Printing Turnouts on a 3-D printer

[BG John]

What would be really interesting would be baulk road track and pointwork, especially for broad gauge, but narrow (standard) and mixed gauges too!

[rue_d_etropal]

does look good, but strength is the big issue. Others who have used hard plastics have found them very fragile, and get a lot of failures when fitting rail. Also does anyone know long term effect of UV and other things found on layouts.

[AndyID]

does look good, but strength is the big issue. Others who have used hard plastics have found them very fragile, and get a lot of failures when fitting rail. Also does anyone know long term effect of UV and other things found on layouts.

Strength is not an issue if the parts are printed properly. It takes some effort to learn how to fuse the extruded filament correctly.

 

In the example above I pulled the rails through the chairs and the chairs aligned the rail.

[AndyID]

Andy

 

Looks excellent and perhaps something we all thought not achievable so quickly a few months back. Just to overcome fitting the common crossing together is impressive. How about a slip next 

 

Would be interesting seeing a standard RTR offering next to it just to compare sleeper size/spacing and to see the difference in the geometry of the turnout

 

 

Hi John,

 

I'm going to attempt a crossing next. Some interesting shortcuts can be made, but you'll have to wait and see

 

Unfortunately I don't have any recent RTR turnouts here for a fair comparison. Also, bear in mind that this is made in my funky 1:82 scale. I plan to complete a small section of my layout with printed turnouts and track, ballast it, and see how it looks.

 

Now that I have completed the models for the various components of the chairs (there are fifteen different bits) I should be able to crank out the next turnouts very quickly. If not, all the work I put in will have been a big waste of time.

 

I might even get a (short) train running by Christmas, but I'm not holding my breath.

 

Cheers!

Andy

[rue_d_etropal]

Andy,

I am basing what I say on what others have found using FUD on Shapeways. I had no problems using WSF for some large scale sleepered track, as long as I pre-bent rails for curves. When testing any idea it is always better to fully test something, and that should include effects I have mentioned, which are known problems with some harder plastics. It is not just a case of setting something up 'correctly' and letting it go. In my IT background experience, that quite often results in a failure in the future, and the excuse 'never thought that would happen'.

It is great to try out 3D printing, but if using your own machine it takes a lot of setting up, and maintenance, and then there is the design work on the computer. I find it takes up enough time just designing the items I want printing, and there has been a lot of trial and error, and compromise. I now have a pretty good idea how far I can go with designs. I just don't want people thinking it is an easy process.I no have hundreds of designs, but if I had been using my own printer, not only would some of those designs not be possible, but I would probably only have a handful of successful prints. 

I had trouble finding this thread again, as I had assumed it was under the '3D printing ' section. There is a lot of useful advice there.

[raymw]

I think Shapeways and others use a different 3d printing process, Andy's printer is extruding PLA. PLA, in some ways is better/more stable than ABS, and in a normal domestic indoor environment, I think any long term problems will be little different than for, say, Peco track. PLA is often used as medical implants/structures, and unless you decide to eat your track-work, not much seems to effect it, afaik., (for example the common plastic solvents seem to not work too well).  I've got some printed bits of PLA from a couple of years back, and they still seem the same as when I first made them. The rate at which some folk build/change layouts, 2 years would be plenty, but I see no reason why PLA should deteriorate differently, than say polystyrene or ABS, in normal circumstances. However, i think, (not yet tested) that being based on cornstarch/similar, PLA may be a tasty meal for mice or snails, so may not be the best choice for outdoor track.

[rue_d_etropal]

2 years is not very long, and the person building the layout might have finished with it, but passes/sells it onto someone else. It might therefore be seen as irresponsible using a material withich might break down . By comparison plastic used in Peco points and flexi track seems to last a lifetime. 

Styrene plastic does deteriate , in fact it tends to get hard and brittle, which is not something good for track. Last year there was also a bit of a scare with some track cleaning fluid and Peco track. Peco investigated, and as far as I remember they said it was the cleaning fluid. I have not seen anything recently about it, but have seen adverts for the cleaning liquid.

I remember many years ago when WD40 hit the shelves, and like others I tried in on my locos, not realising then that it could damage the plastic.

I am not trying to stop anyone trying out 3D printing, just saying there might need to be a bit of caution.

[hayfield]

I must admit not knowing too much about the materials used either with home printers or companies like Shapeways, longevity of the products has been my concern. Certainly as time goes by the products will improve.

