@[email protected] @nyrath @FredKiesche

What sort of vehicle should accelerate at 1 g or so, though? You get the benefits of the Oberth effect at much lower accelerations, and designing for lower accelerations provides a lot of benefits. You get reduced structural mass - especially for solar arrays and radiators, lower thruster/engine mass, and usefully better specific impulse.

@isaackuo @sudnadja @nyrath @FredKiesche If you've got a magic reaction drive (unlimited thrust, no waste heat, no fuel mass problem), then you might as well thrust at whatever acceleration is best for the cargo. If that's humans, it would be the *lowest* acceleration to keep the adverse medical effects of microgravity at bay (probably a lot less than 1g but more than 0.1g).

@cstross @[email protected] @nyrath @FredKiesche

Depending on your definition of "unlimited thrust", you could squint and say artificial spin gravity qualifies. It obviously has no waste heat or fuel mass problem.

If your spacecraft spins, the most intuitively obvious thrust axis is parallel to the spin axis, but it's not a no-brainer. For example, with solar electric you probably want the spin axis pointed to the Sun, but you usually want thrust perpendicular to the Sun.

@isaackuo @cstross @sudnadja @nyrath @FredKiesche One problem is that you better get your thrust _EXACTLY_ through the center of spin or your ship starts precessing, further exacerbating the problem. The famous docking sequence at the space station in "2001: A Space Odyssey" for example is a disaster waiting to happen.

@60sRefugee @cstross @[email protected] @nyrath @FredKiesche

It's pretty easy to get the average thrust exactly through the center of spin if you simply spin the entire spacecraft (including the thruster). I actually prefer placing the thruster on the tip of the spacecraft opposite the crew module, because that lets you efficiently perform cyclic thrusts to alter spin axis (and thus, thrust direction).

It's more complex if you need main thrust to be in a different direction than spin axis, though.