Experimental 3D plug track - up to version 244c

[Martin Wynne]

.
A couple of additional menu items will be in 239a, neither of any immediate use:

To save time in getting the crossing chairs out, I shall be starting with the old-style cast "A" chairs with jaws and keys. The later slab & bracket design option will follow afterwards.

I just know this will produce a stream of messages telling me that I have got it wrong, and that the "A" chair should be the slab & bracket pattern. In reply I can post this screenshot.

The option for parallel wing crossings will have to wait until the "chair heaving" function is done, so that the chairs in the exit can be swapped to CC check rail chairs.

cheers,

Martin.

 

[Phil G]

Hi martin,
I for one totally endorse your thinking of doing the cast A chair as the starting point for the vee.
The slab and Bracket in the prototype relies on bolting it all together though the rails, something harder to replicate in resin.
cheers
Phil

 

[Steve_Cornford]

I expect "Chair Heaving" to be an Olympic sport before long

 

[Martin Wynne]

.
Traditionally when building track a 3-point gauge tool is used to apply progressive gauge-widening on sharp curves.

For plug track there are no gauges needed, the gauging is done by the chairs and sockets. So how to achieve gauge-widening on sharp curves?

Simply widening the gauge on the curve is easy -- just change the gauge like this:

But how to transition into that from the normal gauge? The answer is to combine a transition curve with a slew, splitting the template into 2 partial templates, one for each rail. Or preferably 3 templates, one containing the original timbering.

Here we have a sharp curve with an easement transition to straight. It could be all one template, but I have split the straight track as a separate template to make it clearer.

Normally gauge-widening would be maybe 0.5mm at the most in 4mm scale, and usually much less, so barely visible on here. So I have applied a ludicrous amount of gauge widening -- 6mm -- to make it visible here:

Switch off both rails and store the underlying timbers. It is necessary to keep the track centre-line on when there are no rails, otherwise Templot can't draw the timbers:

Now add a slew to the transition. In practice it doesn't need to be exactly aligned to the transition zone. Set the slew amount to half of the gauge-widening -- 3mm -- in this case negative:

That makes a mess of the timbers, so click the hide all button on the shove timbers dialog. Hiding a timber leaves the chairs visible (unlike omitting a timber, with removes the chairs with it):

Store that rail, and then swap to the other one. Change the slew amount to 3mm positive for this rail:

Finally, store that rail, and the end result is:

Plug track with 6mm of gauge-widening, applied progressively as the radius reduces into the curve.

cheers,

Martin.

 

[Martin Wynne]

.
Templot2 update version 239a is now on the server.

Restart Templot to update.

There are quite a few changes, but not enough to get us much further with a full plug track turnout.

Please report anything which looks wrong or a bit odd. Thanks.

cheers,

Martin.

 

[Steve_Cornford]

Thank you Martin. Installed ok.
A welcome diversion after spending a day battling with weeds on the allotment.
Steve

 

[Martin Wynne]

.
If you have already installed the recommended preview applications, the export preview button should just work in 239a.

If not, click this edit button for more info and download links:

More info here:

https://85a.uk/templot/club/index.php?threads/experimental-plug-track-continued.673/post-6678

n.b. To get the 3-D viewer working on raw unfixed STL files direct from Templot, see:

https://85a.uk/templot/club/index.php?threads/experimental-plug-track-continued.673/post-6688

cheers,

Martin.

 

[Phil G]

Hi Martin,
Just a quick question the attached box file is a simple C12 crossover to EM-SF, as generated and expected, the timbers at the cross over don't align. My question, is the intension to shove timbers first until the cross over timbers make sense, then heave the chairs to bring everything back into alignment?
cheers
Phil

 

[Steve_Cornford]

Just to report back that the pre-view function is really neat, so thank you Martin for this add on.

Quick test of chairs only (without a raft)
This was on an HP laptop runnings 64-bit Windows 10 Pro

 

[Steve_Cornford]

Hi Phil,

When shoving timbers using [along] button move chairs with timbers, as [crab] leaves chairs where they are.

