Templot Club forums powered for Martin Wynne by XenForo :

TEMPLOT 3D PLUG TRACK - To get up to speed with this experimental project click here.   To watch an introductory video click here.   See the User Guide at Bexhill West.

  • The Plug Track functions are experimental and still being developed. Some of the earlier pages of this topic are now out-of-date.

    For an updated overview of this project see this topic.   For some practical modelling aspects of using Plug Track see Building 3D Track.

    The assumption is that you have your own machines on which to experiment, or helpful friends with machines. Please do not send Templot files to commercial laser cutting or 3D printing firms while this project is still experimental, because the results are unpredictable and possibly wasteful.

    Some pages of this and other topics include contributions from members who are creating and posting their own CAD designs for 3D printing and laser-cutting. Do not confuse them with Templot's own exported CAD files. All files derived from Templot are © Martin Wynne.
  • The Plug Track functions are experimental and still being developed.

    For an updated overview of this project see this topic.   For some practical modelling aspects of using Plug Track see Building 3D Track.

    The assumption is that you have your own machines on which to experiment, or helpful friends with machines. Please do not send Templot files to commercial laser cutting or 3D printing firms while this project is still experimental, because the results are unpredictable and possibly wasteful.

    Some pages of this and other topics include contributions from members who are creating and posting their own CAD designs for 3D printing and laser-cutting. Do not confuse them with Templot's own exported CAD files. All files derived from Templot are © Martin Wynne.

Messing with FB

Quick reply >

AndyB

Member
For a variety of reasons I've been mucking about with printing FB track and turnouts. This is supposed to be a Pandrol baseplate.

Screenshot (28).png

It's quite experimental and intended to be FDM printed. The missing Pandrol clips will be bits of wire, maybe :)
 
_______________
message ref: 4010
And here is a test print (00). No clips I'm afraid. I might be able to apply a squiggle of filament but I'm not sure it's worth it. Other than a blast of grey primer both sections are identical. These are 12" timbers and the baseplates are the wide ones used in turnouts. The timbers are really thin to speed up the print process.

The outside screw heads have merged into the clip holders. I might be able to expose them a bit better. These were printed with a 0.2 mm nozzle on a Folger printer with tweaked resolution. The image is not magnified so it represents how it might look from normal viewing distance.

DSCN5391.JPG
 
_______________
message ref: 4025
@AndyB

Hi Andy,

That looks good. :)

Is that standard PLA, or toughened, or something else? I found that standard PLA gave better detail, but the rail fixings were too fragile in 4mm scale.

Which slicer software are you using? What settings and temperature? How long does it take per timber using a 0.2mm nozzle?

cheers,

Martin.
 
_______________
message ref: 4026
@AndyB

Hi Andy,

That looks good. :)

Is that standard PLA, or toughened, or something else? I found that standard PLA gave better detail, but the rail fixings were too fragile in 4mm scale.

Which slicer software are you using? What settings and temperature? How long does it take per timber using a 0.2mm nozzle?

cheers,

Martin.
Hi Martin,

It's pla/pha and I agree regular pla gives better results with small details. I'll try a pla print for comparison.

I might have found a solution to the lack of strength with pla. The idea is to make the jaws a loose fit on the rail and offset them slightly alternately in and out. The rail is effectively weaving between the jaws.

Slicer is Slic3r and I'm dictating the layer heights.

Temperature is 205. It should just start to run out of the nozzle just before it reaches temperature.

I'll need to take an accurate time. I think its around six minutes for four timbers but they are only 0.45 mm thick. I would print them 1.2 mm with a very low density fill for real turnouts.

Cheers,
Andy
 
_______________
message ref: 4033
This is actually bullhead. The rail weaving method seems to work well with flat-bottom rail and I wondered if it could also be applied to bullhead rail. Based on a sample of one :) it seems to. The rail is gripped well and it is vertical. Of course all the other bullhead cosmetics bit would have to be added. The advantage with this method is that the chair jaws are not required to flex to support the rail (it's the rail that's flexing rather than the jaws.) Flexing the jaws can break them.

This is printed in PLA/PHA but I think it should work just as well with regular PLA.

DSCN5424.JPG


This visualization might give a better view of what's going on.

BHtest.jpg
 
_______________
message ref: 4185
This is the flat-bottom version above printed in PLA. As usual the camera is very cruel but without any magnification it could easily pass for an injection molding, even more so after a coat of paint :)

The rail-weaving thing seems to work quite well. I have not detected any tendency for the chairs to break when the rail is inserted. I should really make some turnouts now to see if it all works in practice.

