Jump to content
  • entries
  • comments
  • views

3D-printed Broad Gauge carriage





I have already described my halting progress with 3D printing on my other Pre-Grouping blog, with the previous post having been made there on 3rd May.


Since then, I decided to have a go at printing a Broad Gauge carriage of a type I have previously constructed using layered card sides, cut with my Silhouette Portrait machine.  This meant that I had already produced side elevation drawings of my chosen prototype in 4 mm scale.




Thus, my latest ‘build’ was mainly an exercise to see what I could do with my 3D printer, and to combine various aspects of the carriage body into a single operation.  As I thought about it, I realised that all the internal partitions could be included in the print and then I added seats as well.


A feature of early broad gauge carriages was that 1st-class compartments were sometimes divided into two sections, on either side of the centre-line of the carriage.  This resulted in two separate 4-seat compartments on either side of the carriage, with a connecting door between them, to allow the passengers on one side to reach a platform on the opposite side.  I decided to incorporate this feature into my design.


The following illustraion, although not a Broad Gauge compartment, shows several interesting details, such as the cords to hold top hats, mounted under the low ceiling.



As described in my earlier posts, I used Fusion 360 software to develop a 3D model of my chosen prototype on my computer.  My method was to extrude an open-top box from a profile of one end of the carriage, which included a ‘tumble home’ on each side and a curved top over which the roof could be mounted.


The software allowed me to ‘paste’ a 2D drawing of the carriage side onto the computer model and I then used the ‘push-pull’ tool to inset the droplights and to open out the window openings.  I then drew the seat bases and partitions on the floor of the carriage and, again, used the ‘push-pull’ tool to raise these to the appropriate heights. 


I still find that it needs a lot of trial and error to get all these details in their correct places and, when I make a mistake, I find it easier to delete and start again rather than attempting to correct my mistakes.  Perhaps, when I understand the software a little better, I shall find it easier to make such modifications.


Once I had reached the point where it seemed worth trying to make a print, I used my Cura software to ‘slice’ the model into the format needed by my printer.  I could have used the Fusion software to do this but preferred Cura because I had already set this up with the characteristics of my Geeetech E180 printer.


After slicing, the display on my Cura software indicated that it would take about 7 hours to print, which struck me as rather long but I didn’t investigate further.  In fact, it proved to be a fairly accurate estimate.




Once it started printing, I realised the first of my mistakes!  After a previous test, when the carriage sides had turned out too thin, I had set the ‘infill’ to 100% which meant that the seats I had provided in the compartments were printed as solid blocks, which was very time-consuming!  At least it meant that the finished body had a reasonable weight (c.40 g) and was very rigid.  In spite of my profligacy in the use of plastic filament, the cost of materials was still only about 60 p!


My second mistake was that, somehow, I had made some of the transverse partitions much too thin, so that they came out rather skeletal in form.  This can easily be corrected by adding some small rectangles of plastic card, although it has defeated my object of an ‘all-in-one’ construction.


Thirdly, I had made the moulding lines too shallow so that they hardly appear at all on the model.  I must practice making such features, to find suitable dimensions for them to appear in 4 mm scale.


In addition, the fact that I have had several reliability problems in the past meant that I was reluctant to leave the printer unattended for too long at a stretch.  If I can gain more confidence, then it would be preferable to do the printing overnight.  The printing would also have been a lot quicker if I had not included the unnecessary solid infill.  If I had set up for a 25% infill, the Cura software estimated a time to completion of only 4 hours, for the same model.


On the positive side, there is no limit to the thickness of the side panels of the carriage, so there is no need to build up a side from several laminations, as is the case when using the Silhouette cutter.  Also, in spite of my concern, the printing proceeded without a hitch.  After having had some of my early models detach themselves from the base during printing, I was pleased to find that this model was quite firmly attached to the blue masking tape that I had used.




In fact, for a model like this, which is mainly flat surfaces, it would almost certainly have been easier (and a lot quicker) to fabricate the carriage from Silhouette cut panels.  The trouble with my type of 3D printer is that the process takes such a long time that it becomes off-putting to make trial prints.  I had put off starting this model for some time for this very reason.


