Locomotive driving wheels

[Guest]

Interesting. Seems others are considering this path - have been for some time, from what I hear.

 

http://www.rmweb.co.uk/community/index.php?/topic/86072-3d-printed-wheel-centres-to-convert-rtr-n/

 

For myself, I've not long finished doing battle with a corrupted boolean subtraction in something I've been working on. A viable STL file has now been uploaded to Shapeways. More soon.

Edited May 23, 2014 by Guest

[Guest]

So, having wasted all this time learning CAD, I thought I’d share with honourable members of this forum a few images of the iffy driving wheel that has resulted.

 

This arrived this morning. It is not prototypical by any means, being an old file that I simply dumped onto Shapeways to test the technology. I say test because it is 3D printed in steel. For speed, I’ve taken a couple of shots with the webcam, so please excuse the quality. We have a friend who used to be a professional photographer and who does all my wife’s publicity material so hope to get a few really revealing close ups when next we see her (of wheel, not wife!).

 

Honourable members will note that, being made of metal, they are entirely suitable for split-axle pick-up. No, this is not a Road To Damascus conversion on my part; I simply could not see a viable way of using the plastics on offer. Call it manufacturing and constructional expediency. Be assured, I do not like split-axle current collection.

 

 

 

There are print lines in evidence but not to any marked degree. Clean them up a little, slap on some primer, not to mention a final colour, and you wouldn’t even know they were there.

 

Observations.

 

Accuracy. Shapeways give a manufacturing variation for this material of +/-1% of any dimension (and one layer thickness of 0.1mm), but I don’t know if this variation relates to part design or individual orders: in effect, if one were to order six of these wheels, would they all come out the same size but maybe not quite the size intended; or would one get six wheels of varying sizes—each within +/-1% of any dimension (and one layer thickness of 0.1mm) but none of them exactly the same? Honourable members will note that I designed in a hole for the crankpin, and I’m sure I don’t need to point out (assuming the latter to be the case) such variation causing binding of the coupling rods with each revolution of the wheels. Other related issues are available.

 

Wear. How durable is this 3D Printed stuff? Steel it may be but what sort of quality are we talking about here? Tyre wear? The stub axles wearing or are we looking at separate axles here? These are questions that I suspect can only be answered through experience. I’m up for that.

 

Strength. Would this wheel withstand being put in a lathe and turned? The tyre feels rough when a fingernail is scraped along it, it being not perfectly round. This is because CAD always draws curves as a series of flats, which are then translated into the print. With the CAD package I’m using, the maximum number of flats that can be incorporated in any full circle is 200—small flats and a lot of them, granted, but the result is still not perfectly round. The excellent Douglas Self site makes mention of a loco being designed with non-circular wheels—this intended to boost adhesion, apparently, by way of increasing contact area between tyre and rail—but unfortunately for me, this was confined to one prototype only and was not the resounding success its designer doubtless hoped it would be. 

 

This point about roundness can possibly be answered with the next wheel on order. This is minus tyre and made of brass—Shapeways have been very clever with this method, printing the STL file in wax then using that for an investment casting process. Obviously, being minus tyre, the roundness issue doesn’t arise. Images will be posted as soon as it arrives. As an aside, I’ve also had a crack at the cross-section of the spokes, which I was never happy with. A quick CAD lesson with elder son via Teamviewer and I’d mastered spline editing and loft object creation thereof. Easy when you know how. Anyway, the long and the short of it is that the spokes are now much closer to the flattened elliptical shape shown in the images that Jim pointed towards earlier in this thread. Said son is now taking me through the finer points of mesh editing so the spokes can be produced as single parts instead of the three I’m currently using, and also deal with the flare-into-rim issue that I’m still having trouble with. This intermediate shot will show you what I mean.

 

 

It is not easy trying to produce a file from a vague diagram and some (albeit excellent) photos: trying to gauge dimensions is not fun. If I had access to a drawing of the actual wheel, I have no doubt I could—even at this stage—produce a pretty accurate file for printing.

 

So this is where we are at the moment (naysayers note the emphasis, please), with something that is not by any stretch of the imagination perfect but still quietly encouraging. Given that 3D Printing is improving in quality all the time, I will continue to experiment because I really think this is a way forward.

