The Space Shuttle was originally intended to be able to be turned around with little maintenance. But early design decisions and the state of the technology of the time just didn’t allow for it.
Part of the problem is that it was just such a complex machine. The main engines were incredibly complex things that required teardown and inspection after each flight.
Part of it is that the government is risk-averse and insisted on refurbishing or replacing a lot of stuff that technically might not have needed it. For example, all three engines on the orbiter were removed after every flight and fully serviced.
The design problem was that the heat shield on the orbiter was made up of thousands of custom ceramic tiles - every single one unique. Originally they believed that the tiles could just be inspected and reflown many times, but in fact after each flight many, many tiles would be damaged or missing. So between flights hundreds of tiles would have to be laboriously manufactured, the old ones removed and new ones put in place.
The orbiter itself spent about 100 days at the processing facility being cleaned, refurbished, and repaired between each flight.
Next, you had the main tank - this was just thrown away after each flight.
The side boosters - they were not ‘reusable’ in the sense that they could land and re-fly with minimal work. They splashed down in the ocean - a fairly violent event that would damage them in various ways. Then of course they became soaked in corrosive salt water. Reusing the boosters was more like salvage and rebuilding than re-using.
Then once you had all the pieces together, you had to transport them to the launch facility and assemble them all in the Vehicle Assembly Building.
The problem NASA got into is that once you lose your ability for short turnaround times, your flight rate drops and the cost per flight increases, because you still have to maintain the same massive infrastructure, only now you can’t amortize it over as many flights. NASA’s original plan was to be able to turn the orbiter around in a couple of weeks and fly many missions per year. But fixed costs for the shuttle program were about $1 billion per year, so when the flight rate dropped, the cost for flight skyrocketed. Over the life of the shuttle program, the per-flight cost of each shuttle turned out to be around 1.5 billion dollars. Overall, that works out to about $60,000 per pound of payload launched into orbit by Shuttles. On an incremental basis, each flight cost about $450 million, or about $8,000 per pound. Falcon Heavy should cost around $1000/lb. BFR more like $200 or so.
NASA learned all the wrong lessons from the failure of the Shuttle. They decided that reusability wasn’t worth it, so their new rocket is fully expendable. But ultimately, they never changed the core problems that made the Shuttle so expensive - a huge workforce spread across many centers, and a set of political requirements that forces them to use old technology inherited from the Shuttle so all those workers in all those scattered facilities can stay employed. The result is a new big rocket that’s already outdated before it ever flies, and which will only fly once every year or two at a cost of a billion dollars per launch anyway.
Seems reasonably relevant. Here’s the bottom line - Falcon Heavy can lift 53 tonnes into low Earth orbit for $90 million dollars. The block 5 variant will do better. SLS will live 70 tonnes into low Earth orbit for about $600 million dollars, assuming no more cost overruns. And by the time SLS is operational, SpaceX might be close to flying BFR, which may cut the cost to LEO by another significant amount because the entire thing is reusable whereas Falcon Heavy has to throw away everything other than the first stage boosters.
No, first you have to learn to walk. SpaceX set their original sights low, building a small rocket that would service a market for small payloads to LEO. That rocket failed the first three times they tried. If the fourth had failed, SpaceX would have been done. They had no resources to try for anything larger at that time.
Once they figured out how to fly Falcon 1, they discovered that the launch market for small satellites had shrunk, and that there was quite a bit of competition. So they ramped up to Falcon 9. In the meantime, they started to learn how to land a rocket on the ground again. They built a small test vehicle called the Grasshopper, and used it to develop the software and hardware necessary.
At this point, it was still an open question whether reusability could be made to work cost-effectively. There were serious technical hurdles to overcome, such as being able to restart a rocket engine while travelling hypersonically in the wrong direction. But SpaceX did it the right way - rather than commit to a grand design up front like the Shuttle program did, they just kept trying incremental changes until they had it figured out.
