At the AIAA meeting this week in Pasadena, NASA officials admitted that the Space Launch System (SLS) will likely cost half a billion dollars per launch


Please consider donating to Behind the Black, by giving either a one-time contribution or a regular subscription, as outlined in the tip jar to the right. Your support will allow me to continue covering science and culture as I have for the past twenty years, independent and free from any outside influence.

It’s only money! At the AIAA meeting this week in Pasadena, NASA officials admitted that the Space Launch System (SLS) will likely cost half a billion dollars per launch.

That means that after only two flights this rocket will have cost about the same as the entire manned commercial program, from which three different space companies are building three different methods for getting humans into space. After three missions it will cost more, and after four missions it will have cost double. And this is assuming that the half billion dollar “target” number ends up correct.

We can’t afford this. We never could, which is why the Saturn 5 rocket was abandoned, and why the shuttle never fulfilled its stated goal of lowering the cost of access to space and after thirty years was abandoned as well. Instead, we have got to find a cheaper way to do this, and to my mind, competition and private enterprise is the only hope.

Share

45 comments

  • I bet that $500M number is calculated like the fictional marginal cost numbers NASA
    used to give for the Shuttle. If they had, say, four Shuttle flights in a year and
    the Shuttle program cost $4B that year, they would not give the average $1B
    cost but would do a hand-waving calculation of how much it would cost to do
    an extra flight and use that pretend marginal cost number when asked for the
    cost of Shuttle flight. I remember numbers as low as $150M.

    If one is making 100k widgets per year, a marginal cost number is of some
    interest. It is of no interest for low single digit production. It’s the average
    cost that is meaningful.

    Gerstenmaier says the SLS program will support 2-3 flights per year. Ignoring
    the question of where NASA is going to get the money to create 2-3 payloads of
    70-130mT each per year, that means the average cost is going to be at least
    $1.0B-$1.5B per flight if you assume the SLS program will cost $3B per year,
    which is a very conservative guess.

  • Steve C

    This thing is not even intended to fly, it is just make work for paper pushers. They could not even put the effort into coming up with a modern looking design and instead did a 5 minute photoshop of strapping Shuttle SRBs on a Saturn 5. On earlier projects like the National Space Plane, they at least came up with a nice drawing for the millions dumped into the study. Now, they’re just mailing it in.

  • Patrick Ritchie

    $500 million is an interesting number. If we make a leap of faith and assume this is the actual cost per launch what does $500 million a launch represent?

    If it’s for the 70 ton, that’s $500,000,000 / 70,000 kg = $7142 per kg.

    If it’s for the 130 ton, that’s $500,000,000 / 70,000 kg = $3846 per kg.

    $7,142 is less per kilo than all US launch vehicles with the exception of Falcon 9.

    $3,846 is less per kilo than all active launch vehicles, the Falcon heavy and Falcon 9 1.1 are both projected to cost less but have yet to fly.

    Of course Clark is correct and these numbers are most likely the incremental cost of a launch and don’t include the fixed costs. And even then, both the upgraded Falcon 9 and the Falcon Heavy are due to fly long before SLS, making it obsolete before it even gets off the ground.

  • Joe

    Actually Space X launch cost can be better determined by their own press release on the signing of the ISS Cargo Contract:

    http://www.spacex.com/press.php?page=20081223

    They are contractually committed to supply 20 Metric Tons in twelve flights for a total of $1.6 Billion. That comes to 3,667 lbs. /flight, at $133 M/flight that is $36,269/lb. ($79,791/kg.)

    This is what they are willing to actually ‘sign on the dotted line’ for. Makes $7,142 p/kg. look pretty good.

    The overall cost of Space X services is lower because it delivers so much less (and does so – it is becoming clearer as actual contracts are signed – inefficiently). If you are going to have only a de-scoped/unambitious space program then I guess Space X is all you need.

  • Patrick Ritchie

    That’s pressurized cargo vs total payload to LEO. Completely different.

  • Joe

    No actually by the contact that is pressurized and unpressurised cargo (the Dragon vehicle supposedly delivers both). In any case that would not explain a factor of eleven in cost. Not a chance.

  • libs0n

    Joe,

    You can buy a Falcon 9 flight with a Dragon for your specialized space station cargo, and that is the price for that cargo, but you can also just purchase the Falcon 9 itself, without the expense of a Dragon, and get its full payload capacity for just that price, for your non space station cargo needs.

    You are acting as if you can only buy space on the Dragon, and not just the launch vehicle. The ISS contract is for space on the Dragon. Every other usage is for just the launch vehicle.

    SpaceX just got three new launch contracts. Do you think they also purchased the Dragon?

    ” In any case that would not explain a factor of eleven in cost.”

    Yes it does. The Dragon increases the cost per flight, while decreasing cargo capacity to what it can contain within. It’s not that this is hard to understand, Joe, it is that you actively refuse to try to understand things from a non-malicious perspective.

    Let me turn this around to show you what you are doing. Should we judge SLS+MPCV prices for the cost per pound of humans weight delivered to orbit? If the average man weighs 200lbs and there 4 of them, and we will use the price from this thread of 500 million, which doesn’t even include fixed costs or MPCV costs or MPCV fixed costs, then the SLS cost per pound is a whopping $625,000 dollars a pound! Your Dragon cargo costs are quite the bargain.

  • Again Joe, you are matching real contract numbers with fantasy SLS numbers.

    Bigger rockets are suppose to be cheaper, but not when the government does them. SpaceX on the other hand will provide a cheaper big rocket based on components that already have a flight heritage.

    So you imagine the SLS will replace 52 F9/Dragon flights? Where are they going to put the stuff?

