FDM-printed chairs and baseplates

[AndyB]

At 00 scale FDM printed track and turnouts are definitely pushing it a bit although I believe it can be acceptable if we are willing to accept some compromises. At larger scales there is less need to compromise.

This is a D crossing chair for an 0 scale 7.5 turnout printed in PLA+. It's quite sturdy and the "stratification" is almost invisible without magnification (as below).

And the rendering from TurboCAD

STL attached if anyone cares to take a shot at it either in FDM or resin. The CAD drawing is full-size and should scale to anything.

Andy

 

[Martin Wynne]

Hi Martin,

Here you go!

My printer driver complains that the model is not manifold. It did repair it but there seems to be something not quite right towards one end. This was printed using the same settings I used for my test piece. It took about around 15 minutes to print. Some of that could be eliminated if the slicer would stop being too clever.

I managed to insert SMP rail into one side but the other side didn't do so well. It might be necessary to beef-up the inner jaw a bit. The rail is also a bit wobbly but that might be because SMP rail is underscale.

View attachment 5975

Here it is alongside mine. (I should rotate the fill orientation at the printer to make the grain run lengthwise on the sleepers.)

View attachment 5976

Cheers!
Andy

@AndyB

Hi Andy,

Thanks for that. Your FDM results in 4mm/ft with the Templot chairs look similar to mine. From which I concluded that FDM-printing wasn't feasible in 4mm/ft for the chairing. The results from resin-printed chairs are an order of magnitude better, so I have continued with that, and not given much further thought to FDM for the chairs.

Some settings for SMP rail are already available on the export dialog:

But I'm still waiting for someone to tell me how well they work.

The Templot chairs do not contact the rail head or foot on the outside. The rail is held by means of the key against the rail web in the prototype manner. The results from this have turned out so much better than I expected that we have now progressed to the loose-jaws system which makes pointwork assembly so much easier and more enjoyable. No sore fingers!

But I can understand the preference for FDM printing for home use, and your chunkier FDM chairs do look doable, in 7mm scale at least. So I can see a whole fresh strand of Templot chairing options opening up in front of me. I don't know whether to welcome that or give up in despair of ever getting anything finished.

It can't be called plug track because there is no practical way of FDM printing chairs and timbers separately, at least in 4mm scale. It would mean a one-piece timbering base similar to the Finetrax kits. Which means threading rails horizontally, which in turn means breaking the bent wing rails into 2 parts with a join at the knuckle.

Thanks for the files. My TurboCAD won't open your TCW file with the message "no input filter available for this file". I seem to remember we have been here before. But your DXF opens fine, and I can see it is made up of a great many blocks, similar to the Templot DXFs.

I'm not sure what should happen next in all this. I think it is over to you or someone else to progress this FDM strand. Meanwhile...

Write out 100 times:

I will not get distracted by the 3D printers until I have finished coding the crossing chairs for plug track.
I will not get distracted by the 3D printers until I have finished coding the crossing chairs for plug track.
I will not ...

cheers,

Martin.

 

[AndyB]

@AndyB

Hi Andy,

Thanks for that. Your FDM results in 4mm/ft with the Templot chairs look similar to mine. From which I concluded that FDM-printing wasn't feasible in 4mm/ft for the chairing. The results from resin-printed chairs are an order of magnitude better, so I have continued with that, and not given much further thought to FDM for the chairs.

Some settings for SMP rail are already available on the export dialog:

View attachment 5985

But I'm still waiting for someone to tell me how well they work.

The Templot chairs do not contact the rail head or foot on the outside. The rail is held by means of the key against the rail web in the prototype manner. The results from this have turned out so much better than I expected that we have now progressed to the loose-jaws system which makes pointwork assembly so much easier and more enjoyable. No sore fingers!

But I can understand the preference for FDM printing for home use, and your chunkier FDM chairs do look doable, in 7mm scale at least. So I can see a whole fresh strand of Templot chairing options opening up in front of me. I don't know whether to welcome that or give up in despair of ever getting anything finished.

It can't be called plug track because there is no practical way of FDM printing chairs and timbers separately, at least in 4mm scale. It would mean a one-piece timbering base similar to the Finetrax kits. Which means threading rails horizontally, which in turn means breaking the bent wing rails into 2 parts with a join at the knuckle.

