Having acquired some confidence in making 3D extrusions from 2D drawings, I decided it was time to move on and try something slightly more ambitious.
While there are many good things about building engines from sheet brass – after all, the real things are made from metal – there are limitations when it comes to making components with significant depth or complex surface detail. Although it was an interesting challenge to complete the firebox and smoke-box from the BGS Gooch Goods kit for my model of ‘Rob Roy’, it wasn’t an exercise that I was especially keen to repeat, so I decided to try the idea of making these components as 3D-printed items.
Having already explored the possibilities of ‘hybrid’ construction techniques, I decided to try making the boiler assembly of ‘Tantalus’ from a combination of brass and 3D-printed parts. I’ve found already that there are many advantages to using 3D printing for making components of a model, rather than attempting an overall print. The design process can be broken down into small steps (remembering Mikkel’s dictum on ‘how to eat an elephant’ ) and each part can be printed quite quickly. When a model takes several hours to print, there is a strong disincentive against doing too much experimentation but, when print times are down to a few minutes, it’s easy to find out what does or does not work, with the tools that I have.
Although my 'Geeetech E180' printer had me tearing my hair out at times, often due to inadequate documentation, I am now finding it to be an easy-to-use and non-temperamental machine, which fits very well into the domestic environment. I also note that, having checked recently on the web, there is now greatly improved documentation available, including a proper explanation of the bed-levelling process.
The overall precision seems very good, for a low-cost machine, and the main limitation on detail seems to be the 0.4 mm diameter of the extrusion nozzle, which means that prototype details smaller than about 1”, cannot be represented in 4 mm scale.
So, tackling my ‘elephant’ one bite at a time, I start with the boiler:
Standard Goods Boiler
These early engines had remarkably large boilers for their time. By the time of the final lots of the ‘Standard Goods’, the boiler barrel was 11’ long by 4’ 6” diameter. I obtained a length of 18 mm diameter brass tube as the main component, which I shall fit into holes in the back of the smoke-box and the front of the firebox, respectively. Boiler bands will be added as narrow strips cut from 5 thou brass sheet.
Standard Goods Smoke-box
I decided to 3D-print the smoke-box in two parts – the main body and the front plate. I extruded the main part of the box from a 2D drawing of the end-on profile and opened out an 18 mm diameter hole to accept the end of the boiler tube.
I raised a ring of rivet heads around the opening for the boiler, as seen on the prototype. These were on the limit of what could be resolved by my printer, so they appear more as a ‘suggestion’ than an accurate representation of each individual rivet. I also extruded the back ends of the cylinders and the gland for the piston rod, with a hole to accept a 1 mm diameter rod, leading to the cross-head.
When I came to the front plate of the smoke-box, I realised that there was an important difference between the front of ‘Rob Roy’, which I have already modelled , and the Standard Goods. In the ‘Waverley’ class, the cylinders were almost horizontal but, in the goods engines, the cylinders were inclined, to allow the connecting rods to pass over the front axle on their run between the cross heads and the cranked driving axle. Although books, such as the RCTS series, provide plenty of information about boiler and cylinder dimensions, they remain silent on the angle of inclination of the cylinders. I measured the angle, from a detailed drawing of ‘Pyracmon’, as 7 degrees.
This angle has to be reproduced on the front face of the cylinders which are contained within the lower part of the smoke-box wrapper. Setting up this angled section presented a new challenge on my learning curve about 3D-modelling with ‘Fusion 360’.
My solution was to start with a flat front plate, matching the profile of the main section of the smoke-box. I soon realised that, while there are plenty of drawings showing the side elevations of locomotives, other views, including the head-on appearance, are much harder to find. At the period I am modelling, these engines had a bottom-hinged door of rectangular shape, with rounded sides. There were two hinges at the bottom of the door and three clamps at the top to keep the door tightly closed.
Although I could not find a drawing of the front of a Standard Goods engine, I found that Mike Sharman’s book of Broad Gauge drawings included one of the front of ‘Lord of the Isles’ (1851), taken from the Locomotive Magazine 1950, p.197, Fig.5. This provided me with many details, including the line of rivets around the edge of the front plate.
