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 Some of the earlier pages of this topic are now out-of-date.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.

Experimental Plug Track: 3D-printed, CNC-milled, laser-cut

Quick reply >
pps, so my conclusion is that both parts of a 2-part switch chair are "owned" by the stock rail rather than the crossing rail.
@Steve_Cornford

That's correct Steve.

I detect someone keen to get started. :)

I need to write at length about all this, and make some videos. Maybe later today.

I've been reluctant to spend too much time on docs yet, because it's quite possible that everything will change in 236c, or 236d, or 250z ... :)

Here below is the BOX file for the switch block chairs used to create the sample EM C-switch which I posted earlier. You will see that it contains 4 partial templates and 40 chairs. Enough to make up 2 switches. I cloned 3 of these on the Mars build plate. The spacing of the chair rows was too wide for convenient with the Xuron snips and wasteful of resin. It was just a test print, I will do them differently in future.

exp_c_sw_chairing.png


exp_c_sw_chairing1.png


These are all the 2-part chairs (easily identified by having screws/bolts on one end only).

Note that the stock-rail part chairs are all identical (apart from the randomising of the keys and screw heads).

n.b. 40 chairs is probably too much for the Formware online STL fixer, which will probably time out. As I posted earlier I'm happy to do the STL fixing for folks on my local copy of the fixer. Or you could split it into fewer chairs at a time, of course.

So much to write about. More later.

cheers,

Martin.
 

Attachments

  • exp_28sep22_em_c_sw_chairing.box
    44.7 KB · Views: 67
_______________
message ref: 5009
Not so much keen to get started, more keen to learn more about using Templot so that I have the skills necessary to get started in earnest and join in the experiments when practical. It is great to be learning new skills at out age!

I thought it sensible to subtract 5mm all round from the build plate size to calculate a maximum raft size, is that about right?

You have 4 rows of chairs, what would you recommend as the optimum distance between rows?
Oh and for that matter between columns?

By waste of resin I assume you mean the resin used in the "raft"?

So your boxfile contains a raft segment containing 2 sets of the 2-part chairs, interlaced for efficient use of the raft space, and then having output the .STL file and "fixed" it, you clone the segment 3 times to make up a bigger raft containing 4 segments for printing.

It would be convenient if there was a way of adding some text to the raft to indicate the spec. eg EM-B-LH.

But dont let me distract you from chair design with questions that I ought to be able to find the answers out for myself.

I have noticed the way you are representing the outline of the soleplate in trackpad, and as you have stated before not often modelled.
 
_______________
message ref: 5010
I thought it sensible to subtract 5mm all round from the build plate size to calculate a maximum raft size, is that about right?
@Steve_Cornford

Hi Steve,

I don't think you can have a full-size skate/raft like that. It will create too much suction against the FEP film as the build plate lifts, possibly damaging the film, and not fill with resin below it properly as it drops back down for the next layer.

I have found the maximum skate size for consistent results is about 1500 sq.mm for each skate. More than that and you get some ominous suction noises from the printer. Advice from other 3D printing users welcome.

Then for ease of use in removing the chairs from the skate with the Xuron snips, it's best to have only two rows of chairs on each skate.

I have also found that on the Mars 2 Pro the print quality drops off at the ends of the plate, presumably due to uneven illumination levels. If this results in half-formed bits of chair left in the bottom of the tank it is bad news -- it means you will have to empty and clean the tank and filter the resin (or discard it) before you can do more printing, which is a lot of faff. Much easier to keep back from the ends of the plate, accept a lower yield from each print and be ready to go again without any fuss. No doubt the print quality is fine for bigger chunkier models.

The result looks something like this, with several skates cloned in the Chitubox slicer program, each separated by a gap. This produces a 100% chair yield:

index.php


Those skates were 15mm wide for ordinary S1 chairs, but you would need wider for longer P slide chairs, etc., say 20mm wide. It might have been possible to get one more on the plate, but that might be getting a bit close to the ends. Also they could be longer, or split into two shorter ones width-wise on the plate, with a gap between them. There is so much experimenting (and writing) still to do.