    Do I see a lightbulb yet? No? How much will it cost per kg. if the SLS takes less that 70mt to the I.S.S. at a time? Things need to be the right size to make sense. NASA’s own reports say with fuel depots they don’t need SLS… for any mission.

    The F9 is real and the government can work whatever deal they can. Competition brings the price down. There is no competition yet; but when there is the SLS isn’t going to be it. Boeing and private companies will be. That’s when you’ll see prices come down. SpaceX would be idiotic to lower their price until that time.

  • wodun

    I would rather see NASA spend $2b a year on payloads and $1b a year in F9 launches than see them spend $3b a year and have no launches. Waiting X years for SLS doesn’t make sense. It would be even better if NASA didn’t build the payloads but said, this is what we need and how much we will spend, who can make it for us?

    I am sypathetic to those that want a SHLV but I’m not convinced we have a need for one right now. Sometime in the future? Maybe.

  • Joe

    ken anthony says: Posted September 13, 2012 at 1:35 PM

    “Again Joe, you are matching real contract numbers with fantasy SLS numbers.”

    Forgive me, but in the last several months I have had people on the internet telling me that the Falcon 9/Dragon would (specifically) deliver 13,000 kg (kg not lbs) to the ISS for $27 Million a launch (and then specifying that was $950/lb). The fact that within weeks the actual numbers turned out to be 37 times higher has made me somewhat cynical about Space X supporters throwing around insults about other peoples “fantasy numbers”.

    I will give you some “fantasy numbers” with which I am familiar. A study was done based on the Side Mount configuration SDHLV which happens to have the same basic 70 Metric Ton payload capacity as the smaller SLS.

    The study involved cargo supply to the ISS. It used a delivery system of a “strong back” and payload faring to deliver three MPLMs at a time to the ISS. This is not the most operationally efficient way to do this, but it used virtually all already existing hardware it was considered the least risk, fastest, cheapest way to get the job done. The MPLMs could carry 13 Metric Tons, but I know certain people claim it could only carry 9 Metric Tons so I will use the lower figure.

    For a 70 Metric Ton gross payload 27,000 kg could have been delivered to the ISS. If that $500M/flight figure could be achieved that would be $18,518/kg (23% of the Space X number).

    Even if the unsubstantiated assertion above that the SLS launch cost would be $1.5 Billion that would still make the SLS 30% cheaper than Space X.

    Now make all the snarky comments you want about “fantasy numbers”, but (if you want to have any credibility) it would be a good idea to look into that 3700% Space X cost increase first.

  • Patrick Ritchie

    Agreed, 11x sounds off.

    Took a deeper look at the contract. Emphasis mine.

    At the time of award, NASA has ordered eight flights valued at about $1.9 billion from Orbital and 12 flights valued at about $1.6 billion from SpaceX.

    These fixed-price indefinite delivery, indefinite quantity contracts will begin Jan. 1, 2009, and are effective through Dec. 31, 2016. The contracts each call for the delivery of a minimum of 20 metric tons of upmass cargo to the space station. The contracts also call for delivery of non-standard services in support of the cargo resupply, including analysis and special tasks as the government determines are necessary.

    http://www.nasa.gov/offices/c3po/home/CRS-Announcement-Dec-08.html

    So the 20 metric tons is a minimum and the contract also includes “non-standard services”. If 20 metric tons is the minimum, it begs the question: what is the maximum?

    http://www.nasa.gov/pdf/636362main_FS-2012-04-014B-JSC%20COTS1-pager.pdf
    http://www.spacex.com/downloads/dragonlab-datasheet.pdf

    The above disagree, NASA says 6,620kg and SpaceX says 6000kg. Regardless the 1,666 kg per flight is less than the published amounts by up to 4x. If we use the 6000kg number we get: $133,000,000 / 6000kg = $22,166. Still much higher than the theoretical max of the Falcon 9.

    But if we’re comparing the theoretical max we also need to include the weight of the empty Dragon: 4200kg. $133,000,000 / (6000kg + 4200kg) = $13,039/kg. That’s about 2.5x the theoretical max. This seems reasonable given SpaceX will be building a brand new Dragon for every flight as well as returning cargo + any ‘non-standard services’ that NASA requires.

    *Trying to determine what the actual cargo per flight is going to be is difficult. The COTS 2+ hauled up 620kg, but we know they had to be light due to all the maneuvers they had to perform, so that’s no help. The first CRS mission is probably going to be our best indication, but I can’t find any details on the cargo manifest.

  • Patrick Ritchie

    See above for my thoughts on the SpaceX numbers. Given the operational advantages + return capability they compare favorably to the MPLM scenario you outline. Although given it never flew it’s hard to know what the actual costs would be…

    Another MPLM scenario is the Shuttle itself. I’m going to cheat in it’s favor a bit and use 2001 to determine the cost of a Shuttle flight.

    2001 Shuttle budget: $3,118.8 million
    2001 Shuttle flights: 6 (4 of which flew an MPLM)

    Cost per flight: $519.8 ($672.8 in 2012 dollars)

    The max of ever flown on the MPLM is 12,748kg (STS-126), the actual amounts flown in 2001 are listed below:

    STS 102 – 10,213kg
    STS 100 – 8,811kg
    STS 105 – 9,467kg
    STS 108 – 9,228kg

    Average: 9,429kg

    $672,800,000 / 9,429kg = 71,354 $/kg of cargo

    Ouch. Let’s use the max from STS-126.

    $672,800,000 / 12,748kg = 52,776 $/kg of cargo

    Now this really isn’t a fair comparison as the Shuttle also carried up other payloads as well as up to 7 astronauts. We can use the Soyuz cost of ferrying an astronaut to ISS to give the Shuttle a rebate for that portion of the flight.