BTW, I'm happy to share any of my models if anyone wants to give it a shot.

DSCN5435.JPG
 
_______________
message ref: 4207
Last edited:
The Karlgarin 82 has a slightly wider foot than Peco 83, just enough to make it difficult to fit to Peco Pandrols. I think I've detailed the dimensions of the Karlgarin rail before, Martin said it seemed quite close to prototype.

Head is 1.10mm, foot is 2.00mm, web is 0.70 (I think)
 
_______________
message ref: 4217
Last edited:
Martin said it seemed quite close to prototype.

Head is 1.10mm, foot is 2.00mm, web is 0.70 (I think)

@Stephen Freeman

For what prototype is that? For what scale? If 4mm/ft it looks like some American heavy rail. If 7mm/ft it might do for some narrow-gauge.

If you quote me, please include the context. :)

For UK BS-113A FB rail in 4mm/ft scale the dimensions are these:

Head width: 2.3/4" == 0.92mm (same as bullhead)

Foot width: 5.1/2" == 1.83mm

Web thickness (at mid-height): 25/32" == 0.26mm

Rail height: 6.1/4" == 2.08mm = 0.082" - code82

cheers,

Martin.
 
_______________
message ref: 4219
Thanks Stephen.
Foot width for FB-109, BS-110A, BS-113A rails is all 5.1/2" which scales to 1.83mm in 4mm scale, so 2mm is not too far out.
The problem with existing supplies of Code 82/83 is not the height (code number) but the head width. Do you know the head width of the Karlgarin rail?
cheers,
Martin.
 
_______________
message ref: 4221
_______________
message ref: 4222
Head width is 0.044" for 82/7.
0.053" for 100/7
0.056" for 125/7

They were designed for 7mm scales and are very accurate for those.
The profiles are a good match for NSWGR rails.

Matt M.
 
_______________
message ref: 4223
Head width is 0.044" for 82/7.
0.053" for 100/7
0.056" for 125/7

They were designed for 7mm scales and are very accurate for those.
The profiles are a good match for NSWGR rails.

Matt M.
I know they were intended for NG but there isn't (IMHO) a good enough code 82 otherwise. I normally use code 83 C&L code 82 used to have a very thin head, good if he's changed it but doubt it. Code 83 has a head comparable to Bullhead 75
 
_______________
message ref: 4224
Last edited:
Hi Stephen, I can't comment on the 82/7 as it wasn't what I needed at the time but they are
the widths listed by Karlgarin.

The 100/7 and the 125/7 were decent compromise for most the 31 flat bottomed rail sections
in use in 1936 on the NSWGR when modelling in 7mm. The 125/7 is a little narrow on the foot
when you get to 100lb rail but not badly so.

Back in 2010 when I was corresponding with Richard McLeish he did point out that one
of the reasons for having these sizes of track produced was to overcome the narrow heads on
a lot of what was available when used in 7mm.
Also using the high-Ni nickel silver as the material of choice.
But they were the best compromise of head and foot widths for height that would cover
a good number of prototype rail sections that he could come up with.

Matt M.
 
_______________
message ref: 4225
The depth of the base is also an important dimension and it's not so easy to measure, and then there are the tolerances in the rail itself. Typically the available values are only nominal. There can be significant variations between samples. Another thing I've noticed is a certain amount of asymmetry which can alter the track gauge depending which direction you feed the rail into the clips.

Weaving the rail helps to accommodate some of these variables and prevents the clips from being over stressed but the clips really have to be designed for a specific rail supplier and that's likely to require a bit of trial and error with the clip design using actual rail samples.
 
_______________
message ref: 4226
Last edited:
Another thing I've noticed is a certain amount of asymmetry which can alter the track gauge depending which direction you feed the rail into the clips.
@AndyB

Hi Andy,

That also applies to the Plug Track chairs and bullhead rail. I suggest making a note of which way round you are using the rail and keeping to it. Then after testing, a final gauge adjustment can then be made if needed:

chairs_gauge_adjust.png


There is always more to everything than meets the eye. :)

cheers,

Martin.
 
_______________
message ref: 4227
Hi Martin,

That's a good idea.

I'm also tempted to make go/no-go gauges. I could try printing them but I suspect they would really have to be in steel and that could be a bit tricky :)

Andy
 
_______________
message ref: 4229
These are regular PLA prints with a blast of grey primer. They represent wide clips for turnouts on 12" timbers. The timbers are very thin - only three 0.15 mm layers. I do that just to speed up the prints as I'm only really interested in the clips at the moment but it might be possible to glue thin printed timbers with clips on plywood or other timbers.