For example -- do we really need a solid skate? Narrow ribs between each chair support pyramid would save a bit of resin and might allow larger skates? Or would the printing be worse because of "cupping" effects? Would it break up when being removed from the plate? The current skate design has a 0.5mm lip all round to assist in getting a scraper under it (hence the term skate or skate-raft).

Time for a boiled egg. :)

cheers,

Martin.
 
_______________
message ref: 5011
1665331800752.png
Here you go, you certainly deserve one!

I will plan on having about 30mm border on the build plate (for Mars 2 Pro) and keep skates as narrow as possible and separate them, rather than consolidate into one large skate per print.
Looks like a library of skates would be a good starting point.
Thanks for the feedback,
Steve
 
_______________
message ref: 5012
It would be convenient if there was a way of adding some text to the raft to indicate the spec. eg EM-B-LH.
@Steve_Cornford

Hi Steve,

You can do that now, sort of, although it needs some work on the sizes:

skate_label.png


That's using the 7-seg labels intended for the FDM timbering bricks.

It's limited to 0-9 and a few letter characters, so it would need to be an ID code number. You just need a notebook by the printer for the codes (and write the full description in felt-tip on the back of the skate as soon as it's dry).

I will look at adding it properly for resin printing.

Or you could edit the STL in CAD to add proper text / font labels to the model.

cheers,

Martin.
 
_______________
message ref: 5013
I did try using Fusion360 to add some text, managed to add some text and extrude it, but not succeeded in adding it to the correct face/plane, but I will persevere.
Would be useful for those owners of resin printers that plan on producing chairs to swop for timbers withose owners of laser-cutters or FDM printers.
In the meantime will use your suggestion of writing on the back, or have pre-labelled plastic storage boxes.
As it is i managed to label my boxfile as LH when it contained LH. Doh.

Just hope you dont run out of boiled eggs before cracking the remaining chairs :) ,or you cease getting enjoyment out of your development hobby.
 
_______________
message ref: 5014
Just hope you dont run out of boiled eggs before cracking the remaining chairs :) ,or you cease getting enjoyment out of your development hobby.
@Steve_Cornford

Hi Steve,

The development mojo comes and goes occasionally, but I've been messing with Templot one way or another for over 40 years, so I doubt I will lose interest now. :)

The real sweat is in writing the docs and explanations, which I do find tiresome and exhausting at times, because it is so important not to make mistakes. I don't want folks to end up wasting time or materials because I got it wrong, or being disappointed with their results after building and laying track. Everything I write has to be tested and double-checked -- I can no longer rely on my memory for the details as I used to do.

Here's some stuff from the archives which you may not have seen:

WEST MERCIAN EM GROUP NEWSLETTER, SEP 1980

" Our last meeting, held at Broadwas Village Hall on 7th Sep 1980, was
one of the liveliest we have had. Ten members were present and
among matters discussed were
.....
[snipped]
....
....A most interesting development was then described by Martin Wynne.
Neville Fairbairn had asked Martin for help in setting out a turnout from
a curved road. From this request Martin had gone on to devise a system
for enabling any turnout to be drawn out on 1mm graph paper by even
the most inexperienced.

Given the radius of the main line, the crossing angle and switch type,
Martin can supply the X and Y co-ordinates of the rail running face at
every chair position. From this data it is easy to construct a turnout
template, indeed Martin did it in about twenty minutes, explaining as he
went. Thus the days of track plans being juggled to suit the templates
available from commercial sources are over - pointwork can now be
made to suit the site as was done in full-size practice.

Needless to say the figures for the various offsets are produced by
a micro-computer, using a program written by Martin. This service
will be available soon from 85A Models. This must be one of the most
interesting developments in modelling in recent years."

