3-INCH TRACTION ENGINE – THE BOILER
Jan-eric’s latest road engine project, started last month, focuses on the heart of any traction engine, the boiler, and as ever employs some unusual methods...
Atraction engine boiler performs all the same functions as the boiler in a railway locomotive, with one significant difference: it is a very important structural component of the engine. A typical locomotive boiler is basically just ‘sitting there’ on the frame, with only a few points of support – in fact, the rear support of a loco boiler should always be flexible, so that thermal expansion doesn’t strain the boiler or the frame.
On the other hand, the boiler in a traction engine attaches to the main massive parts of the engine, including the cylinder, the cab with all its heavy gearing and coal bunker, and of course also the front end, which consists of the smokebox with its saddle, and the front axle with the wheels and suspension attached to it. For this reason, and since all the forces affecting the front wheels will be absorbed by the boiler, a traction engine boiler needs to be very sturdy.
Unusual material
I was lucky to still have enough copper-nickel (‘Cuni’) tubing left over from my previous locomotive projects to be able to construct this boiler. I obtained this Cuni material almost two decades ago, when a shipbuilding company sold it to me at scrap prices – this firm had earlier used Cuni for heat exchangers in the ships it built, but had changed over to more modern, composite materials.
Cuni as a boiler material appealed to me because its tensile strength is practically identical to mild steel, and being corrosion-resistant, it would therefore not need the extra thickness that is always included in steel boilers – usually double the thickness calculated for strength, allowing for half of it to rust away before the boiler needs to be scrapped.
This extra for corrosion is of course not needed in pure copper boilers, but since copper is weaker than steel, the actual thickness specified is usually quite the same. I used a common formula to calculate the required wall thickness for a Cuni boiler, and found that 3mm (a tad under ⅛-inch) was quite enough, giving a safety factor of at least six for the boilers I was contemplating.
Thus, I stocked up on all the Cuni that I could obtain, and so far, including the one for this traction engine, I have constructed four large and one small boiler. The latter was my first experience in TIG, ‘Tungsten Inert Gas’ also called GTAW, ‘Gas Tungsten Arc Welding’ of a boiler. I built the tiny boiler as a test of this welding method.
The procedure, making a vertical boiler for a small stationary engine, worked out well; the construction was described in the March and April 2018 issues of EIM. The TIG technique proved more successful (and much cheaper) than silver soldering, which is the standard method of building a pure-copper boiler.
Silver soldering is not a process recommended for Cuni unless special precautions are taken – I had quite a setback during my first effort to build a Cuni boiler for my Ten-wheeler loco by silver soldering; large cracks developed due to ‘intra-granular penetration’ of the solder.
Material warning
A note of caution – I am not suggesting that you should build Cuni boilers for your own projects, for several reasons – primarily because this material is not recognised as a standard boiler material by clubs and associations. In any jurisdiction where official boiler testing is mandatory, this could certainly cause problems!
Since we have no clubs or miniature boiler standards or hobby inspectors in Finland where I live, I will have to ensure that my boiler is safe all by myself. This boils down (very good – Ed...) to two important considerations: first, a design that is safe and can take the mechanical strain imposed on it; and second, hydro-testing the boiler to ensure that it can stand the steam pressure – in fact, the test will be performed at no less than two and a half times the normal operating pressure.
I’m telling you all this just so that you can get an idea of how I built the boiler for my traction engine project, not as instructions on how you can
build a boiler! Besides, Cuni is very hard to obtain these days – and horribly expensive! It is akin to ‘Monel’ metal (also a copper-nickel alloy), which is sometimes used for making fittings in full-size as well as model boilers.
You have other options; you can build a boiler in steel, welding it with approved filler materials, or, you can use pure copper, using either silver soldering or oxy-acetylene brazing to join the parts. Tig-welding may be possible, but since copper is a very good heat conductor, extremely high amperage is necessary for welding, thus hardly possible with hobby equipment. It is always a good idea to talk to some experienced boiler constructor or inspector before starting a project, in order to ensure that there will be no errors in planning or construction that may preclude the boiler from being approved for use.
Starting the work
All the Cuni material I obtained all those years ago is in the form of round tubing, from 10mm (25/64-inch) to 220mm (81/2-inch) in diameter.
Scaling down from full size, the boiler diameter for this project is 170mm (611/16-inch), and one piece of tube I had was 160mm (619/64-inch) – perfect for the purpose, considering a 5mm allowance for an insulating boiler wrapper. I cut the tube to length, and also cut away part of it at the firebox end with the help of my trusty angle grinder (Photo 13).
For the rest of the boiler, I obviously need other shapes, including several more flat pieces for the firebox. Cuni is very ductile, even while cold, but even more so when it has been annealed by heating it to dull red. Photo 14 shows how a piece cut from the largest tube is heated with my largest propane burner (consuming some 2 kgs of gas per hour) before being flattened with a big hammer on an ‘anvil’ of sorts – in fact, the same heavy piece of I-beam on which I heated the tube served as the anvil.
With enough pieces of flattenedout Cuni material, I could plan the firebox and the front tubeplate of the boiler. A CAD sketch helped me get the dimensions right, and a laser printout on self-adhesive label paper helped me drill the holes for the tubes in exactly the right position.
Photo 15 shows a preliminary CAD drawing of the boiler at top, and the tubeplate and firebox front with the printouts attached to them. Under the plates you can see part of a printout of a hind wheel in the same quarter-scale, while at left is a quick sketch of the differential drive on the hind axle, which has been drawn on 5mm squared paper – but I’ll return to that in a later article.
