BAD NEWS 3D printed spoked driver centres go eccentric

[Brian McKenzie]

Disappointingly, the driving wheels of two locomotives, using 3D printed centres, have become distorted over two years, and will need replacing with brass castings.

Uneven mass in the wheel centres (large counterweight opposite small crank boss) seems to settle to a new equilibrium within the press fit of the tyre, causing the axle hole to shift off centre.



The difference measures 0.0035" T.I.R. at the same position in all wheels. Although not a lot, it looks awful as this eccentricity is highlighted as the side and con rods rotate.

This is disappointing, after the care taken to bore axle holes after clocking up the wheels.



Hopefully, this shouldn't be a problem with wagon wheels and any others with symmetric shape.

It's worth noting that not all injection moulded wheels are immune from a similar predicament, as some plastics (e.g. glass-filled Acetal) have anisotropic shrinkage (i.e. they contract at a different rate out from the gate, compared to the direction perpendicular to it).

-Brian McK.

 

[simond]

Brian, this is obviously frustrating. What were the 3D prints made from? Resin or FDM? Which compound?

thx
Simon

 

[Lawrence Boul]

I was rather surprised that you tried this Brian. It raised my eyebrows when I first saw it, and it quite shattered my illusions of your infallibility

I'm not surprised things moved a bit if it is a uv resin. I'm reasonably happy using that process for wagon wheels, which are obviously symmetrical, but I wouldn't be game to use it for loco divers. I take the view that I want loco mechanicals to last and I'm extremely risk averse in that area.

The good news is that you know I'll cast them for you!

3D prints in a castable resin come out very nicely with low shrinkage and pretty good concentricity. We in 9mm raw casting, Wa in 3/16" (RP25-088 tyre, and no surface clean up by the look of it) and Ab in 3/16" with the tyre profile you defined for us a very long time ago. I probably should have wiped the gunge off the Ab wheels before taking the image.

 

[Brian McKenzie]

Simon, SLS printed in China from Somos® Taurus material.

A Brutalised Barclay

Barclay's #1270 of 1912 started life as an 0-4-0ST on a 3' 6" gauge forestry line in New Zealand. It proved rather heavy for the track, so was hacked about and extended in length to accommodate pony trucks at both ends as a 2-4-2ST. It operated in that configuration until 1924, until sold on...

www.westernthunder.co.uk

Not too much of a problem for me, except that rim insulation might show - as the tyres are flush with these wheel centres - with no overhanging front lip to hide it.

Thanks Lawrence, I just ordered brass castings moments ago before reading your message. In the years before my OE (1972-9) got through 740 nickel silver wheel castings by long gone Alloy Productions, Wellington. They disliked doing n/s as it meant having to light up the Oxy Acetylene. Another 1500 wheels or so since, mostly gone offshore.

In those early days, a form tool and a Unimat SL did the business. The castings were driven by the pins in these brass adapters, which mounted to a steel mandrel threaded on the lathe spindle. The lathe's cross slide thread eventually wore out, and a larger diameter LH leadscrew was screwcut. A piece of this was made into a tap (as shown below).


-Brian McK.

 

[Osgood]

……
3D prints in a castable resin come out very nicely with low shrinkage and pretty good concentricity.
….

Lawrence, I’m probably at the back of the class here suddenly paying attention, but by ’castable resin’ I take it you mean a ‘lost resin’ equivalent of lost wax for printing sacrificial patterns for use in casting?

I was at a printers recently arranging scanning and 3D printing of a replica inertia starter from a WW2 truck using an original as a pattern, and was shown a large multi-finned cylinder head from a 1930s motorcycle that had been printed from a scanned cracked original using powdered aluminium - where each particle was fused by a laser upon deposition, thereby effectively creating a ‘printed casting’!
Also some basic flanged aluminium pipe fittings using the same process, so I guess cost might not be prohibitive?

I imagine far more spacially complex and intricate fine ‘castings’ could be created using this 3D printing /fusing process than within the limitations of lost wax casting (although those wheel centres you show are a work of art for sure!).
Also wondering whether other materials can be used in this process?

 

[Lawrence Boul]

I've got some tooling a bit like that. I now 3d print the holding side of things with a model derived from the wheel. The resin will not be as strong as your pins but as it's an exact fit the bearing area is maximised. And it's easy.

I buy my taps though...

 

[Lawrence Boul]

Lawrence, I’m probably at the back of the class here suddenly paying attention, but by ’castable resin’ I take it you mean a ‘lost resin’ equivalent of lost wax for printing sacrificial patterns for use in casting?

I was at a printers recently arranging scanning and 3D printing of a replica inertia starter from a WW2 truck using an original as a pattern, and was shown a large multi-finned cylinder head from a 1930s motorcycle that had been printed from a scanned cracked original using powdered aluminium - where each particle was fused by a laser upon deposition, thereby effectively creating a ‘printed casting’!
Also some basic flanged aluminium pipe fittings using the same process, so I guess cost might not be prohibitive?

I imagine far more spacially complex and intricate fine ‘castings’ could be created using this 3D printing /fusing process than within the limitations of lost wax casting (although those wheel centres you show are a work of art for sure!).
Also wondering whether other materials can be used in this process?

More or less Tony. If you have a resin printer then the principle is straightfoward. A bit trickier in practice but not a lot. Limitations to the lost wax casting applicable to modellers were a) you needed a pattern, b) it needed to be mouldable and c) it needed to be castable - meaning the cavity had to fill and the investment structure needed to resist the casting forces. With a sacrificial resin a) is easy, b) no longer applies, but c) is a bigger deal as it is very easy to push the limits. Shrinkage and distortion are much less than with moulded wax.