Kind words there from the late Roy Miller. What would Roy have made of Templot and 3D printing today? But 20 minutes poring over 1mm graph paper! With my eyes now 42 years older I can understand why there were so few takers at the time. :)

p.s. Roy was the originator of the 00-SF track standard, which he called "EM minus 2".

cheers,

Martin.
 
_______________
message ref: 5015
_______________
message ref: 5017
oops...

The brick colour panel has lost its click!

Sorry about that, will be fixed in 236c.

In the meantime, brick colours can be set via the menus.

Martin.
 
_______________
message ref: 5018
Hi Martin,
Since upgrading to 236b (i think thats when it started) when performing a 3D export chairs only that includes the slide chairs that are located on the soleplate (timber S1) I am getting some extra 3d shapes exported. Would they be fragments of a soleplate?

Or is this something I am doing wrong. I have attached a basic example.
 

Attachments

  • lh_b_slide-chairs.stl
    6 MB · Views: 65
  • lh_b_slide_chairs.bgs3
    979 bytes · Views: 59
  • lh_b_slide_chairs.box
    18.3 KB · Views: 58
_______________
message ref: 5022
I then amended the template using shove timbers, then omitted timber S1, performed the export to .STL and the extra blocks have disappeared, so i think they are linked to timber S1.
.STL as exported and fixed version.

As a workaround i will just create a skate containing S2 to S6 slide chairs for now for my skate library.


Regards Steve
 

Attachments

  • lh_b_slide_chairs_omit-s1.stl
    5.6 MB · Views: 73
  • lh_b_slide_chairs_omit-s1_fixed.stl
    4.1 MB · Views: 67
_______________
message ref: 5023
@Steve_Cornford

Hi Steve,

Yes, I noticed the soleplate bug earlier. Too many bugs are getting through lately! :(

Just omit the S1 timber.

That will lose you two P slide chairs.

But all the P slide chairs on all templates are identical*. You can make yourself a batch of them by overlapping a couple of F switches (without the S1 timber for now):

raft_p_slide_chairs.png


In 4mm scale, reduce the track gauge to say 8mm so that they are close together on the timber, to save space. The above will give you a raft about 90mm x 16mm (1440 sq.mm).

n.b. But note that EM slide chairs are not the same as P4 slide chairs, including the plugs and socket sizes, so they are not interchangeable.

p.s. after a bit of thinking I have changed the term "skate" to "raft" throughout. Nothing is ever fixed for two minutes when it is experimental. :)

236c will be on the server soon because there are several other glitches to sort out, including the soleplate bug. I have also added the 7-seg labelling for the rafts, although deciding where to put it might be tricky. It may be necessary to extend the size of the raft just for the label.

*apart from the randomised heads on the chair screws, and the randomised nut rotations on the stock rail bolts.

cheers,

Martin.
 
_______________
message ref: 5025
.
ID label now on a chair raft:

raft_label.png

It's done using the same functions as the timbering brick labels. Same restrictions apply:

Must begin with a # character -- which becomes a leading dash, so you know which way up to read it. 7-seg characters, so 0-9 and a limited range of letters only.

Text can now be made smaller to fit somewhere on the raft:

raft_label_text_size.png


If no space for it, an extra raft rectangle could be attached to contain it.

in 236c soon. :)

cheers,

Martin.
 
_______________
message ref: 5041
What software are you using to display that image?
@Steve_Cornford

Hi Steve,

TurboCAD Deluxe 2015

Can probably still be found with a bit of Googling.

I did recently try a trial download of the current TurboCAD Deluxe, but it's quite expensive now. The 3D rendering has changed to a more photo-realistic style which requires higher spec graphics hardware. Without it, the results are worse than in the old 2015 version, so I'm sticking with that for now. It's running fine on Windows11.

edit: see also: https://85a.uk/templot/club/index.p...3d-printed-cnc-milled-laser-cut.229/post-5575

In truth I do very little CAD work. It's mostly used as a DXF viewer, as above.

cheers,

Martin.
 
