Icarus wrote:

How much could the Shuttle decelerate with its own engines, given an
unlimited supply of fuel, before it encountered significant
atmospheric drag? I was thinking that if they built a (relatively)
cheap and cheerful rocket for the sole purpose of getting rocket fuel
into orbit, the Shuttle could use that for its descent, and the same
orbiting refueling stations could also be used for spacecraft leaving
Earth orbit, making (say) missions to other planets cheaper and
quicker.

Let's look at it this way. Suppose that the space shuttle could just
"magically" lose all of its orbital speed in an instant. The pilot
throws a switch and, presto, the spacecraft no longer has any
horizontal speed. What happens next? What does happen is that the only
component of motion remaining is the ship's downward acceleration
towards Earth; It falls like a rock. That's no good, because this alone
will also subject the ship to a significant air speed with high
temperature. (Not as high as with the re-entry trajectories that are
actually used. But high enough to be a problem still.)

So you see it's not strictly a problem of having enough fuel to slow
down. Slowing too rapidly can actually be counterproductive. The
spacecraft needs a way to travel slowly both horizontally and
vertically. It might be conceivable at least hypothetically to deliver
a large tank of propellant to the space shuttle, which could then
retrofire its motors to slow it waaayy down horozontally. But to keep
it from falling too quickly vertically would require a high *upward*
thrust for the entire trip down. There's just no way to do that. The
mass ratio required would be impossibly high.

Of course, as far as the Shuttle is concerned, the easiest solution
would be just to make its heat shield 100% reliable...

Probably the best thing would be to abandon the whole space shuttle
paradigm of carrying both cargo & people all in one package. It'd be
much safer, much more reliable, to send them up separately on
conventional rockets. The Russians have been doing it this way for
decades, and it works very well, and more cheaply. They built an entire
space station (in fact, several space stations) by boosting the
components and then rendesvousing crews with them later, in the Soyuz.
The Soyuz is a very highly optimised configuration for a disposable
orbital ferry. It's designed so that the ablative heat shield is a
minimal size, to keep its mass down. It works nicely. I think that in
45 years there's never been a genuine failure of an ablative heat
shield, either Russian or American.

-Mark Martin