Jump to content
 

Printing Turnouts on a 3-D printer


AndyID

Recommended Posts

  • RMweb Gold

Hi Andy,

 

Is it necessary to print plastic timbers? Would it be possible to lay out plywood timbers on a template or jig, and then print the chairs onto them? This would not only look good, it would save a lot of printing time?

 

regards,

 

Martin.

Link to post
Share on other sites

Hi Andy,

 

Is it necessary to print plastic timbers? Would it be possible to lay out plywood timbers on a template or jig, and then print the chairs onto them? This would not only look good, it would save a lot of printing time?

 

regards,

 

Martin.

 

Hi Martin,

 

Yes, I think it's possible to do something along those lines, although I'm probably too lazy to do it that way. I like to be able to launch the printer and watch TV (or visit RMWeb) while it does all the "heavy lifting" :)   I only keep an eye on it while it's printing the first layer. That's the most critical part of the process, but my new printer gives me very few problems.

 

Obviously it would require careful registration (horizontal and vertical) between the timbers and the printed chairs, but that should not be too difficult. I'm not sure if the PLA would adhere all that well to the plywood, but it might be OK. I'll try a quick test to see.

 

If it does not adhere an alternative might be to print the chairs on a very thin web to interconnect them, then cut away the web after the chairs were glued down.

 

Best,

Andy

Link to post
Share on other sites

  • RMweb Gold

Thanks Andy.

 

What is the polymer you are using? Is it possible to obtain in sheet form? Timbers could be cut from it, and the hot chairs would presumably adhere well if they won't stick to plywood.

 

What timings are we talking about? How long does it take to print 1 timber? 1 chair?

 

regards,

 

Martin.

Link to post
Share on other sites

If going down the route of gluing chairs to sleepers, then may as well mass produce standard chairs, similar to ccl, etc.  If you are doing other things, then time taken printing does not matter much. I remember when windows used to take long enough to boot that I could make a cup of tea, then it became annoyingly quicker, had to sit watching it for 30 seconds, whatever.

 

If the pla does not stick too well to bare wood, then maybe an intermediate coat of varnish may help. But, i think Andy's current method is the best, a very simple process wrt assembly, apart from threading the rails (which will still need doing, whatever system was used).

 

All you need, Martin, apart from a printer, is to add a few 'logic statements' to your software.

Link to post
Share on other sites

Thanks Andy.

 

What is the polymer you are using? Is it possible to obtain in sheet form? Timbers could be cut from it, and the hot chairs would presumably adhere well if they won't stick to plywood.

 

What timings are we talking about? How long does it take to print 1 timber? 1 chair?

 

regards,

 

Martin.

 

Hi Martin,

 

It's fairly slow. At the moment as I'm more concerned about the quality I have not been paying a great deal of attention to the speed, but I think the complete turnout above took about one hour in total.

 

The simplest way to speed up the process is probably to use a double extruder with a small diameter nozzle for printing the details and a larger nozzle for bashing out the timbers. I'm limited to a layer thickness of 0.2 mm at the moment. The timbers are 1.2 mm thick, so it takes six layers. That could be cut in half with a larger nozzle. It likely the XY motion could be sped up for printing the timbers too, but it's probably still going to take about half an hour to print a reasonable size turnout. I could also print both pieces in one pass but as I'm still developing the chair models I usually just print one piece at a time.

 

The material I am using is PLA. I don't know if it's available in sheets - might well be.

 

(I see Ray just posted. He has a lot more experience with different techniques/materials than I have.)

 

Best,

Andy

Link to post
Share on other sites

All you need, Martin, apart from a printer, is to add a few 'logic statements' to your software.

 

I keep trying to tell him that, but I think he's ignoring me. Don't know what the problem is. I mean, how difficult could it really be?

Link to post
Share on other sites

  • RMweb Gold

I keep trying to tell him that, but I think he's ignoring me. Don't know what the problem is. I mean, how difficult could it really be?

