Jump to content

An Ugly Duckling - 3




If my last post was about ‘making choices’, the subject of this one is definitely ‘rivets’. These earlier engines seem to have been covered in the things so, thank goodness, 3D-printing software tools have come to my aid in reproducing them all. In fact I only had to draw one or two and all the rest were produced by tools such as ‘pattern on path’ which instantly created long rows of the things, following the contours of the surface on which they are placed. There must be at least 350 rivets on the tank surface alone.



‘Aurora’ Tank with added rivets


Compare this illustration with the one in my previous post, where I had just completed the basic tank structure! It is the sort of job that could have been a nightmare, when using traditional construction techniques, but was relatively simple with 3D-modelling software – at least with ‘Fusion 360’, by means of its ‘pattern on path’ commands.


I consulted several different drawings and found discrepancies between all of them. A consistent failing was in the width of the outer cladding of the smokebox, which was insufficient to surround the 24” diameter cylinders. I have widened the wrapper around the cylinders, as on many other engines of the period, although this does not appear on the drawings I have.


The cross-section of the tank also varies between drawings although all agree that the tank is wider at the sides than the space above the top of the boiler. This is different from the full length tanks on, for example the Bristol & Exeter engines, where the tanks appear to be concentric with the boiler.


Apart from that, the modelling and printing of the major components has gone extremely smoothly. I thought I might have a printer problem, since the print head was very slow to warm up, when I first used it after a break. My printer is unusual in that the head is part of a separate, plug-in module that contains the head itself, with its heater and thermocouple. I wriggled it a little in its socket and all was well again. I suspect that the heater draws quite a large current, so there will be a substantial power drop if there is any resistance at the electrical connector.


Following my usual practice, I printed the main body of the engine as three major components and a couple of minor ones. The Firebox, Smokebox, and Tank were all designed to fit around a length of brass tubing, which forms the boiler and provides the main structural component of the model. As on the prototypes, one only sees the ‘cosmetic’ outer skin. The two minor components are the backplate and the front of the smokebox.


3D-printed Main Components


I’ll not go into much detail about the construction of the main components, since my methods have been described in earlier posts. Note, however that I have included a filler cover for the tank and holes to provide mounting points for the chimney, sand-box, and safety-valves cover.


I shall single out the front of the smokebox for some more detailed comments. This small component could easily have been combined with the main smokebox in the 3D drawing but could then prove difficult to print, with my fused deposition printer, which needs a flat surface to start from and no major overhangs. The inside of the smokebox needs to be a clear space to accommodate the brass tube, which forms the ‘spine’ of my model.


The front of the smokebox can be made very thin so that, when laid flat on the bed of my printer, the complete job prints in only about 6 minutes. Once having copied the profile from the cross section drawing that I had used for the smokebox itself, I then added details to the front surface as described below.




‘Aurora’ Smokebox Front


After sketching the outline in ‘Fusion 360’, I extruded the sketch to create a flat plate, 0.5 mm in thickness. Using this surface as my drawing-plane reference, I first drew two pairs of concentric circles to represent the cylinder covers. I also drew one small circle to represent one of the bolt heads and then used the ‘pattern on circle’ to create a ring of 16 bolt heads for each cylinder cover. I find it best to extrude the deepest parts first, so that the drawings for other parts remain visible on the surface. So, I selected all the smaller circles and extruded these by 0.5 mm.  I then selected the annulus between the inner and outer large circles and extruded this area by 0.25 mm.


Next, I tackled the smoke box door. This is made up of a series of three-point arcs and straight lines, to create the outer rim of the door. Then I sketched one of the door latches and used the ‘copy/move’ command to replicate the other catches at the required locations and orientations.similarly, I drew the outlines of the two hinges at the bottom of the door. As before, I extruded the drawings of the latches and hinges by 0.5 mm and the rim of the door itself by 0.25 mm.


For the rivets around the edge, I created a single rivet as a ‘New Body’ at the top of the smokebox front and then used the ‘pattern on path’ command to create replicas all around the edge.  I selected the ‘symmetrical’ and ‘align to path’ options, to spread the rivets evenly in each side and at a constant distance from the edge. The distances that I quote are all the result of trial and error, a process that is greatly facilitated by the very short print times of components such as this.


I made the backplate in exactly the same way, although this is thicker to allow for curved fillet round the edge, which was polished brass in the prototype. I sketched and raised the firebox door and the minimal controls


The end result, with the components all mounted onto a brass tube (visible through the chimney aperture), is shown below:




Complete boiler assembly with components mounted on brass tube


Next, I shall turn my attention to designing and constructing a suitable chassis.



Edited by MikeOxon
Restore images

  • Craftsmanship/clever 12


Recommended Comments

  • RMweb Gold

Uh-oh, nearly missed this, thinking it was your previous post.


That boiler assembly looks very good, in fact almost edible. Is there any filing/sanding when you join the components together, or do they just fit straight away?

  • Like 1
Link to comment

My secret is out!!! yes, I'm a chocoholic!


Unfortunately, the taste of PLA is not great, so I shall stick to my usual brands :)


At present, the parts are straight from the printer, Mikkel.  They do slide together pretty closely on their brass tube but I shall probably do some surface fettling before I take it all to the painting stage.


If I were to use this method for a later GWR tank engine, I'd be able to swap over from saddle tanks to panniers just by lifting off the relevant components.  I've already got a couple of spare tanks, after re-prints to improve the appearance of the rivets.


At present, I'm working on the frames and it's useful to be able to 'test assemble' the parts within the computer, before finalising the design for printing.



  • Like 2
  • Interesting/Thought-provoking 1
Link to comment
  • RMweb Gold
8 hours ago, MikeOxon said:

If I were to use this method for a later GWR tank engine, I'd be able to swap over from saddle tanks to panniers just by lifting off the relevant components.  I've already got a couple of spare tanks, after re-prints to improve the appearance of the rivets.


Thought provoking. I'm imagining a whole range of mix-and-match saddle tanks using 3D printed parts on Bachmann 57xx/8750 chassis.


Edited by Mikkel
  • Like 1
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...