Printing Turnouts on a 3-D printer

[mightbe]

It is quite good-looking..the only thing I can fault is the attempt at a wood grain on the sleepers, which to my eyes would be better off being left out. A few seconds with a metal pick would probably yield a better result. 

 

Also I'm quite intrigued by the idea that you've built in pockets to make misalignment of the wing rails nearly impossible. Could we see some closer shots to illustrate?

 

Perhaps the easiest solution for power the wing rails and the vee is to attach a dropper to each one very quickly (and using heat sinks) and then combine the four wires into one before heading over to the polarity switch.

 

Quentin

[trustytrev]

Hello Andy

               That is an impressive result you have acheived. People manage to solder droppers onto rail with plastic chairs and timbers from the likes of C&L/Exactoscale. With a bit of practice and care the same should be possible with printed components. If it was possible to print these things in ABS the results would be weather proof and virtually indestructable. I agree with Quentin with regards to the droppers wingrails and V's.

trustytrev.                                                              

[AndyID]

Hi Quentin and Trev,

 

The pseudo-woodo effect you see is quite accidental     I was experimenting with different infill patterns, and that sample happened to have a diagonal infill. It should be possible to instruct the printer to fill the top layer of the timbers to 100%, but I have not been worrying too much about that so far.

 

Here's a reject sample that has a linear infill pattern, but I think it should be possible to produce something a bit better.

 

 

Here is the close-up (cruel!) of the vee with the rails retracted. Nothing very fancy about it, but it seems to do the trick. There is a similar solid piece between the check rails and the stock rail. If they are not soldered together, I'd be inclined to glue the vee rails in place as the last step in the assembly.

 

 

 

I'll do some experiments to see if it's possible to solder droppers on without the whole thing turning into a blob of molten plastic. In terms of reliability, solder attachment would be the first choice. 

 

Cheers!

Andy

 

Edit: Speeling Error

[AndyID]

Later:

 

Silly me! I just realized there is no particular reason why droppers could be soldered to the far ends of the vee and wing rails before they are slid into place. However, I'll attempt to solder them with the rails in place on a reject base and see what happens

[AndyID]

Looks like, if you are very quick, you can solder without too much damage. The vee is soldered (pay no attention to the poor filing). The plastic has melted there slightly, but I don't think it would cause a problem.

 

There is also a dropper. The chairs don't look too good, but that's mainly because they weren't very good in the first place.

 

I will probably try to attach droppers at the rail ends before assembly. In situations where that is not practical, I'll try weaving the rail through a shortened rail-joiner that has a dropper attached to it.

 

 

[AndyID]

Going a bit tangential for a minute (another thread perhaps?) I need to make quite a few "offstage" points for the storage loops on my layout. How they look does not matter in the slightest, but they will have rather tight radii and I'm thinking of making the facing ones "swing nose" (or similar to Wrenn or old Hornby O points). I suspect I'll be able to crank out the bits on the printer fairly quickly. Come to think of it, the trailing points will likely have sprung blades and isolated frogs. Short length isolated frogs should not be too difficult to print either.

 

Speaking of insulated frogs, how about insulated turnouts? Practical battery operation with radio control can't be too far off, even in OO. I don't think it's too ridiculous to consider printing the entire turnout, including the rails. It would actually be a lot simpler than trying to print functional chairs. I might give it a shot just to see how bad it looks.

 

Another idea is to print the bottom half of the rail with the base and glue/clip a metal rail head on top.

 

OK, I'm rambling!

[martin_wynne]

Another idea is to print the bottom half of the rail with the base and glue/clip a metal rail head on top.

 

Another idea is to locate the rails in a jig on the printer and print the base around them.

 

This is called insert-moulding and is the way some Peco track is made.

 

Martin.

[28XX]

The inserts would obstruct the extruder nozzle.

[Joseph_Pestell]

Going a bit tangential for a minute (another thread perhaps?) I need to make quite a few "offstage" points for the storage loops on my layout. How they look does not matter in the slightest, but they will have rather tight radii and I'm thinking of making the facing ones "swing nose" (or similar to Wrenn or old Hornby O points). I suspect I'll be able to crank out the bits on the printer fairly quickly. Come to think of it, the trailing points will likely have sprung blades and isolated frogs. Short length isolated frogs should not be too difficult to print either.