_______________
message ref: 5043
.
236c is now on the server. Close and restart Templot to update.

cheers,

Martin.
 
_______________
message ref: 5044
Hi Martin, many thanks for the update.
My Windows 11 Updated ok, then managed to add a label to raft and here is the fixed result:-
1665650224890.png

Still experimenting at this stage, think I will make raft a bit smaller.
Then just need to decide on a labelling strategy.
Probably need a spreadsheet and a boiled egg.
~4-E-001-LH might mean 4mm, EM, number = row in spreadsheet, LH left hand, but for Right Hnad might have to use FH.
F for Facing , is that the same as Right Hand (in Uk at least)?
Row 1 in spreadsheet would have description column as "EM Left Hand B switch 2-part chairs, 2 sets"

I am assuming the following.
In 4mm scale, S1, S1J, L1 chairs are common to OO-SF, EM, P4

Chairs common to OO-SF and EM are P slide chairs, 2-part switch chairs LH, 2-part switch chairs RH

P4 needs its own P slide chairs and its own 2-part switch chairs LH and 2-part chairs RH

All good experience with Templot.
Steve
 
_______________
message ref: 5046
@Steve_Cornford

Hi Steve,

Wow, that was quick. You must have had a boiled egg for breakfast. :)

I shall try adding a few more 7-seg characters, for example RH could be lower-case as r h.

Don't assume too much commonality of chairs. S1, S1J and L1 should be common everywhere (for the same rail section and scale). I want to add an M1 to that list, and also the bolted half-chairs. Possibly the jaw-less saddle chairs. (M1 is a narrower bridge chair which fits on 10" sleepers.)

P slide is common for 00-SF and EM, but 00-BF, EM-SF and P4 will each differ.

Stock-rail part of 2-part chairs will also be mostly common everywhere but not always. For example the stock rail part of check rail chairs will be the same as the switch chairs in all except EM-SF and P4. In fact the 2-part check rail chairs are going to be a challenge in P4.

The switch-rail part of the switch block chairs should be common for all 4mm for the same hand and size of REA switch. But not necessarily for non-REA switches, where 00-SF might differ from EM for example. I want to add the straight switches to the available options.

And it will be all change for GWR switches and chairs -- but that's still for the future.

In 7mm scale so many modellers are using code 125 rail (which represents the lighter 85lb/yd rail) that to look right I should do the S2, L2 and M2 chair variants for it. For now I hope they would use the C&L code 131 rail with the existing chairs.

p.s. if you don't have an STL viewer, the Chitubox software (included with the Elegoo printer) is quite good:

chitubox_viewer.png


cheers,

Martin.
 
_______________
message ref: 5047
I have been watching this thread eagerly since this and sister threads started. Hands up initially I was very sceptical mainly down to some of my experiences with early 3D printing, plus I just thought it would be an impossible hill to climb

Thanks to Martin and James I now have seen and used the system of 3D printing and laser cutting for sleepers, timbers and chairs.

All I can say is WOW,

The system is far better than I expected and any lingering doubts I had have been more than satisfied. I must say the 3D printed chairs are far better than their injected moulded cousins.

Using the system it would be easier to remove the chairs in single lines, something which Martin referred to recently

1092.jpeg


I have fudged the timbers PL1 to PL4 as the chairs had not been finished. Martin asked me to try it with the latest C&L rail, which fits it like a well made glove. It works with some old EMGS or P4 rail I have and I have some exotic brand of rail PSM !! which fits but is a little on the tight side (may actually be a benefit). Martin told me the chair were a tight fit into the Timbers, I thought they were fine

1091.jpeg


James Walters kindly gave me a laser cut P4 ply turnout base, as of yet I have not tried using it.

If it was a plastic base I would be tempted to build a hybrid turnout using the parts that are available now and supplementing those that are not designed with Exactoscale parts.. I know this goes against the grain and perhaps defeats the object, but its a way of getting there quickly.

