Transterrestrial Musings

Comments

> First, a mission that could have been performed
> with a single Shuttle launch will now require
> at least two, and perhaps three flights of a
> still-expensive expendable (probably on the
> order of a hundred million per flight). One to
> deliver the OSP, providing delivery/return of
> the crew, and one or two to deliver the payload
> that the Shuttle would normally carry in its
> payload bay.

I think you can get a good back-of-the-envelope estimate by looking at the cost of current systems. For unmanned logistics, one would presumably use an Ariane Transfer Vehicle type spacecraft (@ $50M+ per copy) plus an EELV-H ($140-170M). This combination could deliver a 9 metric ton payload to ISS or about the same capability as the Shuttle's. The manned Orbital Space Plane missions would probably cost $200-300M/flight according to NASA's Space Transportation Architecture Study. You need both vehicles to do the same job as ISS, and the required launch rate is about 2x. E.g., if one assumes the new system must replace four annual Shuttle cargo/crew transfer missions, four manned OSP crew transfer plus four unmanned cargo missions will be required.

On paper, this option does seem cheaper (8 x $250M = $2B) but the OSP has a significant drawback since it can't return large payloads from ISS and the development cost of a winged mini-shuttle would be very high. This would increase the cost of operating ISS, since experiments and other pieces of equipment no longer would be refurbished on the ground. An ISS International Standard Payload Rack costs as much as $100M, so this is a significant issue. Wales larrison has posted a very interesting analysis on this, basically concluding it's a wash.

The OSP does have a significant advantage if it's got a crew escape system, though.

I think the cheapest option would be some sort of pseudo-commercial Soyuz deal plus additional ATV/HTV cargo missions. The development cost would be almost zero in that case, and there are some interesting possibilities for privatization as well.

MARCU$

Posted by Marcus Lindroos at September 19, 2003 10:00 AM

The manned Orbital Space Plane missions would probably cost $200-300M/flight according to NASA's Space Transportation Architecture Study.

What do you want to bet that doesn't include amortization?

Posted by Rand Simberg at September 19, 2003 10:15 AM

In the article where you state that NASA risks irrelevancy if they ignore the demand for public space travel, are you suggesting NASA should fly tourists on the OSP?

Posted by B.Brewer at September 19, 2003 03:49 PM

No, I'm suggesting that we need to develop an overall space policy in which public space travel is not just considered, but encouraged, and figure out what role NASA might play in that (e.g., simply buying thousands of tickets into space, using the ones it needs, and selling the rest on the market). OSP as planned doesn't make sense under any scenario.

Posted by Rand Simberg at September 19, 2003 04:09 PM

The OSP does have a significant advantage if it's got a crew escape system, though.

A true sign of a mature orbital delivery vehicle will be that it has no "escape system", just as no commercial passenger aircraft does. Escape systems are for experimental vehicles and warcraft; commercial vehicles are simply robust enough to handle almost all emergencies (and those they can't handle are just the very occasional losses).

That this isn't nonsense was demonstrated by the DC-X recovery and return-to-launch-site when it had the hydrogen explosion and fire. But NASA is again going the wrong way...

Posted by Troy at September 19, 2003 08:41 PM

>> The OSP does have a significant advantage if
>> it's got a crew escape system, though.

> A true sign of a mature orbital delivery
> vehicle will be that it has no "escape system",
> just as no commercial passenger aircraft does.

Maybe -- but the \current\ baseline for comparison remains the Shuttle, which is a dangerous vehicle since there are no ejection seats or capsules.

MARCU$

Posted by Marcus Lindroos at September 20, 2003 09:26 AM

The Shuttle's technological and political difficulties are no reason to penalize new designs, I think.

Besides, if you're talking escape systems the baseline better not be the Shuttle -- with that design there are major portions of the launch envelope from which there is no practical method of emergency escape, due to the solid boosters, side-mounting of the Orbiter on the stack, and the placement of the engines on the Orbiter rather than the External Tank. (For reentry, of course, an "escape system" is phenomonally difficult to implement -- which is why none exists, or has ever existed, on any manned spacecraft... despite more than half the fatalities occuring on reentry. But that's another argument.)

Posted by Troy at September 20, 2003 10:01 AM

> The Shuttle's technological and political
> difficulties are no reason to penalize new
> designs, I think.

Again, the answer to your claim is "maybe." But I do think any ELV launched system such as the Orbital Space Plane will need an escape system.

MARCU$

Posted by Marcus Lindroos at September 20, 2003 11:22 AM

> The manned Orbital Space Plane missions would
> probably cost $200-300M/flight according to
> NASA's Space Transportation Architecture Study.