It was around this time that Musk realized that true reusability changes everything. For example, we are used to the idea of small rockets taking up small payloads and big ones taking the large ones. When you have to throw away the rocket every time, this makes sense. And so there are a wide range of rocket sizes available on the commercial space launch market. But if the entire vehicle is completely reusable, then why not just make one great big rocket and fly it for all missions? If your launches only cost fuel and maintenance, you can launch a 300 lb satellite in a vehicle that could have taken 100 tonnes, and still beat the competition. And then you don’t have to maintain multiple assembly lines, your inventory costs go down, and lots of good synergies come your way.
As for the other variants of Falcon 9, they were all incremental improvements done to improve reuse, lower cost, increase payload, etc. For example, the latest variant of Falcon 9 uses super-cooled ‘densified’ liquid fuel, which gives the rocket much better performance. That too had to be learned the hard way, and they blew up a Falcon 9 on the pad by accident while learning. Later versions also got the grid fins which make steering easier, and lots of other small changes.
Pricing theory is complex. If they charge just the same as, say, ArianeSpace, what are they bringing to the table for the customer? And risk has a price. SpaceX hasn’t been around long enough yet for people to fully trust their rockets - at least not as much as they might a Delta III or something, so SpaceX has to offer a discount.
Also, by setting the price low they can ensure lots of sales, increasing their flight rate, which decreases per-flight costs. There are economies of scale involved. This wouldn’t matter if the launch market was fixed in size, but it’s not. By offering substantially cheaper access to space SpaceX can grow the market and pave the way for even bigger rockets and more launches in the future.
Elon Musk is never short of PR ideas. For example, he just made a cool $10 million dollars selling flamethrowers with the name of the ‘Boring Company’ on the side. Leave it to Musk to come up with a way to promote a company such that the people you are promoting to happily pay you for the privilege.
The next big mlestone we will likely see from SpaceX (other than more Falcon heavy launches and routine Falcon 9 launches) will be the launch of the spaceship half of the BFR. Musk says their aspirational goal is to be able to do test hops with it by the end of next year. This thing is huge - the size of a full rocket in its own right - and will have the ability to take off and land again by itself. That should be pretty cool to see.
After that, the full BFR is supposed to fly within four years, but knowing Musk’s inability to project timelines I’d say it’s more like 5-6 years down the road. But if that rocket does what it’s supposed to, it will change the rocketry business forever. It will also change how we think about space and what can be done there. When access to space is so cheap that a single mining company could consider launching missions to the asteroids, or a movie studio could finance a picture in space, everything changes.
I’m sure it is. I am old enough to remember the aftermath of Apollo, when half the boys in my school said they wanted to be astronauts. It gets into the culture, too. Part of our ‘we can do anything’ attitude in the 70’s came from the space program. It was not uncommon to hear, “If we can put a man on the moon, we can do X”. I’m a firm believer that these sorts of things do a lot more to lift Keyne’s ‘animal spirits’ and invigorate an economy than they are given credit for.
Here’s hoping that Musk, Bezos and other people pushing the frontiers give today’s kids the same feeling.
They have been consistently late in developing and producing new designs, just like every other space company (or almost every large new engineering project). But their track record for actually launching commercial payloads on developed hardware is excellent. Half of all commercial space launches in the world last year were conducted by SpaceX. With Falcon 9 they seem to have managed a steady rhythm of launches on time and at the promised cost.
Yes, Falcon Heavy was 7 years late. And Falcon 9 has had its teething issues and delays. So what? Any comparison of timelines with other companies has to take into account just how much innovating SpaceX is doing. Along with keeping up with their launch manifest they upgraded their main rocket many times, pioneered many techniques like densified fuel and booster re-use, etc. Frankly, their performance has been astounding, even considering the missed milestones.
Or look at it this way - it’s taken North Korea, with all the resources of a state, decades of rocket development to manage to fling a rocket partway across the globe and crash in the ocean. In the meantime, just one of our many billionaires built a rocket and launched his own car to the asteroid belt for yucks. And now he’s going to sell access to space for 1/10 of what any government or large aerospace company ever managed.
You still haven’t provided anything beyond your own gut feeling that contradicts my claim that F1+F9 1.0 cost $390M to develop.