All I can say is Martins a Genius, if there was a Nobel prize for Railway modelling he deserves it
 
_______________
message ref: 5071
@Hayfield

Hi John,

Thanks for the kind words, I'm glad you liked trying the samples of plug track.

I think a Nobel prize might be a bit over the top -- a boiled egg would do fine. :)

A few points:


The idea is that you adjust the chairs setting to fit best on your rail, after making some trial chairs on your 3D printer. The default start-up setting is just what works best with my rail and printer and resin. In practice I fully expect users might need to adjust the setting to suit their own circumstances. I have in mind a 3D printer calibration function similar to the paper templates, where you 3D print a selection of chairs and choose the one which fits your rail best. So much still to do.

The chairs need to slide along the rail easily as you align them to the paper template, but then grip the rail enough to stay in place while you position the rail over the sockets. Fortunately there is a little bit of flexibility in the resin to assist this.

My test rail has been some C&L rail which is about 10 years old. I have now been given some more recent C&L rail to try. The chairs are just a fraction loose on it, so I will try tweaking the default setting just a fraction tighter in the next program update.


If you try a hybrid construction using Exactoscale chairs in place of the ones I haven't done yet, bear in mind that all the plug track rails are vertical. Whereas the Exactoscale chairs are intended to cant the rail at 1:20. I have never believed this really works around the special V-crossing chairs, but to the extent that it does you might need to fudge the difference.

I don't know what adhesive would be used to attach injection-moulded chairs to the 3D-printed timbering base, perhaps superglue. For 3D printing and laser-cutting you can omit the unwanted sockets from the relevant timbers by unticking these boxes:

chair_switch_offs.png


Those sockets are completely wrong anyway, I haven't looked at them yet for the proper chairs.


To use the existing chairs in a laser-cut plywood base it will need to be at least 1/8" (3mm) thick to suit the chair plugs. To use them with 1/16" plywood you would need to change the settings for a shorter plug depth. This makes the chairs quite a bit more fiddly to align over the sockets. It's also likely that you would need to change the chair/socket fit settings for a tighter fit.


cheers,

Martin.
 
_______________
message ref: 5073
Hi Martin,
I have just laser cut two B7 switches for a scissor crossover (00-BF 16.5mm), 1.6mm ply thickness (to match Peco bullhead rail)
B7 switch test 00-BF 16.5mm.PNG
I'm now trying to export the 3D stl file for the chairs to print on my Formlabs resin printer. I'm sorry if you have told me this previously but, I have looked through the various threads and can not seem to find the answers. I'm able to export the chairs as an stl file but,
a) is it possible to reduce the chair supports to 1.6mm so they do not protrude below my 1.6mm ply timbers?
b) how do I bunch-up/compress the chairs (timbers) so the resultant stl file is not so large/widespread?
B7 templot.PNG

Terry
 
_______________
message ref: 5095
Hi Terry, suggest read from page 22 of this topic where Martin gives us a clue to using partial templates etc, then group & copy that set and shift by say 5mm to interlace the second set, as bunching not yet available, you need to move the two parts of the two part switch chairs apart so that they print separately
 
_______________
message ref: 5096
Last edited:
I'm now trying to export the 3D stl file for the chairs to print on my Formlabs resin printer. I'm sorry if you have told me this previously but, I have looked through the various threads and can not seem to find the answers. I'm able to export the chairs as an stl file but,
a) is it possible to reduce the chair supports to 1.6mm so they do not protrude below my 1.6mm ply timbers?
b) how do I bunch-up/compress the chairs (timbers) so the resultant stl file is not so large/widespread?
@Terry Downes

Hi Terry,

I know nothing about laser-cutting, but that timbering base looks a bit over-cooked? It gives the impression that press-fitting the chairs into it would break the timbers apart? Are you intending to use smaller plugs and glue the chairs in place instead?