: What do you want to bet that doesn't include
: amortization?

Clearly, development costs are not included. But I suspect a ballistic capsule would be quite cost competitive vs. the Shuttle even when you factor in the cost of developing the new manned system. For example, ESA estimated its ballistic Crew Transfer Vehicle would cost about $2B (vs. 6B for Hermes, which incidentally "only" exceeded its development budget by 20%).
---
Anyway, Rand -- how do you feel about replacing the Shuttle with off-the-shelf manned (=Soyuz) and unmanned spacecraft...? It seems to me as if any "privatized" ISS transportation system would try to keep development expenses to a bare minimum. E.g. the Automated Transfer Vehicle is to be operated on a quasi commercial basis by Arianespace. I would say the least expensive resupply option would be to use existing ELVs and spacecraft rather than the Shuttle. The United States would not even have to develop any new spacecraft; it could simply adapt existing ones for use with the Delta IV and Atlas V. For example, the internationals would give the U.S. cargo spacecraft free of charge while the Americans would launch them, also for free.

Of course, NASA (and the United States in general) would never give up its independent manned spaceflight capability, e.g. by relying exclusively on privatized Soyuz for manned access to the Space Station. But I still think a large ballistic capsule (capable of returning International Standard Payload Racks to Earth) could be a cost effective alternative to the Space Shuttle.

MARCU$

Posted by Marcus Lindroos at September 20, 2003 11:41 AM

Anyway, Rand -- how do you feel about replacing the Shuttle with off-the-shelf manned (=Soyuz) and unmanned spacecraft...? It seems to me as if any "privatized" ISS transportation system would try to keep development expenses to a bare minimum.

Frankly, I'm indifferent. I don't think we need a Shuttle replacement. I don't think that we need the ISS, and wouldn't shed a tear if it fell into the Pacific. I don't, in fact, think that we need anything resembling NASA's current manned spaceflight program, but it will lumber on, because there are too many jobs at stake.

Posted by Rand Simberg at September 20, 2003 12:36 PM

> I don't, in fact, think that we need anything
> resembling NASA's current manned spaceflight
> program, but it will lumber on, because there
> are too many jobs at stake.

By the way, I don't agree with the following quote (from the techstation.com article) at all:

"Let's be generous and assume (improbably) that NASA will actually reduce the work force at the Cape and in Houston currently devoted to maintaining the Space Shuttle and training astronauts (it will be hard to do that, because of Congressional pressure to maintain the jobs)."

There is nothing "improbable" about this at all, if you check the history of human spaceflight since Apollo. Every manned project to date has had to make do with less (in terms of manpower and funding) than its predecessor. For example, Apollo itself employed about half a million civil servants and contractor personnel at its peak whereas about 100,000 engineers were working on the Shuttle during the peak development period in the late 1970s. It seems the Space Station workforce has been about 50,000 including international partners.
---
If STS is mothballed, I am sure the system that replaces it will be less costly and manpower intensive unless the White House and Congress somehow decide to increase the space budget.

MARCU$

Posted by Marcus Lindroos at September 20, 2003 01:34 PM

Every manned project to date has had to make do with less (in terms of manpower and funding) than its predecessor. For example, Apollo itself employed about half a million civil servants and contractor personnel at its peak whereas about 100,000 engineers were working on the Shuttle during the peak development period in the late 1970s. It seems the Space Station workforce has been about 50,000 including international partners.

But the Shuttle workforce remains huge (otherwise it wouldn't have a three billion dollar annual budget). I don't know if the manned space program employs fewer or more than it did twenty five years, ago, but there's no way to tell with the information provided.

If STS is mothballed, I am sure the system that replaces it will be less costly and manpower intensive unless the White House and Congress somehow decide to increase the space budget.

Only if they come up with some other project to employ the current personnel.

Posted by Rand Simberg at September 20, 2003 02:22 PM

> But the Shuttle workforce remains huge
> (otherwise it wouldn't have a three billion
> dollar annual budget).