    7 astronauts * $60,000,000 = $420,000,000

    ($672,800,000 – $420,000,000) / 12,748kg = $19,830/kg of cargo

    Much better. The only ‘problem’ is you have to send up 7 astronauts every time AND make 6 shuttle flights a year locking up ~$4 billion a year of the NASA budget. If you think going to the ISS 4 times a year + 2 other missions to LEO is the best use of $4b then the Shuttle was actually a pretty good ‘deal’.

  • Joe

    I actually basically agree with the $19,830/kg of cargo figure (in fact derived a very similar number myself by somewhat different analysis – probably a sign it is a good estimate). It is also 25% of the Space X cost.

    However, it would not require four ISS flights/year. Space X contract is for 20 metric tons in three years and Orbital Sciences has another 20 metric tons in three years contract, note it is for $1.9 Billion making its cost $95,000/kg (making the average cost for the two contracts $ 87,500/kg). But that also means that NASA is saying its cargo up mass to the ISS in that period is only an average of 13.3 Metric Tons a year. That could have been handled by two ISS utilization flights per year, which would have synchronized well with ISS crew exchanges. The delivery of astronauts is a feature not a bug, it not only allows crew exchanges for the same price, but would supply up to three EVA’s to support ISS maintenance.

    This would have to be done of course (as would any such use of an HLV of any design) as part of a bigger overall program. That is the reason I said in my first post on this subject: “The overall cost of Space X services is lower because it delivers so much less (and does so – it is becoming clearer as actual contracts are signed – inefficiently). If you are going to have only a de-scoped/unambitious space program then I guess Space X is all you need.”

    Having thought about it more if your program is to be as de-scoped/unambitious as ours has become you do not need Space X/Orbital Sciences either. Probably makes more economic sense just to buy services from the Russians.

  • Patrick Ritchie

    I suspect (but cannot confirm) your number is still too high for SpaceX, not sure about Orbital, hopefully we’ll know sometime next year.

    Given how much both NASA and SpaceX say the dragon can carry, the real number is likely to be closer to $25-35k / kg. Of course, we won’t know for sure until CRS-1.

    I honestly don’t know enough about ISS operations to know what the optimal resupply scenario is. Is it multiple small flights? Infrequent larger ones?

    I doubt using Progress makes more economic sense, given the experience of giving the Russians a monopoly on crew I suspect they would bleed you dry on cargo if they were the sole supplier. Assuming they charge the same per flight as a manned Soyuz as they do for a Progress-M and they max out every flight on cargo at 2,600kg I get $69k/kg and zero return capability.

  • Joe

    Not trying to be argumentative but a link to Space X statement is in my first post on this subject:

    20 Metric Tons for $1.6 Billion ($80,000/kg).

    A link to a Washington Post article on the Orbital Sciences contract is below.

    http://www.washingtonpost.com/business/capitalbusiness/orbital-readying-for-resupply-mission-to-space/2012/08/24/5cb35bf4-eb12-11e1-b811-09036bcb182b_story.html

    20 Metric Tons for $1.9 Billion ($95,000/kg).

    That makes the average for the two contacts $87,500/kg.

  • Coastal Ron

    The Shuttle was never capable of keeping humans in space any longer than two weeks, so it’s ability to sustain the ISS from a crew standpoint was pretty much zero. Only Soyuz spacecraft can act as long-term lifeboats, and they have a six month limit on how long they can stay in space.

    What that means is that regardless how many people came up on a Shuttle, you still had to pay the same amount to have Soyuz spacecraft at the ISS. The only value the Shuttle provided was a short bump in the number of people at the ISS, but it did not affect the cost of supporting crew at the ISS.

    As to mass to orbit comparisons, I know that Joe likes to compare pressurized SpaceX CRS prices with unpressurized SLS cargo prices – truly apples to oranges comparisons. Of course he never wants to compare Falcon Heavy prices, even though it is far closer to reality than the SLS, and has already been purchased by a customer.

    Falcon Heavy would deliver mass to LEO for $2,415/kg, which is 63% the cost Patrick Ritchie estimated for the 130mt SLS. What the price difference doesn’t include is the $0 the U.S. Taxpayer is paying for Falcon Heavy development, and the $30B the U.S. Taxpayer is paying for SLS development. Where is the incentive for the U.S. Taxpayer to pay $30B and still not be as inexpensive as Falcon Heavy?

    Oh sure, someone will say “Falcon Heavy can’t carry Fictional Payload X”, but isn’t that the real problem here – there are no known payloads for even one SLS flight, much less 2-3 flights per year for decades into the future.

    As to the $500M/flight estimate, as Clark points out, they really don’t know. It’s a figure provided to politicians for political reasons, and until a GAO audit validates it, it’s fiction.

  • Patrick Ritchie

    That’s an interesting point on needing the ISS lifeboat to be a Soyuz. and really hurst the shuttle in terms of cost. Do you know what the plan is for the lifeboats, when / if Comercial Crew gets going? Is it safe to assume that the whatever spacecraft is used it would also be the lifeboat?

  • Joe

    “As to mass to orbit comparisons, I know that Joe likes to compare pressurized SpaceX CRS prices with unpressurized SLS cargo prices – truly apples to oranges comparisons.”

    That subject was addressed in the back and forth with Patrick Ritchie above, might want to read it since there is a lot of good information there (if that is actually what you are interested in).

    “Of course he never wants to compare Falcon Heavy prices, even though it is far closer to reality than the SLS, and has already been purchased by a customer.”

    That is because the Falcon Heavy does not (as yet anyway) exist. I am sure you will be willing to tell everyone that it will have incredible prices, just as before the contract announcements it was said that: the Falcon 9/Dragon would (specifically) deliver 13,000 kg (kg not lbs) to the ISS for $27 Million a launch (and then specifying that was $950/lb). Then, when the contracts were released, the actual cost turned out to be 37 times greater.