[Horsetan]

....We have a friend who used to be a professional photographer and who does all my wife’s publicity material so hope to get a few really revealing close ups when next we see her....

This made me smile, if only because of the ISIHAC angle

[Guest]

This made me smile, if only because of the ISIHAC angle

 

A radio show fondly remembered, Ivan, though it nearly led to my undoing. Being the proud holder of a Class 1 LGV licence (or whatever the hell it's called, these days) and being in a slack period in my freelance work, I signed up with an agency and did a little trucking for a while, mainly night driving. I remember I'd just started a shift, a Newbury-Birmingham-Bridgend-Newbury triangle run with a 32-ton artic strapped to my back, and was listening to the show while going up the A34. Just passing Oxford and Humph closed the show (from Blackpool or some such seaside town) with the words "Now the lovely Samantha and I will be going for a traditional blow along the front." I was laughing so hard, I nearly came off the road. Would have looked good on the accident report...

Edited June 23, 2014 by Guest

[Guest]

Interesting news and progress, David.

 

I like the idea of printing in wax, but personally would prefer to receive just that so the waxes can be inspected and checked for any flaws or imperfections, and then arrange to have the lost wax brass sourced directly by myself. This is what people like Alan Harris did, and Mark Wood does, for larger scales with cast iron. Cast iron makes less sense in S, as do many techniques and materials from 0 and upwards - adds support to my personal conviction that S is the "largest of the small", rather than the "smallest of the large".

 

Keep up the work: I suppose the ultimate question is whether or not you could make (including learning the techniques) the required number of different patterns in the time it takes to learn the necessary 3D CAD skills. Obviously, if those skills will be used elsewhere, then the "balance point" in the trade-off of "man versus machine" (or C vs. Si!) shifts.

 

 

Thank you, Simon, I appreciate that. I understand your preferences about sourcing your own castings. Different modellers have different methods and each is used according to taste. I don't know if you're aware of this but Shapeways now offer a wax master printing service. Details here. Either way and with the technology and services available to us, the future looks exciting.

 

CAD techniques are very transferable. What I'm learning now can be used to simplify some of the bodywork files I'm working on. The flare on the spokes, for example, is just a solid cylinder converted to an editable mesh then squeezed through one plane, given a curved taper and bent a little to accommodate the rim. This same technique can be used for, say the base of a funnel, to taper/curve it then wrap it round the smokebox. This is not to everyone's taste, I realise that and would not presume to suggest that this way is the only way. It just works for me. That's all that can be said for any method.

 

The thing I like about 3D Printing is that once the file is uploaded, it's there and can be used to make any number of the same item for any number of people. This is certainly not scratchbuilding, true,  but it is very convenient when talking about repeat items like buffers, axleboxes, etc. Personally, I like that idea. 

 

[Adrian]

Interesting news and progress, David.

 

I like the idea of printing in wax, but personally would prefer to receive just that so the waxes can be inspected and checked for any flaws or imperfections, and then arrange to have the lost wax brass sourced directly by myself. This is what people like Alan Harris did, and Mark Wood does, for larger scales with cast iron. 

 

That is an option Shapeways is currently testing.

 

http://www.shapeways.com/materials/wax

[Adrian]

It's why we have the parts we have in the SSMRS range: who wants to make 4 dozen identical buffer stocks by hand?

 

How the patterns are made is largely irrelevant.

 

I am intrigued by the steel: be interesting to see how that turns.

 

Yes the steel production is interesting. According to shapeways

 

Steel is printed by depositing a liquid binder onto a bed of steel powder one layer at a time. The product is then removed from the printer and infused with bronze. While the product is being transferred from the printer to the infusion chamber, it exists in a delicate "green state"

 

So you end up with approx 70% steel powder "glued" together with 30% bronze.

[Guest]

Yes the steel production is interesting. According to shapeways

 

Steel is printed by depositing a liquid binder onto a bed of steel powder one layer at a time. The product is then removed from the printer and infused with bronze. While the product is being transferred from the printer to the infusion chamber, it exists in a delicate "green state"

 

So you end up with approx 70% steel powder "glued" together with 30% bronze.