Thanks for the files. My TurboCAD won't open your TCW file with the message "no input filter available for this file". I seem to remember we have been here before. But your DXF opens fine, and I can see it is made up of a great many blocks, similar to the Templot DXFs.

I'm not sure what should happen next in all this. I think it is over to you or someone else to progress this FDM strand. Meanwhile...

Write out 100 times:

I will not get distracted by the 3D printers until I have finished coding the crossing chairs for plug track.
I will not get distracted by the 3D printers until I have finished coding the crossing chairs for plug track.
I will not ...

cheers,

Martin.

Hi Martin,

Sorry for any possible distraction but you might not want to abandon FDM just quite yet. Tonight I printed that crossing chair at 1:81 and it looks very good. All I did was scaled it down. The proportions remained the same.

Despite the benefit of keys, particularly at crossing, I had ruled them out in the past because PLA is too feeble but the print in PLA+ does seem strong enough, possibly even stronger than resin.

I'll do some more experiments to see if it's really practical.

Cheers!
Andy

 

[Stephen Freeman]

This looks interesting, I have an FDM printer, still not used in anger, I really must find time to finish setting it up.

 

[AndyB]

My slicer program (Slic3r) adds some elements in the fill layers that I'd rather not have so I downloaded the latest Prusa slicer to see how it behaved. The prints were horrible. I couldn't get anything like the results I was getting with Slic3r.

Just to be on the safe side I tried another print with the previous Gcode from Slic3r that had worked so well. The print was horrible!

I powered down the printer and tried the Slic3r Gcode again. Perfect print this time.

It looks like the Prusa code is setting a value in the printer that overrides the Gcode from Slic3r and as well as the Gcode from Prusa itself. I think it has to do with the amount of filament being extruded because the bad prints were overloaded with filament. It probably also explains why the Prusa prints were so bad. I suspect it's something in the "custom Gcode"

I'd like to be able to use Prusa because it is supported whereas Slic3r isn't. More investigation required.

 

[AndyB]

My slicer program (Slic3r) adds some elements in the fill layers that I'd rather not have so I downloaded the latest Prusa slicer to see how it behaved. The prints were horrible. I couldn't get anything like the results I was getting with Slic3r.

Just to be on the safe side I tried another print with the previous Gcode from Slic3r that had worked so well. The print was horrible!

I powered down the printer and tried the Slic3r Gcode again. Perfect print this time.

It looks like the Prusa code is setting a value in the printer that overrides the Gcode from Slic3r and as well as the Gcode from Prusa itself. I think it has to do with the amount of filament being extruded because the bad prints were overloaded with filament. It probably also explains why the Prusa prints were so bad. I suspect it's something in the "custom Gcode"

I'd like to be able to use Prusa because it is supported whereas Slic3r isn't. More investigation required.

Problem solved. It looks like the Prusa slicer thought it was driving a Prusa printer . I reran the installation and the G-code it generates is much simpler now. The Slic3r print is still a bit better but that's probably just a matter of fine-tuning the Prusa slicer to get the same results.

Like Martin I really need to focus now on creating a full set of crossing chairs.

 

[Martin Wynne]

My printer driver complains that the model is not manifold. It did repair it but there seems to be something not quite right towards one end.

@AndyB

Sorry Andy, I forgot to run that STL file through the mesh-fixer. I have now replaced the attached file yet again. 3rd time lucky?

Martin.

 

[AndyB]

@AndyB

Sorry Andy, I forgot to run that STL file through the mesh-fixer. I have now replaced the attached file yet again. 3rd time lucky?

Martin.

Thanks Martin.

The STL is ok now. I printed it and the chairs on one side are working well but the ones on the other side are falling apart the same as with the previous print. I suspect they are not quite the same for some reason. I can see that the periphery of the chair layers are not consistent after slicing. Some of them do not follow the correct profile.

The overall dimensions of your chairs are not all that different from mine so I'm pretty sure they can be printed on an FDM printer. Can you post it, or preferably just a chair, as a Turbocad or DXF file? (I'm on TC2020). I'd like to rebuild the assembly in Turbocad and see it that makes a difference.