From this drawing and various photographs, I was able to make a 2D drawing of the main features of the front plate. I then used the ‘push/pull’ tool in ‘Fusion 360’ to raise these details above the background surface. This left me with the problem of adding the 7° angle to the lower part of the front plate. Trial and error led me to the following method, although there may well be other ways of achieving my aim..
I split the front plate horizontally, immediately below the smoke-box door. I then rotated the lower part by 7°, using the ‘Move’ tool in ‘Fusion 360’. Next, I used the sketch tools to draw a triangle between the original plane and the inclined plane. I extruded this triangle across the width of the front plate to fill the gap between the inclined section and the original front of the smoke-box. After carefully aligning the lower part with the upper part, I used the ‘join’ command to re-combine the two parts into a single body. I illustrate the steps below.
Because the front plate is thin, it only took 5 minutes to make a test print. Sadly, this demonstrated that all the fine rivet detail, which I had laboriously added, was too small to print clearly. I decided to make the individual rivet heads larger, to give the printer a chance, but this also required me to increase the spacing between the rivets. Rivet counters look away – it mean that the total number of rivets had to be reduced . I decided that this was an appropriate compromise in 4mm scale, as the presence of rivets remains obvious, although the number is incorrect.
Once I had printed these two parts of the smoke-box, I checked the fit of the front onto the main body, as shown below:
In general, I was pleased with the overall appearance, although the printer did not reproduce the smaller details, like cylinder cover bolts, on the inclined surface as well as the details on the flat surface. There are also visible steps in what should be a smooth slope.
Standard Goods Firebox
The firebox on the prototypes was a large but simple round-top design. All that was necessary was to make a 2D drawing of the cross-section of the firebox and then extrude it to the required length . I opened out a hole of the diameter of the boiler tube (18 mm), so that the boiler could be fitted into the front of the firebox. I also provided a small hole for fixing the safety valve cover in the appropriate location.
One of the most awkward features to model by traditional methods is the curved ‘fillet’ between the firebox and the boiler. This is very simple, however, when using ‘Fusion 360’ - just a mouse-click on the ‘fillet’ command!
When designing models for 3D-printing, I find it is useful to pause for a while, to think about other details that it is appropriate to add. In this case, I realised that it was easy to add the spectacle plate to the rear end of the firebox. I left the back of the firebox flat and printed a separate part to form the backhead,
The various driving controls and the fire-door are mounted on the backhead of the firebox. I made this as a separate part by again extruding from a drawing of the cross-section of the firebox. As with the front of the firebox, I used the ‘fillet’ command in ‘Fusion 360’ to produce the curved rim to the backhead – this will be given a polished brass finish, as on the prototype.
As I mentioned before, in the context of the smoke-box, it proved difficult to find a drawing of the backhead but, once again, Mike Sharman’s book of drawings contained one of ‘Lord of the Isles’ (1861). This showed the great simplicity of the controls on engines of this period.
The regulator is placed centrally, near the top, while on the left are ‘try-cocks’ for checking the water level in the boiler, together with a circular gauge. Apart from the oval fire-door, that’s it!
I made drawings to represent these components and then extruded them from the face of the back-plate, using ‘Fusion 360’. I realise that this is how things were done back in the days of Tri-Ang models and that applying separate fittings is now the norm. I thought, however, that it was useful to use the 3D-printer as a simple method for reproducing these items. They can always be replaced by something better, if I decide to refine the model at a later date.
Again, I was pleased with the appearance, which conveys the ‘basic’ nature of these early engines.
My method of constructing the boiler from an assembly of separate components has proved to be a quick and easy way to build a locomotive! Most of the time, of course, is spent on preparing the drawings while the printing of each part only took a few minutes. This was significant for me, because it meant I could try out ideas, without having to wait for hours to see if they worked.
I show the complete ‘kit of parts’ below and am pleased to report that the firebox and smoke-box fit firmly over the ends of my brass tube, which will add substantial weight to the whole assembly.
I’m sure I could add many more refinements but I intend to turn my attention back to the chassis for my next session, so that I can then assemble the complete engine, with any ‘extras’ I think of along the way.