This isn’t new, but I don’t think I linked it at the time. Eric Berger reports on the people working SLS:
May turns the cost issue around.
“My question would be, how could we afford not to do this?” May asked. “Great nations explore. Great nations push their boundaries. And this country has continued to the limits of what we know and learn for a generation, and I think we’ve got to continue to explore.”
And in the larger perspective, he argues, SLS does not cost that much. NASA spends about $1.6 billion a year building it, less than 9 percent of the space agency’s total budget, he said, which is itself less than one half of one percent of the federal budget.
“I think it’s a relatively small amount of money to set the leadership for the world in space exploration,” he says.
Count the number of logical fallacies in just those four grafs.
“Big is not bad,” May says. “If you really want to colonize somewhere you’re going to need to send a lot of stuff over there. We humans are very needy. Just think about the water, or just the t-shirts. A lot of little things is not the way to do it.”
The need to launch lots of water is just about the worst argument for a big booster.
This isn’t the 15th century. We have already discovered the Moon, and Mars, and Ceres, and Jupiter, and on and on. Not only that, we’ve mapped them, studied them, dissected their innards, gauged them, roved them. We’ve cataloged their histories, we’ve viewed their sunsets and their volcanic eruptions.
Which is to say, the era of going somewhere in person just to see that it exists, just to map the land and catalog the fauna is over. Long over. We do all that robotically now. We’ve DONE all that robotically ALREADY.
So saying that exploring the solar system at great cost is still worthwhile is rather silly. What we need is colonization. Even within the context of human facilitated exploration, that fundamentally requires a colonization infrastructure. Exploration superior to what a robot could do requires the presence of humans for long periods of time (months if not years), and that requires de facto colonies. And that’s aside from the necessity of off Earth colonization, which exists on a priority scale beyond most other human endeavors ever undertaken in history.
But we’re not going to colonize anywhere if it costs a billion dollars a ride as it did with Apollo and as it will for the foreseeable future with SLS. It should go without saying that such exploration is not sustainable.
Spending lots of money on exploration and colonization is worthwhile, even in the multi-billion dollar a year range. But spending so much money in exchange for very little results is always a mistake.
I hope that Elon’s big fucking rocket has greater payload and is launched before the SLS just to show how ridiculously bullshit the whole scam is.
Robin,
Personally, I hope that long before Elon has wasted much real money on a BFR he decides that reusability, depots, and good high-performance upper stages make more sense than super massive rockets built for a mission where the planets are only properly aligned once every two years.
~Jon
Very few details of the rockets or the MCT architecture has been released as of yet, so it’s perhaps a bit too early to make judgments about it. My personal hunch, based on some fair amount of evidence, is that it’s a scaled up version of a common core design similar to the Falcon Heavy. The upper stage of which would be integral with a large, manned spacecraft capable of interplanetary travel, landing on Mars, and returning to Earth, and also capable of transporting roughly 100 tonnes of cargo, supplies, and crew towards Mars. In principle every part of such a design could be reusable. And my understanding is that each trip would be an opposition class mission (visiting Mars and returning within one synodic period). Perhaps making use of some sort of fixed infrastructure on Mars for manufacturing LOX/Methane (which would be sensible).
Such a system could be used similar to the way crew rotations on the ISS work. One crew heads out and lands on Mars; the MCT is refueled, serviced, and inspected; during the appropriate launch window the MCT returns the previous crew to Earth; and so on.
Meanwhile, you could build proper upper stages for the “BFR” and operate it commercially during “off-peak” times, amortizing the fixed costs of the thing.
Overall it could realistically be a very inexpensive and capable way of colonizing Mars. Is it superior to other architectures? I’m not sure. Personally I like the idea of orbital propellant depots and orbital assembly, but I can envision a reality where enormous *reusable* rockets are potentially cost effective.
