Could micro-technologies make a live steam N-scale steamer technically and economically feasible? A national lab has already demonstrated a working steam engine that you can see only under a microscope, and that can be produced, thousands at a time, on a silicon wafer using ordinary silicon manufacturing techniques. It vaporizes water and produces a measurable output. So I dream. Could this technology be scaled up from the micro world to the mesa-world, i.e., one or two (or more) magnitudes of order? It would still be very small--about the size of specks of dust. An alternate approach would be to take thousands of these micro-machines, and synchronize them into one bigger machine. The "coal or oil" needed to power these machines is abundantly plentiful from the rails--12 VDC at even one amp would be like drinking from a fire hose. So the tender could hold nothing but water--real water, though it would have to be incredible pure. So we have a working steam engine, whether it's one, or a few, or a thousand micro-components working together with one purpose--to create energy. How do we use this energy? Do we try further micro-engineering to create real cylinders, with micro-valves? It's been done at a microscale, but now we're needing it on a much larger scale. A purist would insist, but here we get into even newer applications of micro-, mesa- and minature technology. Possible? Probably. Economical? Probably not. Could we be satisfied with using steam to power a micro-generator? Probably not--why not use the electricity from the rails to directly drive an electric motors, which is what we do now. Today's scheme would be more efficient, I think. This is where my dream bums out. I dream that I can create the steam in a package small enough to live in an N-scale boiler, with more than enough space for weight. That is, much less space than today's motors and drive-trains. But I can't figure out how to transfer that energy to a drivetrain, either by prototypical or other means. Any ideas?