 

[Martin Wynne]

Hi Martin,
Just a quick question the attached box file is a simple C12 crossover to EM-SF, as generated and expected, the timbers at the cross over don't align. My question, is the intension to shove timbers first until the cross over timbers make sense, then heave the chairs to bring everything back into alignment?
cheers
Phil

@Phil G

Hi Phil,

When you shove the timbers the chairs will move with them (and re-size/re-position as necessary). (But as Steve mentioned, NOT if you use the crab function. Don't use the crab shove function with plug track.)

For the conflicting timbers on a different template, use the hide outline option instead of omit. Hiding the outline will leave the chairs behind, which will then be captured by an extended timber on the opposite template.

It will only be necessary to use the as-yet-non-existent heaving functions in cases where the chairs conflict and need to be changed. Until then, with a bit of shoving and adjusting it will often be possible to avoid any conflicts.

Here for example I extended a couple of timbers and captured the chairs from the other template. Unfortunately there is a conflict between two S1 chairs from different templates:

That was fixed by changing the exit on the lower template from sleepered to timbered. That causes Templot to generate more turnout-road chairs and fix the conflict by swapping one of them to an L1 bridge chair. Hiding that timber outline then allowed those chairs to be captured by the other template:

With this result, and the L1 and S1J chairs now happily living on the extended timber:

It is possible for a timber to have up to 12 chairs, 4 of its own and 8 captured from other template(s). I'm hoping this will be enough to provide for even the most complex formations.

There is so much still to explain, and further changes no doubt needed. It will be so much easier to explain everything when I have finally got a full turnout done. Or at least, I hope it will.

cheers,

Martin.

 

[Martin Wynne]

.
Made a start on 239b today.

This is the old-style cast A chair in progress (called AA in Templot):

There are rather more discrepancies from the prototype than I would like. For a start the base shouldn't be rectangular. But at this stage I just want to get something usable as soon as possible, it has taken much too long to get this far. I can come back and refine the designs later. Also the later slab & bracket variant.

Ignore the rendering of the rail nose. That's something else I need to get back to eventually.

Now to complete the nose spacer blocks...

cheers,

Martin.

 

[Terry Downes]

Looking good Martin. I agree with your philosophy, functionality first, then aesthetics.

 

[Phil G]

Hi Martin,
just a question, does the AA chair grip the Vee? Or does the Vee just lie on top of it? And your going to use the nose spacer blocks to hold the vee in place?
phil,

 

[Martin Wynne]

Hi Martin,
just a question, does the AA chair grip the Vee? Or does the Vee just lie on top of it? And your going to use the nose spacer blocks to hold the vee in place?
phil,

@Phil G

Hi Phil,

Can I draw your attention to the words "in progress" in the first line of my post?

cheers,

Martin.

 

[Phil G]

Hi Martin,
Of course you can, still I think as a question, from a very curious chap, its valid,
The follow up questions would be, do the AA chair and the nose block fix to each other? Or more of a wedge or sandwich fit round the rail nose?
Of course, I am not trying to waste your time here. Given a choice of progressing the chairs or answering my questions, please carry on with your project, as that will also answer the questions in the fullness of time
Phil

 

[Martin Wynne]

@Phil G

Hi Phil,

I'm sorry if that reply was a bit curt.

I'm reluctant to make definitive statements until I have finished the design and made some actual chairs. Ideas don't always survive contact with reality.

I'm intending that the vee nose point rail will "hook" into the grip on the main side only (the other side having been filed flat). My thinking at present is to have a recess in the seat into which a very small bead of epoxy putty can be placed before assembly. That stuff takes 24 hours to harden, so wouldn't add much time stress to the assembly process (an important point for me). Whatever, it needs to be a gap-filling type of adhesive. How to mix such a small bead economically is a bit of a puzzle at present -- some sort of 3D-printed roller thingy might be next on the list of assembly tools. The same stuff would then be used in the AB or BB chairs to hold the splice rail in place against the point rail. In the end it might be found that the assembly is firm enough without needing any adhesive.

