for smallish items is lost wax investment casting - you can get 3D printable waxes, which allow the production of really accurate (~0.1mm) bronze castings, whilst you can use 3D printed PLA for larger items at a slightly lower quality.

 

Bradley Manor is another loco with newly cast cylinder blocks using poly patterns. There is a report with pictures of the actual castings in the March 2020 online update.

https://www.erlestokemanorfund.co.uk/march-2020-update/

The poly patterns themselves weren't shown in the update but were pictured in a recent 6-monthly newsletter. I hope the EMF won't mind me reposting here.

 

I wonder to an extent if it doesn't matter (within reason) which technology is used, but that enough of it is done to develop an understanding of it.

What would happen if you fabricated a cylinder block then filled up the (non-working, obv) voids with poured metal? Would that give you greater resillience and better thermal properties, or just the worst of all worlds?

 

My guess would be solid block s weigh more, and expand /contract more with nowhere for that force to go except the internal spaces ( ie cylinders) and external faces ( which mate with other components) that remain..

 

It does seem puzzling that the 4709 people had so much trouble finding someone to cast their cylinders when new cylinders have been cast recently for several other locos.

 

Might size come into it? The castings for the 47 are a tad on the large side compared to say inside cylinder 0-6-0T.

 

I heard that polystyrene contains carbon, which becomes incorporated into the steel and affects its strength.

 

With any new technique there will always be people who tell you why it won't work. No doubt there are subtleties that have to be learned, but the process has been in use in the motor industry for a good few years now. As for carbon, I've seen a reference stating that a typical mould is around 2.5% polystyrene, the rest air. The actual carbon content will be less than that, of which the vast majority is going to evaporate or burn off. My uninformed guess is that anything left will mostly be on the surface and that few metallurgists will be losing much sleep over the possibility of significantly altering the composition.

 

A minor factor in that is that the foundry everyone had been using up until and including 7802 is no longer in business, the day after the arrival of the 4709 patterns I kid you not....!

Foundries understandably are reluctant to take on items of perceived high risk, in addition to that there are even less with the capacity to cast something like a GW cylinder, once sifted through those that don't want the risk and don't have the capability there are very few fish left in the pool. With the demise of Shakespeares it has been a long difficult search to find foundries even willing to talk about it, let alone have a go. Patterns have been delivered to one, further meetings held, only for them later to back out. I've had involvement of some degree with every set of GW cylinders cast or planning to be cast post 6880 using the polystyrene method so have gained a degree of understanding by now.

To successfully cast from a polystyrene pattern there needs to be an understanding of the process and its limitations, the more intricate the item the less chance of success, therefore at the drawing stage you need to look at 'beefing up' the cylinder and leaving a good machining allowance, following a traditional pattern approach won't work.

Then you need to orientate the pattern in a particular way so that any little defects from residue, which can not be avoided, end up on surfaces that are non critical. A polystyrene pattern of course is not as rigid as a wooden one so care in particular needs to be taken when moulding and packing where there are large flat thin surfaces as it would be easy to distort said faces. Just chucking a pattern at the Foundry and saying 'cast' isn't going to pass muster, they need to understand what it does, where critical faces and surfaces are and where they can cut holes in the pattern to allow access for moulding to be able to successfully carry out their side of the process



Picture of 3850's cylinders last week, just prior to them being assembled for the final time after a 'dry assembly' run. You can see even after a healthy machining allowance a few minor defects, but these are in areas where they are of no consequence thanks to the orientation of the pattern.



Moulding underway.

Since Shakespeares were lost, who had perfected the process, two cylinders have failed during the casting process in different locations. There is a risk, the risk needs to be understood from the outset because obviously in the event of a failure you lose the pattern, but with the pattern a fraction of the cost of a traditional item you can afford to have a second go with plenty of change, if the Foundry is prepared to try again of course! Boro Foundry who cast the 4709 cylinder which I visited to inspect last week were very proud of themselves (rightly so) and very much have had a 'can do' attitude throughout the process whilst many others thought best to steer clear.

https://www.4709.org.uk/4709blog/4709s-first-cylinder-cast-satisfactorily-at-boro-foundry

4709 blog link.


https://www.facebook.com/DinmoreManorLocomotiveLtd/

DMLL link, more pics of 3850's new build front end within, no FB account or login needed to view.

Hope this is of interest,

Mike

 

That's the kind of thing I had in mind when I made my post which started this sub-thread.

Thanks for the insight from the coal-face.

Noel

 

I think the simple reason is that a gentleman with the necessary time and skills offered to produce the patterns at no cost other than materials, as it was not an urgent item at the time it was agreed.
Certainly the report on progress and description of the work makes it a very interesting project on its own. Clan Project has used several different methods, wood, poly and 3D sand-printing so far.

 

Having just re-read O.S. Nock's Locomotive Monograph on them, and always having been very interested in them, I'd love to see a pair of Caledonian Railway Dunalastairs.... For comparison, either a rebuild Drummond 4-4-0 or a Non-Superheated Dunalastair I, and a superheated Dunalastair IV.
The free-running of the Dunalastairs was legendary.
It's surprising that they were withdrawn as soon as they were, leaving the later Pickersgill variants to soldier on until the early 1960s, when the Pickersgill Locomotives were known not to be as free-running.
Maybe even a Pickersgill 4-4-0 would be good, as they had thicker frames - but with a re-think regarding valve events perhaps?

There's something about Caley locos with smokebox wingplates to me though - Along with the NBR's Glen Douglas, they just look "Right".

Richard.

 

Is it this locomotive you thinking of Richard? https://www.railsimulator.net/wcf/attachment/40063-20180521001955-1-jpg/
If so I`m agree with you that they are beautiful locomotives and the livey as well. I have also a soft spot for the Caledonian cardean class

Knut