Thanks,
Andy

 

[Martin Wynne]

The overall dimensions of your chairs are not all that different from mine so I'm pretty sure they can be printed on an FDM printer. Can you post it, or preferably just a chair, as a Turbocad or DXF file? (I'm on TC2020). I'd like to rebuild the assembly in Turbocad and see it that makes a difference.

@AndyB

Hi Andy,

Here you go. Although you could easily generate it yourself in Templot.

It's in 00-SF. I changed the rail-fit to SMP, but I have had no feedback about how well the SMP dimensions in Templot actually work. For anyone else printing this, please note that C&L rail won't fit. Any custom rail section can be set in Templot.

Sleepers 1.2mm thick. Any thickness can be set in Templot. Splints 0.6mmx1.2mm. Any size or position can be set in Templot.

If you just want the chairs, you can easily delete the sleepers and splints in TurboCAD. They are in separate layers.

DXF from Templot.
DXF from TurboCAD 2015
TCW from TurboCAD 2015

If converted to STL, before printing it will need fixing via the online service at: https://www.formware.co/onlinestlrepair

cheers,

Martin.

 

[AndyB]

@AndyB

Hi Andy,

Here you go. Although you could easily generate it yourself in Templot.

View attachment 6006

It's in 00-SF. I changed the rail-fit to SMP, but I have had no feedback about how well the SMP dimensions in Templot actually work. For anyone else printing this, please note that C&L rail won't fit. Any custom rail section can be set in Templot.

Sleepers 1.2mm thick. Any thickness can be set in Templot. Splints 0.6mmx1.2mm. Any size or position can be set in Templot.

If you just want the chairs, you can easily delete the sleepers and splints in TurboCAD. They are in separate layers.

DXF from Templot.
DXF from TurboCAD 2015
TCW from TurboCAD 2015

If converted to STL, before printing it will need fixing via the online service at: https://www.formware.co/onlinestlrepair

cheers,

Martin.

Thanks Martin.

I thought I'd be able to repair the model by adding all the bits to make a single ACIS Solid but TurboCAD won't let me do that.

This is the problem I'm seeing and I think it could account for why some of the chairs are not holding up. At the 1.5mm slice the peripheries of the chairs are inconsistent.

Cheers,
Andy

 

[Martin Wynne]

This is the problem I'm seeing and I think it could account for why some of the chairs are not holding up. At the 1.5mm slice the peripheries of the chairs are inconsistent.

@AndyB

Hi Andy,

I'm not too clear what you are trying to do in TurboCAD, but I'm seeing the same strange variations in the Cura slicer, on the same chairs:

Which is a bit of a mystery, because the inserted blocks in the DXF are identical. Apart from the random key positions, but this is happening below the keys.

It seems to be happening in the mesh fixer. Here is the original exported unfixed STL from Templot. This is the cross-section function in the 3D-Tool preview:

And after fixing, some of the chairs gain an internal hole. I can only think that this must be due to rounding effects in the dimensions. I will look to add an integrity overlap between the chair jaw and the rail seat:

So I changed the settings in the Cura slicer to fill internal holes, which seems to have solved the problem:

Do you have that setting in your slicer?

This is the cross-section of the STL re-exported from Cura after filling the holes:

I don't know what to make of all this. I can't do a test FDM print because currently I don't have a machine set up with a 0.2mm nozzle.

In any event, I'm not minded at present to get distracted from the plug track, which requires separate resin-printed chairs and there doesn't seem to be any such issue with the slicing for those.

cheers,

Martin.

 

[AndyB]

Thanks Martin.

I have downloaded Cura and it does seem to resolve that problem but it gets completely confused and fills in the rectangle between the timbers

It doesn't have that problem with my STL. The difference is that my model is a single entity. All the bits are added together. I thought I had managed to do that with your model but TurboCAD sort of choked on it due to the file size. I give up!

I think the only way I can get it to work is if I reconstruct your chair from scratch in TurboCAD which does seem a bit pointless.

Cheers!
Andy

 

[Martin Wynne]

I have downloaded Cura and it does seem to resolve that problem but it gets completely confused and fills in the rectangle between the timbers

@AndyB

Hi Andy,

That's the result of the hole fill setting.

I'm struggling to see why the fixer inserts a hole at random. There just isn't anything there in the exported STL that I can see. All the chairs are identical:

Still looking.

cheers,

Martin.