 

Speaking of insulated frogs, how about insulated turnouts? Practical battery operation with radio control can't be too far off, even in OO. I don't think it's too ridiculous to consider printing the entire turnout, including the rails. It would actually be a lot simpler than trying to print functional chairs. I might give it a shot just to see how bad it looks.

 

Another idea is to print the bottom half of the rail with the base and glue/clip a metal rail head on top.

 

OK, I'm rambling!

 

The big question with insulated crossings (or indeed entirely plastic track) is the durability of the plastic.

[AndyID]

Another idea is to locate the rails in a jig on the printer and print the base around them.

 

This is called insert-moulding and is the way some Peco track is made.

 

Martin.

 

Hi Martin,

 

You seem to be up early!

 

I don't think that's possible/practical with a typical 3-D printer. Even if the software was set up so that the printer's extruder nozzle didn't collide with the rail, the nozzle's aperture would not be able to get close enough to the places on the chairs where it had to deposit material.

 

Cheers!

Andy

[AndyID]

The big question with insulated crossings (or indeed entirely plastic track) is the durability of the plastic.

 

Hi Joseph,

 

Yes, that is a concern. However, PLA is really quite tough/hard stuff. It's not easy to cut it with a knife - a saw or a file works much better. And there are other materials, ABS for example, which is a lot more heat resistant.

 

Unfortunately, my low-end printer is not up to ABS in it's current form, but it might be a better option.

 

Cheers!

Andy

[hayfield]

Andy

 

Looks very good indeed, 70 degree solder may be the answer with pre tinned wire, might also work for the tip of the vee, just need a very light touch on top and sanding flat. The wing rails could be attached to the switch rails by cast C&L/Exactoscale cast metal fishplates, Touch of low melt solder will do the trick, also at the join bonding wires from the switch to the stock rails will sort out the electrics

 

Would be nice if the solid flangeway of the wing rails could be gapped, flat sleepers would look much better.

 

As I previously said, a great proving ground for the idea. The suitability of the material and quality of printing will be a future progression as will be design tweaks. For instance you could print this part in 3 pieces, first break perhaps near the tip of the vee to enable soldering, next at the knuckle at the bend of the wing/closure rail occurs, though you may have proved these unnecessary. 

 

Well done, how about a diamond crossing 

[AndyID]

Andy

 

Looks very good indeed, 70 degree solder may be the answer with pre tinned wire, might also work for the tip of the vee, just need a very light touch on top and sanding flat. The wing rails could be attached to the switch rails by cast C&L/Exactoscale cast metal fishplates, Touch of low melt solder will do the trick, also at the join bonding wires from the switch to the stock rails will sort out the electrics

 

Would be nice if the solid flangeway of the wing rails could be gapped, flat sleepers would look much better.

 

As I previously said, a great proving ground for the idea. The suitability of the material and quality of printing will be a future progression as will be design tweaks. For instance you could print this part in 3 pieces, first break perhaps near the tip of the vee to enable soldering, next at the knuckle at the bend of the wing/closure rail occurs, though you may have proved these unnecessary. 

 

Well done, how about a diamond crossing 

 

Hi there John,

 

Thanks for the input on LMP solder. I'll try to track some down here and give it a shot.

 

Not sure I agree about the solid flangeways. It's a question of form over function, and I think in this case function is more critical. Unless you look vertically down on the turnout, it's almost impossible to see the infill piece. After ballasting and painting, I'm pretty sure very few people would even know it's there.

 

Anyway, this is all about personal choices, and that's what 3-D printing allows. At this point I think I've demonstrated what is referred to in the US as a "proof of concept". I'm naturally a lazy person, and I'm always looking for ways to reduce repetitive effort. I'm also a cheap (economical) person, and I've always been reluctant to fork over the amount of money manufacturers ask for points! (Like for 50+ years  )

 

That's why this appeals to me. The added benefit is that it might allow me to crank out all sorts of turnouts from Templot with minimal effort. (Thank you Martin!)