  • Joe

    Patrick,

    There was a time when the US was developing a Crew Return Vehicle (CRV) it was defunded due to (what else) budget cuts. It all gets back to how small you want your space program to be. Sadly the country appears to have decided (or at least been maneuvered into accepting) that it is to be very small.

    Also note that some of the folks now leaning on the lifeboat argument have previously argued that the ISS did not need a lifeboat capability.

  • Vladislaw

    20 tons is the minimum amount. The dragon can move 13,000 lbs of combined cargo per launch. It also has a DOWN CARGO capability… how much is THAT worth a pound?

    They have the capability to move more cargo than the minimum. They have the ability to move 36 tons of potential down cargo…

    NASA was the one that insisted that SpaceX use a new capsule for every flight… that added how many millions to each flight.. 50 – 60 million?

    You are really using the straw men today.

  • Vladislaw

    where would they store that much cargo on the ISS? lol .. that is laughable.

    Where is the money for this imaginary 70 ton cargo carrier for the SLS that would be delivering all this cargo to the ISS and where is all this cargo going to be stored on the ISS?

    What is the development costs for this 70 ton cargo carrier? Is that development cost factored in on the cost per pound?

  • Joe

    “where would they store that much cargo on the ISS? lol .. that is laughable.”

    In the MPLMs that would be berthed to Common Berthing Mechanism Ports on the ISS. Not only not laughable but checked out in detail.

    “Where is the money for this imaginary 70 ton cargo carrier for the SLS that would be delivering all this cargo to the ISS”

    It has not been proposed for the SLS, so for the moment nowhere.

    “and where is all this cargo going to be stored on the ISS?”

    You just asked that question (and had in answered) two sentences above. Might want to calm down.

    “What is the development costs for this 70 ton cargo carrier? Is that development cost factored in on the cost per pound?”

    For the Side Mount configuration Study, yes to both questions.

  • Joe

    The way contracts work is the following:

    The contractors (Space X/Orbital Sciences) are obligated to deliver 20 Metric Tons to the ISS.

    From their point of view on the negative side:
    – They will be paid in increments as the flights proceed.
    – But if they for whatever reason failed to meet the 20 Metric Ton requirement they would face penalties (as in they might have to pay back money they had already received)

    From their point of view on the positive side:
    – If they meet the 20 Metric Ton requirement they receive the full contract amount ($1.6/$1.9 Billion respectively).
    – Any further ISS Deliveries would cost extra (at some new, presumably higher rate). Space X says as much itself in the first paragraph of its press release: “The firm contracted value is $1.6 billion and NASA may elect to order additional missions for a cumulative total contract value of up to $3.1 billion.”

    Therefore the unit costs remain unchanged by all this discussion:
    – Space X: 20 Metric Tons for $1.6 Billion ($80,000/kg)
    – Orbital: 20 Metric Tons for $1.9 Billion ($95,000/kg)
    – Average for the contract: $87,500/kg

    Those are numbers for comparison.

  • Coastal Ron

    Patrick, all of the CCiCap Commercial Crew vehicles will be able to be used as lifeboats for at least 6 months, just like Soyuz.

    However since each of them can also carry up to seven passengers, it will allow NASA to increase the current staffing level of the ISS from six to seven without needing more lifeboat vehicles standing by. It is already planned that one of the Dragon CRS flights will bring up a new docking port for the commercial crew vehicles in it’s unpressurized trunk section, likely a NASA Docking System (NDS) design, which will be installed on one of the CBM’s that Dragon cargo, Cygnus and the HTV use.

  • Coastal Ron

    Joe said “That is because the Falcon Heavy does not (as yet anyway) exist.

    Using your standards, the SLS doesn’t exist either. Heck, taking your attitude even further, the money to fully develop the SLS doesn’t exist, and no money for any use of the SLS exists either. Talk about fictitious.

    By contrast, the Falcon Heavy is made up of three Falcon 9 rockets, and the Falcon 9 rocket has already flown three times, so it’s not a big leap to see how the Falcon Heavy is far more in existence than SLS. Also the Falcon Heavy exists enough that a customer signed a contract to use it, and likely paid $Millions as a launch deposit. That means customers believe that the Falcon Heavy does exist – I haven’t seen anyone do the same for the SLS (i.e. commit money to use it), have you?

    As to your talk about costs Joe, you keep getting confused by third party chatter. There are only two numbers that I look at:

    1. What SpaceX advertises, because everyone, including future customers, can see what they say they are selling their services for. If they are perceived as engaging in bait & switch, then their potential customers will complain publicly or just not buy from them. With the backlog SpaceX has, that doesn’t seem to be the case.

    2. What they have signed contracts to do.

    What you also ignore is that even though SpaceX will change their prices for the same reason everyone does, the same happens with government contracts too. You may not see it in the tiny corner of the engineering world you live in, but companies change their prices with the government all the time. NASA may think the SLS will only cost $500M today, but that assumption is acknowledged to be only a guess, and one that is dependent on the unneeded SLS flying 2-3 times per year.

    There is an article over at Aviation Week that talks about the supply chain problems Boeing and Lockheed Martin are having with the procurement of the electronic components for the SLS and MPCV. The procurement is far more expensive and far more drawn out than they originally anticipated, and they think it could affect the test schedules for the SLS and MPCV. They wouldn’t mention that unless it was going to be a big cost impact, so we are already seeing that whatever costs they have assumed previously are going to rise.

    In contrast, SpaceX recently lowered their pricing for Falcon 9 from $59.5M for their current version of Falcon 9 (9mt to LEO) to $54M for Falcon 9 v1.1 (13mt to LEO).