 

[AndyB]

Hi Martin,

Yes it was the hole fixer. I thought I had tried with it turned off but I was getting a bit tired.

Here's another clue. This is from the latest Prusa slicer. The top of the rail seat is being treated as a top layer in some chairs but in others it's still an infill layer.

That's three different slicers

Cheers,
Andy

 

[Martin Wynne]

@AndyB

Thanks Andy.

Is that before or after using the mesh repair/fixer?

cheers,

Martin.

 

[AndyB]

This one - FTFT5_sleeper_test_00sf_no_sockets_fixed.stl

but I should double check to make sure Prusa didn't fix it too.

 

[AndyB]

I've attached it here.

 

[AndyB]

I realize all of this may not be relevant to resin printing but It's probably worth getting to the bottom of it in case it shows up there too.

Meanwhile I've made a very basic chair model by extruding a profile of your chair's jaws and key. I'll plonk it on some sleepers and make a print. I'm mainly interested to see if it's possible to print chairs from Templot in FDM that are robust enough.

 

[Martin Wynne]

@AndyB

Hi Andy,

I agree that it needs a solution for the resin printing too.

I have changed the order of inserting DXF faces into the file for the rail seat. This shouldn't make the slightest difference, but it seems to have solved the problem -- the fixer seems to be happier, no holes in the cross-section for any of the chairs I have looked at so far:

No idea why previously most of the chairs were as above, but some gained holes after fixing.

I will post a fresh file in a few minutes.

cheers,

Martin.

 

[Martin Wynne]

@AndyB

Hi Andy,

Here is a modified file. Mesh already fixed so your slicer shouldn't need to fix it again. The only change is in the order of inserting the seat faces into the DXF file. Why that should make any difference I don't know. I can't find anything wrong with it and Cura has sliced all the chairs identically (for 0.4mm nozzle):

It's for 16.2mm gauge and SMP rail. Hope it prints ok.

cheers,

Martin.

 

[AndyB]

YES! That fixed it. Gauge is right on the money at 16.2

The grip on the rail is might be a little tight for FDM but the jaws are holding up OK. To be honest, after a bit of paint I'd be surprised if anyone would know the difference between them and resin without a loupe.

Thanks for sorting it out,

Cheers!
Andy

 

[Martin Wynne]

@AndyB

Hi Andy,

That's great! The rail-fit can be easily adjusted in Templot if necessary.

But we need to know WHY that change fixed it. All the chairs were identical DXF blocks, so clearly it was hitting some edge condition in the mesh fixer for some chairs to be ok and some not. If we don't know what that edge condition is, we could easily fall foul of it again.

To be honest, after a bit of paint I'd be surprised if anyone would know the difference between them and resin without a loupe.

The obvious next question is where can folks buy an FDM printer off the shelf which will do this? You said that you made several modifications to your own printer. Do you think it will be possible to FDM-print separate chairs? If not, vertically assembled plug track is not possible. Even if it is possible I very much doubt that separate loose jaws and slots would be. In which case I have to re-think my ideas for the crossing chairs and restore the option to split the chairs into separate half-chairs for plugging.

Alternatively, forget vertical plug track assembly entirely and replicate the Finetrax kits in FDM for horizontal rail threading in a one-piece timbering base. A lot of folks might be happy with that for single turnouts. But it would have to be a separate strand of development from plug track. I can do the Templot code for that easily enough, enjoy it too. But the repeated EXPLAINING of yet another option on various web sites and forums is a nightmare and I don't think I'm up to it.

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin,
Would it be worth reporting this problem the the mesh fixer authors by supplying them with the two unified files, the one that produced random holes , and the one that worked ok?
Steve

 

[Martin Wynne]

@Steve_Cornford @AndyB

Hi Steve,

I had thought of doing that. I have a paid-for copy of the Formware slicer which presumably includes some user support. They will want to know where the files came from because they have never seen anything like it in 100 years! It will take quite a bit of explaining about bullhead track chairs and DXF files created using a 1980s version of DXF. I can only do so much explaining, explaining, typing, typing, day after day. And here I am doing it again.

