Let’s talk about the SpaceX attempt to soft-land on its barge-landing system. Here’s a video of the unsuccessful, but still very exciting landing attempt:
Apparently, for the descent trip back after stage separation, they use an open-loop hydraulic system to control a set of the steering vanes. In this case, open-loop means that they use the kerosene from the fuel as a hydraulic medium for actuating the vanes. SpaceX said that they didn’t carry enough margin to control the system, leading to the crash. Initially, I bought that explanation. Now that I’ve had a few days to let the thoughts marinate, what bothers me is that if the kerosene ran out to control the vanes, why was the engine still at thrust during the crash? Also, at the then-low descent speed, what would the vanes have to do with touchdown?
I’m thinking some other factor or set of factors caused the crash.
It is much less embarrassing to say that the crash is the result of a technical problem that you can throw engineers at than to admit that an assumption of good weather and calm seas are a given at landing time. If it were the latter, it was a failure of basic planning, and they can fix that either by being more circumspect on good landing conditions on the ocean, or pressing hard to get land landing approved by the FAA.
What could be even worse to admit is that the actual landing control authority you thought you had wasn’t really there. In other words, a basic level of advertised capability didn’t pan out. That would be much harder to recover from as a program.
Don’t get me wrong, I really don’t buy into the SpaceX vs Orion competition meme. I wish Elon Musk and his team all the luck in the world. Really and truly, the outbound expansion into the solar system and beyond isn’t going to be the purview of government forever, it just can’t be. Accordingly, the desire to have a successful fly-back booster is a good economic one, and government never equals good ROI. As fuel is relatively cheap compared to hardware*, and if the hardware can be truly refueled and reused, then it makes sense for SpaceX to try to soft land their first stage. I can certainly understand the motivation.
*If they want to loft more mass to orbit, then it becomes a surcharge to the customer to pay for the booster. Otherwise, you just limit the mass of useful payload to allow for the landing fuel margin. That is, you use the booster in a non-flyback mode to trade fuel mass for useful payload mass. For any given flight, then, SpaceX has to assume some risk that they lose a booster, but under nominal flights the booster is already paid for, or the customer pays a premium to replace it.