    It’s not the simple world that you think it is Joe.

  • Coastal Ron

    The better question is what would they do with so much at one time? It’s obvious you don’t do the shopping in your household Joe.

    Part of the reason for having a constant flow of deliveries is that it provides for “fresher” supplies, and it also allows scientific supplies to be delivered closer to when they are actually needed. Delivering 150,000 lbs of supplies once a year doesn’t make sense, and neither does using something the size of Shuttle Sidemount or the SLS to deliver massive amounts of supplies before they are needed.

  • Joe

    Coastal Ron says: Posted September 14, 2012 at 12:45 PM
    “Using your standards, the SLS doesn’t exist either. Heck, taking your attitude even further, the money to fully develop the SLS doesn’t exist, and no money for any use of the SLS exists either. Talk about fictitious.”

    Actually the conversation between Patrick and myself was more about the Shuttle Ron. You know the active system that was shut down to make room for the ‘Commercial’ system that turns out to be more expensive.

    I note that above you brought up the lifeboat issue. I think it is bogus but hypothetically removing the advantages of crew exchange from the equation Patrick’s cost analysis of Shuttle Utilization Flight costs is still very interesting:
    $672,800,000 / 9,429kg = 71,354 $/kg of cargo
    Ouch. Let’s use the max from STS-126.
    $672,800,000 / 12,748kg = 52,776 $/kg of cargo

    Using the higher figure (to give you every advantage) the cost for Shuttle Utilization Flight cargo is $71,354/kg

    Using the contract numbers for Space X and Orbital Sciences their cost to supply the same service are:
    – Space X: $80,000/kg (12% more expensive)
    – Orbital Sciences: $95,000/kg (33% more expensive)
    – Average for the two contracts (the one that is really the most pertinent): $87,500/kg (23% more expensive)

    The point that any dispassionate observer would note is that the new system (in all three cases – and throwing all advantage to your side of the discussion) was developed in large part at tax payer expense and is more expensive than the system it was intended to replace by double digits.

    “It’s not the simple world that you think it is Joe.”

    Thank you for the advice Yoda, but I do not think the world is simple at all. However this issue is as simple as what I just described above.

  • Joe

    “The better question is what would they do with so much at one time? It’s obvious you don’t do the shopping in your household Joe.”

    This gets truly absurd. The supply portion of the study was run by the ISS logistics people. You know the one that actually handle day to day ISS operations. Do you think they might know a little bit more about that sort of thing than you Ron?

  • Coastal Ron

    Over the life of the Shuttle, the average cost per flight was $1.2B, or $1.5B if you amortize the development cost (the more accurate number from a taxpayer standpoint). NASA now even agrees with these figures.

    For supporting the ISS, the Shuttle was useless in actually keeping crew at the ISS any longer than two weeks, so therefore it can not be considered as an alternative to Soyuz or any of the Commercial Crew vehicles, all of which can keep crew at the ISS for up to six months.

    So beyond assembly (which is now complete), the only other function the Shuttle could be used for in supporting the ISS was delivery of pressurized cargo. For that NASA used the MPLM, and the MPLM had a capacity of 9mt. Using the historical pricing above, that means the lowest average price the Shuttle could deliver pressurized cargo to the ISS for would be $165,346/kg ($75,000/lb). Wow!

    The $3.1B CRS contract, as you noted above, values the cost of the total 40mt of cargo at $77,500/kg. So no, from a cargo support standpoint the Shuttle, using pricing that NASA agrees with, was double the price of Commercial Cargo.

    And there is even more upside for NASA in the future. On the current CRS contract SpaceX will be flying Dragon less than half full, since it could carry up to 6mt of pressurized cargo, but is only required to average 1,667kg/flight (20mt over 12 flights). SpaceX is also required to use a new Dragon for each flight, and not reuse one that has previously flown. Because of both of those factors, which were likely prudent when they took a risk with SpaceX initially for the CRS program, there is a strong possibility that $/kg will drop even further on the follow-on CRS contract.

    All you need are the correct facts to see the truth, and the truth here is that the Shuttle was not the least costly way of supporting the ISS (or delivering cargo to LEO either). Not only that, it was not shut down in favor of Commercial Crew, since Bush/Griffin were planning on the Ares I/Orion combo to come online by that time.

    The Shuttle program was shut down because there was nothing critical enough to use it for after ISS assembly to risk lives – and that was the reason it was shut down. After losing 40% of the Shuttle fleet and 14 deaths, it was too dangerous to keep flying.

  • Joe

    Coastal Ron says: Posted September 14, 2012 at 9:05 PM

    “Over the life of the Shuttle, the average cost per flight was $1.2B, or $1.5B if you amortize the development cost (the more accurate number from a taxpayer standpoint). NASA now even agrees with these figures.”

    You are trying to throw in the sunk cost including developing the Shuttle and amortize them over the flights and add all that to the operating costs. It’s an old accounting trick for trying to overestimate operating cost. But if you are trying to use it you really should not have linked to a Wall Steer Journal article titled “”As Shuttle Sails Through Space, Costs Are Tough to Pin Down” as your source. It kind of under cuts your alleged point.

    Patrick Ritchie used actual operating costs (adjusted for inflation) and actual MPLM cargo numbers from a representative year for his analysis. That is the only way to get a comparison.

    “For that NASA used the MPLM, and the MPLM had a capacity of 9mt.”

    Again Patrick used actual cargo numbers. Additionally if you are going assert that “and the MPLM had a capacity of 9mt” you should not link to a source article that clearly states the module can carry 10 tons. It kind of under cuts your alleged point.

