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

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
I mentioned that I wasn't happy with the socket-fitting code and might re-write it. That's now done.

As always when you make improvements, you can't understand why you didn't do it that way in the first place. :)

The sockets have been freed from the template which generates them, and now have a separate existence where they can be captured by any shoved timber which surrounds them.

This means that existing BOX files containing partial templates and shoved timbers in complex formations can now be used unchanged for both 2-D and 3-D timbering exports. There is no need for the clunky joins in timbers which I previously suggested for 3-D files.

That moves the dial significantly back towards FDM printing from CNC milling. I'm glad I have both machines and don't need to choose.

Here you can see I have been doing some timber shoving on a crossover to create some long timbers:

new_sockets2.png


Notice for timbers E2, etc., I have hidden the outline to leave only the chairs. These can then be shoved along the rails over the long timber which has been extended from the other turnout.


new_sockets1.png


A couple of things to notice. I omitted one rail to make a partial template, as a test to make sure it didn't upset the timber as it did before.

You can also get creative -- for that one chair the gang had to move it sideways a bit because the screw holes in the timber had stripped, and needed to be filled and re-drilled. :)

I still have some work to do on the trackpad view to ensure the chairs are always visible over the timber infill.

And the result is (ignoring the still missing special chairs):

new_sockets3.png


Notice the moved chair. Also the missing rail. Those timbers are now happy to have only 3 chairs instead of the usual 4.

I'm quite pleased with this. It needs more testing before I release 232a for trials, but I can now see a clear way forward which was cloudy before. And then back to the special chairs.

cheers,

Martin.
 
_______________
message ref: 3054

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
I'm going to delete some of the earlier posts about making joins in shoved timbers for FDM -- before Google indexes them for 100 years. If you need to reference any of it, do it now.

edit: now deleted

That's the thing about experimental stuff -- even the posts about it are experimental. :)

Martin.
 
_______________
message ref: 3057

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
Another day, another button:

bgnd_chairs_option.png


Having the chairs showing on all background templates makes such a hit on the screen response for a large track plan that you will want to turn them off except when actually timber shoving, or chair heaving, or creating timbering bricks.

You can turn them on and off by delving into the menus, but I thought an easily accessible switch would be helpful.

cheers,

Martin.
 
_______________
message ref: 3065
I usually come by once a week to catch up on this thread and every time I do I am amazed at your productivity Martin! Thanks so much for putting in such a tremendous effort. I think this is genuinely the future of the hobby for a large subset of people.

I assume one of the "larger" bed resin printers (such as the planned Elegoo Jupiter) would be just as good or better than an FDM printer here?

 
_______________
message ref: 3072

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
I assume one of the "larger" bed resin printers (such as the planned Elegoo Jupiter) would be just as good or better than an FDM printer here?
@murphaph

Hi Phil,

Thanks for your kind words.

You are adding another option for the timbering bases -- FDM printed, laser-cut, CNC milled, and now resin printed!

My first thoughts re the Jupiter for track building would be:

At that cost you are entering laser-cutter cost territory, and paying for a build height way beyond anything needed for track. But great for a one-piece water tower in 0 gauge. :)

With such high-resolution it might be possible to print the ballast at the same time?

My one doubt would be about the ductile strength of the resin. The intention for Plug Track is that the chairs are a firm interference press fit in the timbers. That might cause resin-printed timbers to fracture. If not immediately, maybe after a few years UV-exposure embrittlement? Like the chairs they would need a good coat of paint to protect them. Or be made a looser fit for gluing instead. A lot would depend on the choice of resin -- there are (expensive) engineering resins which might be needed.

I need to try printing some timbers on my little Elegoo printer and doing some tests. On a small printer the timbers could be bunched-up for printing, and maybe located in a FDM timbering fret (3D template) for accurate gauging and alignment. There is already an option for that:
index.php


So many options, ifs and buts. At some stage someone has got to write all this up. I'm not looking forward to that. :( I still haven't finished the timbering brick topic, but it all applies equally to a resin-printed brick:

index.php


More: https://85a.uk/templot/club/index.php?threads/extracting-a-timbering-brick-from-a-track-plan.295/

cheers,

Martin.
 