 

What is important is that perhaps in the future printed turnouts will be available, no doubt in 00 gauge (4 mm scale) and with a choice of chairs/clips, bullhead or flatbottom etc. It could be from a company or printed at home. Unlike injection moulding where large set up costs are involved 3D printing will make items like this very much cheaper and/or with a larger range and options

[AndyID]

Simon,

 

Your information is interesting, but it is also anecdotal. (I also find it quite entertaining that FUD is a well-known acronym for Fear Uncertainty and Doubt.)

 

Anybody reading through this thread is hardly likely to rush out a buy a 3-D printer and think they'll be printing their own turnouts a week later. It's pretty obvious that I'm conducting an experiment. That said, it seems to be turning out considerably better that I thought it might, but the proof of the pudding will be in the eating, or in this case, the running of the trains.

 

I don't think it's fair for you to discourage others from printing their own models, particularly when you don't own a printer yourself. Mastering a 3-D printer is actually much simpler than learning to use, for example, a lathe or a milling machine, but many modellers have taught themselves how to use lathes and milling machines.

 

A cynical person might infer you are looking for reasons to point out that this is a bad idea because you have a vested interest in discouraging people from printing their own models. I hope that is not the case.

 

Andy

[AndyID]

 

This is something I tried recently just to see if it was possible. I created the model and went straight to print. This was the first (and only) print.

 

It's actually remarkably strong, and quite flexible. It was printed as one piece, folded into a square and glued along one corner.

 

I've never used Shapeways, but I'd be a bit surprised if they would accept this model. Anyway, I posted it hereto demonstrate the kind of thing you might experiment with if you have your own printer.

[Joseph_Pestell]

I must admit not knowing too much about the materials used either with home printers or companies like Shapeways, longevity of the products has been my concern. Certainly as time goes by the products will improve.

 

What is important is that perhaps in the future printed turnouts will be available, no doubt in 00 gauge (4 mm scale) and with a choice of chairs/clips, bullhead or flatbottom etc. It could be from a company or printed at home. Unlike injection moulding where large set up costs are involved 3D printing will make items like this very much cheaper and/or with a larger range and options

 

Unless the printers and materials can be made to work a lot faster, turnouts made this way are not going to be cheaper  than mass-produced injection moulded items. But the total flexibility of design that it offers makes it more than worthwhile. Valid price comparison would be with a C&L kit.

[Joseph_Pestell]

DSCN2521.JPG

 

This is something I tried recently just to see if it was possible. I created the model and went straight to print. This was the first (and only) print.

 

It's actually remarkably strong, and quite flexible. It was printed as one piece, folded into a square and glued along one corner.

 

I've never used Shapeways, but I'd be a bit surprised if they would accept this model. Anyway, I posted it hereto demonstrate the kind of thing you might experiment with if you have your own printer.

 

With nothing to compare it against, I  can't quite make out the size of that for certain. But it looks as though you have mastered a 4mm lattice post signal.

 

Would it not be possible to print that in its final form rather than having to fold it?

[AndyID]

With nothing to compare it against, I  can't quite make out the size of that for certain. But it looks as though you have mastered a 4mm lattice post signal.

 

Would it not be possible to print that in its final form rather than having to fold it?

 

Hi Joseph,

 

Yes, it's supposed to be a lattice post. That sample is probably more like 0 scale, but an 00 versions would probably be quite convincing.

 

I don't think it would print very well in final form. The diagonal elements would have no support while being printed - a bit like trying to build a staircase from the ground up with no additional supports. Also, as they were printed as a series of steps "glued" together, they would be weaker. Printing them flat produces a series of full length strands of filament "glued" together in a sort of bundle. (BTW, I've printed some "close to scale" tiebars for the points. They are strong and flexible enough for the task, but I have not been able to work out a method of reliably connecting them to the point blades.) 

 

I'm not sure it would be possible to produce the lattice with injection molding either. It would probably have to be made in at least two pieces. I think Martin will know.

 

One advantage of the PLA material is that you can actually fold a thin section and it won't break. It's pretty tough stuff. The trick is to make sure the extruded strands are properly fused together. Like any process, you have to adapt the production technique to the characteristics of the medium you are working with.

 

Cheers!

Andy

[hayfield]

Unless the printers and materials can be made to work a lot faster, turnouts made this way are not going to be cheaper  than mass-produced injection moulded items. But the total flexibility of design that it offers makes it more than worthwhile. Valid price comparison would be with a C&L kit.

 

Joseph

 

My comment was made with the view that the machinery and materials will improve in the future, once these machines become common place the costs will tumble. You may well see printing shops spring up, the benefits of this technology not only spreads across many other model making hobbies but for domestic use as well. Instead of going to the model or hardware shop we pop down or order online the item we require

 

I guess the likes of Shapeways have started the process, but will be far more common place and cheaper in the future, I would guess the printing franchises are monitoring these technologies with a view of offering a service. A bit like taking your roll of film in for processing, you will take in a file on a stick

 

As Andy has said you can do a piece of work which parts of the code can be transferred to different piece. A "B" switch can be used for several different turnouts,  by altering a code you can change to the opposite hand, as you can see he has cracked the common crossing, again you can swap hands using the same code, in fills for diamonds and slips will follow etc.