    “The $3.1B CRS contract, as you noted above, values the cost of the total 40mt of cargo at $77,500/kg. So no, from a cargo support standpoint the Shuttle, using pricing that NASA agrees with, was double the price of Commercial Cargo.”

    You make the assumption that the proposed contract extension would be for another 20 MT and that is not stated anywhere in the source article. In fact new rates (that would probably be higher) would have to be negotiated. In any case for the third time out of three your supposed sources do not backup your assertions.

    Your “analysis” is therefore (at best) useless.

  • Coastal Ron

    Joe, “9mt” (9 metric ton), or 9,000 kg, is equal to 19,841 lbs. 10 tons (10 x 2,000) is equal to 20,000 lbs.

    As to Shuttle pricing, budgets are never used to determine pricing, so only by looking at what was actually spent can you determine actual cost. This is especially true when the budget numbers used are for the year of the launch, but the actual production of the hardware was made in the years prior to the year of the launch. And YES, taking into account development costs, which was only 20% of the Shuttle cost, is a valid way to evaluate whether it is worthwhile building something. Still, if you use the $1.2B price/flight the Shuttle is still more expensive than Commercial Cargo.

    Getting back to whether development costs should be used for determining “value”. If I told you with a little more money we could lower the flight cost of the SLS to $10 Million/flight, wouldn’t that be great? But then if I told you we would have to spend $1 Trillion to do that, it would change your calculation for deciding whether it would be worth it.

    Right now the calculation for the SLS is to spend $30 Billion to develop it, and there are no funded payloads for it after it becomes operational. If it turns out that the U.S. decides to cancel the program after one flight, the cost of that one flight won’t be the supposed $500M, it will be $30B plus $500M.

    Hiding the development cost is an accounting trick, not the other way around.

  • Joe

    Coastal Ron says: Posted September 15, 2012 at 8:28 AM
    “Joe, “9mt” (9 metric ton), or 9,000 kg, is equal to 19,841 lbs. 10 tons (10 x 2,000) is equal to 20,000 lbs.”

    Ron, if you assume that the NASA PAO website really makes those distinctions fine. But, since they are your trusted source why did you not use them for your Shuttle launch Cost estimate as well:

    http://www.nasa.gov/centers/kennedy/about/information/shuttle_faq.html

    Q.How much does it cost to launch a Space Shuttle?
    A.The Average cost to launch a Space Shuttle is about $450 Million a mission.

    Instead you went to a Wall Street Journal Article that may have used the $1.2 Billion figure, but never said NASA “agrees with these figures”. You however asserted that NASA did. A cynical type (not a naïve sort like myself – who as you have kindly pointed out does not realize how complicated the world is) might reach the conclusion you are trying to “cook the books”.

    Anyway let’s use your trusted source:
    $450 Million per launch/9,000 kg per launch = $50,000/kg.

    That makes Space X $80,000/kg contract price 60% more expensive than a Shuttle Utilization flight (even when all the other Shuttle advantages are left out to avoid further argument/distraction from the key point). If you like those numbers better, fine by me.

    The rest of your post amounts to word salad. The operating cost in any period divided by the cargo delivered is the unit cost for such a program.

    From past experience I know you will keep dragging this thing out by the internet posting rule “He who posts last wins.” But this is it for me. Anyone who believes I could not handle your next set of “facts” will just have to think so.

  • Joe, SLS will remain fantasy number until it is operational (which may never happen.) Deal with it.

    SpaceX numbers come from an operational vehicle. As far as what others may claim, I’m not responsible for someone else’s claim.

    In theory, bigger rockets should be cheaper. SpaceX has plans for bigger cheaper rockets as well. But reality often differs from conjecture. Just look at the claimed costs of the shuttle compared to reality.

    Frankly, none of this matters because there is one irrefutable way to determine which is the better system.

    Let the consumer decide in open competition.

    Anytime the government is involved it distorts open competition. Score one for you.

    But competition remains the only proven method of lowering costs.

  • Hey all,

    The discussion here has been very interesting, but see my latest long post on Behind the Black:

    “Prices, Demand, and SpaceX”
    http://behindtheblack.com/behind-the-black/essays-and-commentaries/prices-demand-and-spacex

    We now have some solid information about SpaceX’s prices, from one of its biggest competitors!

  • Coastal Ron

    Joe, the capacity of the MPLM is a pretty simple thing, so using the capacity NASA has provided is fine. What NASA spends though is another matter.

    As to the $450M NASA number, it is not usable because it is non-specific. Does it include only direct costs, or also indirect? What flight rate did that assume? How do they account for service costs when the Shuttle wasn’t flying? Too many unknowns in that simple answer.

    For instance, the last full-up set of contracts for the Shuttle that I found showed that the External Tank cost $173M/each, a set of SRM’s (without being assembled into SRB’s) cost $69M/set, and United Space Alliance (USA) was charging $97M/month (regardless if the Shuttle flew or not). Since Shuttle flights resumed after Columbia, the most flights in any one year was five, and from 2006-10 they averaged about four flights/year. That means just USA’s costs would equal $291M/flight over that period, and when added to just the ET & SRM costs that equals $533M/Shuttle flight. So no, I don’t believe that generic $450M number on NASA’s website.

    As to the WSJ article, apparently you didn’t read it. If you did, you would have seen that they were talking about a study that Professor Roger Pielke Jr. had started back in the mid-90’s and updated through the end of the Shuttle program. And apparently you missed this part of the article:

    Adding to the confusion, NASA also has released an inflation-adjusted figure, despite its preference for a figure representing cash outlays. That number is even higher than Prof. Pielke’s: about $211 billion.

    That NASA hasn’t updated the webpage of a program that has now ended is no surprise. You were probably the only one to visit it in a long time.

