This will be yet another ambitious mission for SpaceX. One of the things that strikes me about this stuff is that NASA would test some new feature, then test it two more times to be sure that it working wasn't a fluke, and then go for the next feature. SpaceX has been testing multiple new features on every flight, which has led to some really rapid advancement in their capability. If this thing successfully does its "pretend" landing on the water I am sure they will push for the next one to land back at the pad.
NASA would design a new feature, send it through half a dozen design reviews, apply a few pounds of paperwork, test it on the ground, test fly it (maybe) once, and then deem it "correct" and fly it as-is forever. Manned.
This happened with STS and it's happening with SLS. If you can believe it, the SLS/Orion schedule is this:
1. This year, the first Orion will fly unmanned on a Delta 4. For two orbits.
2. Three years later (2017) the second Orion will fly unmanned on the very first launch of SLS--around the moon.
3. Three years later (2021), the third Orion will fly manned, on the second launch of the SLS, to rendezvous with a captured asteroid.
4. After that, they hope someone somewhere will come up with additional payloads for SLS and/or missions for Orion. Know anyone who needs a giant rocket at $1B a pop? Have them call NASA. They're targeting one launch per year, starting 2022.
It's completely mad. I wouldn't ride on the second flight of a rocket designed by God Himself.
Apollo, granted, was at a completely novel scale. But still: one test, all up? Then three years? I won't be able to watch.
By the time SLS-1 gets off the ground (even if I grant them their schedule) SpaceX will probably be launching a reusable Falcon Heavy (at about 2/3 the payload of the SLS Block I) every other month, for $200M or less. And they'll have, oh, about 100 F9 core-launches under their belt.
The SpaceX plan sure looks like a better idea to me. For anything. Other than spending lots of money.
By the time SLS-2 is picking up their asteroid, Elon will probably be watching from frickin' Mars.
I repent using the broad brush against NASA. It's a very diverse organization, and above I mean the "big ticket vehicle design" side of NASA. And the core of their problem is that the whole program was basically designed by Congress. I'm sure they're doing their best to deal with the absurd hand they were dealt.
NASA efforts seem to be most sane in inverse proportion to the amount of attention paid to them by Congress. Much of what NASA does is science and small-scale engineering that Congress doesn't know or care about, and that's usually done pretty well and efficiently. But the more visible parts really make up for that in their absurdity.
(Even the more agile parts aren't exactly lean and mean, of course. Project Morpheus, which is running an Armadillo-derived rocket, is a very agile program for NASA, and has more than 25 people full time. Armadillo at its peak was, what, half a dozen?)
> NASA would design a new feature, send it through half a dozen design reviews, apply a few pounds of paperwork, test it on the ground, test fly it (maybe) once, and then deem it "correct" and fly it as-is forever. Manned.
Traditionally, until the Shuttle anyway, NASA has been more cautious. Apollo had 2 unmanned Saturn V launches and 4 unmanned Saturn 1B launches, 10 unmanned Saturn 1 launches, and 5 Little Joe II launches.
Beyond that, Apollo 8, 10, 11, and 12 were place on free-return trajectories for safety, and 7,8,9,10 were all essentially test missions for aspects of 11.
To be fair they attempted a soft water landing three launches ago with CASSIOPE, though without the legs. They are basically attempting this for a second time, and the addition of the legs will probably improve their chances (with CASSIOPE, the first stage began spinning around the vertical axis shortly before reaching the water, causing an engine cutoff. The legs are expected to improve the aerodynamics of the returning first stage to prevent this from happening again.)
I was the Launch Systems Lead for one of the primary payloads on this flight and was surprised when they announced this but excited to watch it happen (or attempted). Almost every Falcon/Dragon flight is innovating in multiple major ways which is NOT how Aerospace normally works, but why many of us we chose to work in the industry and a refreshing change.
CRS-3, by the way, will make good use of Dragon's unique down-mass capability. It will carry up and return three sample return freezers, and swap in a replacement for a malfunctioning space suit (bringing the bad one back for examination.) The bad suit (or EMU) is not the one that famously leaked back in July, interestingly enough, but the station's fourth suit that's been out of commission for a while.
I still find it hard to believe that it is energy-efficient to carry fuel all the way up and then all the way back down only to spend it in the last few meters of the decent. I'm sure they've run the numbers and decided it is but it is the most surprising part of this approach to me. Does anyone have an idea of how much fuel is spent in the return phase?
Not all that much, really. An almost-empty rocket stage is very light, and it doesn't take much to slow it down from terminal velocity (which can be rather low if you fall right) to 0. In fact, one of their main problems seems to be throttling down an engine enough that it produces less thrust than the weight of the thing.
I once did some rough numbers to guess that the landing propulsion will take 10-20s with 1-2 of the motors, using 5-10% of the fuel at launch. Hair-raising, but well within the capabilities of today's sensors, actuators, and computers. The extra fuel is not so bad on a first stage, which is not so weight sensitive as higher stages. And it may be about what it would re-enter with anyway, given reserves. In which case, it's a pretty good deal. A pinpoint 5mph dry landing is a lot nicer than a 30mph wet one, which it what you'd get with parachutes, and it only costs you the weight of legs, since all the other engine and guidance stuff has to be there already.
Energy efficiency also needs to consider how much energy you need to build the vector. It looks quite likely to me that reusing the vector would be more energy efficient too.
A coworker of a friend was recruited by SpaceX to help design the rockets themselves. He turned it down. Hes a product designer but damn I didnt know product designers could also be qualified to design rockets. Maybe it was just recruiter-speak but that's an incredible opportunity.