I think I will first do a bit more delving into the Templot-produced code. The chairs are not absolutely identical (which Formware would no doubt notice) -- the rotations are random on 3 separate screw-heads and the keys are offset by random amounts. The latter might affect the auto-support calculations which might feed back into the slicing/repairing functions. Also I notice that there is a Formware program update available which I must first install and test for the same results. The golden rule is that you don't ask for help unless you are using the latest version of any software.

Andy, did your slicer produce the same random holes? Or was it only on files fixed using the Formware mesh fixer?

cheers,

Martin.

 

[AndyB]

@AndyB

Hi Andy,

That's great! The rail-fit can be easily adjusted in Templot if necessary.

But we need to know WHY that change fixed it. All the chairs were identical DXF blocks, so clearly it was hitting some edge condition in the mesh fixer for some chairs to be ok and some not. If we don't know what that edge condition is, we could easily fall foul of it again.



The obvious next question is where can folks buy an FDM printer off the shelf which will do this? You said that you made several modifications to your own printer. Do you think it will be possible to FDM-print separate chairs? If not, vertically assembled plug track is not possible. Even if it is possible I very much doubt that separate loose jaws and slots would be. In which case I have to re-think my ideas for the crossing chairs and restore the option to split the chairs into separate half-chairs for plugging.

Alternatively, forget vertical plug track assembly entirely and replicate the Finetrax kits in FDM for horizontal rail threading in a one-piece timbering base. A lot of folks might be happy with that for single turnouts. But it would have to be a separate strand of development from plug track. I can do the Templot code for that easily enough, enjoy it too. But the repeated EXPLAINING of yet another option on various web sites and forums is a nightmare and I don't think I'm up to it.

cheers,

Martin.

Hi Martin,

I don't think you should change direction at all. Plug track will definitely be superior. If anyone wants to see if they can "get away with" FDM prints I'm happy to help them. The printer mods I made (if they are really necessary) should not be beyond most modellers, particularly those that want to construct their own turnouts.

If your chair models can be printed without the plug (maybe they can already?) it should not be too difficult to add them to timbers in CAD, should anyone be keen enough to go in that direction.

Now get back to those crossing chairs

Cheers!
Andy

 

[AndyB]

Andy, did your slicer produce the same random holes? Or was it only on files fixed using the Formware mesh fixer?

Can't remember I'll go back to the original and try it again.

 

[Martin Wynne]

Hi Martin,

I don't think you should change direction at all. Plug track will definitely be superior. If anyone wants to see if they can "get away with" FDM prints I'm happy to help them. The printer mods I made (if they are really necessary) should not be beyond most modellers, particularly those that want to construct their own turnouts.

If your chair models can be printed without the plug (maybe they can already?) it should not be too difficult to add them to timbers in CAD, should anyone be keen enough to go in that direction.

Now get back to those crossing chairs

Cheers!
Andy

@AndyB

Hi Andy,

Sure the chairs can be exported without the plug -- just untick the box on the export dialog. But for printing they are going to need some form of support.

I wasn't thinking of changing direction. Just adding yet another option to the Templot 3D saga. The thing is -- resin printers, like laser-cutters, are messy and not for everyone. At the very least they require their own workspace away from living areas.

Whereas an FDM printer using PLA is very user-friendly and can chunter away happily in the corner of a room or on the kitchen table. For some that makes it a much more attractive option, even if the end quality isn't quite as good. But I suspect that most such users would want an FDM printer which can work straight out of the box. Changing the nozzle to 0.2mm (and back to 0.4mm for most other FDM projects) is about as far as they would want to go. I'm hopeful that my little Kingroon printer might be the answer as a sub-£200 entry to precision FDM printing, but I'm not going to start fiddling with it to print FDM chairs until I've resin-printed some crossing chairs!

More info:

https://85a.uk/templot/club/index.php?threads/3d-fdm-printers.277/post-6398

https://85a.uk/templot/club/index.php?threads/3d-fdm-printers.277/post-6378

cheers,

Martin.

 

[AndyB]

Can't remember I'll go back to the original and try it again.

I think that would be the version you uploaded on the 16th. Slic3r chokes on it so I just tried it with the Prusa slicer. The upper chairs don't seem to be correct but they are all consistent.

 

[AndyB]

@AndyB

Hi Andy,

Sure the chairs can be exported without the plug -- just untick the box on the export dialog. But for printing they are going to need some form of support.