 

It is indeed, hence my comments about durability and machining. I think, though I'm happy to be proved wrong, that the best path would be a 3D Printed wheel centre and separate turned tyre. To that end, I'm looking at sourcing the tyres elsewhere; I already have a couple of places in mind. Once I get my hands on the lost wax wheel centre (Shapeways advise a shipping date of June 27th), I'll have a clearer idea of what to aim for. 

 

 

Edit to add: Shapeways have just informed me that it's done and on its way. This one I am looking forward to seeing.

Edited June 24, 2014 by Guest

[Guest]

A new post in the Shapeways forum today suggests that items printed in steel suffer a high failure rate because of the constructional method, something I did wonder about but ignored. So that option is out (wish he'd posted before I ordered...). A reply says the problem can be solved by 3D Printing by DMLS (Direct Metal Laser Sintering), which sounds impressive and I haven't the faintest idea how it works but Shapeways doesn't offer it anyway - yet! So I await the brass offering with even keener interest.

[Horsetan]

....I think, though I'm happy to be proved wrong, that the best path would be a 3D Printed wheel centre and separate turned tyre.....

Depending on what material you use, do be careful with this. Some types are more brittle than others, and you do need to keep an eye on the eventual fit on an axle.

[Guest]

Depending on what material you use, do be careful with this. Some types are more brittle than others, and you do need to keep an eye on the eventual fit on an axle.

 

Agreed. I think I may have found a way to eliminate the split axle element, though, (which I do not like) and that will help. This is going to be something of an extended experimental phase, I suspect, before anything truly viable results, but I don't think it's impossible. The technology will catch up with me, I'm sure.   

[Guest]

Okay, this is where I start to get impressed with this 3D Printing lark. This arrived this morning. Pictures taken with standard digital camera as close up as I dared.

 

 

 

The spokes are too thin, this I know and is down to the file supplied rather than the manufacturing process. For information, it's done in Shapeways' raw brass option - at least, that's what I asked for - the print being remarkably clean. The faint print lines that can be seen in these images (about three times actual size) will certainly not be seen at normal viewing distances, still less under a coat of paint.

 

Next step? Find a proper dimensioned diagram of an SECR P class wheel (for this is what this intermediate test purports to be) and supply an accurate file to Shapeways. The cost of this one print, be it noted, was about €11. Given the Shapeways pricing structure (basic charge plus cost of material actually used) and out of interest, I uploaded a single file containing six of these beasties. The cost worked out to be a shade under €22, which I consider to be pretty bløødy reasonable. 

 

I'm also looking at this process for making things like buffers. Suggestions for other parts welcome.

Edited June 27, 2014 by Guest

[flubrush]

David,

 

That looks very good.  Are you going to try machining it to see how it reacts?   I remember having problems some years ago with Gibson lost wax brass centres when the brass was too soft to provide a reliable push fit.

 

Jim.

[Horsetan]

Okay, this is where I start to get impressed with this 3D Printing lark. This arrived this morning. Pictures taken with standard digital camera as close up as I dared.

 

HPIM1911.JPG

 

HPIM1912.JPG

 

The spokes are too thin, this I know and is down to the file supplied rather than the manufacturing process. For information, it's done in Shapeways' raw brass option - at least, that's what I asked for - the print being remarkably clean. The faint print lines that can be seen in these images (about three times actual size) will certainly not be seen at normal viewing distances, still less under a coat of paint.

 

Next step? Find a proper dimensioned diagram of an SECR P class wheel (for this is what this intermediate test purports to be) and supply an accurate file to Shapeways. The cost of this one print, be it noted, was about €11. Given the Shapeways pricing structure (basic charge plus cost of material actually used) and out of interest, I uploaded a single file containing six of these beasties. The cost worked out to be a shade under €22, which I consider to be pretty bløødy reasonable. 

 

I'm also looking at this process for making things like buffers. Suggestions for other parts welcome.

 

Will you consider 4mm scale equivalents?

[Hawk]

 

 

I'm also looking at this process for making things like buffers. Suggestions for other parts welcome.

 

This is getting interesting! How does the dimensions on the actual brass wheel compare to the .stl file. Any shrinkage? 