 

What exactly do you need? Templot already creates 3D CAD output in DXF format for the timbers and rails. There is no chairing information in Templot because it is a prototype minefield.

 

I hope and intend to add chairing one day, but it is not a simple task. If chair files are available in wire-form DXF format from elsewhere (are they?) it would be possible to insert them into the DXF output without necessarily integrating them into the full program.

 

regards,

 

Martin.

Link to post
Share on other sites

I keep trying to tell him that, but I think he's ignoring me. Don't know what the problem is. I mean, how difficult could it really be?

 

My programming skills are rather limited (although I did once produce a really neat version of "battleships and cruisers" that ran on a 1K by 16 bit computer. Of course, I had to design and build the computer first - but that's another story.)

 

Oh, I see you just posted Martin.

 

I was being facetious of course, and I think you've done quite enough for us already. I was thinking someone "skilled in the art" (not me!) might be able to knock out a script that would take a Templot template and populate it with desired components.

 

My chairs were created in TurboCAD and can be exported as STL, DXF or lots of other formats. The chairs only need to be oriented and positioned. There is no need to merge them into a single solid or surface. I create an STL that includes all the elements, then run it through Netfabb to plug any holes before it goes to the slicer program that generates the gcode for the printer.

 

Best,

Andy

Link to post
Share on other sites

post-25691-0-00873700-1449293529_thumb.jpg

 

Don't be fooled by the straight webs between the timbers. Both roads are curved.

 

I think this should test the dimensional accuracy of the printing process a bit more than a turnout. If it passes it will be on my layout.

 

Keen observers will notice I had to remove a couple of chairs because they were oriented 180 degrees off in the print. I had to use a file because you can't cut this PLA stuff with a knife.

 

EDIT: Sorry for all the hype.

Link to post
Share on other sites

  • RMweb Gold

attachicon.gifDSCN2528.JPG

 

Don't be fooled by the straight webs between the timbers. Both roads are curved.

 

I think this should test the dimensional accuracy of the printing process a bit more than a turnout. If it passes it will be on my layout.

 

Keen observers will notice I had to remove a couple of chairs because they were oriented 180 degrees off in the print. I had to use a file because you can't cut this PLA stuff with a knife.

 

EDIT: Sorry for all the hype.

 

What hype? I think you are being very modest with your progress on something that could revolutionise railway modelling.

Link to post
Share on other sites

Andy

 

Looks very interesting, nothing to do with the ease of printing or fitting on to the plate, but from a builders view. Perhaps from an interchangeable size view

 

My thoughts would be to separate the module into 3/4 sections. I can see the benefit of printing all in one go though

 

1  The switch

2  The closure section

3  The crossing section (either printed in 2 or cut into 2 after printing)

 

The benefit of this way is that the rails in the common crossing can be formed then slid into position for the builder as can the switch and closure rails. Then the switch/ closure/crossings can be interchanges for differing sizes

 

How did the diamond crossing come out?

Link to post
Share on other sites

What hype? I think you are being very modest with your progress on something that could revolutionise railway modelling.

 

Hi Joseph,

 

The "hype" was a cheap shot at a previous comment :)

 

I'm not sure I would describe this as revolutionary, but if it gives the manufacturers a good nudge, that works for me.

 

Andy

Link to post
Share on other sites

Andy

 

Looks very interesting, nothing to do with the ease of printing or fitting on to the plate, but from a builders view. Perhaps from an interchangeable size view

 

My thoughts would be to separate the module into 3/4 sections. I can see the benefit of printing all in one go though

 

1  The switch

2  The closure section

3  The crossing section (either printed in 2 or cut into 2 after printing)

 

The benefit of this way is that the rails in the common crossing can be formed then slid into position for the builder as can the switch and closure rails. Then the switch/ closure/crossings can be interchanges for differing sizes

 

How did the diamond crossing come out?

