Experimental 4mm COT track

[Richardb]

I'll preface this and echo Martins sentiments, This is an experiment on a experiment. Don't try this at home unless you're happy with well err experimenting. Please don't base your entire layout on what I'm saying here or strain the Templot development team to pipeline optimum results. This is a bit of fun for me and I really want to see how this pans out. Plus, I'll say it again, thank you to the Templot folks for this awesome open source resource.

Additionally, I’m relying heavily on my printer’s capabilities, presets, and system to make this work. For context, I previously owned an Ender 3, and I genuinely believe I would have struggled to achieve a great 4mm print on that machine. I placed zero skill and experience in my result images at the bottom of this post. One day I'll build a Voron with multiple tool heads, for now I just want to print things.

With that said, let’s get the show on the road!

----

Last week, I mentioned my trial run with COT in 4mm. While it printed fine using the stock presets in Bambu Studio, I saw potential for improvement. My initial experiment used a 0.4mm nozzle with the slicer’s highest quality preset for that nozzle. I then hypothesised that using a 0.2mm hardened steel nozzle with the highest quality preset, 0.06mm layer high quality preset might yield even better results. After 5 hours and 30 minutes of printing, I was thrilled with the outcome. This is my very first set of hand-built points, and I’m really looking forward to starting the filing process. To build confidence in the setup, I’ll probably print some plain track next, just to prove everything works as expected.

Now, let’s recap why I’m experimenting instead of strictly adhering to the scale recommendations in the COT dialog window. It’s not because I disregard the hard work Martin and the team has put in, thats not what I want to convey. Rather, I’m intrigued by developments in tabletop 3D printing. My other hobby is tabletop gaming (40k, for those interested), and I follow several YouTubers who print with both resin and FDM. Traditionally, resin is used for miniatures, while FDM handles terrain, vehicles, and larger pieces. Recently, however, some hobbyists have shifted to FDM for miniatures due to the logistical challenges and health concerns associated with resin printing. For example, someone with a family member sensitised to resin, or a young couple living in a flat who can’t share space with a resin printer, might find FDM a more viable option. Others simply don’t want to invest in or learn a second system after heavily committing to FDM. Here’s a great video exploring this shift. FDM printing minis

I personally relate to all three scenarios. I have asthma, limited space, and a significant investment in FDM printing. Resin printing would require setting up a dedicated area in my damp, concrete shed, using PPE, and adopting strict procedures to avoid long-term health risks—not to mention learning an entirely new system. Additionally, if my commercially available printer (Bambu Lab X1C) is capable of producing the chairs with passable detail with a 0.4mm nozzle, why would I want to run two systems?

So, where does plug track fit into this, and why experiment with COT despite the guidance in Templot? Like many of you, once someone highlights the difference between finer standards and commercially available track, you can’t unsee it—let’s set that aside for now. I think many of us have taken that journey already. Now lets consider our track building options, since the introduction of plug track and COT, a few things have evolved: FDM printers have become far more capable and affordable, making high-quality results accessible to more people. Filaments, particularly enhanced PLA, have improved, as have methods for storing them. Slicers, too, have become significantly more powerful, with better presets.

So, why COT instead of plug track? COT offered the quickest way to print OO-BF without needing supports or a bespoke chair raft. It also requires less post-processing and, most importantly for me, no resin. I can print accurate track to my hearts content and try new things without blowing a hole in my modelling budget. And, what I'm really looking forward to is dabbling in other scales, again, whilst not blowing my entire budget all in the comfort of my living room

For my modelling purposes, it’s absolutely perfect. Here are my results:

 

[Hayfield]

Richard

In 7mm scale I am completely hooked on COT track, simply for the reasons you have stated

ow you have done what I thought was still sometime away. Logically I thought the quality of home printers and materials will develop in the future to enable 4mm COT track to be possible. but not so soon

The prints do look good. But can you thread rail through the chairs. I the gauge affected etc

Would be good to see a model with rail attached

Well done

John

 

[Martin Wynne]

a 0.2mm hardened steel nozzle with the highest quality preset, 0.06mm layer high quality preset might yield even better results. After 5 hours and 30 minutes of printing, I was thrilled with the outcome.

@Richardb @Hayfield

Hi Richard,

That's amazing. Far better than my experiments with a 0.2mm nozzle 5 years ago on a Marlin-based printer (BIBO) which were in effect unusable.

Just to add that if you are using 100% COT, i.e. no hybrid plug-in chairs, you don't need the thick timbers. You could reduce the timber thickness by half which would reduce the printing time significantly.

The default thickness for 4mm timbers is 3.36mm. If you reduce it to 1.6mm to match most 4mm flexi-track, the print time would likely be between 2 and 3 hours. You could reduce it even more to match SMP Scaleway flexi-track which is 3/64" = 1.2mm timbers.

p.s. for COT there are some still very experimental adjustments to the rail section in progress (including for 7mm), subsequent to recent feedback. I will write more about this soon.

cheers.