Whatever, the point and splice rails will be fixed separately, no need for a pre-assembled vee. A filing jig will be printable if desired to ensure the angles exactly match the template. I'm hoping no soldering will be needed, because it wouldn't need much heat to loosen the press-fit plug in the PLA socket. Dropper wires/connector tab thingies are attached to the underside of the rails before assembly. With a 3mm timber base there is plenty of space for such gubbins to be hidden under the ballast.

I've abandoned my previous ideas to have separate clip thingies which engage into the timber base because it won't work with laser-cut plywood. The prototype spacer blocks will now be separate cosmetic inserts.

I have previously posted methods to create a proper solid vee nose by bending and double-filing. That won't work here because we need the rail web to run all the way to the nose on one side (as in the Finetrax kits). Filing the blunt nose removes most of the overhang.

I wasn't going to write any of this stuff until I have an actual crossing on the bench in front of me. But since you ask...

But there is still plenty of time for the whole thing to go horribly wrong. As I keep saying, this is an experiment.

cheers,

Martin.

 

[Phil G]

Hi Martin,
Thanks for the detailed reply, very much appreciated, your thoughts sound very doable to me, I fully understand its very much a work in progress, and also acknowledge, Its really your hobby and we all gain because your so open with it. As such there is no such thing as a bad idea, that is simply a learning in action.
I maybe a minority of one here, but I find how your approaching or planning to approach this, and then watching the progress, almost as interesting as making track.
There is no doubt in my mind the vee will be the hardest nut to crack, especially as your looking for a solution that works with both laser cut timbers (currently my preferred option) or FDM, (something I have not ruled out however). Which I think is totally amazing to be honest.
please keep up the great work.
would I be right in assuming the 1.5 vee jig can be very easily adapted to convert it into a jib that is made to whatever angle a user of Templot creates and will in the final analyse be Templot turnout dependant, clearly if we wish to create turnouts with the same cross vee there is no need to make a second jig of the same angle.
phil

 

[Martin Wynne]

would I be right in assuming the 1.5 vee jig can be very easily adapted to convert it into a jib that is made to whatever angle a user of Templot creates and will in the final analyse be Templot turnout dependant, clearly if we wish to create turnouts with the same cross vee there is no need to make a second jig of the same angle.

@Phil G

Thanks for the kind words Phil.

The current 1:5 filing jig was just a proof of concept. The intention is that for any template you can click a button and get the STL for a filing jig to match the V-crossing angle. Likewise for K-crossing angles. They need an FDM printer of course.

But it all takes time. The jigs are quite a long way down the list, although they might get closer to the top if there is a general clamour for them. Also on the list is all the track furniture -- fishplates, switch anchors, spacer blocks, etc. Also the infill pieces for the 3P and EF chairs. The CD chair is going to be the trickiest, and will also require a loose insert.

Also still on the list are the side grooves for the dummy slider below the stretcher bar, and how to deal with that in plywood I don't at present have any answer for. Over to you?

cheers,

Martin.

 

[Phil G]

Hi Martin,
Clearly there is quiet a a list of things to progress, good on you for that. I would have thought however, from a practical perspective, almost as soon as the chairs are complete, there will be a strong need for Templot selectable vee filing jigs to be an option. As failing that, we will all be forced to make 1:5 crossing angles to complete turnouts. I would have thought some possible issues and/or wise instruction words will be required to expand to a full range of turnout crossing angles. As things to understand or interpretations of how to make it work will arise as more varied permeations are tried.
phil,

 

[Martin Wynne]

@Phil G

Hi Phil,

I take your point. But bear in mind that at present folks are filing vees for all sorts of angles without any aid from Templot, so the jigs aren't necessarily as urgent as you are suggesting. Let's get the chairs done first, and then have time to stop and think.

cheers,

Martin.

 

[Phil G]

Hi Martin,
That's is a very valid point, however I would think somewhere between 90 to 95% of people currently filing vee crossings go on to solder these together. As such the solder will allow for a considerable latitude, in shall we say filing angle error. If I understand your previous post correctly, the better the filing of the vee the less work any adhesive has to do.
So the vee jig maybe more critical to the plug track concept, than maybe first thought. That said its your project and I am not trying to de-rail (pun intended in this case) your process plans in anyway.
I do 100% agree with you chairs must come first however.
Phil,

 

[Martin Wynne]

@Phil G

Hi Phil,

OK, chairs, then jigs. I had thought that the K-crossings should come next, but as you say precise filing will be an important part of successful plug track, so perhaps the jigs will be quite urgent.

