starship-design: Linear EMP motor (name needs work)

[Steve VanDevender <stevev@efn.org>]

Connor writes:

 > So once I was reading about an idea for propulsion, one where the sun
 > and the earth were connected by a long aluminum bar. So the idea is,
 > the sun emits an electromagnetic pulse for a minute, and it takes
 > eight minutes to get to the earth. When it reaches it, the earth is
 > pulled foreward in the pulse for a minute, and the whole structure
 > moves foreward. This happens again and again, and the whole thing
 > gets going pretty fast. By the way, this is hypothetical.
 > 
 > So this isn't really practicle for interstellar travel, because it's
 > so big, and I don't think the sun does that :). But I was thinking
 > that you could keep the apparatus in a big tube of liquid sodium. The
 > speed of light in liquid Na is only 52 ft. Per second, so with a
 > fancy computer it would probably only have to be 25 feet long. How
 > does this sound to anybody?

I don't get it.  If the Earth pushes or pulls on this rod, then the
compression or extension travels back along the rod at no greater than
the speed of light and pushes or pulls back on the Sun, and there is no
net motion.  Remember conservation of momentum?  There's also no such
thing as a perfectly rigid instantaneously-transmitting rod.  Matter is
held together by electromagnetic interactions, so any tap on one end of
a rod propagates at at most the speed of light along it to the other
end.

Basically, there is no way to induce motion that violates conservation
of momentum; if something moves one way, something else has to move the
other.  This can get rather subtle, in the case of propulsion that
involves emitting electromagnetic or gravity waves in one direction so
that the drive system goes in the other, but there is always _something_
that gets thrown out the back or pulled on from the front.