In any event, at least SpaceX isn’t dumb enough to put all their eggs in one basket. They’ll continue to operate F9-R commercial regardless of their Mars colonization plans, and if the BFR/MCT design turns out to be dumb they’ll have a mature operational fallback.
SpaceX’s “BFR”, also known as MCT, stands a good chance of launching before SLS – at least, SLS block 2, the 120 ton to LEO version. MCT is supposedly 10 years away, so let’s say 15. SLS block 2 initial capability is supposedly in 2032, so even if we accept that date, it’s 18 years off.
MCT, as currently described, is beyond big. Very, very big. It uses the same general design as Falcon; 9 main engines, with a 3 core heavy version (27 main engines). The single core version, with 9 engines (Raptor, at least 1 million pounds thrust) exceeds the power and capacity of Saturn 5 or SLS in takeoff weight and throw weight to LEO. Each of the cores is 10 meter. (for comparison, SLS core diameter is 27 feet, Saturn 5 is 33 feet). However, the saturn 5 was tapered, MCT does not look like it will be. Each core will have 9 engines (raptor) of at least 1 million pounds, so it’ll have more thrust, and thus higher launch mass, than Saturn 5. That’s for single core. 3 core would have in excess of 27 million pounds of thrust (Saturn 5 was 7.6 million).
Payload to LEO? I have utterly no idea, but as a truly wild guess, I’ll go with the payload ratio of the Saturn 5. 118 tons to LEO, 6.6 million pounds on the pad) that’d be a ratio of 25 to 1. (25 pounds on the pad = 1 pound in LEO)
If we assume the low end estimate for a single core MCT of 9 million pounds (pound-force) thrust, and use the same thrust to weight of 7.6 lbf thrust, 6.6 million pounds weight) Saturn 5, we get an on-the-pad weight of 7.83 million pounds for MCT. If we use the 25-1 pad weight to LEO ratio of Saturn 5, that’d give us a very wild guess of single core MCT payload to LEO capacity; 313200 pounds, or 142 metric tons. (SLS block 2 is supposed to be 130 metric tons to LEO).
However, if we’re looking at the 3 core version., and use the Falcon Heavy vs. Falcon 9 payload differential (13.5 tons F9, 56 tons FH, a ration of 4.14) that’d ballpark the MCT 3 core at 1.3 million pounds (587 metric tons) to LEO. No wonder it’s nicknamed BFR.
My personal guess; MCT (like heavy lift in general) only makes sense if it achieves low cost per pound to orbit. *if* they can do that, then it might make sense. Now, the only way I see it being possible to achieve low cost per pound with an enormous rocket is both reusability and a decent launch rate to amortize the fixed and development costs. I think they’d need commercial customers for earth-orbit launches to do this. Right now, there is a market for a number of 300,000 lb payloads to orbit. The only little drawback is that number is currently zero. That could well change if FH delivers on its promised low cost per pound (private space stations, fuel depots, etc) but it’s not there now. In 15 years, maybe. But lacking that, or some way of making frequent Mars launches pay, I don’t see MCT being economically viable, for many of the same reasons SLS isn’t.
Personally, if, and only if, MCT can deliver low price per pound, I hope to see it fly. It would indeed be an awesome sight.
First, as I understand it, “Mars Colonial Transport” is an umbrella term that applies to SpaceX’s deep-sea vehicle as well as the Earth-to-orbit transport.
Second, while fanboys believe Elon is infallible, Elon himself does not believe that. If you look at the history of SpaceX, you will see that his plans have evolved over time. His current plan for recovering and reusing Falcon 9, for example, bears very little resemblance to his original plan (just parachute into the ocean, rinse off the saltwater, and fly again). It would not surprise me at all if the design of Falcon 9R changed again, perhaps even before it enters operational service. Elon knows that — that’s why he’s doing testing.
Similarly, I would not assume that the Mars Colonial Transport will look anything like the very preliminary conceptual designs Elon talks about today. Take all of those numbers with a very large grain of salt.
