The trouble is that no-one has ever assembled a multi-stage spacecraft in orbit. This is not likely to be easy. Provisioning, yep, that could be OK, but fuelling? Trying to transfer fuel at cryonic temperatures in orbit woudl be a seriously non-trivial task. maybe lofting a full tank which is tehnbolted into place? Grand idea, hope those seals are OK. Of coruse, you could do QA checks, but if anything fails them, what are you going to do, de-orbit them and loft the whole thing again? Better to send more or less everything ready-built and quality-checked on Earth up in one direct throw, no?
Whilst I'm generally supportive of Zubrin's postion on how best to get to Mars (see 'The Case for Mars'), and so far as I can see his assesment of the mass required may be OK, one thing I am a tad dubious about is the amount of living space the astronauts would have. That said, perhaps something as simple as docking with a Bigelow Aerospace BEAM unit to add living space might be all that's required, if the journey is to be done in freefall. It'd certainly be safer in a BEAM than in a traditional 'tin-can' vessel, although of course a 'tin-can' type is what's needed for passing through atmospheres. Better yet would be if the actual Mars ship can be tethered to the spent previous stage and teh whole assemblage spun slowly to give at least minimal g for teh astronauts to live in. Which is also a concept that hasn;t yet been tested.
It was the concept of having to construct a large spaceship in orbit that sunk the notion of going to Mars directly after Apollo, along with some fairly silly notions about the types of trajectories and length of stay on Mars (basically, least total mission time was aimed for, which resulted in least time on Mars, maximum time in freefall and maximum radiation dose partly due to a Venus 'fry-by' on the way home).
But anyway, it boils down to that if we want to get to Mars soon, then we need to avoid the mission relying on stuff that we haven;t done in orbit before. Which, so far as I can see, means docking one, maybe two things to another thing is ok, resupply is OK, but refuelling is a no-no. Otherwise it could be decades more before we send anyone to Mars, what with testing how to construct ships/refule ships in orbit multiple times first, not to mention testing tethering systems for 'artificail gravity'. I would be utterly delighted to be shown to be wrong.
Biting the hand that feeds IT