Martin.

 

[Richardb]

The prints do look good. But can you thread rail through the chairs. I the gauge affected etc

Would be good to see a model with rail attached

Well done

John

Cheers John, You are very welcome.

That's my intent I'm hoping to thread some rail through tomorrow, I'll let you know how I get on. I'll make short section of plain track to and if I can get to grips with bricks, I'll try and get a curve pushed out.

 

[Richardb]

The default thickness for 4mm timbers is 3.36mm. If you reduce it to 1.6mm to match most 4mm flexi-track, the print time would likely be between 2 and 3 hours. You could reduce it even more to match SMP Scaleway flexi-track which is 3/64" = 1.2mm timbers.

p.s. for COT there are some still very experimental adjustments to the rail section in progress (including for 7mm), subsequent to recent feedback. I will write more about this soon.

cheers.

Martin.

Thanks Martin,

I'm definitely on a he who dares journey and I'm absolutely leaning on the technology however, its a good essay into what a 3d printing newbie with some decent tools can get away with and the capabilities of Templot (albeit experimental). I'll bear those measurements in mind the next time I print some more track, which will hopefully be tomorrow.

 

[Martin Wynne]

That's my intent I'm hoping to thread some rail through tomorrow, I'll let you know how I get on. I'll make short section of plain track to and if I can get to grips with bricks, I'll try and get a curve pushed out.

@Richardb @Hayfield

Hi Richard,

There are some changes in the pipeline for the COT settings on the rail section dimensions. It's very likely that the rail won't fit too well at present in 4mm scale. Or 7mm scale. Try with this setting both ticked and unticked:

I will write again when I have had a chance to try a 0.2mm nozzle on the Neptune 4. Don't assume that the results you get now correspond to what will eventually be possible. As always you are all jumping ahead too fast, I can't keep up.

https://www.amazon.co.uk/gp/product/B0CPT2XD3V

p.s. what filament did you use?

cheers,

Martin.

 

[Richardb]

@Richardb @Hayfield

Hi Richard,

There are some changes in the pipeline for the COT settings on the rail section dimensions. It's very likely that the rail won't fit too well at present in 4mm scale. Or 7mm scale. Try with this setting both ticked and unticked:


View attachment 13111


I will write again when I have had a chance to try a 0.2mm nozzle on the Neptune 4. Don't assume that the results you get now correspond to what will eventually be possible. As always you are all jumping ahead too fast, I can't keep up.

https://www.amazon.co.uk/gp/product/B0CPT2XD3V

p.s. what filament did you use?

cheers,

Martin.

Understood Martin, I am greatful you allow people to play with the outputs.

Stroke of luck, I had those boxes ticked, printing some plain track right now but bed beckons. I should be able to get stock rails threaded tomorrow and the bit of plain track. Quickly grabbed some code 75 BH I was given by a MRC member a few years ago. Have to say it slides in quite well, I thought it would have been a bit trixkier. The 0.4mm nozzle piece at 0.16 layer line was pretty stiff, the finer 0.2 nozzle at 0.06 seems a doddle without being unacaptably loose that tp cause running issues, I guess proof is in the pudding. I'll report back tomorrow

I don't have any gauges unless there are stl's knocking around but I'll try and get some vernier measurements tomorrow.

I used eSun PLA+ I like to print with + because it is dimensionally stable. An accuracy of 0.03mm is good enough in my book. https://esun3dstore.uk/products/pla-pro

 

[Richardb]

There are some changes in the pipeline for the COT settings on the rail section dimensions. It's very likely that the rail won't fit too well at present in 4mm scale. Or 7mm scale. Try with this setting both ticked and unticked:

Quick update, the straight track was easy to thread with code 75 BH, I don't know who the manufacturer was- I was gifted 2 metres by an MRC member a few years a go and just got round to using it.

Template drawn with 5.56.a and used 3d Builder for mesh repair. This time I used the 0.2mm hardened steel nozzle again but used 0.08 layers and high quality pre-set in Bambu Studio- I didn't touch anything else. In Templot I reduced the timber height as suggested by Martin and ensured the 3-D for FDM boxes were all ticked (thanks Martin). Printing time was 2 hours, which is fine. Scale is the same, OO-BF for use with RTR stock. Edit: filament is e-Sun PLA+.

I'm pleased to report a Hornby 20t mineral wagon passes along it easily. Threading the rails was easy peasy with no need to clear the chairs with a pointy bit of rail, noticeably easier than a similar piece printed with the 0.4 nozzle- which makes sense, the lines are thicker.