 

Hi John,

 

I was thinking along the same lines and I think that might be a way to simplify things a bit.

 

I'm still trying to work out a good way to thread the diamond. I think I have it now, but I'll probably change my mind :)

 

Andy

Link to post
Share on other sites

When you creat a curved point or crossing, how do you align the sleepers and chairs on the templot template. Do you do it by eye or is there a process?

 

Hi Tim,

 

I output the template as a 2D DXF from Templot and load it into TurboCAD. Then I can snap on the corners of the sleepers/turnouts and extrude them into 3D solids.

 

I'll try to post some pix tomorrow. It's getting a bit late here!

 

Andy

Link to post
Share on other sites

My initial thoughts would be to have the bases in 5 sections

 

Cut at the knuckle of the common crossings, and have the central sleeper separate

 

I am thinking you can then bend both stock rails and all 4 wing/closure rails, with the closure rails being in 2 parts. You could make the stock rails in 2 parts but the angle should be bent, not acute. If made in 2 parts then the rail ends in the centre need rounding off 

Link to post
Share on other sites

Hi Tim,

 

I output the template as a 2D DXF from Templot and load it into TurboCAD. Then I can snap on the corners of the sleepers/turnouts and extrude them into 3D solids.

 

I'll try to post some pix tomorrow. It's getting a bit late here!

 

Andy

Thanks, I'd considered that would be a possibility after I posted. Would be interested to see how you align the chairs too.

Link to post
Share on other sites

  • RMweb Gold

I'm still trying to work out a good way to thread the diamond. I think I have it now, but I'll probably change my mind :)

 

Hi Andy,

 

Is the prototype method out of the question? Drop in the rails, insert separate keys?

 

Perhaps snap-off keys on the end of a stick. Then a drop of solvent or capillary cyano to secure them.

 

Slaters do separate keyed track parts in 7mm scale.

 

Martin.

Link to post
Share on other sites

  • RMweb Gold

I'm still trying to work out a good way to thread the diamond. I think I have it now, but I'll probably change my mind :)

 

 

 

I don't think that it should be that difficult. But you will need to thread it right through from the ends. Just a matter of doing it all in the right order.

 

Or printing the base in three parts could help as suggested by John.

Link to post
Share on other sites

Hi Andy,

 

Is the prototype method out of the question? Drop in the rails, insert separate keys?

 

Perhaps snap-off keys on the end of a stick. Then a drop of solvent or capillary cyano to secure them.

 

Slaters do separate keyed track parts in 7mm scale.

 

Martin.

 

Martin

 

I thing if the plastic is strong enough threading the rails through in the correct sequence with sleeper bases separated where the rail bends are ,the sleepers then can be thread from both ends on to a bent rail. The keys can then be printed on the chairs

Link to post
Share on other sites

  • RMweb Gold

the sleepers then can be thread from both ends on to a bent rail.

 

Hi John,

 

I must be missing something because I can't see how that works. For the wing rail knuckle bend for example. You could do the first rail by sliding on the timbers from each end, and then bending the flare angle on the end of the rail (although strictly the flare occurs within the B chair).

 

But how do you then do the other wing rail?

 

Puzzled.

 

Martin.

Link to post
Share on other sites

Martin

 

I am hoping that the chairs have a bit of strength, the base starts with the A chair position onwards. Bend both wing/closure rails including the flares, with the closure part being about 3 sleepers long, slide one rail on, then the next. Now slide the 3 timber second part on (X,Y & Z chairs). just hoping the chairs are flexible enough (as C & L check rail chairs are) not to distort whilst the rails are threaded. I think that is what Andy did for the common crossing. The vee having to be slid in place then soldered

Link to post
Share on other sites

Andy to get back to the i3 printer ( maybe this should be in another thread)

 

Would you recommend buy the frame   separately (and all the bits separately )  and spending  money on a high res extruder , or would you go for a complete " genuine " kit like the Prusa ( about 600 euros ) 

 

 

dave 

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...