 

Yes the technology is in its infancy, but it will not be long before we see further progress. What will be interesting is how will the likes of Peco react

[AndyID]

A teaser for John (Hayfield)

 

 

[hayfield]

Andy

 

Very cunning, there was me thinking that you made the base in 2 parts (which you have) then pre-formed the rails and threaded each half from opposite ends.I am afraid not liking non metallic rails

 

Now where is that slip ?

[AndyID]

Joseph

 

My comment was made with the view that the machinery and materials will improve in the future, once these machines become common place the costs will tumble. You may well see printing shops spring up, the benefits of this technology not only spreads across many other model making hobbies but for domestic use as well. Instead of going to the model or hardware shop we pop down or order online the item we require

 

I guess the likes of Shapeways have started the process, but will be far more common place and cheaper in the future, I would guess the printing franchises are monitoring these technologies with a view of offering a service. A bit like taking your roll of film in for processing, you will take in a file on a stick

 

As Andy has said you can do a piece of work which parts of the code can be transferred to different piece. A "B" switch can be used for several different turnouts,  by altering a code you can change to the opposite hand, as you can see he has cracked the common crossing, again you can swap hands using the same code, in fills for diamonds and slips will follow etc.

 

Yes the technology is in its infancy, but it will not be long before we see further progress. What will be interesting is how will the likes of Peco react

 

Hi John,

 

You may well be right. I see it as a trade-off between tooling costs and time-to-money. The quality is not as good as an injection molded part, but it may well be sufficient for a lot of users, and the potential for flexibility is enormous.

 

A reasonable printer costs about $400, but it could recover its capital outlay in a matter of months, if not weeks. They are so inexpensive that someone could buy a large number of them for the cost of a single injection mold.

 

The preliminary print I just posted mainly reuses components I created for the turnout. Because I got a bit ambitious it still needs a bit of tweaking, but it should be done tomorrow.

 

Cheers!

Andy

[AndyID]

Andy

 

Very cunning, there was me thinking that you made the base in 2 parts (which you have) then pre-formed the rails and threaded each half from opposite ends.I am afraid not liking non metallic rails

 

Now where is that slip ?

 

Hi John,

 

The turnout was made in two parts because the printer bed is not long enough to handle the entire turnout in one piece. The models are just a collection of various "bits". You grab the appropriate chair and glue it to the timber. I'm doing something very similar in CAD.

 

At the moment it takes me about 45 minutes to take a Templot template and populate it with the timbers and the appropriate chairs from my "chair library". There are a couple of other automated steps to consolidate it for the printer, but that's about it. I'm sure there is plenty of room for improvement too. (Maybe we could talk Martin into adding a 3-D turnout assembler to Templot?)

 

The "not entirely metallic check rails" are a bit of an experiment. Personally, if I can get away with it, I will use them because it saves a lot of time (why buy a dog and bark yourself?) but I'll wait to see how they turn out. It's not a problem to remove them either. I simply added them to a model that accepts metal rails.

 

No slips planned for this stage of my layout I'm afraid, although I do have an idea about how to make them that will probably horrify a lot of people

 

Cheers!

Andy

[jonny777]

This is a fascinating and exciting thread.

 

Thanks Andy for all your experiments and photos.

 

I can envisage club members buying a 3D printer between them and printing all manner of track geometry for their layouts.  And maybe as an exercise in trackwork appetite whetting, here is a photo of the western approach to Edinburgh Waverley in 1970,

 

 

[hayfield]

Hi John,

 

You may well be right. I see it as a trade-off between tooling costs and time-to-money. The quality is not as good as an injection molded part, but it may well be sufficient for a lot of users, and the potential for flexibility is enormous.

 

A reasonable printer costs about $400, but it could recover its capital outlay in a matter of months, if not weeks. They are so inexpensive that someone could buy a large number of them for the cost of a single injection mold.

 

The preliminary print I just posted mainly reuses components I created for the turnout. Because I got a bit ambitious it still needs a bit of tweaking, but it should be done tomorrow.

 

Cheers!

Andy

 

Andy

 

To be hyper critical, yes the chairs are very basic, but so are Peco's clips in 00/H0 turnouts. The detail of the parts will come, just look at some of the parts now available on Shapeways.

 

Good to know that you have a reusable library so items can be easily copied.