[Phil Copleston]

David,

 

A suggestion: Rather than trying to print a driving wheel centre direct, wouldn't it be better to use this 'rapid prototyping' process as originally intended, and use it to produce a lost-wax master? This is a tried and tested process, and would then give you the opportunity to clean up the printing ridges, fill any blemishes and correct any errors, before sending the master for casting in brass. This is the route Maurice Hopper and I are taking to produce some Shanks 3ft dia. loco wheels.

 

Trying to skip straight to finished driving wheel centres using the 3-D printing process seems to put too much faith in an outcome over which you have little control.       

[Guest]

Gentlemen, apologies for the delay in replying. There was a major historical re-enactment event in Sønderborg yesterday, and my wife and I were at it all day (another one for you there, Ivan). 

 

David,

 

That looks very good.  Are you going to try machining it to see how it reacts?   I remember having problems some years ago with Gibson lost wax brass centres when the brass was too soft to provide a reliable push fit.

 

Jim.

 

Thank you, Jim. I appreciate that. Machining will be very much on the cards but not with this example. The spokes are too thin and I can just see them buckling with the first lathe tool being applied to the rim. I'm uploading a modified file soon and will take it from there.

 

Shapeways list the brass they use as being made up of 80% copper, 15% zinc and 5% tin. I've done a quick search to find the properties of brass with that combination but have turned up nothing. Is there a metallurgist amongst us who could offer comment? While I'm here, is push fit the only method available for fitting centre to tyre? I have heard of a sliding fit being used in conjunction with the tried and tested Araldite-cured-in-the-oven process. I believe split-axle fans use or used to use something similar on the axles they wished to insulate, and it seemed to offer no problems over breakage, etc.

 

Will you consider 4mm scale equivalents?

 

This had never occurred to me, Ivan. And frankly, I'm not sure it would be possible at the moment. Shapeways sets a minimum wall thickness for this material of 0.6mm, and the rim is already on that limit so a straight scale reduction can't happen. Having said that, honourable members will recall I said that I'd uploaded an old file. Looking at how thin the spokes turned out, I went back and checked the file to find that the spoke thickness turned out at about 0.4mm, so how I managed to get that one past the Shapeways (often strict) checking system is anyone's guess. 

 

Refinement and resolution are improving all the time, and I'm going to stick my neck out here and say that it will not be too long before 3D Printing this in 4mm is possible.

 

This is getting interesting! How does the dimensions on the actual brass wheel compare to the .stl file. Any shrinkage? 

 

Hi. Hawk. The answer is yes. I've just checked and the STL file gives a diameter of 15.88mm. Running a vernier caliper over the print, I find a diameter of 15.7mm. This can be easily compensated for with a slight re-scaling of the file.

 

David,

 

A suggestion: Rather than trying to print a driving wheel centre direct, wouldn't it be better to use this 'rapid prototyping' process as originally intended, and use it to produce a lost-wax master? This is a tried and tested process, and would then give you the opportunity to clean up the printing ridges, fill any blemishes and correct any errors, before sending the master for casting in brass. This is the route Maurice Hopper and I are taking to produce some Shanks 3ft dia. loco wheels.

 

Trying to skip straight to finished driving wheel centres using the 3-D printing process seems to put too much faith in an outcome over which you have little control.       

 

I agree, Phil, this is certainly another way. For my own part, the wheels are going to be primed, rubbed down and painted anyway so such few blemishes as there are will largely disappear. Or as I've heard it said "Paint covers a multitude of sins."

 

If this is a question of faith then yes, I do have faith in this system. As noted above, refinement and resolution are improving all the time and will continue to do so. I'm sure Shapeways would be the first to acknowledge that they're not using the highest state-of-the-art machines, even though the quality of the finished product is still pretty damn good. I look to the future with interest and some considerable measure of confidence.

[Guest]

I do value members' comments in this thread. I want to make it clear, though, that in pursuing this 3D Printing path, I am not seeking to replace scratchbuilding but rather examining an alternative. Not all of us have room for a workshop. I certainly don't, much as I would like one. I do, however, have a computer, the Internet and access to some rather excellent services. Each of us, I'm sure, will follow the path best suited to us.

 

I'd like to put this print I received on Friday into a little perspective - Phil, your valid comments about print lines put me onto this. I've beaten my digital camera over the head and it's obliged me with another couple of shots, the first now showing the wheel centre at about five times actual size.

 

 

The second shows it in my own fair hand (ignore the drainage works in the background).