I wasn't thinking of changing direction. Just adding yet another option to the Templot 3D saga. The thing is -- resin printers, like laser-cutters, are messy and not for everyone. At the very least they require their own workspace away from living areas.

Whereas an FDM printer using PLA is very user-friendly and can chunter away happily in the corner of a room or on the kitchen table. For some that makes it a much more attractive option, even if the end quality isn't quite as good. But I suspect that most such users would want an FDM printer which can work straight out of the box. Changing the nozzle to 0.2mm (and back to 0.4mm for most other FDM projects) is about as far as they would want to go. I'm hopeful that my little Kingroon printer might be the answer as a sub-£200 entry to precision FDM printing, but I'm not going to start fiddling with it to print FDM chairs until I've resin-printed some crossing chairs!

More info:

https://85a.uk/templot/club/index.php?threads/3d-fdm-printers.277/post-6398

https://85a.uk/templot/club/index.php?threads/3d-fdm-printers.277/post-6378

cheers,

Martin.

Hi Martin,

Looks like the X/Y resolution on your printer is half a thou. That ought to be enough.

Andy

 

[Martin Wynne]

Hi Martin,

Looks like the X/Y resolution on your printer is half a thou. That ought to be enough.

Andy

Hi Andy,

If you believe that with a rubber belt you will believe anything.

Martin.

 

[AndyB]

Hi Andy,

If you believe that with a rubber belt you will believe anything.

Martin.

That would be the nominal resolution based on the minimum step angle and gearing. Stretch etc will affect accuracy/repeatability. If I remember correctly it's about the same resolution that I have.

 

[Martin Wynne]

Stretch etc will affect accuracy/repeatability.

Plus temperature, humidity and the price of kippers.

It also relies on micro-stepping the stepper motor, which is an imprecise function at the best of times. Not to mention friction in the slides, inertia of the table and print head, lubrication, etc.

If we can get below +/-0.05mm ( 2 thou) without a leadscrew and zero-backlash nut I shall be well-pleased. It's what I've based the plug track dimensioning on, and have managed to achieve with care and some adjustments to the jerk and acceleration settings in Cura. The biggest unknown is repeatability. If it works today, will the same settings produce the same result next week?

The big surprise was how much difference my utility to edit the backlash out of the gcode for the Minibo made. The results for that little child's 3D printer were spot-on. So there's a lesson there for fine-tuning the whole process once we can actually get started. There is so much still to learn and find out, but we can't do a damn thing until we can print a full turnout. I'm getting there as fast as I can.

cheers,

Martin.

 

[AndyB]

Plus temperature, humidity and the price of kippers.

It also relies on micro-stepping the stepper motor, which is an imprecise function at the best of times. Not to mention friction in the slides, inertia of the table and print head, lubrication, etc.

If we can get below +/-0.05mm ( 2 thou) without a leadscrew and zero-backlash nut I shall be well-pleased. It's what I've based the plug track dimensioning on, and have managed to achieve with care and some adjustments to the jerk and acceleration settings in Cura. The biggest unknown is repeatability. If it works today, will the same settings produce the same result next week?

The big surprise was how much difference my utility to edit the backlash out of the gcode for the Minibo made. The results for that little child's 3D printer were spot-on. So there's a lesson there for fine-tuning the whole process once we can actually get started. There is so much still to learn and find out, but we can't do a damn thing until we can print a full turnout. I'm getting there as fast as I can.

cheers,

Martin.

So why are you stalling posting stuff like this?

 

[AndyB]

All good points but it's not helping to get those chairs finished

 

[AndyB]

Hi Martin,

Are you creating unique jaws for every crossing angle just like the real things were cast?

I confess to cheating a bit. I realized there is very little difference in the "twist" angle over a large range of crossing angles. For those angles my chairs are all assembled using a pair of common jaw models that are twisted at an angle at the center of that range.

The jaw rail faces are properly aligned with the rail for each crossing angle but the jaw end faces may not be precisely parallel with the center line of the complete chair.

It's a useful approximation that works for my assembly method. Probably not relevant to your method but I thought I'd share it just in case.

Cheers!
Andy

 

[Martin Wynne]

Are you creating unique jaws for every crossing angle just like the real things were cast?

