Wagon CAD generator

[Rabs]

I'm currently writing a parametric CAD model for 4-wheel GWR wagons in OpenSCAD. I'm using my sparkly new copy of "GWR goods wagons" by Atkins, Beard and Tourret as the main reference.  I'm mainly writing this for my own use to churn out some wagons on my 3D printer (http://www.rmweb.co.uk/community/index.php?/topic/66521-and-who-says-that-home-3d-printing-cant-do-detail/page-5) but I was wondering if this might be of interest to others.  This isn't an advert - I'm not selling anything but I'm just wondering if this is a tool others might be interested in.  However, I am curious if such a tool might be worth an informal exchange of beer tokens.

 

 

(example here is a partially complete chassis and underframe for an Open to O21, adjusted as necessary for a model in N (1/148).  This is an early version.  When finished it will include all the features below and all the rivets and bolts to hold the various bits together.)

 

The code takes a list of inputs and spits out a finished wagon 3D design.  I'm currently working on the chassis module and the list of parameters for this is below (to give you an idea).  Modules to build bodies for for Opens and Vans will come later.

 

Underframe chassis parameters:

• model scale
• minimum wall thickness (the software is designed for 3D printing)
• buffer height
• wheel diameter
• wheel gauge (model, in case the model isn't finescale)
• wheelbase
• length over headstocks
• chassis width
• headstock width
• headstock ends tapered/chamfered? (yes/no)
• solebar height (vertical height)
• solebar width
• solebar section thickness
• solebar type (options are "wooden", "channel" and "bulb")
• underframe panel length (for cross bracing and scantling)
• scantling design (options are "plain", "GWR reinforced", "RCH reinforced" and "post 1939") - not sure if I will bother implementing this as nobody sees the scantling on a model

• axlebox type (options include "OK F", "RCH" "RCH heavy", "GWR round", "grease")

• axleguard type (options are "crowned" (standard W-iron) or "2 part" (double-V type))

• spring length

• number of spring leaves

• spring mount type (options are "plain", "J-hanger", "suspended link" and "eye-bolt"

• brake type (options are "simple lever", "Thomas", "DC1", "DC2", "DC3" and "Morton", with vacuum fitted variants where appropriate)

• brake either side (yes/no)

• buffer type (options TBD - will include 'plain hole' for fitting buffers separately)

• coupling type (options TBD - will include prototype couplings and model couplings)

(as an aside - can anyone see anything I've missed in this list to fully define a 4 wheel chassis?)

[Tricky-CRS]

I think others will be interested, I certainly will be.

[Miss Prism]

This is a very useful development, Rabs.

The six coupling base plates (bible fig 23) would be a useful addition, as would a couple of styles of brake rod loops. Solebar holes (figs 360 and 417, although I can't read the actual dimension) could be useful. Triangular body-support gussets option? Door bangers option (for opens etc) - the style was similar, but the banger spacing varied. Most vehicles also had triangular gussets between the bottom of the solebar and the bottom of the headstock channel - the length of this gusset seems to have varied (fig 239 shows an untypically long-ish one). Some of these details you might understandably consider a bit too silly for 2mm scale.

Some dimensions/shapes you might find useful (done for the proposed revamp of the Coopercraft 4mm underframes) can be found here.

The headstock should be U-section, like the solebar.
 

Your brakeblocks look a bit rudimentary at the moment, but brakes is a whole new ballgame. Enough for now!

 

 

 

 

[Rabs]

Thank you, useful additions to the list.

Don't read too much into the image in the OP - it's very much a work in progress.  Quite a few of the components are just rough outlines so that I can check that the script repositions everything correctly if I change the parameters such as chassis width or wheelbase.  Once it's all working I'll add more detail.

For my personal use I'll be aiming at including all details visible in 2mm, but once the code is working adding more detail for 4mm is relatively simple (if time consuming).

[Wild Boar Fell]

Hello Rabs,

Looks a superb piece of work to me, certainly sounds like it would be a very useful tool, will it build the files as STL's?

Is OpenSCAD Free by the way?

Regards,

Wild Boar Fell

[Rabs]

Yes and yes:

http://www.openscad.org/

 

I was introduced to it by Etched Pixels (Alan), who I think uses it for some of his coach side designs.