But it means users of plywood will need an FDM printer?

cheers,

Martin.

 

[Phil G]

But it means users of plywood will need an FDM printer?

Hi Martin,
I can't disagree with that comment, The best I have got is, maybe there is an option for our Templot club community to pool resources. I say that somewhat reservedly, as being on the other side of the world does not help my comment.
I would be happy to assist with FDM printed jigs, but the practicality of the high postage cost will make that a non viable an idea.
Phil,

 

[Martin Wynne]

I would be happy to assist with FDM printed jigs, but the practicality of the high postage cost will make that a non viable an idea.

@Phil G

Hi Phil,

It might be possible to create the jigs as small inserts at different angles which fit in the main clamping block. Once folks have a main block, that would make postage a lot less. It would also make it a lot faster for anyone to print the inserts for a batch of crossings if they already have a main block.

But there is always something to think about -- small inserts make it more difficult to add a visible label marked with the angle size.

cheers,

Martin.

 

[Michael Woods]

Sorry - I must be doing something stupid here but cant see what it is......I have upgraded and rerun the STL print for my point chairs.....I am now getting the chair bases but not the top.....any clues as to what I might be doing wrong?.....done this a few times and always worked before so I must be having a memory lapse somewhere....

 

[Martin Wynne]

@Michael Woods

Hi Michael,

You have loaded a file from version 238a into the latest update version 239a.

While everything is still experimental for plug track that doesn't work. Each update version is a separate entity for plug track. I keep trying to say that plug track isn't actually ready to be used in anger.

To solve the problem, click this menu item:

Then click these buttons:

1. click one of the buttons on the left. Most likely the bottom one if you are assembling track by sliding rail through the chairs.

2. click the modify group to match button.

You will be asked if you want to select all your templates into the group. The answer is group select all.

Then also click the apply changes button for your future templates in 239a.

Try printing again. If it is still not working, click the generator > rebuild all background templates menu item.

cheers,

Martin.

 

[Michael Woods]

Thank you - let me give this a go............................experimental or not this is really good......

 

[Phil G]

Hi Martin,
Do you think its possible to create an FDM holding block and then resin print the inserts?
FYI the cost of Postage from NZ to the UK is 22.5 pounds for a 50mm cubic package weighing less then 0.5Kg
so that's doable I guess.

Phil,

 

[Martin Wynne]

.
The AA chair is looking promising, so I think it is time to fire up the Mars printer and see what happens. Fortunately it is a bit cooler in the 3D printing den today.

This is a left-hand V-crossing. The bottom view is looking from the far end back towards the nose. You can see the recess in the rail seat into which a speck of epoxy putty (or whatever) can be placed with a cocktail stick before assembly. It certainly will be a speck if you compare it with the size of the rails. I doubt it will be possible to mix such a tiny amount without wastage -- it would be a good idea to have some other job on the go in the workshop for which the rest of the mix can be used. 2-part epoxy putty is available from model shops as Milliput and from DIY places for plumbing repairs. Other slow-set gap-filling adhesives may yet be found preferable. Or maybe none at all will be found necessary.

The end of the point rail will be filed to a flat on one side (on the right here in the bottom view), so no grip into the rail web is possible on that side. Obviously the point rail can't just be dropped down into place vertically. The idea is to "twist and hook" it into place, and then push it forward until it is wedged firmly between the two spacer blocks -- possibly breaking through any flashing between them in the process. How well that all works I am yet to discover. It could be "back to the drawing board" before teatime.

cheers,

Martin.

 

[Phil G]

Looking good Martin,

 

[Martin Wynne]

@Phil G

Hi Phil,

You are quite right -- having the proper matching filing jigs will make a massive difference. This is 1:5 in EM:

The junior hacksaw blade is intended to assist the camera in focusing -- it sems to have sort of worked. Otherwise my camera is hopeless at trying to focus on the translucent resin.