The piece achieves a nominal 16.5mm measurement between rails along its length, measured using verniers. Plus, its square and the chairs and detail printed really well. EDIT: I should say I set the verniers up and then used them using a feeler gauge technique. i.e. I did not stick them in the track, expand to 16.5 thus a false reading. Verniers are manual and accurate to 0.02mm.

Next step is to catch up on Bricks and build an arrangement consisting of a curve, straight and point to test a loco and my limited track building ability. I'll hopefully report back in a week, we're putting the decs up and my mother is due down, so duty calls.

All experimental of course and I'm really interested to see if the consistency can be maintained and to find the limitations of 4mm COT.

Have a great Sunday.

 

[Martin Wynne]

@Richardb @Hayfield

Inspired by Richard's results, I'm having another go at 4mm COT track on the Neptune4. The last time I tried, using a 0.4mm nozzle and my usual timbering profile, the results were disappointing. The chairs were fragile and easily broken when attempting to thread rail through them.

So I have now ordered a 0.2mm nozzle, to try again. In the meantime I thought I would try the trick of fibbing to Cura that it's a 0.2mm nozzle without actually changing the physical 0.4mm nozzle. The physical nozzle diameter is surprisingly unimportant, it is the calculated 0.2mm flow rates and axis movements from the slicer which mostly determine the finished result.

But a difference of 200% in nozzle size is asking a bit much, so I wasn't expecting much and I was pleasantly surprised by the results.

This is EM, printed with 0.2mmm nozzle settings in Cura with a 0.4mm physical nozzle and 0.04mm layers. I also set a negative -0.1mm horizontal expansion to get accurate dimensions with the larger physical nozzle. I reduced the timber thickness to 1.4mm and the web thickness to 0.6mm. I also ran the Gcode through the Templot backlash correction utility.

Print time for 4 sleepers was 26 minutes with the usual Neptune4 speed settings:

Straight off the printer with just a quick flick with a toothbrush to remove a few stray strands. The rounded corners on the sleepers seem to be an optical illusion in the camera, not at all evident on the actual print. The resin is the same Sunlu PLA+ wood colour that I used for 7mm COT -- looking darker because no sun today for photography.

The rail was a tight fit in the chairs and required some force to push it through, but no chairs broke in the process and they seem much stronger than before. It's looking promising, but I need to adjust the rail fit before posting any settings. And also try the actual 0.2mm nozzle of course.

edit: Another print just off the printer with changed settings in the backlash correction. Looking better, with a better rail fit. For someone who really, really, doesn't want to change the nozzle on their Neptune4, this just might be a doable option.

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin,
Is it possible ( or indeed desirable) to iron the timber tops before printing the chairs on them?
Steve

 

[Martin Wynne]

Hi Martin,
Is it possible ( or indeed desirable) to iron the timber tops before printing the chairs on them?
Steve

@Steve_Cornford

Hi Steve,

If you switch on ironing for all exposed top surfaces, it will iron the tops of the timbers in between and around the chair positions before continuing to print the chairs above them. It will do the same unnecessarily for the timber webs, splints and connector clips, but you can't have one without the other.

If you switch on ironing for the uppermost top surface only, that's fine for plug track. But for COT track it would skip the timbers and iron just the tops of the keys!

Hopefully one day Cura (free) will have similar options to Simplify3D (definitely not free), where you can specify different settings for different heights within the model. It would be possible in Cura by merging several Gcode files, but a lot of faff to do that for every print.

But looking at the results without ironing for a 0.2mm nozzle, I'm not sure that ironing is really needed. Generally I have found it to be less important on the Neptune4 than on the Kingroon printers -- presumably the higher printing temperature allows the top surface to merge into a uniform layer more easily.

There is so much still to try and learn before we find the optimum solutions for 3D track.

cheers,

Martin.

 

[Richardb]

@Martin Wynne

Fantastic result with your print and I'm intrigued with your approach in using a 0.4mm nozzle set to 0.2mm in the slicer, thats actually a curious way of going about it.

I need to read up on backlash correction because I'm not sure what it is and why we need to correct it.

Intrigued to hear how you get on with the 0.2mm nozzle.

 

[Steve_Cornford]

Hi Richard,
How did the loose jaws turn out?
Steve

 

[Martin Wynne]

Fantastic result with your print and I'm intrigued with your approach in using a 0.4mm nozzle set to 0.2mm in the slicer, thats actually a curious way of going about it.

@Richardb

Hi Richard,

It's just a kludge method of getting improved detail when you don't want the faff of physically changing a nozzle, and then changing it back again. Generally the fine detail improves quite a bit, but dimensional accuracy isn't so good.

I need to read up on backlash correction because I'm not sure what it is and why we need to correct it.