I suggest you don't change the height of the chair support pyramids. They are set at 3mm tall, and anything less makes it awkward to use flush cutters in removing the chairs from them.

You need to change the depth of the plug part of the chair for your thinner plywood:

short_plugs1.png


The plug part of the chair comprises three sections.

The first section below the chair base is parallel-sided and a bash interference fit in the sockets at each end of the socket.

The second and third sections provide a taper lead-in, so that the chairs can be located loosely in the sockets before being pressed home.

short_plugs.png


Make sure you have set your model scale/gauge, and then click the chair / socket fit... button.

Press the Enter key about 16 times until you get to the above settings 1, 2 and 3, corresponding to the drawing above. They are stored internally in prototype dimensions but shown there at the model size (hence the importance of having your model scale/gauge set correctly in the control template).

The default dimensions showing there are for 4mm/ft scale and intended for use with 1/8" (3.2mm) timbering thickness.

For 1/16" timbering I suggest a first trial would be to divide all three of them by 2. Then it will likely need some further trial and error changes to get the best results.

The total plug depth needs to be less than the timber thickness to allow for any rough or broken remains on the bottom of the plug after snipping from the support.

More info in this topic: https://85a.uk/templot/club/index.php?threads/the-book-of-plug-track.529/

cheers,

Martin.
 
_______________
message ref: 5097
Hi Terry, suggest read from page 22 of this topic where Martin gives us a clue to using partial templates etc, then group & copy that set and shift by say 5mm to interlace the second set, as bunching not yet available, you need to move the two parts of the two part switch chairs apart so that they print separately
Hi Steve, yes, that answers my chair bunching query. Many thanks.
 
_______________
message ref: 5098
Hi Martin,
You mentioned "In fact the 2-part check rail chairs are going to be a challenge in P4."

If a 2 part chair is problematic, would if be worth considering a one part chair as is the case with the Exactoscale P4 check chair?

1. form the check rail to the desire shape (using the templot printed template)
2. slide on the set of check rail chairs ( CCL, CC, CC, CC, CCR) onto the check rail.
(2a. optional) plug the check rail asssembly into a temporary jig with slightly looser sockets.
3. form the stock rail to the desired curvature (using the templot printed template as a guide)
4. Slide the stock rail into the check rail chair sub-assembly.
(4a optional) remove from temporary jig.
5 Add remaining other chairs to stock rail.
6 plug the stock rail assembly into the destination timber sockets and tamp home.

This is very similar to the process that can be implemented when utilising the Exactoscale check chairs, then gluing the completed assembly down gauging in the first instance from the the check rail, but of course in plug track no gauge is necessary as you have already proved with your system.

Perhaps it would be possible to design them as a 2-part chairs but with an option to have zero gap between the two parts so that when the resultant .STL file is "fixed" and sliced they are printed as a 1 part chairs.

Steve

ps On the Scalefour turnout templates they also list L1CCL and L1CCR chairs. Where are these used?
 
_______________
message ref: 5118
@Steve_Cornford

Hi Steve,

Thanks for your thoughts.

I said a "challenge", not "impossible", so I don't want to give in without even trying. :)

I've been surprised at how tough even quite thin parts can be in the "ABS-like" resin, so it's possible the 2-part chairs might work ok even for the narrow P4 flangeways if handled carefully.

If not I have several other ideas to try, such as filing the check rail foot a little narrower on one side. Or a 3-part assembly with the middle bits inserted last (doable for P4 with shallow flanges) by sliding them in from the end of the rail. Or even separate loose keys for the check rail might be doable if we print some sort of holder to handle and insert them -- we could then have a 1-part chair on the stock rail, just like the prototype.

I don't want to deviate from the "plug every bit of rail in separately from above" mantra if I can possibly avoid it. That's because I always have at the back of my mind what will be needed for complex formations such as slips and tandem turnouts, parallel-wing crossings, sharp curves with continuous check rails, etc. The prototype drops the rails into place, it doesn't slide them in, so if we copy that we should be able to replicate any prototype formation.