 

 

As you can see, it's not big. And like I say, at normal viewing distances, all you see is crank and spokes. I'm okay with it. 

Edited June 29, 2014 by Guest

[flubrush]

Shapeways list the brass they use as being made up of 80% copper, 15% zinc and 5% tin. I've done a quick search to find the properties of brass with that combination but have turned up nothing. Is there a metallurgist amongst us who could offer comment? While I'm here, is push fit the only method available for fitting centre to tyre? I have heard of a sliding fit being used in conjunction with the tried and tested Araldite-cured-in-the-oven process. I believe split-axle fans use or used to use something similar on the axles they wished to insulate, and it seemed to offer no problems over breakage, etc.

 

David,

 

I've just checked and the brass I use is clock brass,  which is CZ120 and it has a 59% Copper and 38% Zinc makeup.  The other 2% is lead which is there to aid easy machining.  I get the feeling that the high copper content of the Shapeways brass might render it a bit soft and it might be a bit of a cow to machine - like copper is.    I think I remember that milk is a recommended cutting oil.

 

Simon has covered a lot of the points on push fitting.  I use the method since I want a good current path from metal wheel centre to axle for split axle pickup and it also means that I can machine the complete wheel on a mandrel before fitting.   Other methods that might be possible for wheel fitting is to use cyanoacrylate adhesive,  which has become common in the larger modelling scales.  However,  this glue does require a clearance between wheel seat and axle to work properly which may give problems when fitting.  It might be possible to bore a short length of the wheel seat to be a good fit on the axle and machine the rest of the seat to a larger diameter to give clearance for the adhesive so that the wheel can be concentric with the axle.

 

Another method used in larger scales is the two part axle with one part having a stub end which fits in a socket in the other part.  The axle parts are force fitted in the wheel centres and the two axle parts are locked in position by a taper pin.   I haven't researched whether the taper pin sizes available might work with the 1/8" axles used in S scale.   there might also be problems with this method in providing split axle insulation,  especially in the driven axle with the gear wheel.

 

Yet another method might be to use the method that the 2mm scale people use,  where the wheels have a short spigot or axle fitted and these are a force fit into an insulated muff.

 

Jim.

[Hawk]

Your brass spoked wheel is a solid proof of concept for using this process for driver centers.

I´m currently struggling to improve my CAD skills sufficient to model spoked wheels, and your postings are a great inspiration!

 

My plan is to make my centers slightly oversize, and then use a small sandblaster/air eraser to remove any stepping or regular patterns from the wax masters. Then all surfaces that need optimal precision will be machined on my manual lathe.

 

I have tried Shapeways´ brass process for a coupler pocket, and while impressed by the resolution, I am annoyed by the regular pattern:

 

 

To give an idea of the size of this part, the square holes in the pockets are 3mm X 3mm.

 

Looking forward to see more on your experimenting!

[Guest]

 

Yet another method might be to use the method that the 2mm scale people use,  where the wheels have a short spigot or axle fitted and these are a force fit into an insulated muff.

 

 

 

As you'd probably guessed, Jim, that was actually my first intention, hence the stub axle on the back of the wheel. But then the question of turning this stuff reared its ugly head and that idea had to be tossed out the window. Looking at the rest of your excellent post, I think what we're looking at here is the materials catching up with our needs, and I'm seriously considering contacting Shapeways to ask what their plans for the future are in this direction. The only thing that stops me is commercial sensitivity getting in the way of a worthwhile answer. Might be worth a try, though.

 

Your brass spoked wheel is a solid proof of concept for using this process for driver centers.

I´m currently struggling to improve my CAD skills sufficient to model spoked wheels, and your postings are a great inspiration!