@AndyB

Hi Andy,

Thanks for your thoughts. But I'm not too sure what you mean there?

Each Templot chair is generated programmatically to suit the location of the rail/timber intersection in the template. If you set a crossing angle of 1:7.38 that's what you will get a chair for. There is no such prototype chair. It won't be unique if you create a second identical template. But otherwise it very likely will be.

Of course you wouldn't set a crossing angle of 1:7.38 for a single turnout or a simple crossover. But when you start designing double junctions, outside slips and the rest, usually to fit in tight model spaces, that angle might be exactly what you need.

I have copied the REA prototype designs as far as I can with regard to the details -- diameter of screw heads, length of keys, overall size of chairs, etc. But we can't copy the prototype exactly because:

1. The prototype has a fixed range of standard crossing angles. Where a non-standard angle is needed, the nearest standard chair is used. A period of low-speed running-in under traffic then wears off any rough edges. Models are not capable of doing that, so we have to create a chair which fits the rail alignment exactly from the start.

2. Most model gauges have over-scale flangeway gaps, so the prototype chair sizes and geometry don't fit.

3. Prototype rail is inclined at 1:20 (except check rails). It is next to impossible to build model pointwork with inclined rail. For example the knuckle bend in the wing rail needs to be made around a conical mandrel. Curved inclined rail needs to be held down with heavy fixings (the rail head will be stretched longer or shorter than the rail foot along a curve). For these reasons I decided that plug track would have vertical rail throughout -- just as we used to build in rivetted plywood days. But that in turn means changes in the chair design.

4. Most model wheel flanges are deeper than the prototype. This means that extra clearance is needed in the chair designs for the flangeways. In particular it means that where the prototype uses an inside-keyed chair, that won't be possible in the model.

5. Most models require switch openings wider than the prototype 4.1/4", which affects the design of the slide chairs and switch heel chairs.

All of which means the prototype REA designs can only be used as a guide, but I hope the finished chairs look the part -- even in the larger scales where chair details become more visible.

cheers,

Martin.

 

[AndyB]

Hi Martin,

Yes, as you are generating the chairs programmatically it's not relevant. (My method is to assemble the chairs in CAD from a set of common components in situ.)

My shortcut lets me construct crossing chairs for turnouts from 7 to 14 using only one pair (RH and LH) of jaws that are "twisted" 3 degrees from normal (which is correct for a 9.5 turnout). The jaws are properly aligned with the rail during assembly. The end faces of the jaws ought to be parallel with with the edges of the chair's base but in my case, depending on the particular crossing angle, they could be out of parallel by as much as one degree positive or negative.

Hope that makes sense

Cheers,
Andy

 

[Martin Wynne]

(My method is to assemble the chairs in CAD from a set of common components in situ.)

@AndyB

Hi Andy,

Can you clarify what you are trying to do? If you load the DXF or STL from Templot into CAD you will already have the chairs you need. There is nothing needing to be assembled from "a set of common components in situ" in CAD. Are you wanting to modify the Templot chairs to make them more suitable for FDM printing -- I thought you already established that it could be done with the existing chairs?

I'm concerned that newcomers reading this will imagine they need to get involved in CAD to use plug track.

There is no CAD involved anywhere in Templot plug track.

No CAD program is used to create it.

No CAD program or CAD skills are needed to build it.

I keep saying this because I know there is a perception otherwise for 3D printing and I can't seem to get the point across. The only thing I use CAD for is as a test viewer to see what Templot has generated. No user needs to do that -- the exported STL files can be previewed in the excellent 3D-Tool viewer program (free).

I think we must be at cross-purposes somewhere.

cheers,

Martin.

 

[AndyB]

Hi Martin,

Yes, it has NO RELEVANCE AT ALL to the use of your plug track which requires ABSOLUTELY NO CAD work. Sorry if it might be derailing this thread. I only mentioned it because I thought you could conceivably be able to use the information to reduce the amount of work you have to do to create the crossing chairs, but that does not seem to be the case.

Please carry on! (I'll shut-up now )

Andy

 

[Martin Wynne]

@AndyB

Thanks Andy.

I will move these posts to the other topic.

Done. Moved here.

cheers,

Martin.