I discovered the difference it can make on the children's "minibo" printer, but mainly for geared stepper motors. More info:

https://85a.uk/templot/club/index.php?posts/3228

Currently it is intended only for Gcode from Cura, it might not be wise to try it on files from other slicers. It is also important to get the settings right -- wrong settings is worse than no correction at all:

cheers,

Martin.

 

[Richardb]

Hi Richard,
How did the loose jaws turn out?
Steve

I personally think they are ok, I'll let you know at the end of the week how the build goes. Sorry for the quality, I'm using my S24.

 

[Martin Wynne]

@Hayfield @Richardb

Spare nozzles for Neptune 4 now arrived.

However, after much tinkering I have decided not to do a physical swap to a 0.2mm nozzle on my Neptune 4. I don't think it is something which I can recommend as part of Templot COT track. The Neptune 4 is clearly not intended for rapid nozzle changes -- there is too much faff involved, and too much risk of damage to fragile components and connections. If you want easy nozzle swaps it would be better to get a different printer, such as the Bambu Lab printers (which are far more expensive -- there is a reason the Neptune 4 can now be had for only £152).

If you do want to swap a Neptune 4 nozzle, here is the official Elegoo video. But if you have problems please ask on a 3D printing forum rather than here:

Neptune 4 Nozzle Change

Instead, for Templot 4mm COT track I shall concentrate on finding the optimum settings to use on a Neptune 4 printer using the standard 0.4mm physical nozzle.

The new spare nozzles can go into stock for maintenance and breakdown, if needed.

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin,
I had to perform a nozzle change on my Kingroon and found it reasonably straightforward, but is was just a 0.4mm for a 0.4mm. I sthere more involved than just the physical part swap.
Steve

 

[Paul Boyd]

@Hayfield @Richardb

Spare nozzles for Neptune 4 now arrived.

However, After much tinkering I have decided not to do a physical swap to a 0.2mm nozzle on my Neptune 4. I don't think it is something which I can recommend as part of Templot COT track. The Neptune 4 is clearly not intended for rapid nozzle changes -- there is too much faff involved, and too much risk of damage to fragile components and connections. If you want easy nozzle swaps it would be better to get a different printer, such as the Bambu Lab printers (which are far more expensive -- there is a reason the Neptune 4 can now be had for only £152).

If you do want to swap a Neptune 4 nozzle, here is the official Elegoo video. But if you have problems please ask on a 3D printing forum rather than here:

Neptune 4 Nozzle Change

Instead, for Templot 4mm COT track I shall concentrate on finding the optimum settings to use on a Neptune 4 printer using the standard 0.4mm physical nozzle.

The new spare nozzles can go into stock for maintenance and breakdown, if needed.

cheers,

Martin.

Hi Martin,

When I had a blocked nozzle on my Anycubic printer, it was surprisingly easy to just swap in the whole hot end that was provided as a spare. If that spare had a 0.2mm nozzle, the implications are obvious - just swap the hot ends!

Is it not possible to do something similar with the Neptune?

A real life analogy - I still play records but some are very old and benefit from a conical rather than elliptical stylus. Changing the stylus is a faff but swapping the whole headshell complete with stylus is easy!

Cheers,
Paul

 

[Richardb]

When I had a blocked nozzle on my Anycubic printer, it was surprisingly easy to just swap in the whole hot end that was provided as a spare. If that spare had a 0.2mm nozzle, the implications are obvious - just swap the hot ends!

This is what I did, I swapped the whole hot end assembly, 3 harnesses and 2 screws then updated 1 setting on the printer and that's it.

My ender 3 KE had a similar process to the Neptune which was another nail in the coffin to get rid of it.

Instead, for Templot 4mm COT track I shall concentrate on finding the optimum settings to use on a Neptune 4 printer using the standard 0.4mm physical nozzle.

I am really interested in your experiment, be nice to not need to interfere with the machine.

If you want easy nozzle swaps it would be better to get a different printer, such as the Bambu Lab printers (which are far more expensive -- there is a reason the Neptune 4 can now be had for only £152).

On the subject of Bambulab the A1 has an ultra simple nozzle swap. The A1 is coming in at approx £300 which is still expensive for a starter however, the pathway to printing successful prints is simpler. I think I am biased now, which probably isn't helpful, that ender 3 still makes my blood pressure go up just thinking about it haha!

 

[Martin Wynne]

Is there more involved than just the physical part swap.

@Steve_Cornford @Paul Boyd @Hayfield

Hi Steve,

No, but it involves removing 6 tiny screws with 2 different tools, and disconnecting some delicate thin wires in a tiny connector which is tricky to get at. Ideally you then need an adjustable spanner, but even in the official video they use a rusty pair of thin-nose pliers instead. And a 6mm socket/spanner to remove the actual nozzle. To be done with the nozzle hot after removing the filament. If you don't hold the heater block very firmly with the pliers, and in the right place, there is a risk of damaging the heater element or the thermistor connections. Then wait for it all to cool down and then put it all back together. See the official Elegoo video:

Neptune 4 Nozzle Change

All doable if necessary for maintenance, but not sensible just to swap nozzles to and fro between different jobs.