I am expecting that the 3D printer will come into its own, when I make the chassis for this model.  I shall be able to use a suitable thickness of material, to make strong side frames, and will include spring and axle-box details as part of the process.


That will be my next project ...


  • Like 11
  • Informative/Useful 3
  • Craftsmanship/clever 3


Recommended Comments

Thank you, DrDuncan.  I have found it interesting to explore the internal layout of these early carriages.  It seems that the inner doors between the 1st class compartments slid on rollers.


The 3D printed structure is much more rigid than the models I fabricated from plastic card - probably because I did not laminate enough Silhouette-cut panels when making those.  I feel I could safely let my 3 year old Grandson handle this one!


  • Like 1
Link to comment

Will you be offering these for sale?  You certainly have at least one interested party here,  when I get around to broad gauge printing I surmise you'll be the chap to send the files on to

Link to comment

Hi Killian.  I'm afraid I just do these things for my own interest.  Most of the work is in preparing the files for printing so, if you're able to do this, there's not much extra effort involved, except in buying a printer - and these are improving in quality and falling in price all the time.  At heart I'm an experimentalist and would get bored very quickly if I started repeating things I've done already :)


Thank you for looking in - I'm pleased that you've clearly liked reading about my rather slow progress.



Link to comment

Would this work in 2mm scale? I guess that the partitions, ends and sides would need to be thicker than you have designed it. 



Link to comment
  • RMweb Gold

Well done Mike.  You did not print a roof or will you do that separately in future?  I was not thinking of printing the roof in place but maybe on four sprues from the corner.


You do not seem to need the supports that I have seen on other printed 3D models, and it seems to work flat where the others seem to have to be printed at an angle.  Is that a factor of the printer, the plastic, both or what?

Link to comment
5 hours ago, timatheronwood said:

Would this work in 2mm scale? I guess that the partitions, ends and sides would need to be thicker than you have designed it. 



I would assume so.  I intended to make my walls 40 thou (1 mm) thick but must have made some mistakes because some partitions printed too thin.  As I indicate in my reply to ChrisN, I am adopting a "suck it and see" approach.

  • Like 1
Link to comment
3 hours ago, ChrisN said:

Well done Mike.  You did not print a roof or will you do that separately in future?  I was not thinking of printing the roof in place but maybe on four sprues from the corner.


You do not seem to need the supports that I have seen on other printed 3D models, and it seems to work flat where the others seem to have to be printed at an angle.  Is that a factor of the printer, the plastic, both or what?

Thank you Chris.  Since the roof is a simple arc type, I shall just use plasticard or brass for the roof.  Yes, I could have included some fixing points if I had thought that far ahead!


As an aside, I'm thinking of replacing the rather rickety clerestory roofs on some of my older builds with 3D-printed versions.  The trouble is that there are so many things I could do, that it's hard to know where to start.


As you have observed, experienced 3D printers seem to use all sorts of supporting structures and print at curious angles.  They must spend almost as long cutting all these extra pillars out, as it takes to print them!  There are probably good reasons for them but I am just following my nose. 


A simple box structure was my starting point and the floor is flat on the printer bed.  I was concerned about window apertures but the printer seems to have filled these with fine filament and has continued above them successfully, with no sagging into the gaps.  More complex shapes with overhanging areas will, of course, need more thought and I suspect this is where the support requirements will arise.


Wider windows may well cause problems - perhaps I shall find out in due course!  The 'slicer' software has options to add support structures as it sees fit, so I have left it to its own devices. 


As I mentioned In my post, I had inadvertently left the 'in-fill' setting at 100%, so my seats are completely solid and took (literally) hours to print.  If I had used 25% (say) the printer would have automatically produced a honeycomb structure, using much less filament and therefore printing ,much faster.


I'm afraid I'm not good at reading manuals (no RTFM) but just try things and see if they work.   As I gain more experience, I may find better ways of doing things but I'm quite pleased with my results so far. I'm simply enjoying playing with a new toy :)

  • Like 1
Link to comment
  • RMweb Gold

Thank you Mike for a very informative answer.  I am not sure I will ever get a printer but it does give me a feel for how they work.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Create New...