The Air Force is growing concerned.
I suspect that ULA has been so focused on competing against SpaceX on price that it’s ignored the need to compete on tempo. The fact that SpaceX reuses its boosters means that its launch cadence is not constrained by manufacturing capacity. I think that ULA made a bad decision in assuming that launch rates would not be dramatically increasing when doing its cost/benefit analysis on reusability.
As I pointed out in the Starliner thread: a snippet from a 2016 interview reveals ULA’s thinking at the time when Tory Bruno focused (over conservatively) on cost and what it would take to recover cost assuming launch pacing remained the same.
The flaw has remained baked-in for a long time.
The Vulcan re-use architecture assumes ULA can churn out rocket bodies and first stage tankage at a rate competitive with the ability of SpaceX to inspect, restack and refuel. Who do you think will win that one?
As if competing against Falcon 9 and its expendable 2nd stage isn’t bad enough. What if SpaceX decides at some point to downscale the fully resuable Starship to replace Falcon 9? Or maybe offering rideshares on the existing SS/SH would be enough to co-opt the entire launch market? Of that I’m skeptical. But like filling an outside straight hand in poker, SpaceX has outs.
But like filling an outside straight hand in poker, SpaceX has outs.
Speaking of poker, Doyle Brunson, died one year ago today. He was 89.
I don’t see SpaceX down-scaling Starship. They’d much sooner increase production and decrease unit costs.
A bigger threat is emerging competition: Rocketlab will fly Neutron next year, which is just a little less capable than F9 (but with growth potential baked-in) and it may be cheaper to operate than F9.
And there’s Relativity and Firefly working hard on medium-lift vehicles not much further behind as well.
Vulcan is a very good rocket. But it is at high risk of becoming obsolete very quickly indeed.
I’m all in favor of other companies working to develop reusable rockets. I’m also a big fan of Rocket Lab. One challenge any competitor is going to have is launching at low costs while recouping their development investment. SpaceX has recouped their Falcon 9 R&D costs many times over. While we don’t have any real insight into their costs, it’s very likely SpaceX could cut their launch prices by a large percentage and still be profitable. There’s no market pressure for them to lower their prices any more at this time. They’re already less expensive than just about everyone else. When they do have one or more competitors, they’ll lower their prices to maintain market share.
They’re already less expensive than just about everyone else. When they do have one or more competitors, they’ll lower their prices to maintain market share.
To a point. The limit would appear to be the non-resusability of the upper stage. If you are competing on the cost of upper-stage replacement then you are on an even playing field.
Even then the reuse architecture of Starship/Superheavy addresses that as well. But the size is hard to justify for small sat operation or quick launch on demand.
If your expendable upper stage is cheaper to produce than the Falcon 9 2nd stage then you have something to compete with.
SpaceX got there first using government money and by subsequently expanding the commercial market rapidly recouped their cost. The market is a little more mature, but there seems to be room for additional players. As they say, interesting times.
This “SpaceX got there using government money” that I see over and over again is annoying. I’m not sure it’s true.
Where are all the government contracts that SpaceX supposedly got its money to start off with from? What percentage of SpaceX revenue has come from government contracts vs. private and when? Most of the current launches I see are for things like Starlink.
Development of Falcon was Musk putting his own money in, to the point where he almost ran out. Or so the public story is. I’d like to see a contradictory story.
Has the government given some contracts to SpaceX? Yes, reluctantly, using much more onerous terms than for ULA or Boeing. SpaceX has performed much better than those two. Perhaps the government can learn from that and impose more onerous terms on those two.
But then SpaceX doesn’t have the government/company lobbyist revolving door down yet. The other companies have a much longer head start on that after all.
It’s well established. If you don’t trust Wikipedia try Britannica instead:
https://www.britannica.com/topic/SpaceX
NASA funding got both Dragon and Starliner going. Much more went to Boeing / Starliner than did SpaceX / Dragon.