The Kingroon is easy -- just heat it up, hold the heater block with an adjustable spanner and unscrew the nozzle.

This young man manages to do it on the Neptune 4 without any dismantling, but still uses a (less rusty) pair of pliers because without dismantling there isn't room for a proper spanner:

Which is fine if the old nozzle is not too tight. In practice it very often is, and if the block slips suddenly out of the pliers there is a real risk of damaging the heater or connections.

All of which is fine if anyone wants to try it. But there is no way I'm going to recommend it as part of building Templot 3D track, and then be obliged to help folks who run into trouble in the process.

Originally I suggested that FDM COT track is only for 7mm, with resin plug-in chairs for 4mm. I'm no longer so sure, I think FDM 4mm COT track is reasonably doable if you are not too fussy about the fine chair detail. But I'm going to support it only for use with a standard 0.4mm nozzle using the Cura slicer. For any other options folks will have to sort out their own methods and settings, and I won't be able to help.

Unless or until I get a Bambu Lab printer, when I might have a rethink. But I have no plans for that at present.

cheers,

Martin.

 

[Hayfield]

I think doing major surgery on our printers goes against the Plug/COT track ethos of keeping it simple. Just look at how I and others failed to read Martins Vee filing jig instructions have ended up with misunderstandings. Simply we all come from different levels of understanding and perhaps misreading or forgetting one or more steps in the process

There will be those with a level of knowledge/practicality that others do not have, so perhaps they can upgrade their printers. But this must be seen as an advanced level that most should avoid.

Simply we must wait until printers with the ability to print 4mm COT track are available to all at a modest cost.

 

[Martin Wynne]

Simply we must wait until printers with the ability to print 4mm COT track are available to all at a modest cost.

@Hayfield

Hi John,

You can do it now on a Neptune 4 as it stands, no modifying needed.

It just needs a special profile which I will post a first experimental version of later, and an extra click in Cura which I will explain.

Whether FDM 4mm COT chairs are good enough is for users to decide. They are quite a long way behind 4mm resin chairs in terms of detail, but would probably pass muster when painted. Using the Sunlu wood-colour PLA+ they are as strong as 4mm moulded chairs, but not as indestructible as the 7mm COT chairs with the Sunlu. They need a sharp switch blade pushed through them to "condition" them for the final rail.

Watch this space.

cheers,

Martin.

 

[Hayfield]

Martin

Thanks for that information. you have described what I was erring to far clearer than my poor attempt. No doubt improvements to the quality of printing of both domestic printers and filaments will in time make detailed 4mm COT track available to all, but as you have said strength may be an issue

But as always great news and another step forward

John

 

[Paul Boyd]

I think doing major surgery on our printers goes against the Plug/COT track ethos of keeping it simple. Just look at how I and others failed to read Martins Vee filing jig instructions have ended up with misunderstandings. Simply we all come from different levels of understanding and perhaps misreading or forgetting one or more steps in the process

There will be those with a level of knowledge/practicality that others do not have, so perhaps they can upgrade their printers. But this must be seen as an advanced level that most should avoid.

Simply we must wait until printers with the ability to print 4mm COT track are available to all at a modest cost.

Hi John,

I take your point as far as swapping hot ends for COT track goes, but it’s certainly not major surgery and is something that at some point you will almost certainly need to do when your nozzle gets blocked.

Cheers,
Paul

 

[Richardb]

Hi all,

Using a Bambulab X1C with 0.2mm nozzle, 0.14 layer lines in Orca slicer using printer pre-sets, no manual messing with the slicer.

Quick update on the crossing vee jig for clamping Code 75 bullhead rail. I started with the suggested groove depth of 0.98 mm per half (combined depth: 1.96 mm), but it was too loose. Adjusting it to 0.9 mm per half (1.8 mm total) improved things, but the rail still wasn’t held securely enough.

I’ve now reduced the groove depth to 0.85 mm per half (1.7 mm total), and I’m optimistic this will provide the firm grip needed. I’ll share results after further testing. As always, remember that printer tolerances can vary, so fettling might be necessary to get the best fit from your device.

Cheers,

 

[Richardb]

Morning all,

Thanks to C&L, I’ve got my hands on some Code 75 rail and rail joiners! This is a brand-new process for me, so I’m fully expecting to make a few mistakes along the way. That said, I feel confident and empowered to approach it without just blundering through. Time to dive in and see where this takes me!