 

I'll keep chugging along and apply some of it on my own layout, but I'm hoping others might find this sufficiently interesting to take it to the next level. I can provide a limited amount of support to those who want to do that.

 

And now, to bed.

 

Cheers!

Andy

[martin_wynne]

The inserts would obstruct the extruder nozzle.

 

Maybe not if the track is printed upside down?

 

Martin.

[Joseph_Pestell]

Maybe not if the track is printed upside down?

 

Martin.

 

I don't think that would work. But what if one stopped the print part way through, placed the rails, and then restarted?

[AndyID]

I don't think that would work. But what if one stopped the print part way through, placed the rails, and then restarted?

 

I don't see how that could work either

 

You'll still run into the problem of trying to position the extruder where it cannot go. If the extruder was something like a hypodermic needle with an omnidirectional positioning system it might work, but there would be no way to conduct sufficient heat to the tip to keep the plastic liquid. The extruder nozzle has to be a big chunk of heat conducting metal e.g. brass.

 

I suspect there would be another problem when the hot extruded material came in contact with the metal rail. It would immediately cool down and harden and I doubt if it would adhere to the rail. This will probably result in a mess of loose plastic. The printing process relies on the new plastic heating the previous layer sufficiently to fuse with it.

 

I've lost sight of what problem this is trying to solve. Is there some major objection to sliding the rail through the chairs?

 

If that's the issue, it might be simpler to make the chairs entirely cosmetic. The rail could be dropped in from above and glued in place. The key side of the chairs could still be made to engage with the rail's web, but the inside (wheel flange side) would not overlap the base (foot?) of the rail.

[AndyID]

Here's a "drop-in" chair, although "roll-in" might be a better description. The key side of the chair is unmodified. It's still pressing against the rail's web and it is actually locating the rail in the correct vertical position.

 

Those skilled in the art could probably add a small blob of some sort of adhesive/filler to replace the material on the inner side of the chair.

 

EDIT: Or, the inner side of the chair could have a vertical "flap" that could be heated and bent over to hold the rail in place. Hmmmm??? I wonder how that would work. Might give it a shot.

 

 

 

[28XX]

Drill the bottom of the rail and solder in a length of fine wire. Form the wire to run tight to the bottom of the rail along its centreline

 

Mould the chairs with a slot in the bottom to correspond with the wire position

 

Lead the rail into the chairs, at the point where the dropper is to exit guide the wire under the sleepers.

[Grovenor]

Here's a "drop-in" chair, although "roll-in" might be a better description. The key side of the chair is unmodified. It's still pressing against the rail's web and it is actually locating the rail in the correct vertical position.

 

Those skilled in the art could probably add a small blob of some sort of adhesive/filler to replace the material on the inner side of the chair.

 

EDIT: Or, the inner side of the chair could have a vertical "flap" that could be heated and bent over to hold the rail in place. Hmmmm??? I wonder how that would work. Might give it a shot.

 

 

DSCN1674.JPG

 

DSCN1675.JPG

Perhaps think a bit more like the real thing, keep the chair complete but leave the key out of the print, then you can drop the raqil in and add real wood keys to hold it.

Regards

Keith

PS. I don't see an objection to sliding rails in if you can do it, you seem to have managed very well with the wing rails and checkrails in the crossing above, puzzles me how you got the bent rails to slide in.

[AndyID]

Perhaps think a bit more like the real thing, keep the chair complete but leave the key out of the print, then you can drop the raqil in and add real wood keys to hold it.

Regards

Keith

PS. I don't see an objection to sliding rails in if you can do it, you seem to have managed very well with the wing rails and checkrails in the crossing above, puzzles me how you got the bent rails to slide in.

 

Hi Keith,

 

I actually tried that at one point, but I was not able to make the key side of the chair well enough defined and sufficiently robust. However, I've been adjusting the printer settings and the chairs are a lot stronger now. Maybe I will take another shot at it. The problem is there are an almost infinite number of things I could attempt, but then I'm not getting on with my layout! 

 

It's not entirely clear to me how I got away with sliding the wing rails in either. I presume the rail was able to bend enough without taking a set, and I suppose there was a bit of compliance in the plastic base. Next time I try it I'll pay more attention.

 

Cheers!

Andy

[AndyID]

3-D printing is a tricky business.