 

 

Thank you for that, Hawk. I knew I was on the right track with this one. CAD has been something of an uphill battle for me, too. When I started out, I was limited to standard primitives (boxes, discs, tubes, etc.). This was one of the first things I modelled in CAD, a simple wagon:

 

 

As you can see, everything squared off, no taper to the buffers or tyres. And everything placed by dragging the mouse: it's not immediately obvious but the buffers are not an equal distance from the wagon centre line. Only when I got more into it did I discover a keyboard entry system that allowed me to create/place/angle things very accurately indeed. But I persevered. Late last year, I completed this:

 

 

No, I haven't lost my head: this is not for modelling but is rather guidance for an illustrator to work to for a series of children's books I'm working on. But the point is, it's much more sophisticated, more complex, more everything. It took time to get to that stage but I did it. The stuff I'm learning now - mesh editing and suchlike - is really only a variation on what has gone before. The learning curve is now almost a stroll up a gentle slope. I also find a lot of help on youtube or in 3D forums. For example, I was talking last week with my elder son and asking how I could turn a 2D element into 3D. He said simple, I need to create a loft object. He couldn't run through it with me there and then, he had a client to deal with, but we'd hook up later that evening via Teamviewer so he could show me what needed to be done. Well, now I knew what the process was called, it was a simple matter to search youtube and find out for myself. When that call took place later, I was able to tell him how it was done. He took me through some of the finer points that hadn't been covered in the video lesson and I was set up.

 

Half the battle, I find, is simply knowing what the terminology means. Spline editing, parametric deformers, boolean subtraction - one just has to find out what these things are, what they do. After that, it's pretty much plain sailing. I wish you well with your CAD. I find it frustrating as hell sometimes but I'm glad I didn't give up.

[Guest]

The plot thickens. The forum thread on Shapeways that I alluded to earlier has some further discussion and clarification. The thread was titled Harding (sic) 3d printed steel, and the failures appear to have occurred because this stuff is not only actually pretty hard but also somewhat brittle, and was probably being ask to do things it was never intended for. The title was therefore rather misleading, and the diagnosis of the problem and suggested remedy a little erroneous. Rather than go into further depth here, I direct honourable members to the thread itself:

 

http://www.shapeways.com/forum/index.php?t=msg&th=19722&start=0&

 

The long and the short of it is that the steel option is back on the table. I cannot think that anything we railway modellers would ask of it would push it into breaking. The only issue, as I see it, is that of machining. Tests will be undertaken forthwith.

[Guest]

As suggested earlier in this thread, I earlier today e-mailed Shapeways to ask about the possibility of a harder grade of brass being offered in the future. This is their reply, shared with full permission:

 

Adrian Rivera (Shapeways)

Jul 17 17:06

Hey David,

Thanks for writing in. I'd be happy to give some more info about this.

Currently we do not offer nor are we launching a brass material with an increased zinc content.

We do however offer a Castable Wax material this may be an option for your experiments that would allow you to create 3D printed models, but you'd be able to control the properties of the metal you'd want to cast it in.

You can read about our Castable Wax material here: https://www.shapeways.com/materials/wax?li=nav

You are more than welcome to share our answers, we're here to help and try to answer any questions you might have. Also please be sure to pass on your feedback to our Materials forum, just as I will pass it onto our Materials team.

Cheers,

Adrian
Customer Service Agent
www.shapeways.com

 

A helpful and informative answer. I see what they're getting at but going down this path removes one of the aims of this experiment, which is to make these driving wheels available to any who want them. And no, wax masters produced in this way can't be made generally available after uploading: if honourable members care to follow the link, they will find that wax models are supplied only to the model owner.

 

Let's see what the machining test on the steel version turns up.

[Phil Copleston]

David,

 

But this "Castable Wax material" that Shapeways are suggesting you use is what a number of us have advised all along - that you exploit the 3D-printing process as originally intended to make masters for lost-wax brass castings.

 

And of course these can be made available to all - except that you (or someone) has to buy them first and send them out to purchasers. T'was ever thus. You can't just dial-up and email precision things like loco wheels.

[Guest]

 

And of course these can be made available to all - except that you (or someone) has to buy them first and send them out to purchasers. 

 

- which defeats the object of having them available to any who want them. As it says on the Shapeways website, This is a Maker Material - Only available to the model owner., so it would have to be me that buys them (no one else can) and sends them out to interested parties. Given the stated fragility of this material -

 

Wax is very fragile, and should be handled very carefully. Even with our custom padded packaging, Wax products sometimes break in the mail after we ship them to you.

 

- I'm not sure two journeys through an international postal system is really such a good idea. As for "dialling up and e-mailing precision things like loco wheels", I've never suggested that the precision will come from the printing process alone. Why do you think I'm having machining tests done?