However, some sort of disposable jig which fits the rail head using clips or double-stick tape, and is assembled upside down with the rails and chairs before plugging into the timbering base in the middle of a complex formation (all in one go), is an idea still to be tried. This would allow for 1-part multi-rail chairs. If done as an FDM print and maybe softened with a hair dryer it could be snipped through and/or pulled away from the rails afterwards. Just thinking aloud.

The L1CC chairs are the short stubby 11" wide chairs for check rails where the standard check chairs won't fit. Within tandem turnouts for example, and on sharply curved turnouts needing a continuous check rail. There are dozens of these special chairs for specific situations. I will get them all done eventually -- no doubt just as the prototype scraps the last bit of bullhead track, and there is no-one left who can remember it. :)

cheers,

Martin.
 
_______________
message ref: 5120
Hi Martin,

I think that bullhead rail use, will see us out, there are still areas down here in Devon and Cornwall, where it's being replaced with newish rail.
 
_______________
message ref: 5126
Or even separate loose keys for the check rail might be doable if we print some sort of holder to handle and insert them -- we could then have a 1-part chair on the stock rail, just like the prototype.
@Steve_Cornford

p.s. Steve,

When I wrote that I had completely forgotten that I have already implemented this:

loose_pegged_jaw.png


(I'm not sure how I'm going to get this project finished without a functioning brain. :( )


index.php


Which might be doable for the check rails. I have never tried printing it, but I will do before I go any further on the crossing chairs, just to see what happens. There might be just enough "give" in the resin for the key to clip down under the rail head. But it must end up tight against the check rail to set the check gauge accurately against the inside jaw. (For the check chairs there would be 2 chair screws).

See previous post:

https://85a.uk/templot/club/index.p...3d-printed-cnc-milled-laser-cut.229/post-2243

cheers,

Martin.
 
_______________
message ref: 5128
@Steve_Cornford

p.p.s.

Much easier to handle, and a more robust fixing, if the loose peg extends down into a socket in the chair plug:

loose_pegged_jaw_deep.png


I'm thinking this might actually work, even in 4mm scale. :)

Martin.
 
_______________
message ref: 5131
I can see more tick boxes arriving shortly

I took my boiled egg for a walk today. Sat in a churchyard to eat it. There was a small stone obelisk in one corner.



THIS OBELISK WAS PLACED HERE ON MIDSUMMER DAY 2000 TO MARK THE THIRD MILLENNIUM

BENEATH LIES A TIME CAPSULE FROM THE 20th CENTURY

THE CAPSULE SHOULD BE ADDED TO AT THE END OF EVERY CENTURY HEREAFTER

I couldn't help wondering if the Plug Track would be finished in time. :)

Martin.
 
_______________
message ref: 5164
loose_jaw_key.png


Well this looks very promising: :)


pin_slots_plinth_plug.png


loose_jaws.png



In the slicer. Chitubox says the total cost of resin was 14p (40 chairs + 20 spare loose jaws):

loose_pinned_slicer.png



4mm/EM. As usual, almost impossible to photograph, and by the time they came off the printer there was no outdoor light left:

loose_jaws.jpg


The rail is dropped into the chair, just like the prototype, and then the loose outer jaw + key is inserted on a rectangular pin to lock the rail in place. The pin flexes as the key passes the rail head and clips under it. The rail cannot be lifted out because it is locked under the opposite jaw.

Not as fiddly as it looks to assemble with the aid of tweezers. We could 3D-print a holder tool for the loose jaws to aid assembly -- Blu-Tack holds the rail in place until the loose jaws are inserted. The rail is held quite firmly once the loose jaw is pushed home, it's not immediately obvious that it's not in ordinary chairs. The pin part is quite flexible and none were broken in the process.

It needs a few design tweaks. The top of the pin support pyramid is too small -- they all printed ok, but before UV-curing they were very delicate and several were lost in the washing process.