If you want to claim that some small portion of that $390M rightly should be counted against NASA–up to an absolute maximum of $100M, since that was all the CRS money SpaceX got before the first F9 flight–then more power to you. That’s just like your opinion, man. Even at that, the fact remains that SpaceX developed two rockets for an order of magnitude lower price than what NASA thought they could achieve, and with the majority of the investment coming from Musk and other private investors.
Dragon is a completely different subject, and you bringing it up makes it look like you’re deliberately confusing the topic. Nobody [li], SpaceX least, ever claimed that Dragon was purely privately funded. It is very clearly designed as a vehicle for ISS missions. And yet, even at that–and from your own link–SpaceX put in over half the money themselves. That is more than you can say that just about any other government contractors, who won’t lift a finger unless they can guarantee themselves a profit regardless of what else happens.[/li]
I’m sure that if you try, you can dig up some misleading or incorrect news stories that somehow imply otherwise. There is no more reason to believe them than claims that Bat Boy voted for Obama.
It should be noted that the FH that launched was vastly more capable than what was bandied about initially. In fact, the current single-stick F9 is almost as capable as the initial FH–22 tons to LEO instead of 25 tons.
So while there have certainly been real delays in the program, counting the full delay time against SpaceX is absurd. Much of it was just a function of them not needing it due to F9 upgrades. It would have been dumb for them to launch it before the F9 upgrade program was nearly complete (which it is finally, with Block 5 on the horizon).
The Onion really made a major mistake with that report. It is illustrated with a photo of one of the smaller rockets instead of the Falcon Heavy! I sure hope someone got fired for that…
Because I’m tired of beating my head against a wall, trying to understand why you think a public/private venture on a capsule would cost 20% more than a purely private investment on a rocket; and that it is just pure coincidence that your oft-cited cost for he private development of the rocket just happens to match the total value of the NASA contract; not to mention that the NASA contract allowed for the development of the rocket and capsule.
I mean, there’s a very simple explaination for all this. But each time I’ve talked about it, I’ve gotten more and more convoluted reasons why Elon is amazing and did most of it himself with his own two hands.
No, the capsule cost 119% more ((396+850-390)/390 = 2.19). And the reason is that it’s just a very difficult problem. SpaceX had to develop a new set of engines (the Draco thrusters) which–while simpler than the Merlin engine–also uses a toxic propellant and required all new systems and ground support for that.
The capsule, while not man-rated for launch, is man-rated with respect to the ISS. When berthed, it becomes a part of the station, being a temporary room addition that the astronauts can enter and leave at will. It has basic environmental controls for temperature and atmosphere.
Dragon returns to Earth on a parachute and is recovered, requiring development of heatshield material (PICA-X, which was derived from NASA’s PICA material, but still needed separate qualification). Not to mention a new set of guidance systems and ocean recovery support.
Speaking of the guidance hardware, that was also all new and required additional robustness. SpaceX stuck with off-the-shelf components but with massive redundancy to achieve the necessary reliability in space.
A rocket’s guidance systems can afford some inaccuracy; no one cares if it’s off by a few hundred meters or a few meters per second. A capsule that berths with the ISS doesn’t have this luxury–it needs centimeter-level accuracy. And of course zero chance at anything going wrong in a way that would risk the lives of astronauts.
I could go on–but really it should be no shock that Dragon cost far more than a basic two-stage kerosene rocket.
Your implication that the $390M F1+F9 cost has anything to do with the $396M number for COTS is really just tinfoil hat level. It doesn’t make sense even in conspiratorial terms.
Here’s a trivia question: who were the presenters for the live stream? I assume they were SpaceX folks; they didn’t seem to be regular on-air talent. Were they all a bunch of moonlighting rocket scientists? That’d totally fucking rock!
Yep, they’re all technical employees of SpaceX. For example, Lauren Lyons is a SpaceX mission integration engineer. I appreciate their somewhat awkward but genuine demeanor. There was a moment during the the Heavy launch where Lyons can barely contain herself (just after the booster landing, IIRC). Their enthusiasm is infectious.
They’re fairly regular. The older guy, John Innsberger (sp?) is almost always there. Of the rest, there are just a handful in rotation. No more than a half-dozen total as far as I’ve seen.