Jump to content

Adding Skins to Carriages

MikeOxon

286 views

One idea leads to another; in this case, I have tried extending the idea that I showed in an earlier post of adding cladding to a brass-tube boiler by 3D-printing an outer sleeve.

 

1382905321_CladdingonBoiler.JPG.8e554fab8f79e2cf48f0087a7c10b362.JPG
3D-printed boiler cladding

 

During my early ‘learning curve’ with 3D-printing (i.e. about a year ago), I made some broad-gauge carriages, as described in a short series of blog posts.  Printing the complete carriage as a single task had several advantages, such as including internal partitions and seats, but also created difficulties in making window openings and other features on the vertical sides. For the window openings, I had to include support structures which proved quite difficult to remove without leaving rough edges around the windows.

 

BG_1857_3Dprint.jpg.7a5a38a37d5b0c90fb5b2123e685f383.jpg
3D-printed Broad Gauge Carriage

 

There is the additional point that each print took about 8 hours to complete, which is a serious deterrent to much experimentation - my preferred method of working.

 

As a result, the completed carriages lacked detail, so one solution that I explored was to use my Silhouette cutter to add outside frames to plain carriage sides, as shown in my Pre-Grouping blog

 

My new ‘cladding’ idea, however, is to add a thin skin to the previously printed carriage body. This ‘skin’ can carry the additional detailing to represent external mouldings and other fittings. My initial experiment was to make a ‘skin’ only 0.5 mm (20 thou) thick, which could be laid flat on the printer bed. With this method, cut-outs for windows could be made with no requirement for any support structures, while slots in the surface could be made to mark the edges of doors. External mouldings could also be extruded, by simply increasing the thickness of the skin in the required locations – again with no need for any support structures. The other major advantage is that a detailed skin of this type prints in a few minutes, so it is easy to correct mistakes and make improvements, without committing to extended time-scales.

 

As usual, there were some lessons to be learned. I made the initial drawings of the carriage side in ‘Autosketch’ and then transferred these to ‘Fusion 360’ in DXF format. I placed the drawing on the horizontal (XY) plane and then extruded the main panels to a depth of 0.5 mm, excluding the window areas and door edges and mouldings. After this first extrusion, I selected each of the door edges and extruded these by only 0.25 mm, so that they were recessed below the main surface. In a similar way, I extruded the mouldings by 0.75 mm, to raise them above the panels.

 

BG_1857_2nd_model.jpg.1603f1beb165466d3a3b789c8cddabaa.jpg

3D-model for side overlay

 

For the louvres above the doors, I raised each rectangle and then applied the ‘chamfer’ tool to achieve the required angle of each louvre individually.

 

1762195677_BG_1857_Louvres640x480.jpg.bd5ce7cec7cb64f95e3e55315ce59d66.jpg
chamfered louvres above window

 

My first print, which took about 9 minutes, showed that some of my details, especially the raised mouldings, were too fine to appear in the final print. In fact, I should have used the pre-view facility in the ‘Cura’ slicing software, which would have warned me that this was going to happen.

 

It is all too easy to draw details that are too small to be realised by my printer, in which the print head has a diameter of 0.4 mm. It is interesting to note that narrow slots (0,25 mm), as at the edges of doors, do appear but similarly-sized raised features do not appear. Modifying the drawing meant that I had to get to grips with the drawing tools in ‘Fusion 360’, which are broadly similar to those in ‘Autosketch’ but with some key differences.

 

One important aspect when creating drawings in a 3D program is that you have to decide in which plane the drawing is required to appear. This can sometimes be tricky, especially after the initial drawing has been extruded to create new surfaces above the original plane. The first action, when entering ‘drawing mode’ is to select the drawing plane. By selecting the top surface of the carriage side, I could draw new profiles for the arc-shaped mouldings at each end of the carriage. I could then extrude these new sections by 0.25 mm above the top surface, to match the original mouldings (which had been extruded by 0.75 mm from the original drawing plane). It is essential to keep a clear head when making these modifications!

 

For the straight side-mouldings, I used a different method by selecting the vertical edges of the moulding, where they rise above the carriage side. The ‘Move’ tool can then be used to move these edges in order to increase the width of the mouldings to a printable size.

 

26060565_BG_1857_mouldingwidthl640x480.jpg.2ab22b59238cf0998f9cdb02d9aef58d.jpg

Changing the width of a moulding

 

After making these changes, I transferred the 3D model to ‘Cura’ and this time, I remembered to use the ‘Preview’ mode, to check that the details would actually print! I also set the line-width to 0.3 mm, even though my print head is 0.4 mm diameter. This can improve the smoothness of very thin panels like these. I suspect this is because the nozzle is circular, so that a rectangular grid of lines benefits from slightly closer spacing between successive passes of the print-head.