SpaceX however didn’t get to where it is today on NASA or DoD money alone. Far from it. Also SpaceX had the sue the Air Force at one point to get it to reform its procurement procedures. It was settled out of court. IIRC.
This “SpaceX got there using government money” that I see over and over again is annoying. I’m not sure it’s true.
Look this is not my day job. Falcon 1 was privately funded starting with $100M of Musk’s own money I presume mainly from what he made off PayPal. Plus money SpaceX raised from private investors. If Falcon 1 Flight 4 had not reached orbit, Musk himself has admitted SpaceX likely would have ceased operations.
But Falcon 1/Kestrel was not the rocket/engine combo that got SpaceX where it is today. Without COTS money I’m not sure how far Falcon 9/Merlin would have gotten.
AFTER the success of Falcon1/Flight-4, NASA and DoD money started to flow. NASA COTS then later the CCDev program, getting full blown going with the looming demise of the Space Shuttle put the accelerator pedal to the floor for SpaceX. If you won’t take my word for it, maybe you will Hans Konigsmann.
Falcon 1 was Merlin as well as Kestrel. By the time SpaceX was ready to go full-out on Falcon 9, it already had the Merlin in-hand. And, yes, SpaceX took NASA money at that point to develop F9 and Dragon. But that money had to be matched by new funds privately raised. SpaceX raised the necessary additional funds.
And it’s not as though NASA had a lot of choice in the matter. Some people still talk like NASA’s COTS contract to SpaceX was some form of gift. Far from it. The Shuttle’s days were already numbered. Ares 1 and Orion were supposed to replace Shuttle for ISS crew transfers, but couldn’t also haul sufficient freight – thus COTS. SpaceX came through. As money is fungible, it’s entirely fair to say that SpaceX developed F9 on its own dime and used the NASA money to develop Dragon.
Dick wasn’t OSC Antares also COTS money IIRC?
Yes, it was. And Cygnus as well. Orbital was pulled in as a substitute COTS participant after Rocketplane-Kistler failed to raise its required matching funds and was cut from the program.
The Vulcan without the SRBs is less capable to both LEO and GTO than a Falcon 9 doing a RTLS mission, which is the lowest cost option.
The Centaur V throws away two RL10-C engines at $20 million each, so there’s no way they can compete with a Falcon 9 on price while saddled with a $40 million upper stage engine cost. And even with 6 SRBs (at I think $7 million each), it can only launch about half as much to GTO as the Falcon Heavy.
So I’ve read that Vulcan’s cost ranges from $100 million (no SRBs) for sub-Falcon 9 performance, to $140 million (6 SRBs) for half a Falcon Heavy’s performance.
My take is that they went with the wrong architecture. Their core stage lift-off thrust is 1.1 million lbsf, versus Falcon 9 at 1.7 million lbsf. That puts them in a performance deficit that they’re try to cover using SRBs, which aren’t cost competitive. A larger core with three BE-4s would’ve given them 1.65 millions lbsf at liftoff, with the higher ISP of LOX/methane staged combustion. If the BE-4 costs $6 to $7 million each, that’s $18m to $21m for the core engines, versus the two engine, six booster design which is $12m to $14m for two BE-4s plus up to $42m for the six GEM-63XLs, so 54 to 56 million dollars for engines instead of 18 to 21 million.
But what really kills it commercially is expending a pair of expensive RL10-C engines. The second stage engines cost three times more than the first stage engines. Financially, the should’ve added more cheap engines to the first stage so they could get rid of one of the RL-10s, or go with a much cheaper but lower performance upper stage engine, perhaps something like a Merlin.
Good data. Thanks for that George.
But what really kills it commercially is …
…giving the Air/Space Force exactly what they wanted. A rocket dedicated to THEIR payloads, for a price that’s easily paid with “Other People’s Money™”. Since nobody else will want to pay that.
They (The Air Force) has spent the money
How much lift have they gotten?