 

[rumball_d]

Morning all,

Thanks to C&L, I’ve got my hands on some Code 75 rail and rail joiners! This is a brand-new process for me, so I’m fully expecting to make a few mistakes along the way. That said, I feel confident and empowered to approach it without just blundering through. Time to dive in and see where this takes me!

Hi All,
It's interesting seeing the results shown on this topic as I'm also using 4mm COT rather than plug track due to not having suitable space for a resin printer and having trouble finding a commercial print house that would produce an chair rafts in resin.

I was pleasantly surprised by the quality of the initial trial pieces of 4mm COT and with a little tweaking to the print profile I'm going to produce track bricks this way.

The examples below are from a Bambu Labs X1-C printer and Anycubic PLA filament.

0.4mm nozzle/0.08HQ profile

0.2mm nozzle/0.06HQ profile

code 75 BH rail is a good fit to the chairs and the gauge is spot on.

 

[Martin Wynne]

and having trouble finding a commercial print house that would produce an chair rafts in resin.

@rumball_d @Terry Downes

Hi,

Welcome to Templot Club.

For resin chairs, see: https://kosmik.co.uk

cheers,

Martin.

 

[Martin Wynne]

Instead, for Templot 4mm COT track I shall concentrate on finding the optimum settings to use on a Neptune 4 printer using the standard 0.4mm physical nozzle.

After much experimenting with 4mm COT on the Neptune 4, I have been unable (so far) to create COT bases I'm happy with and of sufficient quality to be usable for the full range of complex track formations. A short length of plain track is all very well, but doesn't get you much further for building an entire layout. So for the time being 4mm COT is on the back burner and won't be in 556b.

Although I have already a house full of 3D printers, I shall look at getting one with an easy swap to a 0.2mm nozzle, such as the Bambu Lab printers, in the new year. If 4mm COT is doable, it needs to be properly supported in Templot, and the only way to do that is to get one and see for myself. A more expensive FDM printer would be offset by not needing a resin printer.

edit: see below: https://85a.uk/templot/club/index.php?posts/15936

cheers,

Martin.

 

[rumball_d]

@rumball_d @Terry Downes

Hi,

Welcome to Templot Club.

For resin chairs, see: https://kosmik.co.uk

cheers,

Martin.

Thanks Martin, for Templot, the welcome and a really useful link!

 

[Richardb]

Thanks Martin, for Templot, the welcome and a really useful link!

Hello, fantastic looking results by the way! I'm in the same position as you with an X1C too.

So for the time being 4mm COT is on the back burner and won't be in 556b

Thanks for looking into this Martin, it is appreciated. I have forked the Templot5 repo and if I get stuck I don't mind contributing. Tis' open source afterall.

To be fair, the results are fantastic and I'd personally consider 4mm COT an MVP.

If it doesnt work out I guess I need to break it to the wife I now model 7mm

 

[Martin Wynne]

To be fair, the results are fantastic and I'd personally consider 4mm COT an MVP.

@Richardb

Hi Richard,

You have rather lost me there -- I had to Google MVP = minimum viable product.

Given that Templot isn't a product -- it's just my hobby -- I'm not too clear what you are suggesting?

Or maybe it is a product now that it's open-source and not me alone?

Whatever, if anyone wants to make 4mm COT from Templot they are of course free to do that using the various existing settings and options. But 556b won't have any specific pre-sets, defaults or options supporting 4mm COT. Unless someone else adds them, that is.

When/if I get a suitable printer, I will be able to determine what those pre-sets and defaults need to be, and I will then be able to add them.

cheers,

Martin.

 

[Richardb]

Morning Martin,

You have rather lost me there -- I had to Google MVP = minimum viable product.

Apologies for the slight miscommunication earlier. From my limited understanding of web development, an MVP refers to the most basic version of a product that can be released to allow for iterative development. It's an agnostic term that applies to both open-source and commercial software. What I’m trying to say is that what you've created is absolutely brilliant. Aside from a few printer limitations, which is to be expected when printing ultra fine detail at this scale, it looks like it can handle generating 4mm track. I think you've done a tremendous job

I'll give it a go with a slightly larger project once I've taught myself track building, Christmas leave is around the corner which means its time to do some modelling

 

[Martin Wynne]

@Hayfield @Steve_Cornford

FDM 4mm COT track

This is straight off the printer with just a light scrub with a nail brush to remove a few wispy strings.

Neptune 4 with the standard 0.4mm nozzle. This is the "medium" timber thickness, matching current C&L and Peco flexi -- height to rail top is 3.94mm - 3.97mm. Time for the above test piece was 67 minutes. So a full "medium" turnout in maybe 2 hours.