People locally have done wagon underframes in printed aluminum out of China. Costs are not bad, but I think the resolution is quite low. But I've no experience so don't take my word for it.

This is a 1:64 motion bracket/slidebar/cylinder end casting (Ab class). Not perfect, but very good. You can see the lock nuts on the slide bar mounting. The holes are cast in. Where you run into trouble is if fragile bits of investment break off. That is, of itself, problematic but what is often worse is those broken bits end up as a void somewhere - usually in an inconvenient spot. Castings like this make modelling much easier as there is less assembly and part fit should be excellent.

 

[Tim Watson]

Glad to see you on WT, Lawrence. Your horses are doing sterling work on Copenhagen Fields…

Tim

 

[Mikemill]

Hello Lawrence

You say these castings are resin, but they look like brass, am I missunderstanding the process?
If it is brass I have some 3D printed backhead fittings that I would like to have cast in brass, could you help?
Thanks

Mike

 

[Lawrence Boul]

Hello Lawrence

You say these castings are resin, but they look like brass, am I missunderstanding the process?
If it is brass I have some 3D printed backhead fittings that I would like to have cast in brass, could you help?
Thanks

Mike

They are brass. You print in resin and use the print as a sacrificial pattern. 'Lost resin' if you like.
In principle I can facilitate castings for people. In practice what I get often needs quite a bit of work and it's already quite an expensive process. And I'm in NZ, so shipping is significant. So maybe. I don't want to hijack the thread, so here's an alternative approach:

1:64 Locomotive cab detailing – NEW ZEALAND FINESCALE

 

[Nigel Cliffe]

Lawrence's castings look very good.
There are plenty of companies in the UK and other parts of the world who can do the same with "lost resin" or "lost plastic" casting. The "lost bit" produced on a 3D printer, so shouldn't be necessary to pay a fortune in international shipping to-and-from NZ. Many will offer two related services: print from supplied sacrificial elements, or they produce the sacrificial elements on their printers from your CAD files.
As Lawrence indicates there are little nuances in how casting works, so not every abstract shape can cast easily, but there are usually solutions.

All the driving wheels for the 2mm Scale Association are done with "lost resin" casting to create brass centres.


Back at Brian's 3D printed plastic centres. Its highly likely that the choice of 3D printing plastic is relevant to the problem experienced, and a different plastic might offer better stability.

 

[garethashenden]

How tight was the original fit when the tires were pressed on?
When I've used 3d printed centers I've turned them to be a sliding fit and then held them in place with Loktite. I was thinking at the time of the resin's brittleness, rather than its movement with time. The thought of pressing the tire halfway on and having it shatter was unappealing. The type of resin, how it was cured, and how its been stored since could also effect its subsequent movement.

 

[-missy-]

There are plenty of companies in the UK and other parts of the world who can do the same with "lost resin" or "lost plastic" casting.

Hello Nigel.

I would love to hear any suggestions on them. I can only think of one, maybe two in the UK that seem willing to help 'us' out.

Julia.

 

[Ian@StEnochs]

A wee bit worrying.

I have a set of 3D printed centres on my current project. They are only 6 months old and were a sliding fit in the tyres. I had a groove printed on the outside face of the centre and a corresponding machined one on the inside of the tyre, fixed with epoxy hopefully making a permanent joint. The axle holes were bored with the wheel in the lathe. I painted them too just in case oil or flux would react with the plastic. Early days but. I will keep an open mind.

For my last project I did get lost wax brass castings made but in that case the caster used the print to make a conventional silicone mould to make the waxes.

 

[Nigel Cliffe]

I would love to hear any suggestions on them. I can only think of one, maybe two in the UK that seem willing to help 'us' out.

Search for jewellery casters.

 

[-missy-]

Search for jewellery casters.

Oh.

Sorry, I thought you might have some personal reccommendations.

Julia.

 

[Jordan or Plymouth Mad]

I’m probably at the back of the class here suddenly paying attention,

Nah, you're further forward than me, certainly - some of the technical stuff in this thread might as well be in Swahili or Japanese as it sails waaay over my head...!!!

 

[Dog Star]

...some of the technical stuff in this thread might as well be in Swahili or Japanese as it sails waaay over my head...!!!

Me too Jordan... such a pity because Brian McK seemed such a decent chap when I met him at the most recent MMRS show.

 

[Brian McKenzie]

How tight was the original fit when the tires were pressed on?
When I've used 3d printed centers I've turned them to be a sliding fit and then held them in place with Loktite.

Yes, light interference fits were involved, but Loctite 680* was also used in grooves in the plastic, at both tyre and axle fits.

I'm cautious using Loctite 6xx series, as it exerts tremendous pressure when set - but as the plastic was enclosed in a metal rim - I expected it to be OK. The unbalanced proportions of the plastic wheel centre have probably re-settled, causing the axle bore to go off centre.

An all plastic wheel, Loctited to a steel axle is in danger of splitting. A work colleague Loctited screws threaded into plastic electrical enclosure boxes - and the boxes all split.

Additional sets of these wheels were made, and fitted into both steel and stainless steel tyres, with axle holes bored - but not yet on axles. I could check the bore of these, but probably won't bother - preferring to get on with an all-metal solution.

( *product numbers vary internationally )

 

[Lawrence Boul]

Oh.

Sorry, I thought you might have some personal reccommendations.

Julia.

Hi Julia, I've tried a few casters over the years, one within the last few months that did not go well. Not all are able/willing to do the sorts of things that we want, and quality varies a lot. It's also a fast moving field just now so things change. Getting a recommendation is a good approach I think.