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

[Martin Wynne]

.
We now have some outer jaws for the switch block chairs:

These 2-screw outers are common to several of the special crossing chairs, and also the check rail chairs.

(They might also be used with the REA 2-screw S1 inners to provide a generic 4-screw plain chair design for those who want that for their prototype.)

cheers,

Martin.

 

[Steve_Cornford]

Hi Martin, you are making excellent progress in your experiments.
I wonder if in the smaller scales using ply timbers it is practical to have two sockets so close together, it might need to be one socket but with the two plugs shaped to meet at their adjoining faces so that when you insert the 2nd plug it presses against the first plug?
Just a thought.
That is an interesting photo. What is the name of the strap between the switch rail and the stock rail?
Is it for electrical bonding or for minimising the longditudinal movement of the switch blade?
Like the soleplate rarely modelled in 4mm scale apart from a wire soldered underneath the rails that is for electricl continuity.

 

[richard_t]

Hi

It's called a switch anchor- I'm guessing it's there to stop the switch blades from creeping.

 

[Martin Wynne]

@Steve_Cornford

Hi Steve,

The strap is a switch anchor. As you guessed It prevents any longitudinal movement between the stock rail and the switch rail, which is important to preserve the geometry of the planing fit against the stock rail at the toe. It's not often modelled, although there are some etched parts available from Ambis. The most prominent part of it in normal model viewing is the bolt and fish-washer in the rail side. I'm hoping to make all these bits of pointwork furniture available for 3D printing along with the chairs in due course. Also the fishplates of course, which I've already been working on. If the rails are plugged in vertically, the obvious consequence is that the fishplates will have to be applied from the side (just like the prototype).

The question of sockets for the double chairs has been on my mind for a long time. I'm hoping to make every single thing an adjustable option, so that you could adjust the size of the socket/plug on every single chair individually if you wanted to. So you might prefer to make the plugs smaller on one or both of the half-chairs. Or larger so that they fit against each other in a single merged socket -- that prevents the first one being a press fit of course, or providing any gauging or alignment until the second one is added. Or put a radius in the corner of the socket similar to the CNC milled sockets, to stiffen the narrow web between the two sockets. Or an angled corner.

As I keep saying there is a long way to go, and a million adjustments and modifications to try out before the project can be "signed off" so to speak. It is also very likely that what works best in one scale will differ from in other scales. At present I'm pressing on (!) with the chairs so that I can at least see a complete turnout before I start making modifications and changes.

An obvious difference in the photo is the hollow spring steel keys used instead of solid wooden keys. But no, I'm definitely not going to try that in the smaller scales. In the larger scales, who knows?

cheers,

Martin.

 

[Martin Wynne]

.
I thought something didn't look right. I was scaling the fillet radii twice.

This is better:

cheers,

Martin.

 

[Stephen Freeman]

There seems to be no stopping you at the moment, I am finding it difficult to keep up with all the improvements to the standard Templot usage let alone 3D printing. I do have an FDM printer sitting in the corner but still no time to play with it and now I have to start saving up for a resin printer!

 

[Martin Wynne]

There seems to be no stopping you at the moment, I am finding it difficult to keep up with all the improvements to the standard Templot usage let alone 3D printing. I do have an FDM printer sitting in the corner but still no time to play with it and now I have to start saving up for a resin printer!

Hi Stephen,

It's not compulsory! And it's probably unwise to buy machines just on the strength of this project -- there is still plenty of time for the whole thing to bite the dust.

Martin.

 

[Terry Downes]

Hi Martin,
I'm trying to set a custom rail size for Peco 00 code 75 Bullhead rail but I'm getting some weird results. I'm pretty sure its to do with the last three variables relating to fish centres & angle.

here are the Peco code 75 bullhead Rail dimensions I have measured

Any advice or guidance would be much appreciated.

Terry

 

[Martin Wynne]

@Terry Downes

Hi Terry,

The fish angle on all UK REA bullhead and jointed flat-bottom track is 1:2.75. Here is the way the rail section is dimensioned:

You can see that the all-important fishing faces are dimensioned from their intersect on the rail section centre-line. The head and foot depths can then be derived if needed.