PhotoGrey.JPG
 
_______________
message ref: 4230
I have re-measured the code 82 with possibly a better tool and comes out at 0.99mm, as regards the foot, not easy to measure but seems to be 0.33mm
 
_______________
message ref: 4237
I have re-measured the code 82 with possibly a better tool and comes out at 0.99mm, as regards the foot, not easy to measure but seems to be 0.33mm

Hi Stephen,

Is that Peco Code 82?
Is 0.99mm the width of the rail head?

The Proto87 Code 70 rail I'm using is:

.0725" tall (Code 72.5)
Base width 0.0681" (1.73mm)
Base height 0.38mm approximately
Head width 0.0337" (0.86mm)

I used a micrometer for the measurements in inches. Based on the rails I have the widths are maxima.

Andy
 
_______________
message ref: 4241
I should point out that to get that amount of detail I did modify my printer to increase the X-Y resolution. I replaced the steppers with ones that take twice as many steps per revolution and I also used smaller drive cogs for the toothed belts. The extruder nozzle diameter is 0.2 mm which is probably close to the lowest practical limit.

The small nozzle helps a lot with detail but it can also substantially increase the time it takes to produce a print, particularly while printing timbers and sleepers. However, I've been able to speed-up printing timbers and sleepers quite a bit, even with such a small nozzle.

I've been printing a test panel consisting of six 00 sleepers 1.2 mm deep and adjusting various parameters in Slic3r (the slicing program I use) and I've managed to get it down to just over six minutes, or one minute per sleeper. That's without chairs/baseplates of course but they take a much smaller percentage of the total time and there's not much that can be done to make them print faster.
 
_______________
message ref: 4394
I'm not sure if this is a problem with my slicer or my printer. I think it's really caused by ambiguity in the protocol but my prints don't quite complete. The last few moves don't happen before the job terminates leaving a small "hole" in the top layer. It's because some G codes are buffered and some take effect immediately.

Rather than mess around with the G code my workaround is to add a very small sacrificial bump in my STL model that's one layer higher than everything else.
 
_______________
message ref: 4401
Rather than mess around with the G code my workaround is to add a very small sacrificial bump in my STL model that's one layer higher than everything else.
@AndyB

Hi Andy,

Your slicer probably includes an editable script for the printer which is appended to the end of each job. If you insert a M400 command, it should cause the printer to wait in a loop until all printing moves are finished. If there is already a M400 in the ending script, try adding a second one to stop the buffering -- see discussion in the online forums.

Or attach your gcode file here, and we might be able to spot the problem.

cheers,

Martin.
 
_______________
message ref: 4402
@AndyB

Hi Andy,

Your slicer probably includes an editable script for the printer which is appended to the end of each job. If you insert a M400 command, it should cause the printer to wait in a loop until all printing moves are finished. If there is already a M400 in the ending script, try adding a second one to stop the buffering -- see discussion in the online forums.

Or attach your gcode file here, and we might be able to spot the problem.

cheers,

Martin.
Thanks Martin.
Yes, I know about the M400 fix but I would actually have to edit the G-code produced by the slicer. My workaround is a one-time fix for any model I create (or maybe I should call it a "hack") :)
Cheers,
Andy
 
_______________
message ref: 4419
Thanks Martin.
Yes, I know about the M400 fix but I would actually have to edit the G-code produced by the slicer. My workaround is a one-time fix for any model I create (or maybe I should call it a "hack") :)
Cheers,
Andy
@AndyB

Hi Andy,

If you edit the end code script in the slicer it will be a one-time fix. The end code is appended to every gcode file:

end_gcode.png


You can set a different end script for each of your printers. Also the start code.

cheers,

Martin.
 
_______________
message ref: 4420
Hi Martin,

Yes, I know about the start and end scripts and I do use them.

The problem is there is a race condition. The printer buffers some commands (like the normal print moves) but not all commands. The slicer puts a G92 at the end of the G-code and G92 is not buffered. That means it can take effect before the printer has finished printing the final layer.

Adding a M400 before the G92 fixes the problem. Putting it after the G92 doesn't so I'd have to edit the G-code every time the slicer generated a new version. I suppose I could change the slicer code or the printer's firmware but it's a lot simpler just to add the bump to the model :D

Cheers,
Andy
 
_______________
message ref: 4421
Back
Top