Totally agreed on all points. That’s why I used a lot of fudge factors, such as “as currently described”.
Personally, I like that they’ve been very willing to change tack when needed (such as abandoning parachute recovery of the 1st stage when that didn’t pan out) as a result of tests – they learn from tests and adapt, which IMHO is the whole point of testing . This also means that anything they say now is indeed to be taken with a large dose of salt.
I’ve been curious: Why do you often say “graph” instead of paragraph? Do the words mean the same thing to you? (I did google first, but the results were ambiguous.)
I should have spelled it “graf.”
Oh, now I see. I thought maybe “graf” would refer to a short paragraph, but apparently not.
Journalism slang includes the term “nut graph” (with a ph at the end of graph) which refers to the key initial summarizing paragraph, and “graf”, (no “ph” on the end) which is short for a regular paragraph.
That’s all for today, but on the syllabus tomorrow, I see we’re going to learn about lede vs lead.
9 billion in expected development costs? Last I heard (it was given during Senate hearings a while back) was 18 billion through 2017 (this includes 6.1 billion for Orion). Then how much per year to run the program when it’s “operational”? That’s one rocket to build every 2 years, plus all ops and staffing costs, etc. (and I’ll bet the latter costs even more than the former).
A NASA internal estimate was for 41 billion in total through 2025, which would include 4 launches. (68 ton launches, so roughly Falcon Heavy class). After that, HEFT claims 1.86 billion “unit cost” inclusive of ops (2010 dollars).
So, let’s crunch some numbers. 41 billion gets us to 2025. (and 4 launches). After that, we’re looking at unit cost plus operational cost. Shuttle operational cost was around 3 billion a year, so let’s give SLS a wild benefit of the doubt and, due to none of it (Orion or SLS or the SRB’s) being reusable, and say 2 billion a year.
Paul Spudis came up with the following, based on assuming a 30 year (like Shuttle) life of the program;
Orion capsule with service module and escape system $1 billion
SLS first stage, second stage and upper stage $1 billion
Annual operating and launch facility maintenance costs $2 billion
1/30 share of development cost $1 billion
TOTAL $5 billion cost per launch
I think he’s off by a good bit, due to the numbers changing since he wrote that. The dev cost is now 41 billion, not 30. Also, it’s one launch every other year, not one per year. So, plugging the new numbers into his format, we get,
Orion capsule with service module and escape system $1 billion
SLS first stage, second stage and upper stage $1 billion
Annual operating and launch facility maintenance costs, $2 billion, so 4 billion per launch
1/15th share of development cost (15 launches over 30 years) $2.73 billion
TOTAL $8.73 billion per launch
This is FAR higher than shuttle (1.4 billion (today’s dollars) in total per launch over the life of the program) and Shuttle was far too expensive. Also, Shuttle was vastly more useful (for just one thing, it flew several times per year, not once every other year.) Shuttle had 135 launches in 30 years. SLS would have 15.
Compare also to Apollo, which was insanely expensive (total program cost 109 billion adjusted for inflation to 2010 dollars). The SLS program, to do an equivalent number of launches (counting only Saturn 5 here (I’m ignoring S1B, Skylab, Apollo Soyuz, and some manned Apollo shots, all of which were included in Apollo costs) so 13 launches total.) runs 8.38 billion per Saturn 5 launch. So SLS is more, but that’s only on the face of it. With Apollo, you had the capability to land on the moon. SLS will have no such capacity, yet it costs more..
Price per pound of an SLS launch? $32998 per pound (assuming block 2 capacity is as claimed.)
The SLS manages the almost impossible; it combines the worst aspects of Shuttle and Apollo, while having far fewer capabilities and uses.
“The SLS manages the almost impossible; it combines the worst aspects of Shuttle and Apollo, while having far fewer capabilities and uses.”