Your dimensions show that the Peco rail is not a proper bullhead, having head and foot of equal depth. I believe they have done this for easier manufacturing, so that the rail can be threaded into the sleeper base either way up, and the machined pointwork parts are not handed and can be used for left-hand or right-hand versions of each item.

If the fishing faces really are as square as you have drawn them, you need to enter a very large number for the fish angle, say 500000.

cheers,

Martin.

 

[Terry Downes]

Hi Martin,
I have been playing around with trying to create chairs which will be compatible with the new Peco code 75 bullhead rail. I know you are working on may aspects of Templot and this is all still experimental so, please do not feel compelled to answer or adress these queries if you are focusing on other elements etc.
Anyway, I have re-measured the new Peco code 75 bullhead rail section using a vernier gauge and adjusted the custom rail settings as shown below.

This has produced pretty good chairs 3D models (3Ddxf/stl) but, I now need to remove some chair material from the lower fixed jaw side as shown below. Is this something I have initiated from within the settings or should I try and fix post-export from Templot within a 3D dxf/stl environment?

Also, the wedge appears to have generated weirdly as shown below. Do you think this is because I have set one of the variables out of scope? or maybe the fish angle of 1:5 is too shallow?

I did check the 'fill below key' box and this proved successful but, I was looking for the button/function/area which enabled the chair jaws to be beefed up. has this been remove or moved or am I imagining I have seen this in the past? I was hoping I may be able to adjust the fixed jaw upwards (not prototypical) and avoid the clash shown above.

regards, Terry

 

[Martin Wynne]

@Terry Downes

Hi Terry,

Many thanks for posting your feedback from the experimental functions in Templot. It's a great help.

Do Peco now supply their code 75 bullhead rail separately? If not it might be better to use a more readily available rail, such as the EMGS/C&L code 75 rail for your experiments. Not least because your dimensions indicate a very poor representation of BS-95R bullhead rail:

There is an obvious bug in Templot on the lower face of the part I call the "grip" (ringed red) which explains this strange feature:

Thanks for finding it, I will get it fixed for the next program update.

This is what the prototype rail looks like (Templot simplified), the wooden key being symmetrical about the rail edge:

Overlaying the two looks like this:

In order to maintain the track gauge, the rail, chair base and jaws are all referenced from the gauge-face of the rail. Your rail is significantly under scale width, which causes significant distortion to the shape and appearance of the key. I will change the top angle on the exposed part of the key to match the prototype 1:2.75 fish angle, instead of matching the model fish angle. This will reduce the excessive height of the key above the top of the jaw.

However, I can't find what is going wrong with the end of your key, it seems fine here in TurboCAD -- which 3D rendering software are you using?

I tried adding some fill below the key, which seems to be working ok:

Have you tried running the STL file through the mesh repair tool at:

https://www.formware.co/onlinestlrepair

It will probably fix the key problem.

I did have a chair "beefing" option in some earlier versions. I removed it because it was causing difficulties for the special switch and crossing chairs, and it was one degree of complexity too many. After trials with my Elegoo 3D printer and "ABS-like" resin I found the chairs were adequately robust, at least in 4mm scale, and didn't need beefing up.

Eventually it will be possible to enter custom chair data in addition to custom rail sections, to create other prototype chairs and/or beef up existing chairs if needed. Sorry we are not there yet though.

Thanks again for posting you feedback.

cheers,

Martin.

 

[Terry Downes]

Hi Martin,
I'm not sure if Peco supply their new code 75 bullhead rail separately but, the only reason I'm pursuing this rail section is because I want to add a new templot scissor/slip crossing into an existing layout which uses the new large radius points and flexitrack. I will look into whether the rail you have suggested can be joined to the Peco rail and decide which rail to utilise in my scissor crossing. I'm still refining this crossing and will try to find a relevant post to discuss timbering etc. separately.

Regarding key generation errors, I was pulling the 3D dxf file into Solidworks but, when I import the stl file the key appears fine (grey image). Also when I import the 3d dxf into TurboCAD it appears fine (coloured image). therefore i believe this glitch was down to the Solidworks import facility.

going forward...
A nice feature in the future would be to export a 3D solid file step, .stp or .sat as well as the current 3d surface formats .stl & .dxf but, this is only required for people like me who like to play. I'm not sure which version of TurboCAD you are using but I believe the Platinum edition (mega bucks!) is the only version which enables 3d solid exports. So, I'll look at exploring converting surface models into solid models via SolidWorks and Inventor.