Usual cruel close-up. It is no good pretending that the 4mm/ft chair quality from FDM is comparable with plug track resin chairs, or the 3D-printed Finetrax kits. But the individual chair screws are visible, you can can make out the ribs on the chair jaws, and the rail bolt on the slide chairs. When painted and viewed from normal distances I think they would pass muster. Note the 8-sided chairs.

In the last few weeks I have been all round the houses on the slicers, and ended up back with Cura -- I have a pile of test pieces inches deep. I assumed the slicers derived from Slic3r (Prusa Slicer, Bambu Studio, Orca, SuperSlicer and others), which have the "height range modifier" function (Cura doesn't) would give the best results (despite their infuriatingly unintelligible interface as soon as you try to create custom profiles). Bambu came closest and fastest, with some neat chairs, but the timbers were scruffy and no match for Cura no matter what I did. The ironing function in Bambu Studio seems to be utterly broken.

So it's back to Cura. The clue might be in the fact that Elegoo provide Cura with their Neptune printers, presumably for a reason.

Several "corrections" (i.e. kludges) are needed in Templot. The rail fit needs a significant clearance to allow for FDM tolerances. FDM printers don't like printing to a sharp apex, so the keys have been chopped down to a flat top level with the outer jaw. The jaws have been beefed up (as for 7mm COT) and for 4mm it was also necessary to beef up the chair screws to make them printable. Getting the inner jaws to clear 00/EM wheel flanges required the jaws to be chopped down, with the grip part removed from the web. I might try increasing the chair base edge thickness too -- some of the chair bases are very indistinct.

Now I must get back to 556b -- poor Steve is still waiting to see his work on the new dialogs released. But I dare not try any merging until I have finished all the current half-finished stuff with multiple changes scattered through a dozen units.

Time for lunch. A nice boiled egg.

Martin.

 

[Martin Wynne]

@Steve_Cornford @Hayfield

Hi Steve,

FDM 4mm chairs

There is now yet another option. FDM-printed plug-in chairs instead of resin (this was already available in 7mm) -- this is 4mm FDM:

It might be helpful to use a slightly different filament colour for the chairs. Use a wooden block on the rail (or a thin-bladed tool) to snap-fit them into the sockets. It needs a bit more "snap force" than the clip-fit resin chairs.

The chairs would be better done with a 0.2mm nozzle. For someone who doesn't want the faff of resin printing, an option might be to get a second Neptune 4 instead, and fit it with a 0.2mm nozzle. To be used only for the chairs, keeping the first one with 0.4mm nozzle for the timbering bases and general modelling.

And now that I've written that I shall have to get one and do it, so that I can provide a suitable 0.2mm Cura profile. It never ends.

These FDM plug-in chairs would also work with laser-cut timbers, of course.

You can also mix integral COT chairs with sockets if preferred. But that means threading rails from the end instead of dropping them in. So not so useful for complex formations.

Any chance of a 3D-Builder view of a resin raft of them?

Any reason for asking for 3D Builder? We normally use 3D-Tool for previews:

The rail seats need further work. But as they are mostly hidden below the rails, I'm leaving it at that for now.

Note that for REA there are only ever two 8-sided chairs at the vee nose. This means that if there is an AB chair, the BB/BC chair (with the wing rail flares) changes from 8-sided to 4-sided. It drove me nuts:

Whereas XN is always 8-sided, whether or not there is an XA chair.

There is more work to do here for the parallel-wing crossings, but one thing at a time.

cheers,

Martin.

 

[Martin Wynne]

After much experimenting with 4mm COT on the Neptune 4, I have been unable (so far) to create COT bases I'm happy with and of sufficient quality to be usable for the full range of complex track formations. A short length of plain track is all very well, but doesn't get you much further for building an entire layout. So for the time being 4mm COT is on the back burner and won't be in 556b.

After experimenting with dried filament and some further tweaks in Cura, I'm now satisfied that 4mm FDM COT track is doable using a Neptune 4 printer with a standard 0.4mm nozzle. Not perfect, but doable. Unlike 4mm plug track, this makes it possible to 3D-print 4mm track which is height-compatible with commercial flexi-track. It needs the glue-on stub jaws for bent wing rails, complex formations, etc. Which are more fiddly than the clip-in resin loose jaws, but just about doable. See:

https://85a.uk/templot/club/index.php?posts/15936

So it will be in 556b after all, and my previous post above is cancelled.

I think this is called a U-turn, although in my case W-turns are more usual.

Martin.

 

[Phil G]

I think this called a U-turn, although in my case W-turns are more usual.

I think I would call it perseverance
well done, the project has gone a full 360 degrees since you stated playing with FDM printers .

One question though, if somebody already has a resin printer and is able to "safely make resin chairs". that could be two big ifs. Would you be now recommending they stay with this approach? Or are you now more of a mind set that 4mm FDM chairs are close enough that there is no need to even bother with resin at all?