[Etched Pixels]

Dumb buffers, brass bearing type and pinpoint axle length I guess would be the ones that come to mind. For couplings peco ones might be useful but may need a block on the headstocks so stick them to. Likewise a block or option for vacuum cylinders perhaps ?). Beyond that I suspect you get rapidly into excess complexity if you try and cover all the weirdness as you get into the longer chassis, or hoppers.

 

Really nice to see - this is precisely what openscad and parametric 3D modelling is supposed to be all about. I use it for the coach shells, although its fronted by some other tools that write the openscad file from a description file.

 

Certainly could be useful although wagons are not my main interest - and there is no shortage of coach stuff left to do yet !

 

Could be very useful though, either as software or if you were producing/selling chassis. Given how fragile FUD from Shapeways is I'm dubious how well it would work with their printing. Maybe in brass if the wallet can cope 8)

 

If you want to get the RCH underframe right and need any detail photos let me know. The Swansea museum has (at least I think it's still there) an Ocean colliery wagon which is tilted up on display at an angle so providing its not moved I can hopefully photograph any needed underside bits.

[Tricky-CRS]

I would have an option for a pad to stick NEM pockets too.

Wild has done that on mine for me and could tell you how he works it out.

 

[Rabs]

All good suggestions - thanks.  My thought is that the tool will be useful even if I have to exclude some of the finer parts from my prints for now.  Anyway, it's only a matter of time before  3D printing improves to the point where it can to the lot (and probably in full colour too!)

On the complexity and number of variants question - My plan is to include features that would be useful for more than one wagon.  If there's something that is only for a single design then I'll not include it in the parametric model and either add the extra parts manually in another CAD program or in the real world.

[RJS1977]

I can see it would also be a useful tool for the TrainSim community, if the files can be converted to the correct formats.

[Etched Pixels]

I would have an option for a pad to stick NEM pockets too. Wild has done that on mine for me and could tell you how he works it out.

 

You can just print NEM pockets into them although for a lot of very short wagons there isn't clearance between the NEM pocket and the axle in N at least. Its really tight getting a Peco coupling into an 1880s wagon !

[Wild Boar Fell]

You can just print NEM pockets into them although for a lot of very short wagons there isn't clearance between the NEM pocket and the axle in N at least. Its really tight getting a Peco coupling into an 1880s wagon !

EP,

Its not just N gauge vehicles that can suffer this problem, the bogies on my crane require the couplings to be slid further forward than I wanted by a mm or two as despite some calculations the NEM tails just fouled the first axle.

Regards,

Wild Boar Fell

[Rabs]

Just for interest, I made my first print using the output of the generator earlier today.  Pictures are on page 7 of this thread: http://www.rmweb.co.uk/community/index.php?/topic/66521-and-who-says-that-home-3d-printing-cant-do-detail/

[Dazzler Fan]

I like that a resin printer is now available for domestic use - ideal for smaller

scales.

We started with NEM pockets, but are now gradually putting them on sprues.
The costs of out sourcing are such that the market is particular with their own

couplings preferences.

 

While the demand is for N scale, it is not a likely proposition for us..

[chris newman]

Just for interest, I made my first print using the output of the generator earlier today.  Pictures are on page 7 of this thread: http://www.rmweb.co.uk/community/index.php?/topic/66521-and-who-says-that-home-3d-printing-cant-do-detail/

Hi Rabs,

              I've been following your progress on the 3D thread and with this as well!! it's getting me in a "lather" over What is possible in N, Do you think it would be possible (when you proved the viability of this) to do coaches--Pullmans--8 & 12 wheelers in their Varieties?? Also a possibility for Loco And DMU cabs!!,MTIA

regds

Chris

[Rabs]

Hi Chris,  I'm pretty sure that any and all of those would be possible.  They aren't my personal area of interest but I'm sure that they can be done.

[Etched Pixels]

Lots of folk already doing coaches partly in 3D print via the commercial services. Right now its (even in N) a bit pricy doing whole coaches in the finer commercial stuff (Shapeways FUD etc) as well as having inadequate surface quality but it works well for smooth sided modern vehicles (polished materials) and combined with etches or vinyl or similar for bogie vehicles.

 

I'd agree it would be a good way to tackle some of the "no two coaches the same" pullman stuff.

 

Alan

[drduncan]

Just come across this thread and I'm interested in learning more....

 

Drduncan

[Rabs]

I didn't realise that I'd left this on the side for so long.  2 years, wow!  Unfortunately life has awkwardly interjected itself into my modelling time and I've not done any more on this since my last post.  I was thinking just the other day though that it was time to get back on with it.