That depends on what your definition of “worst” is. For an engineer, you’re right. From the perspective of a congressman, the calculus changes from “what gets the best value for taxpayer dollars?” to “what gets me the most votes?” And the more senior a congressman is on the science committee, the more it is in his favor to spread the pork to his district rather than getting value for the dollars spent. Case in point, Shelby. Counterexample: Rohrabacher.
Ed, great point – they do see it as a pork launcher, and from that perspective, it works well; massive cost spread amongst many troughs.
I have a medium-strong suspicion that NASA’s internal SLS numbers are not vastly different from yours. And that’s why KDP-C, which would include a program budget that would serve as a baseline for future work, has yet to see the light of day though it’s been eleven months since PDR completion.
I would love to see what NASA thinks the real numbers are (the fact that it’s secret tells me much).
In my wild ballparking, I might have actually understated the issue of support costs; the massive standing army of launch personnel and facilities. I assumed, because it’s not a resuable vehicle and would have a far lower launch rate (by about a factor of 10) that they’d have less to do than with Shuttle and thus be slightly less costly in that area. I’m thinking more and more that I was wrong on that; one of the political motives for this program was to “preserve” the shuttle workforce, so I am growing increasingly confident that any potential savings will be diligently hunted down and eradicated.
If we make this assumption (that SLS support costs will be on par with Shuttle), my estimated cost goes up by one billion a year, so 2 billion per launch, for a grand total of $10.73 billion. And cost per pound? As Dick Eagleson pointed out, SLS block 2 is probably dead, so 93 tons max (205029 pounds) for $10.73 billion gives us a staggering $52,334 per pound to LEO.
> I would love to see what NASA thinks the real numbers are (the fact that it’s secret tells me much).
Oh, yes, the longer NASA doesn’t say what the projected budget is, not even give caveated high-medium-low estimates, the more they say. I was willing to cut the them some slack late last year because of sequestration and shutdown, but it’s becoming time to conclude that Houston, there’s a problem there.
(BTW and OT, what part of Arizona? I’m from Cochise County: http://www.slaughterranch.com/wp-content/uploads/2012/11/sheriff-slaughter.jpg )
Houston is the problem. 🙂
I wonder if they are on track to do an Apollo repeat of cost ballooning… (the estimated Apollo program cost jumping from 7 billion to 20, nearly triple.) Is this what they really mean by “Apollo on Steroids”? 🙂
I’m in northern AZ, about an hour outside of Prescott, the closest city.
AFAIK the $1.6 billion a year number is bullshit. The fact is they have not started bending metal. Heck they have not even decided which engines they will use on SLS yet! $1.6 billion is like the current funding level for doing paper and more paper. Even if SLS cost $1.6 billion a year, which it does not, Elon spent less than that designing Falcon 1, Falcon 9, and the Dragon capsule. Total multiple year investment. Kind of puts those numbers into perspective.
Like some people here I suspect Block II will never fly and the whole SLS project will end as a Potemkin rocket just like Ares-IX.
Fact is SLS is anti-economic. This guy says the economy does not matter but he is dead wrong. Sure exploration is an essential drive but colonization should be the ultimate goal. Any uneconomic colony is not going to survive in the long term.
IF they wanted to do a big rocket it should be done using a modular architecture. Otherwise making a case for spending a lot on R&D for something that is launched every other year is just not going to convince anyone. The fact is the current EELV launchers have a lot of margin to do something like this and increase the launch payload mass. Boeing and Lockheed Martin actually presented possible roadmaps to do it more than once. I can understand that some people want to preserve the capability to build something like the SSME or large solids but SLS is not the way to do it.
SLS Block 2 was predicated on two things:
1) Advanced boosters to replace the 5-segment SRB’s.
2) An upper stage based on J-2X engines.
Both have been cancelled. Block 2 is dead. Block 1 is all there is ever going to be. 93 tonnes to LEO – max.