 

Here is a test piece of a timber with the first couple of layers of some chairs attached (the chairs should all be identical, but obviously they are not.)

 

The finely crafted distressing of the timber is a complete fluke. It's an artifact of the printing process (mind you, it looks quite good from here.)  

 

On the other hand, I am pushing the limits of my rather inexpensive printer, but the results are probably quite satisfactory for my needs.

 

[raymw]

Hi Andy,

 

The problem you are finding is not unusual. It is a lack of precise control of the filament, the usual software will not do what you want.

 

Although this is not the way you want to do it, you would get better visual results by printing the sleepers separately, with one 'resolution', and do the chairs with a finer resolution, perhaps on a sprue for easier handling. Then glue the bits together. That would make it easy to solder on wires, and slide in the rails, etc. I think it would be quicker in printing, and also most likely to assemble. However, if you enjoy the challenge, keep on doing what you are doing, at least you're finding the limitations of this type of 3d printing.

 

Another approach could be to use a few pcb soldered sleepers where required, perhaps with glued-on cosmetic 3d printed cosmetic chairs.  If you were working in gauge1, then you would get far better results, the stringy pattern would be a better match for wood grain, if you could control the direction of the final layers.  Also, if you have a number of parts, if one is screwed up, the whole thing does not need reprinting cf printing the whole thing as one part. If you can get the solvent I mentioned, and a syringe, you'll quickly change your approach, I expect.

 

If you are printing pla, then try and keep the temp/humidity the same from day to day, it may help to enclose the bed/machine, even cardboard could stop the draughts. Don't be embarrassed to buy cheap hairspray, to get the first few layers stuck to the bed.

 

Best wishes,

 

Ray

[unravelled]

The results look very convincing, and I'll be watching the progress with interest. If you can sort a single slip it would be just what I need!

 

I have a few silly ideas to suggest, and hope they might be of some use. If I'm not making myself clear, please say and I'll amplify with some diagrams.

 

If you omitted the half chairs on the outside of the rails leading up to the vee, (those between the arms of the vee), a presoldered vee could be slid in and cosmetic chairs added later.

 

For the checkrails, a variation of your drop in chair would be to make the base without any chair parts between running and check rail. The running rail could then be slid in as usual, with the check rail dropped in to be finally be located with some sort of wedge/key slid or dropped in.

 

Removing other location material at the frog might allow the wing/running  rail sections to be slid in before the vee is located. Alternatively could the wing rail be a  separate piece of rail. Or should the wing (and check?) rails be printed in plastic. This helps with the problem of shiny non running surfaces after track cleaning although it is usually disliked in rtr points.

...

Last one, and potentially very silly. As an extreme case how would track operate with inside and outside half chairs on alternate sleepers? Practically applied I would suggest cosmetic printed other halves, but with clearance so the rail can be dropped in. There is a risk that this might be end up with a wiggle in the rail, but could be worth trying.

 

I hope there's something here which is of use, for all I know you've tried and discarded them already...

 

Keep up the good work

 

Thanks

 

Dave

[AndyID]

All,

 

Thanks for all the ideas!

 

Nothing very exciting to report. I spent the last couple of days adjusting the printer settings and printing out a large number of test pieces like the one above. I think I've now achieved the sort of print quality and consistency I was hoping for. One criterion is that the bolt heads all have to be visible without being enormously overscale.

 

As mentioned above, I'm lazy. Ideally I'd like the printer to do all the "heavy lifting" so that I only have to do a bit of filing and slide in the rails. (Why buy a dog and bark yourself?) I'd prefer not to have to attach any other bits if at all possible. This might not produce the most faithful reproduction of the prototype, but it will probably work for me, and I suspect it would work for others.

 

One of the problems with printing (there are many!) is producing an overhang. The printer will create them, but the lower surface has to be angled up rather than horizontal. The only way to create a horizontal under surface is to put in a bridge layer then cut the bridge after the part is printed. I may have to resort to that, but it's not my first choice.

 

OK, so all I have to do now is perfect the chair design!

 

Cheers

Andy

 

EDIT: Added the dog bit.

[AndyID]

....and, after quite a lot of messing about, I came up with this.