It’s not so much a matter of what they’ve gotten as it will be what they get in the end. If you get my meaning…
They (The Air Force) has spent the money
Correction. We have…
Financially, the should’ve added more cheap engines to the first stage so they could get rid of one of the RL-10s, or go with a much cheaper but lower performance upper stage engine, perhaps something like a Merlin.
It seems ULA had its heart set on a hydrolox upper stage no matter what. So switching fuels by ULA I would guess was a non-starter. Now according to a Reddit post, the older RL-10B comes in at $6.5 million, so if they’d gone with a DCSS derivative as a second stage and added some more cheaper umph to the core stage as you suggest, you’d be (upper end) $21 + $6.5 = $27.5 call it $28M on the rocket for engines.
It would be interesting if BO buys ULA and makes Vulcan-2 with more engines on stage-1 to eliminate the GEMs, and a BE-3U upper stage engine.
Of course, then tankage size would need to change and so on, so not a small modification.
A Vulcan-Heavy has also been suggested as possible.
A Vulcan-Heavy wouldn’t pay off unless they managed to land the boosters, which is back to the real problem with Vulcan’s first stage being expendable.
The Centaur V throws away two RL10-C engines at $20 million each, so there’s no way they can compete with a Falcon 9 on price while saddled with a $40 million upper stage engine cost.
Pardon my ignorance, but I thought the whole point of the RL10-C was a cost reduction over the largely hand-assembled RL10-B?
According to the Reddit link I posted earlier, I’ve read that the expense driver for the RL10 is:
1) Touch/labor: the nozzle/combustion chamber has delicate tubes that are bent by hand to shape and size needed and then hand-brazed together. Injector plate is equally complex. (see the link)
2) Overhead: Mass production significantly reduces cost per engine.
Now the RL10C is supposed to replace a lot of that hand assembly with 3D printing with cost reductions projected to be about 70%. So why would the RL10B come in at $6.5M per unit while the RL10C is at $20M per unit? It must have far more to do with item 2 above than item 1? Does Aerojet project cost savings after large block orders? Supposedly (again according to the Reddit link) the price for RL10B was driven down by a block order of 100 by Boeing up front (presumably for Delta-IV?). The Reddit link also claims that ULA as of 2 years ago made the largest block order of RL10s (RL10Cs presumably) ever. So why so pricey?
Bidenflation.
The RL-10 is a beautiful 20 million dollar engine that should cost 250,000 dollars a piece.
I remember once reading, if mass produced, an RL-10 should be about the same cost as a turbine engine for a helicopter.
The US Air Force should be worried if SpaceX has capacity to pick up the slack because ULA isn’t doing jack.
Except spending money, of course
What does BO get from buying ULA? Launch contracts and some IP they may or may not need?
It would be simpler to let ULA die and then compete with SpaceX on the contracts, assuming BO has a launcher at that point.
Launch contracts, that’s it. But that’s a lot, especially when they’re U.S. Government launch contracts, whose special feature is that they pay huge amounts of money in return for nothing whatsoever.
Thanks.
I guess a profit is a profit but it is a sad way to go.
To be fair to ULA, SpaceX doesn’t have to spread its procurement chain across most of the Congressional districts in the US. But they also lack leadership to be nearly as innovative as SpaceX.
SpaceX changed the game on both cost and tempo, and the true magnitude of the tempo change only came to light in the last year or so…well after Vulcan had its design finalized. Given the past environment that other rockets were developed, it’s fairly impressive that ULA could pivot on one point, let alone two.
There’s no reason to continue Vulcan once New Glenn is flying. Especially once it has that reusable second stage. Musk himself noted a 7m version of Starship would be viable for LEO/GEO/Lunar, and that describes New Glenn/Jarvis. SMART is a technology whose time came and went.
A point that is too often missed is that the final version of Starship/Superheavy is not only cheaper by the kg, but also cheaper by the launch than Falcon 9. Even if you just launch individual GEO comsats with SS/SH, it will make more money than anything else. Since that’s also true for New Glenn/Jarvis, that’s the competition. Maybe Russian and Indian engineers working for Circus Peanuts?