    So let’s summarize:

    – An outside auditor found that the historical cost of Shuttle flights was $1.2B, or $1.5B if the development costs were amortized.

    – NASA now agrees with what that study found, and may even estimate a higher number.

    – The Shuttle cost about twice as much $/kg to deliver pressurized cargo to the ISS than what NASA is paying under the CRS contract.

    I hope that ends your confusion.

  • pathfinder_01

    Joe the big issues with shuttle(or any NASA vechical) delivery to the ISS is that it costs $3-4 billion a year to have the shuttle period. At maybe 12MT(max) each and at about 4 flights a year that comes out to about $65k per MT for a total of 48MT. Sure the falcon 9 delivers less cargo, but ah the Russians have managed to run just fine with Progress(and it only delivers 2MT and mind you with a crew of 3).Not to mention the ISS recivies cargo via HTV and ATV as well.

    In fact the big issues with cargo is that it is often bulky and not massive. Dragon in theory could deliver 6Mt but more likely around 2.5MT per flight. At about $133 million each(or so) that could be as low as about $22K per MT. If you wanted to send 48MT to the ISS, you could do so in 8 flights of dragon for a total of $1 billion(and in reality you would have a combination of both Space X and Orbital). Orbital is actually more expensive per kilo, but the cost to deliver 12MT would still come out less than the shuttle per year(it would come to a little under $1 billion). You would need to deliver more than 36MT a year before the shuttle was cheaper than Orbital(and the shuttl only carried as little as 8.3mt in the MPLM in some flights).

    The shuttle really offered little else to the ISS. It could exchange crew, but could not act as lifeboat. It could host a spacewalk, but ah the ISS has 2 airlocks of it’s own as well as an assortment of robot arms/cranes. It was instrumental in building the ISS but once built it offered little else.

    Although we don’ t know how much commercial crew will cost, an atlas goes for around $200 million a flight(and man rating will not drastically increase that). Depending on how much the spacecraft costs(lets pull a number out of thin air say triple costs total) say $600 million a flight it is still cheaper than the shuttle because you are just paying for the flight not $3-4 billion a year in shuttle costs plus Soyuz costs. Commercail crew will have to get up to around $800 million per flight before the shuttle is cheaper.

    Basically the shuttle was like a pick up truck. You probably need one to build a house(i.e. the ISS) but using it for a daily driver(i.e. Commercial crew) and for small cargo (like picking up groceries, i.e. commercial cargo) is a bad idea. It is cheaper to own a car and rent a pickup truck when needed than own a pick up truck(unless you are using that pick up’s cargo capcity almost daily and the shuttle almost never flew with a full cargo bay on non ISS construction missions). In the case of NASA it means use Space X and Orbital for cargo and if you do need to haul something 20MT up use Delta Heavy.

  • Kelly Starks

    >..The shuttle really offered little else to the ISS. It could exchange crew, but could not act as
    > lifeboat. It could host a spacewalk, but ah the ISS has 2 airlocks of it’s own as well as an
    > assortment of robot arms/cranes. It was instrumental in building the ISS but once built it
    > offered little else.

    Actually it also offers more construction and maintence capabilities then are possible now – and the station was designed asuming it would have a shuttle classcraft to service it.

    >…
    >..Basically the shuttle was like a pick up truck. You probably need one to build a house(i.e. the ISS)
    > but using it for a daily driver(i.e. Commercial crew) and for small cargo (like picking up groceries,
    > i.e. commercial cargo) is a bad idea. It is cheaper to own a car and rent a pickup truck when
    > needed than own a pick up truck

    You forget we had the shuttle and had to contract for the construction of the replacement craft. Hence why total costs for SpaceX to deliver a pound to ISS is projected (even by using SpaceX’s numbers) to cost the gov. 20% more then shuttle. I.E. the $800+ million NASA gave to SpaceX to dev these craft,and the overhead program costs, etc, plus the Falcon/Dragon launch costs.

    So if you already own a pickup, its cheaper to use that as a daily driver then to by a car.
    ……Well actually its about always cheaper to by a pickup truck and drive it daily, then by a similar car and drive it daily. Given the vastly lower purchase costs and maint costs of pickups, which is one big reason they dominate sales.

  • Kelly Starks

    I might trust the buyers reported costs, then a competitors projections.

  • Coastal Ron

    Couple of things Kelly.

    1. I’ve heard the claim that “the ISS was designed assuming the Shuttle would be around to service it”, but no one has been able to detail what can’t be done now that the Shuttle has retired. It was never needed for keeping crew there, since the Shuttle could only stay in space for two weeks. It isn’t needed for logistics, as there are four (and soon a fifth) cargo vehicles, including two (soon three) that can deliver cargo that only fits through a CBM port. There are no plans for more modules, but they could be delivered using existing rockets and a tug module derived from the ATV, HTV or Cygnus. So what CAN’T we do without the Shuttle that we need to do?

    2. SpaceX only received $396M under the COTS program, not $800M. In fact the entire COTS program, including the $32.1M RpK received before being kicked off the program, is $716M. Adding that $716M onto the $3.5B CRS contract for 40mt of supplies equals $105,400/kg, still far below the $133,333/kg figure Shuttle cost even if the development costs are ignored and it is assumed the MPLM is fully loaded with 9mt of cargo. Shuttle cost more, and did not have the ability to drastically reduce costs like commercial companies would be able to do with follow-on contracts (no developments costs needed to be amortized).

    3. Don’t get too wrapped up in the “pickup” analogy, since your extension of it (that pickups cost less than cars) is not only not always true, but the last Shuttle built (Endeavour) cost $2.2B and required an army of people to refurbish it after every flight. Not cheap. Commercial Cargo & Crew vehicles likely will cost far less to build and maintain.

