Rocket Will Be Recaptured in Midair with a Helicopter

Also interesting is the reasoning behind capturing parts of the rocket (and it is only parts) in midair. ULA say this is because the expensive bits are small, and essentially imply that they can get most of the 'wins' with not much effort when compared to SpaceX's "land it on a boat" approach. This may well be true but it seems SpaceX's approach is thinking in wider terms of the re-usability required for frequent space flight, rather than cheap space flight.

I have a keen interest in space flight and SpaceX in particular, so take my opinion with a pinch of salt, but this article reads to me like SpaceX are well over a decade ahead of the ULA, much more forward thinking, and overall the more exciting player in the commercial space-race.

This is how they used to recover corona/keyhole spysats.

They lost one in Mexico.

http://news.discovery.com/space/history-of-space/the-first-r...

Edit:

Uh, the Mexico thing may not be commonly known, it turns out. The chute failed to deploy properly on a mission in '74, bounced off a C-130, and slammed into a field in Mexico, but the payload canister was still intact. It's apparently commonly written up as a UFO sighting called the "Coyame Incident".

Meanwhile, SpaceX will be attempting to land their entire first stage again this afternoon.

Batman even used it in The Dark Knight :)

http://en.wikipedia.org/wiki/Fulton_surface-to-air_recovery_...

I think that the collection process is more involved (concerning number of "moving parts", i.e. requiring a number of independent systems to work in perfect synchronicity) than what SpaceX does. The F9R does have a more complex navigation and guidance system due to the landing option. But apart from the fins and legs[1] (which are of minor complexity), there's no additional subsystem required for the landing capability. Everything's already there and used in other parts of the launch – including major parts of the landing navigation and guidance system. And SpaceX could launch another F9R with the same 1st stage on the same day – ULA would need days, at least.

So I think ULAs approach is over-engineered and does little to improve on the idea of launcher reusabilty.

[1] Actually, the legs may be used before launch in the future, so one less surplus subsystem ;)

From [1]:

Vulcan, slated for its maiden launch in 2019, will be powered by either a pair of exotic Blue Origin BE-4 liquefied natural gas main engines for 1.1 million pounds of thrust or two conventional Aerojet Rocketdyne AR-1 kerosene powerplants for a million pounds of thrust.

I'm also concerned about using fewer larger engines. While this can be more reliable (less moving parts means less chance of failure), it also means there is no engine-out capability to continue the mission. This also means a lower production volume, which means higher cost per engine, all other things being equal (which they aren't).

The word 'exotic' also gives me pause.

At any rate, it should be interesting. I hope they post testing videos!

[1] http://spaceflightnow.com/2015/04/13/ula-unveils-its-future-...

An overview of a 2008 ULA proposal for mid-air recovery of first stage rocket engines: http://selenianboondocks.com/2008/09/partial-rocket-reuse-us...

An article about Hiller's 1965 proposal for using a giant helicopter to recover an entire Saturn V first stage: http://www.thespacereview.com/article/1045/1

Rather than ULA expending the intellectual effort to build the control and guidance systems needed to land the entire rocket as SpaceX does, they took the lazy route and added a few steps to their existing process.

The fact that they're willing to suggest that something like this is innovative is just further proof that the entire ULA is staunchly resistant to change. I'm not entirely sure this system will even have a chance to be used.

Isn't that introducing quite a few potential points of failure? Or was this all thoroughly solved in the 60s and is boring and not an issue?

1.) It's good to see ULA stepping up to the plate and actually innovating. SpaceX will need some competition by then, to keep the industry healthy.

2.) The name, "Vulcan", is pretty badass. Certainly far better than the other, embarrassingly jingoistic candidates they were considering.

3.) The inclusion of innovative NewSpace companies like Blue Origin and Xcor is pretty exciting, even if ULA hasn't actually committed to using their technologies yet.

That said, there are ways in which the Vulcan seems like it was designed by a committee, and doesn't quite seem fully-baked. In particular: the "dial-a-rocket" approach of strapping on solid boosters is at fundamental odds with the economics of reusability.

In an expendable rocket, solid boosters are used because you need a lot more thrust early in the ascent (to overcome gravity losses), and solid rockets are a relatively cheap way to achieve that. (Solids generate a lot of thrust during a short period of time, but their efficiency over long-duration burns is relatively poor, which is why they're generally not used all the way to orbit). You could accomplish the same thing by using more liquid fuel and more engines in the primary booster stage -- but turbopump-driven liquid engines are a good deal more expensive than solid boosters, so it doesn't make sense to do this if you're going to dump them in the ocean after a minute or two.

On the other hand, if you're recovering the engines, then using more of them is absolutely the right thing to do. In that case, the tradeoff isn't between the cost of liquid engines vs. solid boosters; it's the difference between the cost of refurbishing liquid engines (and some extra fuel) vs. solid boosters. In this case, the latter is likely to be more expensive by an order of magnitude. So if your hardware is reusable, then it's better to have one or at most two vehicle configurations, and simply accept that most of the time your vehicle will be oversized for its payload. You also get some additional benefits from doing this, such as greater mass-produceability and engine-out capabilities, etc.

I suspect that the problem is that aerospace engineers are used to doing high-level trade studies using a crude $$$/kg costing for the cost of a rocket -- at least, this is how I've frequently seen them do it. If this is how you think, then you want your vehicle to be no larger than the payload requires -- and that's what the dial-a-rocket approach allows you to do. But it's kind of like assuming that a truck costs $$$/kg, and therefore if you need to carry a load across town, you should build a truck which is as large as you need and no larger. This makes sense if you take it as a given that you will always be building your trucks by hand, and that they will be thrown away after every journey. But if both of those assumptions are false, then so too is the "build the truck to fit the payload" design philosophy.

It doesn't look like ULA has figured this out yet -- but they've definitely realised that there's a fire under their tail, so hopefully they will do so eventually!

"ULA says it's going to try to capture only the most valuable parts of a rocket"

So... how about when we fly NY -> LA in our 777 we only bother to land the passengers in the remaining "valuable" parts of the airplane? See the point ULA? We're trying to make spaceflight just like commercial airline traffic. That's the ONLY way to make it commercially viable. Your cost-plus methods of building expensive as possible single use bottle rockets are over man!!