_______________
message ref: 3073
Thanks for your thoughts Martin. I will keep my powder dry on a large resin printer then and see how things develop. As you quite rightly point out, that's laser cutter territory and a small resin printer + laser cutter might well be a better investment than a large resin printer alone. I was kind of thinking of printing bogie stock in 00 at some stage, hence the "wondering out loud" about the newer, larger resin printers....

I was kind of thinking that resin printers might be better at representing most concrete sleepers with their obviously "uneven" profile, falling in to a valley in the middle. I wonder how well FDM printers can do that? Clearly a laser cutter can not help there as the wood will almost certainly always be flat.
 
_______________
message ref: 3074

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
I was kind of thinking that resin printers might be better at representing most concrete sleepers with their obviously "uneven" profile, falling in to a valley in the middle. I wonder how well FDM printers can do that? Clearly a laser cutter can not help there as the wood will almost certainly always be flat.
@murphaph

Hi Phil,

You are jumping a long way ahead. Concrete bearers are mostly used for flat-bottom track, and I haven't got anywhere near thinking about FB baseplates yet -- beyond the idea of omitting the jaws from a BH chair. :)

Concrete sleepers are not unknown for bullhead plain track:

mount_pleasant_tunnel.jpg


But I don't know of any bullhead pointwork on concrete bearers.

There would be no inherent difficulty in programming the concrete shape for 3D printing. For FDM the issue might be in getting a smooth top surface if it isn't flat -- although for the above weathered concrete look it might be ideal.

For the pointwork timbers I am already planning some surface detail, such as the switch sole plates, which are seldom modelled and often quite prominent on the toe timber:

index.php


cheers,

Martin.
 
_______________
message ref: 3075
Yeah I'm like 5 years down the road in my head already lol. The possibilities really are limitless aren't they. Patience is a virtue as they say and I am patiently following this thread with great interest :)
 
_______________
message ref: 3076

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
Unless you know different, I think I've got the timbering mostly done, at least for REA bullhead. There is a new timber throw setting in the timbering data in 232a, intended for creating L1-chaired waybeams. I will write more about that later.

So on we go. P chairs (slide chairs) next:

rea_gwr_p_slide_chairs.jpg


REA on the left and GWR on the right.

Slightly confusing because of the two, the GWR looks more like the REA S1 chair with a single outer chair screw and double ribs.

For the special switch and crossing chairs, the REA swaps from 3-screws to 4-screws throughout. The GWR sticks with 2-screws throughout, which is a bit puzzling -- understandable if they were through bolts as on their plain track, but in pointwork the GWR used screws from above like everyone else. The REA fixing looks far more substantial to resist gauge spread -- always a risk in switches resulting from the deflection angle in the turnout road.

There is an alternative REA design, the PJ chair, which has a separate spacer block in the rail web, in effect a cast-iron key, similar to flat-bottom slide chairs. That requires a more upright jaw and a longer bolt. It also has a hexagon nut on the bolt instead of a square nut, for easier spannering between the chair-screw heads.

The PJ would make a more substantial resin print. On the other hand as far as I know they were introduced only in the 1950s (anyone?), so those modelling the grouping era are going to be called rude names for having the wrong style of slide chair. :)

On balance I'm minded to start with the original REA P chair as above. If it proves too fragile, we still have the option of swapping to PJ.

So I'd better make a start on it, instead of putting it off by rambling on about it here.

cheers,

Martin.
 
_______________
message ref: 3093

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
p.s. correction :confused:

Oh no it doesn't. Have a hexagon nut. That's only on flat-bottom, for the wider rail foot.

Here's the REA PJ chair:

rea_pj_slide_chair.jpg


Notice that the spacer block is thinner than a normal key, and there is no fresh air below it.

That's because there is no requirement to drop the rail into the chair before kicking it across under an inner jaw.

Martin.
 
_______________
message ref: 3097

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
Another conundrum.

The prototype switch opening is 4.1/4".

For EM and 00-SF that's increased to 5.1/4" recommended minimum (1.75mm, thickness of a 20p coin).