  • Kelly Starks

    >Coastal Ron
    > no one has been able to detail what can’t be done now that the Shuttle has retired. ==

    Many components and subsystems can’t be lifted or installed without it. It has the ability to bring up more tons of cargo at a time, or year then COTS.

    >==There are no plans for more modules, but they could be delivered using existing rockets
    > and a tug module derived from the ATV, HTV or Cygnus. ==

    Actually not, they lack the lift capability, adn the modules would need to be redesigned to be compatible with them. [Also redesigned to be launched on ELVs would increase the weight about 30%, complicating the weight issues.]

    > 2. SpaceX only received $396M under the COTS program, not $800M. =

    http://www.gpo.gov/fdsys/pkg/CHRG-112hhrg66561/pdf/CHRG-112hhrg66561.pdf
    And SpaceX pres briefs count up to SpaceX has receaved $850M up until now. (orbital got far less.) Congress and SpaceX agree on that.
    Per pound
    Shuttle $21,268
    Progress $18,149
    CRS/COTS $26,770

  • Coastal Ron

    Kelly, the largest component the Shuttle lifted for the ISS was the Truss & Solar Arrays assemblies. Those weigh 15,900 kg (35,053 lbs), which is well within the lift capability of Delta IV Heavy, Ariane 5, Proton, and even JAXA’s H-IIB.

    Alll of the components lifted can be installed on the ISS using the Mobile Servicing System (MSS) – no Shuttle Remote Manipulator System (SRMS) required.

    So specifically Kelly, what can’t be done without the Shuttle?

    Regarding cargo, we have alternatives that provide a more frequent stream of supplies than the infrequently flown Shuttle, so that is no reason either, especially when it cost $200M/month to keep the Shuttle program running, regardless whether it flew.

    Regarding future modules that could be added, all of the worlds 20mt lifters can lift the same the mass to LEO that the Shuttle could, and no Shuttle mission lifted more than 15,900 kg of cargo. As to “redesigned to be launched on ELVs”, what are you talking about? Can you detail out what specifically needs to be redesigned, and why a commercial rocket payload interface is supposedly so much more massive than the Shuttle’s? I think you’re making that up, especially since ULA says Delta IV Heavy can lift a 50,000 lb payload to LEO, and they are not including their payload interface as part of the payload.

    We don’t need the Shuttle to build more ISS-type structures in space, or keep the one we have going into the next decade.

  • Coastal Ron

    Kelly originally said:

    the $800+ million NASA gave to SpaceX to dev these craft

    The link you provided about the congressional testimony said “NASA anticipates it will have invested $800 million in the COTS project”, which was talking about the COTS program overall, not just SpaceX. You said SpaceX was given $800M, so even the link you provided shows you were wrong.

    As to the number you quoted for $/lb of cargo, those were numbers that Representative Mo Brooks gave, and Gerstenmaier did not agree with them. Representatives can say whatever they want (and they do), but that doesn’t mean they know what they are talking about. He also seemed to assume $500M for a Shuttle flight, which been proved to be less than half the real cost.

  • MaDeR

    I am surprised there was so much discussion about it with Joe.
    ” The operating cost in any period divided by the cargo delivered is the unit cost for such a program.”
    So, development costs does not count. Just because. He basically admitted he is cherrypicking numbers that he likes to “prove” that Shuttle/SLS was/is/will be cheaper.

  • Kelly Starks

    > Kelly, the largest component the Shuttle lifted for the ISS was the Truss & Solar Arrays
    > assemblies. Those weigh 15,900 kg (35,053 lbs), which is well within the lift capability
    > of Delta IV Heavy, Ariane 5, Proton, and even JAXA’s H-IIB.

    Yes but
    1- A Delta IV, A5 etc CAN’T CARRY THE TRUSS AND SOLAR ARRAYS. Not that they are to heavy, they need support, g limits, power support, etc that is on the shuttle, and not on the ELVs. Hence why converting sats and systems designed for shuttle, to be launchable on the ELVsincreases their weight about 30%.

    2- You need the shuttle to suport construction adninstallation (no the station doesn’t have the systems for that. Why would it?)

    >== infrequently flown Shuttle, so that is no reason either, especially when it cost $200M/month
    > to keep the Shuttle program running, regardless whether it flew.

    And the crime is the companies the coperated the fleet had long proposed ways cut that by a factor of 4. Course that would defeat the purpose.

    Anyway -still a lot cheaper then COTS/CRS.

    > == As to “redesigned to be launched on ELVs”, what are you talking about? Can you
    > detail out what specifically needs to be redesigned, and why a commercial rocket payload
    > interface is supposedly so much more massive than the Shuttle’s? I think you’re making that up, ==

    Not making it up – I outlined some stuff in general – not feeling like looking up a URL to the references tonight.

    >== especially since ULA says Delta IV Heavy can lift a 50,000 lb payload to LEO, and they
    > are not including their payload interface as part of the payload.

    nor the payload support systems (and no I don’t just mean structural support.)

    >== We don’t need the Shuttle to build more ISS-type structures in space, or keep the one we
    > have going into the next decade.

    Those invoved in building, designing, and operating it – are not nearly so confident. Clearly we have no way to build it anymore. nor any program to build anything that could. Even keeping ISS going to 2020 was debateable.

  • Kelly Starks

    >..Coastal Ron
    > The link you provided about the congressional testimony said “NASA anticipates it will have
    > invested $800 million in the COTS project”, which was talking about the COTS program overall, ==

    Nope, read more carefully.

    >.As to the number you quoted for $/lb of cargo, those were numbers that Representative Mo Brooks gave,..

    Its simple Math Ron, from the CBO.

    The numbers are correct.

Leave a Reply

Your email address will not be published. Required fields are marked *