Indeed -- but it seems maintaining even a bare-bones Apollo/Saturn manned lunar program would have cost about twice as much! I note NASA had to shut down the production lines in 1969-70 when the manned spaceflight budget dropped below ~$5 billion at current dollar rates. And the Apollo peak funding was almost $15 billion in today's dollars! [ http://www.abo.fi/~mlindroo/armagddn/NASA_msf.gif ]
---
Apollo reportedly required about 200,000 NASA and contractor personnel for spacecraft & launch vehicle construction, testing, documentation and management vs. 50,000 for the Shuttle in the late 1970s (source: David Baker's SPACE SHUTTLE HANDBOOK from 1978). Back then, NASA expected that the number of personnel at KSC could be reduced to 10,000, down from 25,000 during the peak of the Apollo project. I think the total is higher today, but still less than the Apollo workforce. The big disappointment has been STS utilization; the annual cost of the system is about as low as originally predicted, but those $3 billion only buy you six missions per year -- not 60...
---
I see (source: AW&ST) NASA announced in September 1995 that privatizing Shuttle operations by 2001 might cut annual program costs by $500 million and eliminate up to 40% of the 19,700 contractor jobs and 2,800 NASA positions supporting the Shuttle program. That's significantly less than Apollo in any case. I think the main difference is procurement of new hardware; those Saturn V and Apollo production lines required lots of personnel.

> I don't know if the manned space program
> employs fewer or more than it did twenty five
> years, ago, but there's no way to tell with the
> information provided.

It seems the number was fairly constant during the 1980s, but Goldin managed to cut some Shuttle jobs during the late 1990s.

: I am sure the system that replaces it will be
: less costly and manpower intensive unless the
: White House and Congress somehow decide to
: increase the space budget.

> Only if they come up with some other project to
> employ the current personnel.

Most of the current personnel, yes. Some layoffs will probably seem inevitable even to the Congresscritters protecting jobs in their districts.
---
By the way -- you seem to be saying NASA should simply buy tickets from prospective operators of commercial manned spacecraft. I agree that this might be realistic, since the U.S. currently is paying $3 billion per year to transport a dozen astronauts to and from ISS. I think if the price were set at, say, $2 billion, Boeing & Lockheed-Martin might figure out investing a few hundred million dollars in a manned EELV capsule (e.g. derived from Soyuz or Apollo technology) would be more economical to them... But they will never accept a deal for transporting 1000 passengers/year to orbit for $3 billion...that would be way too difficult. If private industry has to pay the DDT&E cost, I think we would get a very conservatively designed expendable system -- not a fully reusable spaceplane.

MARCU$

Posted by Marcus Lindroos at September 22, 2003 02:25 AM

you seem to be saying NASA should simply buy tickets from prospective operators of commercial manned spacecraft. I agree that this might be realistic, since the U.S. currently is paying $3 billion per year to transport a dozen astronauts to and from ISS. I think if the price were set at, say, $2 billion, Boeing & Lockheed-Martin might figure out investing a few hundred million dollars in a manned EELV capsule (e.g. derived from Soyuz or Apollo technology) would be more economical to them... But they will never accept a deal for transporting 1000 passengers/year to orbit for $3 billion...that would be way too difficult. If private industry has to pay the DDT&E cost, I think we would get a very conservatively designed expendable system -- not a fully reusable spaceplane.

I wouldn't expect either Boeing or Lockmart to bid, because if I were NASA, I would put out a solicitation for 3000 round-trip tickets annually for a million dollars per. OSP would be a non-starter for that order.

Posted by Rand Simberg at September 22, 2003 06:05 AM

> I wouldn't expect either Boeing or Lockmart to
> bid, because if I were NASA, I would put out a
> solicitation for 3000 round-trip tickets
> annually for a million dollars per. OSP would
> be a non-starter for that order.

Would any entrepreneurial organization be able to convince investors and bankers such a system is economically and technically feasible, though...? 3000 passengers a year requires 60 annual flights by a Kankoh Maru or Spacebus-class 50-seat vehicle or 600 flights/year by a bizjet size vehicle. In contrast, the tiny suborbital X-15 only made at most 40 flights per year and all three vehicles had either crashed or were worn out by the end of the 199-flight test campaign...
---
Aren't you raising the bar so high that nobody would be able to submit a successful bid? E.g. the Japanese Rocket Society estimates the Kankoh Maru VTVL SSTO would require a $15-billion development and certification program.

MARCU$

Posted by Marcus Lindroos at September 22, 2003 07:47 AM

Aren't you raising the bar so high that nobody would be able to submit a successful bid?

There's only one way that I know of to find out. But a successful bid would be in the form of an operational system, not a cost-plus contract.

And I don't think that any of your examples are relevant (particularly Kankoh Maru--SSTO is not required, and I don't know what "certification" means in this context), though the most likely candidate is the high-rate bizjet class.

Posted by Rand Simberg at September 22, 2003 08:07 AM

> There's only one way that I know of to find out.

There is an easier way: just do a simple BOTE cashflow analysis while using basic common-sense assumptions...

