Rotating... amateurs. Just use a detonation-based scramjet burner. Bonus: could be air-breathing for a large part of the first stage. The downside would be the

Rotating... amateurs. Just use a detonation-based scramjet burner. Bonus: could be air-breathing for a large part of the first stage. The downside would be the massive instability from the need to keep a supersonic flow matched up perfectly with the detonation velocity. I'd figure stochiometric deviation would be a feasible way to control the detonation velocity, but the control loop would be hard to get right. And it might be necessary to use film cooling with liquid cryogenics to prevent the shock front from melting the engine wall.