 

I would expect a turnout base would be made from 3 sections, the switch, the central section (closure) and switch, the switch will be the same for 3/4 turnouts and the common crossings shared with diamonds and slips

 

One other thought is that there is a chap making point rodding stools and people by scanning then printing. Stunning detail. Now if the chairs can be scanned that must save a lot of design work. easy with standard, bridge and slide chairs, may be very intensive work with crossing and block chairs though 

[rue_d_etropal]

I am not trying to discourage people buying 3D printers. The trouble is that there is a lot of hype out there, and some think a low cost printer will work wonders. If someone , like Andy, wants to experiment, I won't stop them. I know of some model railway clubs who have their own 3d printers. They have found them useful,and show them off at exhibitions, but also point out the limits and the care that needs to be taken. They have also told me that they can not get as much detail as the big professional machines.

At some time in the future, every household will have a 3D printer in the same way many have 2D printers, but I still prefer to get photos printed on big professional printers. For a few years it has been predicted(mainly by sales people) that there will be a 3D printing shop on every High Street. That has not happened yet, but will happen eventually.

I know of schools with 3D printers. At the moment I think that is a bit of a waste of money(apart from being a good way to show how it works), as teaching the 3D CAD skills is more important, and trying to print off decent 3D prints for a class full of enthusiastic students, would probably end in loss of interest.

3D CAD skills are rapidly becoming an essential skill for many jobs, in the same way word processing and other PC office skills are essential now. better to promote 3D CAD skills, and let the creative ideas flow from that.

At the moment i use Shapeways, but at any time in the future I can switch to another company, even my own printer when they are capable of printing my own designs. I chose to buy a good quality 3D CAD software package, which cost less than a basic 3D printer. Apart from samples(quite a lot as it happens), there is then no cost in setting up a small business to sell my designs. There are already thousands of people doing that, and I want to encourage more.

Ideally this discussion should be under the 3D printing section of this forum, as there is a lot of good advice there.

[hayfield]

Simon

 

In one way you may be correct that there could be some additional mileage if it were in a different section, but this is one of those crossovers topics and we are talking about  track. Though not strictly hand built in the sense we know at the moment, but then we are happy for those who wish to add RTR conversions which are also not exactly hand built either

[martin_wynne]

If someone, like Andy, wants to experiment, I won't stop them.

 

Andy will be relieved. He still has several problems to solve, but so far how to prevent you putting a padlock on his workshop door hasn't been one of them.

 

Martin.

[AndyID]

This is a fascinating and exciting thread.

 

Thanks Andy for all your experiments and photos.

 

I can envisage club members buying a 3D printer between them and printing all manner of track geometry for their layouts.  And maybe as an exercise in trackwork appetite whetting, here is a photo of the western approach to Edinburgh Waverley in 1970,

 

 

ed waverley track layout 1970.jpg

 

Thank you for the kind words Johnny. Lovely photo that is too.  (I lived in Edinburgh in the late sixties.)

 

Andy

[AndyID]

Andy

 

To be hyper critical, yes the chairs are very basic, but so are Peco's clips in 00/H0 turnouts. The detail of the parts will come, just look at some of the parts now available on Shapeways.

 

Good to know that you have a reusable library so items can be easily copied.

 

I would expect a turnout base would be made from 3 sections, the switch, the central section (closure) and switch, the switch will be the same for 3/4 turnouts and the common crossings shared with diamonds and slips

 

One other thought is that there is a chap making point rodding stools and people by scanning then printing. Stunning detail. Now if the chairs can be scanned that must save a lot of design work. easy with standard, bridge and slide chairs, may be very intensive work with crossing and block chairs though 

 

Hi John,

 

It's not possible to get much more detail with the sort of printer I'm using, but there are other types of printers that can produce much higher resolution prints. The advantage of the one I'm using is that it's affordable. (Not so different from the price of some RTR locomotives.)

 

To understand its capabilities you can think of it as a computer controlled hot-melt glue gun. The "glue" (plastic) is extruded into a filament 0.25 mm in diameter. Fundamentally that limits the X and Y (horizontal) resolution for details. For example, the bolt heads on my chairs are about as small as it is possible to make them with this additive printing method.

 

OTOH the vertical resolution (Z axis) is almost unlimited. The layers that make up the solid print can be really thin, but decreasing the layer thickness means more layers to print, and that takes more time.

 

The apparent "grain" on the sleepers is more by accident than design. It's produced by the strands of filament that form the top layer. It's probably a bit too rough to represent sleepers that are in good condition but I think it could be improved with paint. It's fine as far as I'm concerned, but some people might not care for it too much.

 

Andy