Once I have a stl file of all the various chairs for a point/template they will be spaced apart beyond the print area of my printer. Will there be a feature to export chairs only AND close up the individual chairs into a condensed area to facilitate resin printing of a whole point in one go?

Terry

 

[Martin Wynne]

@Terry Downes

Hi Terry,

Wow. A fully-chaired scissors crossover and two double slips is going to be a sight to behold.

However, in Plug Track I am still doing the switch chairs. I've given some thought to the V-crossing chairs and check chairs, but so far I haven't even thought about K-crossings and slips. It's going to be a while before I get that far. Until then you will need to design all those chairs yourself unless you use non-plug moulded chairs from C&L/Exactoscale.

Or you could do some kit-bashing on Wayne's Finetrax kits -- he is just now doing the diamond-crossings. Also of course Peco have just announced the availability of their diamond-crossings and slips matching their "large" radius 00 bullhead turnouts. The difference with those options of course is that the sizes are fixed and straight -- FinetraX is only 1:7 so far and Peco is all 1:4.7 for 00. Plug Track will be for any angle, scale, gauge, or radius.

The C&L rail is a much better section than the one you are using. Many users of it are now using the Peco bullhead fishplates with it, so it must obviously be compatible. The smaller rail foot would make it easier to design the inside chair fixings and retain a clearance for RTR wheel flanges.

I have no plans to export the files in any format other than DXF and STL. There are plenty of online conversion tools to create other formats if needed. I don't know why you feel you need them, bearing in mind that the Chitubox and Cura slicer programs work fine with STL files. Likewise with the online STL mesh repair tools -- which are currently needed with the Plug Track export, although I'm hoping one day to export clean STL files which don't need fixing. The 2-D DXF files will also be available in EMF for users without any CAD program.

I'm currently using TurboCAD DeLuxe to create the DXF renderings, although not for any design work on the chairs. That is all done in Templot.

Yes, there is already a button on the DXF dialog to export only chairs in the 3-D files, and several tickboxes to control which chair elements are wanted (click chairs only first before changing them):

Yes, there will be a bunching function to close up chairs (and timbers if wanted) for resin printing. However that's not yet ready -- and it's much more complex than might at first appear, because the chairs are all at different angles. And a paper printed map sheet is needed to indicate which chair is which. Again not simple because a timbering brick may contain some very similar chairs from different templates but which are not actually interchangeable.

cheers,

Martin.

 

[Stephen Freeman]

Sorry to butt in but another vote for Turbocad Deluxe, mine is 2015, which I use for preparing DXF files for etching by PPD, so haven't really used it for 3D except for visualising a chair that was posted some time ago.

 

[Martin Wynne]

I tried adding some fill below the key, which seems to be working ok

@Terry Downes

p.s. Terry,

Just to add that if you use the fill below key option, you need to check that your actual rail foot width matches your custom rail setting. The filler chunk will be in contact with the rail foot, and if the rail foot is too wide the chair is likely to fracture.

The rail fit - fine adjust setting applies only to the key, not to the filler chunk. The key has a taper on each end and a small pressure pad area at the centre. This is intended to aid threading the chair onto the rail, and possibly to wear the pad a fraction as the chair slides into position to reduce any stress on the chair. The filler chunk does not have any such allowance.

Resin-printed chairs are likely to harden and become more brittle with age, especially if not painted, so any residual stress may lead to age-fracture. The jury is still out on this for all resin-printed chairs, not only Plug Track.

Generally I have found that the chairs work well and hold the rail without undue stress if used without the filler chunk. i.e. as the prototype with the rail held only by the key.

cheers,

Martin.

 

[Martin Wynne]

View attachment 3206

There is an obvious bug in Templot on the lower face of the part I call the "grip" (ringed red) which explains this strange feature:

Thanks for finding it, I will get it fixed for the next program update.

@Terry Downes

Hi Terry,

This is now fixed in 233d:

https://85a.uk/templot/club/index.p...date-version-233d-now-available.404/post-3774

cheers,

Martin.

