>I'm not an engineer so bear with me... If you tether two masses and
>winch the tether in, I suppose the total energy of the system doesn't
>change, but the angular velocity goes up, right?
Not quite correct. Angular momentum -- mass times radius times (linear)
velocity -- is conserved. Energy is not: you're putting energy into the
system with your winch, doing work against centrifugal force.
>If you then cut the
>tether, will the masses merely fly out to their previous orbits, or
>will they fly apart?
Hmm, depending on details, the winching could add enough kinetic energy to
the system to exceed the gravitational binding energy. (Escape velocity
rises as the radius shrinks, but conservation of angular momentum makes
the linear velocity rise too, and faster.) Otherwise they'll end up in
elliptical orbits.
>...Then cut the tether at the right
>moment. Does this give an energy advantage over just sticking the
>engines on a single asteroid and making it a big conventional rocket?
As above, you may not need engines. If you do, the energy advantage is
probably small. If you don't, it's considerable, because the winching
will probably be much more energy-efficient. And unlike a rocket, it
doesn't throw mass away. (Well, if you don't count the other half of
the asteroid, which probably doesn't end up anywhere useful.)

Signature
MOST launched 30 June; first light, 29 July; 5arcsec | Henry Spencer
pointing, 10 Sept; first science, early Oct; all well. | henry@spsystems.net