Office of Steam Forum for Model & Toy Steam Gas & Hot Air Engines

My thinking is that within a closed cylinder with immovable walls and top, the expanding gasses will move the piston to lower the pressure without significant regard to crown shape. Old time flatheads often had pistons with a cylindrical bowl cut out of the middle, (think ash tray) but they were flat heads with side valves. At low compression, as was used in pre-ethyl lead gas engines, such as the model T at 4:1, heavy flywheels meant they could idle very slowly.
I think the larger hit and miss engines benefit from scale, and from well oiled plain bearings. Ball or roller bearings are more durable, but not as silky smooth running as a bronze or even Babbitt bearing that is well worn in on a smooth shaft.
A big honking iron flywheel thumped vigorously by a large volume, low velocity push from a mildly compressed charge is like pushing a bus with a car.  Our little engines with comparatively low mass, relatively rattly bearings, larger tolerances per unit of size (.005” isn’t a lot on six inches. It’s a ton on six mm), and relatively small, perhaps more energetic firing as occurs when the resistance to initial movement is less in the small chamber of a light engine, ends up more like hitting the car with a sledgehammer than pushing it. The initial movement of the piston and rod and crank journal doesn’t translate into the same kind of low velocity, high momentum push as the big motor gives its flywheel. 
And once in motion, our parasitic drag is probably higher per unit of momentum, which degrades coasting more quickly. 

I understand what you’re trying to do. I think it won’t amount to much if any repeatable improvement. I like Gil’s method. Frankly I think the slight dishing and significant shortening is a fair compromise in not reducing wall length or material thickness too much. I don’t think the gasses pushing against it care, so long as it’s not a needle, shoving the gasses past the rings instead of moving the piston.

I say this as someone with decades of experimentation with expanding gasses within relatively cylindrical chambers with varying piston shapes at varying rates. It’s accumulated knowledge from designing bullets for handguns with varying burn rates from lightning fast to black powder slow.
 
I think another area to explore might be the Governor components. Slick and repeatable, rather than rough and clattering mechanisms eat far less.