Unfortunately that means a conflict with the chair screw heads:
p_chair_opening_em.png


p_chair_opening_em1.png


Would you rather:

a. snip the tops off the chair screws to clear (they are very small in 4mm scale), or

b. have Templot make the chair longer than scale to suit the chosen gauge, or

c. have yet another tickbox option for that, or

d. something else.

Bear in mind for 00-SF that making the chairs longer than scale will reduce even further the distance between opposite slide chairs than is already caused by the under-scale gauge.

p.s. in the drawing I have rotated the rail to vertical.

cheers,

Martin.
 
_______________
message ref: 3098

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
B or C. Blades overhanging the chair screws look worse than the chair being a little too long! (Not that I model in either of the scales you mention 😀)

I think I agree. But you can write the replies when folks share chairs, and EM slide chairs don't fit in P4 sockets! :)

Martin.
 
_______________
message ref: 3101

Hayfield

Member
Location
Essex
If you look at what Len Numan did with the Exactoscale chair range, for 00 & EM gauges (4mm scale E4CH 301A) he made a slightly larger slide chair (for all P positions) and a P4 slide chair (E4CH 302A).

If anything there should be a 3rd version for standard 00 gauge, as the 4mm scale one is fine for EM & 00SF
 
_______________
message ref: 3107

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
@Hayfield

Thanks John. I was hoping someone might come up with a clever wheeze to reduce the workload, but I knew it would come down to more tickboxes* in the end:

dxf_switch_opening.png


Set automatically from the current gauge/scale, but can be changed before exporting the file.

This dialog is getting out of hand. It will need a full re-design before a full release for Plug Track. :(

*I suspect we shall have to leave behind any hope of compatibility with Windows95, 98, ME. "Out of resources" errors.

cheers,

Martin.
 
_______________
message ref: 3115

Martin Wynne

Admin
Location
West of the Severn UK
Info
.
Enjoy using Templot?
Thanks.

Please do not send requests for help direct to me via email.

Post your questions on the forum where everyone can see them and add
helpful replies.
.
Now that I'm looking at the switch and crossing chairs, I have had to make a decision about something I've been pondering for ages. Should the chairing be prototypical or programmatic? I've settled on programmatic.

What do I mean by that? Suppose you have created an irregular diamond-crossing which fits your track plan nicely, and one of the V-crossings is showing as 1:7.38 angle. Now obviously no such size of V-crossing exists on the prototype -- the nearest size, 1:7.5, would be used instead. And the big 'ammer would be brought into play to make it fit -- a few days running-in under traffic soon removes any rough spots. That is also what you do if building with C&L or Exactoscale chairs, because they don't make 1:7.38 crossing chairs either, although I doubt model running-in has much effect.

To be strictly prototypical therefore, Templot should do the same, and I should implement a corresponding big 'ammer function.

Well I'm not going to -- it's just too much of a minefield of conflicting dimensions and prototype variations. I am hoping to finish this project within my lifetime!

Instead, Templot will programmatically create 1:7.38 crossing chairs for a perfect fit. Only you and I will ever know the difference, no-one else will notice. If that's not something you can live with, -- using Templot isn't compulsory. :)

Another example of programmatic chairing will be for the switch block chairs. Suppose you shove one of the switch timbers along a bit, like this:

shoved_block_chair.png


You know and I know that you should never do that. The block chairs support angled rails, and will fit the rails in one position only. You can't shove switch timbers more than an inch or so under the chairs, and the chairs need to stay put while you do it.

But Templot doesn't know that, and will happily shorten or lengthen the block chair, and adjust the rail angle, as you shove the timber along. Whereas the prototype has only a few fixed sizes of block chair, not an infinite range of sizes.

But in a cramped model design, you may need to shove switch timbers more than a bit, so this unprototypical result might be useful.

Again, I'm creating these chairs programmatically in order to have some hope of chairing the whole range of switches in my lifetime.

But this decision does have consequences. It means the 3D chairs file and timbering brick file will be a matched pair and need to be created at the same time and kept together, and not interchanged with the files for other templates or bricks, even if the templates are nominally the same size.

I'm intending that it will be possible to bunch together all the chairs from a brick for resin-printing, and print a paper chart showing which chair goes where on the templates.

cheers,

Martin.
 
_______________
message ref: 3121
Top