Having a filed angle which exactly matches the chair makes all the difference. The point rail drops in about 2mm back, and then slides forward and stops dead in exactly the right place with the blunt nose in line with the edge of the chair.

But whoever suggested placing a speck of epoxy putty in to a space you can hardly see needs his head examined! I tried it and failed miserably.

It does need something however. At flatter angles the point might wedge firmly in place, but at 1:5 it is too easy to pull the rail back from its proper position. I'm thinking perhaps the Humbrol enamel as a sealant might be the answer. If you are confident you have got everything right first time you could of course use superglue gel or similar. Or maybe a penetrating cyano afterwards.

But I think these chairs are going to work. The amazing little Mars printer didn't flash across between those spacer blocks -- you can just see daylight between them if you hold them to the light. And the loose jaws are a nice positive clip fit.

So all in all I'm pleased with these test prints and confident I can press on with the remaining chairs in the same way.

cheers,

Martin.

 

[James Walters]

Fantastic progress Martin, Well done indeed.

 

[Steve_Cornford]

jolly good progress Martin.
I hope you are enjoying a celebratory breakfast followed by an well deserved stroll in the countryside.
Steve

 

[Phil G]

Well done Martin that looks fantastic,
any chance you could post the stl file you printed? supper keen to get filing
I think your right re shallower the angle the greater the contact area, so it may just wedge in nicely.
24 hour 2 part epoxy applied under the rail vee after assemble may work as the fixing medium?
Phil,

 

[Steve_Cornford]

Hi Phil,
If you don't have access to an FDM printer for the filing jigs you could try getting them printed by a local NZ 3D print bureau.

 

[Phil G]

Hi Steve,
Thanks for the offer, I do however have an older ender FDM Printer had it for years to be hones, got to be honest, I was never overly impressed with FDM Printers, I do agree some of the plastic filaments are harder then resin though. FDM its an option for sure though.

So no problem for me to print the 1.5 filing Jig.
I was not talking about myself when it came to FDM printers, my comments was more aimed at people that only have either an FDM or a resin printer.
Very keen to start building some vee crossings even at 1 in 5 angle, mainly to play around with different glue or bonding options, also keen to see what would happen if you did try and solder the vee in situ? Maybe with a good enough heat sink the resin can withstand soldering the vee. don't know but only one way to find out.
phil

 

[Martin Wynne]

@Phil G @Steve_Cornford

Thanks all for the kind comments.

Phil, I will post the STL shortly. But you are not going to get very far with it for building a crossing until I have done the BB chairs -- that's why there is a gap on the raft. In 239a you would also need to add a socket manually for the AA chair in the DXF.

And the AB chairs come next if I'm working methodically, although those are needed only above 1:10 (or if you click the parallel wing option for shorter angles).

Changing the subject, we're expecting an announcement later today from the Scalefour Society about the Finetrax kits for P4. I'm interested to know how well Wayne has held the tolerances for P4.

cheers,

Martin.

 

[Phil G]

Hi Martin,
Fully understand the STL file will not make a crossing, that is not my intention. I am keen to make a few Vee rails and then play with options to hold the vee rails in place. My gut feel is in the end, if at all possible a touch of maybe low melt solder is the ultimate answer, but that then needs away to protect the resin from the heat.
Phil

 

[Phil O]

@Phil G @Steve_Cornford

Thanks all for the kind comments.

Phil, I will post the STL shortly. But you are not going to get very far with it for building a crossing until I have done the BB chairs -- that's why there is a gap on the raft. In 239a you would also need to add a socket manually for the AA chair in the DXF.

And the AB chairs come next if I'm working methodically, although those are needed only above 1:10 (or if you click the parallel wing option for shorter angles).

Changing the subject, we're expecting an announcement later today from the Scalefour Society about the Finetrax kits for P4. I'm interested to know how well Wayne has held the tolerances for P4.

cheers,

Martin.

Hi Martin,

There were a pair of Wayne's turnouts on display, I didn't give them a close up viewing, but they look very good. They are due to go on sale in early July. B7s to start with and if there's enough uptake, then additional may be forthcoming.