 

[Terry Downes]

.
Sorry there was no update for a while, I've been busy with some electronics work.

But back on Templot today, working on the P chair (slide chair).

I've decided to go with the PJ version of the REA chair:

because it's a bit beefier and stronger on model rail sections. Maybe I could do the older P type later as an option.

View attachment 3035
Outer jaw still to do (you noticed ) but I've got the slide table and stock rail bolt and boss done. The bolt is at 1:20 to match the prototype, even though the model rail will be vertical.

The slide table is 7" wide, which means for REA switches the switch blade tip should align exactly with the edge of the table. Which will be a convenient guide when track building. The length of the chair and slide table is increased from the prototype to suit the model gauge/scale in use where needing an over-scale switch opening (for 00, EM, etc.).

The nut rotation on the stock rail bolt is part-randomised in 15-degree steps (the chair-screw heads are infinitely randomised).

I'm omitting the spring washer under the nut, the screw-thread on the exposed end of the bolt, and the domed end of it -- I have to draw a sensible line somewhere. Otherwise it will take forever and the STL file size will go through the roof.


View attachment 3036

View attachment 3037
Now to get the jaw done.

cheers,

Martin.

Hi Martin,
I've just modelled up the P chair referencing the PermanentWay1928.pdf and it appears shorter than your PJ chair you have modelled. Does this sound correct? I can not find a drawing of the PJ slide chair within the 1928.pdf or am Going blind?

Terry

 

[Martin Wynne]

@Terry Downes

Hi Terry,

The PJ chair was a 1950s revision of the P chair, so not shown on the original 1928 REA drawings.

The PJ has a separate insert spacer in the rail web, instead of a one-piece outer jaw as in the P chair. The original P chair was subject to frequent breakage in heavy traffic. The PJ version is beefed up and stronger -- the jaw has a solid base and does not have to fit the rail section. Also for maintenance, the slide-out spacer means the toe timber and chaired soleplate can be dropped below the rails and replaced in one piece, instead of needing to have the chairs removed and reassembled:

Sorry, I don't have a detailed drawing of the PJ chair, only a small-scale outline drawing. Looking at what I've done with it so far, I think perhaps the jaw could be a bit beefier.

The Templot chairs are not exact scale models for several reasons:

1. in Templot the model rail is vertical instead of inclined. This changes the inner and outer lengths of the chair relative to the rail gauge-face.

2. model rail sections are generally a poor representation of the BS-95R prototype, usually with a significantly excessive web thickness. This affects the design of the outer keys and spacers, and the inner chair jaws.

3. most model gauges have over-scale flangeway gaps (including P4) which affects the special chairs design and the amount of switch opening needed -- the model P chairs are adjusted in slide length accordingly.

4. the prototype chairs are iron castings from smooth patterns having draft angles and fillet radii all round, which if included in the 3D file sends the file size through the roof.

However, if/when I get all this finished, it will be possible to enter custom chair designs in Templot for any prototype -- it will be for the user to make it fit the model rail and deal with the 3D file sizes.

cheers,

Martin.

 

[Martin Wynne]

.
Just bumping this topic. See:

https://85a.uk/templot/club/index.php?threads/kit-building-on-templot-plans.483/post-4519

Martin.

 

[Martin Wynne]

.
It's a long time since I posted any evidence of progress, so here are some chairs on a test brick.

It is all but impossible to photograph the translucent chairs as 3D-printed -- they have to be translucent so that the UV light can cure them all through. They need a scoosh of rattle-can primer to be properly photographed.

EM gauge. Spot the deliberate mistake*:

S1J joint chair, S1 ordinary chair, P slide chairs:

5 L1 bridge chairs and 5 S1 ordinary chairs:

It's proving tricky to get the socket sizing right for the L1 bridge chairs where they are close together. Making them a tight push fit means the first one distorts the base a fraction and makes the second one more difficult to fit.

It's no problem on the CNC-milled base, where the MDF material is soft enough to give, but the FDM polymer is made of tougher stuff.

I've got a few ideas to try. It's important, because there will be similar issues with the special crossing chairs.

*I used one wrong joint chair -- keys have to driven towards the rail joint, because of the fishplate.

Martin.