Fully understanding resin still has its place, be for 2mm and maybe 3 mm scale. Or simply people looking for more fidelity in there track work.
I guess it goes without saying more complex track formations, using chairs not yet implemented would or could be outside the range of the 4mm FDM printed process?
Or are you thinking maybe a 0.2 mm nozzle could open up even this possibility?
cheers
Phil,

 

[rumball_d]

More experiments with 4mm COT.

I'm finding it beneficial to use multiple materials in the print, PLA for the timbers as it seems to produce a better surface finish and PLA+ for the chairs for strength.

note that the above is COT, printed in one pass and that the web has been replaced with a two layer printed raft (defined in the slicer) giving a larger area for tacking to the underlay.

Here's the same technique applied to a turnout.

 

[Phil G]

Hi Martin,
re this comment from me

well done, the project has gone a full 360 degrees since you stated playing with FDM printers .

on re-reading this I realised it does sound very patronizing, That was in no way my intension, I was simply congratulating you in achieving your very first objective. When it came to the 3D printing of model trackwork.
There is no doubt its been a long and winding road, and maybe when you started out the hardware was simply not up to the job it is today. So it would not have worked in those days.
My other comments are genuine, and only really asking you, with hindsight what would you recommend today?
cheers
Phil,

 

[Martin Wynne]

One question though, if somebody already has a resin printer and is able to "safely make resin chairs". that could be two big ifs. Would you be now recommending they stay with this approach? Or are you now more of a mind set that 4mm FDM chairs are close enough that there is no need to even bother with resin at all?

@Phil G @Hayfield @James Walters

Hi Phil,

As usual, I move a few inches forward, and you want to jump yards ahead.

My hobby is tinkering with the computers and machines to see what is possible, what works, what doesn't work. For my own amusement and recreation. What other folks want do with it, if anything, is entirely up to them. I'm not here to make recommendations.

Even if I was, I would be nowhere near ready to make any. So far I have got a pile of 4mm FDM COT test pieces, and I've tried threading rail into a few of them. It's looking good, but I haven't even got as far as building a full turnout in 4mm COT yet.

---

There are three separate considerations, it's confusing if they get mixed up:

​

1. integral chairs on timbers (COT track) -- OR plug-in separate chairs.​

​

2. resin OR FDM printing for each of those,​

​

3. how separate jaws are handled -- clip-in loose jaws with pins (resin only) OR glue-on stub jaws (resin or FDM).​

1. For myself, I'm not much interested in using integral chairs on timbers (COT). Sliding rail into place from one end goes against the grain of my original motivation for 3D-printed track, which is to build track by dropping rails into place vertically.

For single turnouts, sliding rails is doable and matches commercial kits. But I'm not really interested in single turnouts, I want to build more complex bespoke formations and junctions, for which sliding rails into place is impossible or difficult. Plug-in chairs makes it much easier. It's possible to combine COT and plug-in chairs in the same timbering base, but anyone doing that knows what they are doing and won't be reading this.

But I know many modellers do just want single turnouts. Whether Templot is the best way to get them is a moot point. But to the extent that Templot makes it possible to produce them as FDM COT track, I want to provide the options and settings to get the best results.

Also in 4mm at least, COT track (in resin or FDM) makes it possible to match the height of commercial flexi-track. Again, that's of no great interest to me, but I know it's important for many modellers, and it's not possible with plug-in chairs in 4mm scale.


2. resin printing provides the best chair detail. No question -- and I can't see that changing any time soon, although FDM is getting closer. It is up to modellers how important that is to them. Resin-printed COT is iffy (see recent posts), and for most if you want COT it will have to be FDM.

For plug-in chairs the same quality difference applies. But if you don't want the faff of resin-printing and are happy with FDM quality you can produce FDM plug-in chairs using the COT chair settings. This can be for FDM timbering bricks or for laser-cut bases:

That could be a difficult decision in 7mm scale, where the quality difference between resin and FDM is not so obvious.


3. Whichever method you choose, some rails can be assembled only if the outer jaws are applied separately.

In resin, with full-depth timbers, those can be done using the clip-fit loose jaws with pins. It's possible to use them for every chair -- in which case assembly of complex formations is much easier, there is never any need to slide anything onto a rail, and the check and wing rails can have neat prototypical flat square ends.

In FDM, separate jaws need to be the glue-on stub jaws instead. Much more fiddly to handle, and you wouldn't want to use them for every chair. Which means the other chairs will have to be solid jaw slide-on, and the rail ends will need to be chamfered to make that possible.

---

Phil, I've written all that at least once before. I can't keep repeating myself. I switched this site off over Christmas so that I could have a break from writing stuff.

My priority is to get 556b released as soon as possible, and there is still a lot to do.

cheers,

Martin.