For testing these are ordinary S1 chairs. There is no real reason to do those this way, and you certainly wouldn't want to do a whole railway like this.

But it is making me think again about the special crossing chairs. My thought originally was that this might be doable for the narrow P4 flangeways instead of split chairs. But it might be a better approach for the other gauges too, rather than using multi-part chairs. It would certainly be doable in 7mm scale.

It's also pleasingly replicating the prototype in principle. :)

Now it needs yet more tickboxes to control which chairs are done this way, and which normal slide-on.

cheers,

Martin.
 
_______________
message ref: 5178
Hi Martin,
Looking like another step forward and definitely deserving of a boiled egg for Sunday breakfast!

I am using the label facility when 3D printing a raft of chairs in resin. Would it be possible to make the characters a bit more embossed?
By that I mean a bit more extruded from the surface or is that too complicated?
On the FDM timber brick in your most recent post they seem a lot more embossed.
Or is the degree of embossing related to the font size?

Steve
 
_______________
message ref: 5180
Hi Martin,
Looking like another step forward and definitely deserving of a boiled egg for Sunday breakfast!

I am using the label facility when 3D printing a raft of chairs in resin. Would it be possible to make the characters a bit more embossed?
By that I mean a bit more extruded from the surface or is that too complicated?
On the FDM timber brick in your most recent post they seem a lot more embossed.
Or is the degree of embossing related to the font size?
@Steve_Cornford

Hi Steve,

This should be working, although I haven't tested it lately:

label_text_depth.png


Also, like a lot of stuff, it's not yet included in the saved custom data if you change it. That's another job on the list. :)

cheers,

Martin.
 
_______________
message ref: 5181
loose_jaws_closer.jpg


This is the closest I can get with my camera. Although not visible to the naked eye, you can see that the loose jaw is not quite sitting flush to the chair base.

I thought the pin might be bottoming in the slot, but shortening it makes no difference. I thought the rail might be lifting the key, but removing the rail makes no difference.

So it must be some shrinkage distortion. I think it would get lost under the paint, but even so it was annoying me. So I have put a slight saucer relief around the slot in the top of the chair base to accommodate it:


edit: see later below

cheers,

Martin.
 
_______________
message ref: 5183
This is the closest I can get with my camera. Although not visible to the naked eye, you can see that the loose jaw is not quite sitting flush to the chair base. I thought the pin might be bottoming in the slot, but shortening it makes no difference. I thought the rail might be lifting the key, but removing the rail makes no difference.

So it must be some shrinkage distortion. I think it would get lost under the paint, but even so it was annoying me. So I have put a slight saucer relief around the slot in the top of the chair base to accommodate it:

I do write some twaddle at times. :(

After fiddling with the chairs for hours, a light bulb came on. The reason the loose jaw won't sit perfectly flush to the chair base is none of the above. It's because of a design flaw:

s1_loose_jaw_conflict.png

Top view of REA S1 chair on the left. The chair jaw is conflicting with the screw boss. This doesn't matter for the solid chairs -- the mesh repair function deals with it.

We don't actually need loose jaws on the S1 chairs -- or any other chairs which carry a single rail. But someone will want them, so I ought to get it fixed.

All the REA chairs carrying two or more rails are 4-screw chairs, as on the check-rail chair on the right. For which the chair jaw is just about clear -- partly because I have made the chair base a bit thicker on those. It will probably be fine as it stands, but I think I will tweak the jaw sizes just a fraction there too, to be sure it works as a loose jaw when wanted.

That's enough thinking for today. Time for a boiled egg before I make any more mistakes. :)

Martin.
 
_______________
message ref: 5186
.
I have chiselled out some cast iron so that the loose S1 outer jaw can sit flush to the chair base, clear of any conflicts:

s1_outer_mod.png


s1_outer_mod1.png


loose_jaw_s1.png


So now to test it on the printer.

cheers,

Martin.
 
_______________
message ref: 5187
Back
Top