 

All looked good in the pre-view, so I proceeded to another print, which again took only a few minutes to complete.

 

When printing these very thin ‘skins’, the ‘Cura’ software automatically produces frames around the window openings and the edges of the sides, which, in the case of the windows, provides a fair representation of the bolections. The main panels are very thin indeed and peel off the printer bed like plastic tape but I found they were sufficiently robust to resist tearing.

 

BG_1857_removing.jpg.0b20b147b69d1fad4be50713e1d1a027.jpg

Peeling side-skin off printer bed

 

I smeared some PVA adhesive on the original carriage side and carefully laid my new cladding in alignment with the existing window openings, which already had the droplights represented. Since I had previously sprayed the carriage body with red primer, the droplights automatically appeared in a suitable red colour. As early GWR carriages were painted brown overall, the final painting job was very straight-forward. Although diagonal lines from the printing process were visible on the surfaces of the printed panels, these were sufficiently fine to cease being noticeable under a coat of paint.

 

BG_1857_2nd_overlay.jpg.1863ed66c9f0e48e14cbebf07d51899b.jpg

Skin applied to Carriage Body

 

I shall continue to experiment with different parameters for the printing process and, if making new carriages, I would make the original sides a little thinner, to allow a thicker surface skin to be applied without increasing the overall thickness by too much.  I think my present skins are on the limit of what my current printer can reasonably accomplish.

 

Mike

 

 

 

 

  • Like 8
  • Craftsmanship/clever 7


9 Comments


Recommended Comments

A slightly creepy title, but a most interesting development! 

 

Nine minutes is, I believe, next to nothing in the 3D printing world. I agree it would be good if the inner body can be made thinner, in order to avoid too much thickness potentially showing in the windows.

 

Quote

experimentation - my preferred method of working

 

And the most interesting way.  

 

  • Agree 2
  • Thanks 1

Share this comment


Link to comment

Thank you, Mikkel.  'Skin' is a term I borrowed from the computer simulation world where the final decorative layer on an Aircraft or Railway model is termed a 'skin'. Nothing macabre intended!

  • Like 1

Share this comment


Link to comment

David Jenkinson wrote a book in the Wild Swan series about coach building.  His technique was to cut frets from plasticard, overlay these on thin plasticard sides, and then fix the assembled (and potentially painted) sides onto a structural box which comprised the partitions, seats, and floor, and whose sides were cut away so the window edges did not look unprototypical. 
 

I have not tried the approach, but it seems an ideal one for silhouette cutters, and, maybe for 3D printing too - your skins are very similar to his assembled sides.

 

might be worth a look.

 

atb

Simon

  • Like 1

Share this comment


Link to comment

Thanks Simon.  Yes, I've used the laminated approach with a Silhouette cutter and, as I mention in one of the links above, a hybrid, with Silhouette-cut sides on a 3D-printed frame.  This was an experiment to see if thin 3D-printed sides would work and it appears that they do!

 

Mike

Share this comment


Link to comment

I always admire people who think outside the box and try new ideas. Without them the hobby would never advance. Nice one!

  • Like 1
  • Agree 3
  • Thanks 1

Share this comment


Link to comment

Thanks Mike - a very kind remark.  Having had a career as a research scientist, I'm always chasing ideas - some work, some don't but it's all good fun :)

  • Like 1

Share this comment


Link to comment

Excellent idea!  Love to see more of this develop as it is a pity that you can only mostly get carriage kits in brass which is notoriously hard to paint.  It would be brilliant to see them printed with the colours and lining if this was possible.  Very clever stuff.

  • Thanks 1

Share this comment


Link to comment
On 25/06/2020 at 13:12, Mikkel said:

A slightly creepy title, but a most interesting development! 

 

Nine minutes is, I believe, next to nothing in the 3D printing world. I agree it would be good if the inner body can be made thinner, in order to avoid too much thickness potentially showing in the windows.

 

 

And the most interesting way.  

 

The printing time depends mainly on what one is printing and the printer being used. I printed a HO scale Lancashire boiler as a paperweight for a friend of mine, using a printer very similar to Mr. Oxon's. I think it took about 6 hours, but printing with a laser like Shapeways does, using powder, often takes days and sometimes weeks.

 

As always, very interesting work Mr. Oxon.

  • Like 1
  • Thanks 1

Share this comment


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...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.