I started the machining by first using a centre drill (Photo 16). This work was entirely hand-held, as was the drilling of the smaller holes (15 and 18mm – 5/8-inch and 3/4-inch) for which I had suitable drills. While enlarging the holes with consecutively larger drills, holding the plates I used thick leather gloves in case the drill would catch and whip the plate around – once bitten, twice shy, you know...
The two larger holes (for the superheaters) had to be enlarged with a boring head, which necessitated clamping the plate to the mill table, (Photo 17). Note the piece of wood placed under the workpiece – absolutely necessary, so as not to bore into the mill table...
Photo 18 shows the finished plates, and all the fire-tubes soon to be installed. The fact that I received the larger 18mm tubes for free explains their rather ugly oxidised state – they had been stored outside for quite some time, and had acquired their dark patina. However, cleaning the tube ends with a belt sander on the outside, and an abrasive sponge on the inside to a clean, shiny copper surface will enable me to Tig-weld them to the tubeplates without problems. All materials to be welded using the TIG method must be clean of dirt and oxide!
Using the TIG torch, I just fused the ends of the tubes to the tubeplates – with a protrusion of a little under 3mm (1/8-inch), there was enough
“They had been stored outside for quite some time, and had acquired their dark patina...”
material to form a nice fillet around the tube ends, Photo 19. When I built the little test boiler, I found this protrusion to be ideal – less, and there might be voids in the joint, while a larger projection caused the molten material to flow into the tube opening, obstructing it slightly and making it harder to clean after running. Even though I plan to use propane as a fuel, I’m building this boiler so that it can also be fired with solid fuel.
The firebox
After forming the firebox wrapper from a sheet of annealed and flattened Cuni tubing, I welded the tubeplate using an ample amount of filler wire. The filler is also a Cuni alloy, containing a small amount of de-oxidiser which ensures a good weld – Photo 20 shows the seam. Since this joint will be inaccessible once the boiler is finished, it was very important to do it properly!
Next in order were the outer sides of the firebox, to be attached to the boiler tube. Photo 21 shows how I protected the inside of the seam to be joined, using a strip of woven glass-fibre ribbon and aluminium tape. This will prevent oxidation of this side of the weld, which is performed from the other side.
Cuni, just like any other metal, copper and steel included, will form a scale of oxide when heated to melting temperature, unless air is excluded from the weld area. Professionally, this is often accomplished by using an airtight ‘glove box’ filled with an inert gas, usually Argon. This is not feasible in a hobby environment, so the shielding tape (available from welding suppliers) is a good second best.
Photo 22 shows the two main parts of the boiler; at left the inner firebox with the tubes attached, and at right the boiler barrel with the
pre-drilled firebox outer sides. The spacing for the stay bolts is according to calculations, incorporating a six-fold factor of safety before any deformation will take place.
I have not yet drilled the inner firebox for the stay bolts. Their positions will be marked when the boiler parts are fitted together. The firebox sides are also oversize, and will be shortened when exact measurements have been made during a test assembly.
Next, I added the throatplate, also pre-drilled for stay bolts, Photo 23. Note the threaded hole for the blow-down valve – since it is not advisable to silver-solder Cuni, I built up a ‘mound’ of Cuni filler metal to increase the thickness of the plate at this point. That surface was then ground flat, in order to accurately accept and seat a commercially obtainable ball valve. ‘Plumber’s tape’ of PTFE (Teflon) will then make the joining watertight and secure.
My next problem was to figure out a method to attach the boiler to the frame of the engine – this is normally accomplished with the use of hollow stay bolts which enable the builder to use pass-through bolts with nuts on the inside of the firebox.
I didn’t like this idea, so I cut small, square pieces of Cuni plate, welded a short steel bolt to each of them on their back side, and then welded the pieces one by one to the side of the firebox.
Using a plasma-cut piece of steel plate with holes in exactly the same position as in the engine hornplates, I was able to tack-weld these small ‘bolt-plates’ in place. In Photo 24,I have started the work, one plate is welded at left. It is of course necessary to remove the aligning plate in order to weld the small plates all around.
Photo 25 shows all but the central bolt in place (it cannot be attached until the stay bolts are in place). Not a very beautiful solution, but it works;
“Since this joint will be inaccessible once the boiler is finished, it was very important
to do it properly...”
Photo 26 is proof that all the bolts fitted into the holes that were previously cut in the hornplates.
There is one drawback with this method – the boiler cannot be removed from the engine without first disassembling the frame and its gearing. This, I hope, won’t have to be performed very often – again considering the fact that I intend to fire the boiler with propane. A coal-fired boiler might have to be removed from the engine more often for cleaning and possible repairs.
“There is one drawback with this method – the boiler cannot be removed from the engine without first disassembling the frame and its gearing...”
Backhead and fire door
Continuing work on the inner firebox, I had formed a backhead plate which will incorporate the fire door. It was an easy job to mill a hole that will accept a piece of Cuni tubing flattened to a round-cornered rectangular shape, this is being done in Photo 27.
I have marked the approximate position of the cab floor on this piece. The fire-hole opening of course needs to be positioned above the cab floor. A mistake here would spoil the entire boiler! An ordinary end mill cuts Cuni very easily. Again, note the piece of wood under the plate.
In Photo 28, I have welded the inner backhead to the inner firebox. The fire-door tube is also securely attached with a fillet weld. On top of the firebox you see two substantial ‘girder crown stays’ welded in place. They each have three circulation holes, improving the heat transfer from firebox to water. Note that girder stays of this type are not recommended for pure copper boilers, but are acceptable for steel. Since
Cuni is much stiffer than pure copper, I consider using these to be okay in my own boiler.
There is still much more work to do on the engine before I can finish the assembly of the boiler; for instance the attachment of the steam cylinder and the crosshead support is done directly to the boiler shell. Thus, I’ll return to completing the boiler in a later article.