Arizona CJ’s SLS launch cost numbers, unfortunately, are unaffected by the demise of SLS Block 2. So $8.5 billion to launch 93 tonnes vs. $85 million to launch 53 tonnes on Falcon Heavy. That’s a 57:1 difference in cost per pound to LEO. And that’s without considering continuing improvements to the Merlin 1-D engine and adding at least a twin-engined upper stage for Falcon Heavy. By the time SLS is scheduled to fly, FH may well have made up at least half of its current 40 tonne mass-to-LEO deficit relative to SLS Block 1.
Dick, great points! And you’re right, I paid no notice to the demise of SLS block 2 (I wasn’t aware of it before your post, but you’re right!)
BTW, as I recall, the advanced liquid booster design they were leaning to used Russian engines (NK-33), which might be part of the reason they aren’t going through with it.
The “official” line is the J-2X development was “frozen” because the rest of SLS is delayed horribly. But yeah in practice you are correct. Their plan is to save time using a modified Delta IV second stage instead of the originally planned J-2X stage. Which is complete nonsense but whatever. The whole rocket is nonsense. Are they still planning to use the SSMEs as disposable first stage rocket engines? Von Braun is probably rolling in his grave now.
@ Godzilla;
Regarding the SSMEs, the answer is “yes” but in reality its far, far worse than that.
They have 15 existing SSME engines, and think they can build a 16th from spare parts and a few cobbled together bits, maybe. This assumes no damage or test stand failures, and also assumes some handmade bits for the 16th SSME, so I’m going to say they have a total of 15. (they can’t just make more SSMEs, that production line was closed down long, long ago.)
SLS block one uses 4 SSMEs per launch, so they have enough for the first three and three quarter launches with 15, 4 if they get the 16th SSME going and have zero critical failures on any engine (for example, the turbines being bad on one would use the spare set needed for engine 16).
After those engines are gone, they’ll switch to new version of the SSME (the SSME was the RS-25D, and they’re calling the new one the RS-25E). Does it exist yet? Nope. Is the design finalized? Nope. So, how do they know the exact performance specs sans even a finalized design? They don’t. They just used the SSME specs and gave it 111% of the performance of the SSME. What were the grounds for assuming the better performance? Because those were the numbers they needed to make the payload capacity pencil out. They say they can make the new engine simpler and cheaper and better performing than the SSME because it’s not reusable, but I think they are blowing smoke.
I’ve also yet to see hard numbers on how much, and how long, it’ll take to create these new engines. The JX2 was far closer to completion and they’ve effectively canceled it, so how they’ll do with the RE-25E is anyone’s guess.
They’ve basically designed a rocket around an engine that doesn’t exist yet (that 11% performance gain they need really means it’s a whole new engine).
It’s not totally clear to me that RS-25E will materialize. From a GAO report of only two months ago:
My guess is that “a new engine” is RS-25E. As an aside, it’s not clear to me that 3 years vs 8 years should be the controlling consideration, as NASA has not presented a flight schedule that would need new-production engines before 2023. Probably the costs of the two options is more important.
“So $8.5 billion to launch 93 tonnes vs. $85 million to launch 53 tonnes on Falcon Heavy.”
85 million is for the Falcon Heavy when it is not utilizing a cross feed system, where the strap ons pump fuel into the central core. It will only lift about 43 tons configured that way. it will be 125 million for the cross fed version that will do the 53 tons.
The quote by Bolden about Falcon Heavy and SLS was jaw-dropping.
“The quote by Bolden about Falcon Heavy and SLS was jaw-dropping.”
No, it wasn’t. The deal that was hammered out was Commercial Crew, SLS, Orion & JWST. All together – or none at all.
Bolden is simply supporting the deal vocally. If he started saying “Falcon Heavy is also an option”, Shelby & Co. would be holding hearings and trying to cancel CC in a New York minute….
Progress on Falcon Heavy and MCT is completely independent of what anyone from NASA says. At this point….