In last week’s post, I raised the question outlined by the title of this post, and I mentioned the main problem being that a Roman would never consider doing it. Hero’s device in the Great Library of Alexandria is a dinky toy, but nobody would seriously consider that it could do useful work. In my ebook, Athene’s Prophecy, that problem was overcome by Athene telling the protagonist to do it. Easy, yes. Cheating, yes, but in fiction, why not? One problem for Romans is that primitive steam engines have to be very big to do a useful amount of work, or operate at high pressures. Newcomen designed the first one because too many miners were required to bucket water out of mines.
The first question is, how much steam pressure? Actually, the required pressure need not be exceptionally high, because Newcomen’s engine (the first steam engine that did something useful) actually worked by atmospheric pressure. The way it worked was that there was a finely balanced beam, and the steam provided just enough pressure above atmospheric pressure to lift the piston and push the beam. A squirt of water then condensed the steam, and air pressure pushed the piston down, and it was this down stroke that did the work. However, I did not want to simply reproduce the Newcomen engine, so I made the concept use higher pressure so the steam did much of the work, although there was (or will be, in Book 3 of the trilogy) also a steam condensation cylinder.
A major problem then arises: how to join large pieces of metal together? The Romans knew the principle of the bolt, and they made very small ones by soldering wire onto a metal shaft to make jewellery but they did not know how to make them reproducibly. They knew about soldering, and some of their mixtures were of sufficiently high temperature when melting that they were more akin to welding or brazing, they knew about the rivet, and finally, they had a process known as sweating, essentially heating one piece of metal (preferably a pipe) so that it expanded and could slip over a cold piece, then when it contracted as it cooled down, there was a firm joint. My answer was to use a variety, but emphasise the bolt to get the strength, the idea being to join cast pieces through a flange while employing a leather gasket. To make the bolt, I had my protagonist find workmen in Damascus to develop cutting tools similar to that used by plumbers to thread metal. Is that reasonable? I leave that to the reader to decide. The tools have to be harder than what they are cutting, so the bolts were to be made in bronze, and the tools in Damascus steel, which was actually harder than standard steel, the reason being that the local ores had a small vanadium content.
The next issue was, could they make the necessary metal objects. They had developed quite intricate ability at casting bronze, so I assumed they could, given practice. The engine I thought up for my protagonist was in part based on a design for a fluid hand pump that you can see in the British Museum (or at least I saw it there). The concept was that instead of the piston going up and down and pulling and pushing fluid, the steam would push the piston, the cycle being completed by the inertia of a flywheel. They could make a small piston and cylinder, so I hoped they could scale up.
Perhaps the biggest single problem lay in pipes. I have no idea how long a pipe the ancients could have made, so the design had to assume they would be short. The next problem lay in valves. The valves in the hand pump were simple flap valves, which work well enough when the force comes from the piston, which can exert force either way, but steam will hold a flap valve open from the boiler, and force the exit valve the same way. All that will happen is that you have the most complicated kettle exit! So, I suggested valves that operate by slightly rotating a metal cylinder with a hole in it embedded into a pipe, and operated by a rocker arm. Two valves were needed, or a double valve. I opted for the latter, on the basis that now only one rocker arm was required.
Could something designed like that work? I think so, given enough effort, but maybe not in practice. But the point of the story is not to design a steam engine, but rather to illustrate the process of invention, which is essentially a lot of trial and error, and the making of incremental improvements on a principle. Also, of course, this is only a side-issue for the story underpinning the trilogy.