As a first rule of thumb, for a space LV project, you can justify spending about as much on vehicle DDT&E as the annual revenue stream during the operational phase. In other words, you might justify a $3-billion project including the construction of a whole fleet of vehicles, since you have to perform 500+ flights per year to meet NASA's conditions. In comparison, I note today's high performance rocket engines such as the Russian NK-33 and RD-0120 have an estimated life of 20-30 flights as most. Most RLV planners apparently assume the basic vehicle air(space?)frame would be good for 100-200 flights like the Shuttle and X-15 -- no more. NASA keeps saying we need better technology to build a reusable launch vehicle. I disagree, unless you want flight rates in the range of hundreds or thousands per year... What you've just advocated is much more ambitious than what even Gary Hudson had in mind with ROTON.

Of course, this doesn't mean somebody could not eventually build more economical long life rocket engines, cryogenic tankage, thermal protection systems etc.. But if you *start* by demanding an operational life of thousands of flights, I don't think any investor will want to give it a try.

> But a successful bid would be in the form of an
> operational system, not a cost-plus contract.

That's even worse, then:-) I think private investors would be quite willing to accept a Soyuz derivative if NASA promises it will buy 2 billion dollar's worth of crew seats per year, even if the operator is expected to pay the full development cost of the system.

> the most likely candidate is the high-rate
> bizjet class.

Yeah...the Len Cormier approach. I love Len, but his refusal to consider any first generation(!)vehicle design that does not fly fewer than 100 missions per year or cost more than $500/lb probably explains the lack of investor interest...

MARCU$

Posted by Marcus Lindroos at September 22, 2003 09:38 AM

I think private investors would be quite willing to accept a Soyuz derivative if NASA promises it will buy 2 billion dollar's worth of crew seats per year, even if the operator is expected to pay the full development cost of the system.

They may be, but it wouldn't be worth doing.

Posted by Rand Simberg at September 22, 2003 09:51 AM

>> I think private investors would be quite
>> willing to accept a Soyuz derivative if NASA
>> promises it will buy 2 billion dollar's worth
>> of crew seats per year, even if the operator
>> is expected to pay the full development cost
>> of the system.

> They may be, but it wouldn't be worth doing.

To make your plan work, I think you need to somehow increase the size of the "carrot". E.g. by promising to buy $20 billion worth of flights per year as long as the vehicle (or vehicles) is good for 500+ missions a year. Needless to say, the government could never justify such expenses -- but is there some other incentive that might work? Maybe tax breaks or loan guarantees.

MARCU$

Posted by Marcus Lindroos at September 22, 2003 10:27 AM

I don't agree that that large a market is necessary. Try it small first. If no one steps up to the plate, then increase it.

I'm glad to see that you don't disagree with the basic concept though--we seem to be simply arguing about the numbers.

It's like the old joke about the guy in the bar, who asks a woman if she'd sleep with him for a million dollars. When she says she supposes so, he says, "...how about fifty bucks?" She's outraged, and asks him just what he thinks she is. "We've already established that--now we're just haggling over the price."

Posted by Rand Simberg at September 22, 2003 10:51 AM

> I don't agree that that large a market is
> necessary. Try it small first. If no one steps
> up to the plate, then increase it.

I think $20 million per seat (=what Dennis Tito reportedly had to pay) might be a possible starting point -- particularly if the expected winning entry is a derivative of existing manned capsule technology. That's $500M per year, for a permanent Space Station crew of six astronauts. Even if the tickets were priced at $80 million, the American taxpayer would save some money if the Shuttle could be retired and as long as private industry foots the DDT&E bill.
---
Unfortunately, it seems you have to make a development vs. operations cost/difficulty tradeoff. You minimize the former at the expense of the latter, and vice versa. If you want hundreds of orbital flights per year, you have to invest billions in new engine & materials R&D whereas as simple capsule will only support maybe half a dozen very expensive flights but DDT&E costs are much less. Given NASA's relatively modest needs, I think the choice would be a virtual no-brainer. The United Space Alliance partners may well conclude they could pay for a manned capsule, but there would be no good reason for developing a much more expensive and riskier fully reusable spaceplane unless NASA (or somebody else) somehow could guarantee an annual market worth tens of billions of dollars! I note Boeing and Airbus don't build airliners unless they have a firm commitment from the customer and they are absolutely sure they can build the aircraft.

> I'm glad to see that you don't disagree with
> the basic concept though--we seem to be simply
> arguing about the numbers.

Yes we are. I also disagree with your apparent conclusion that a cheaper ELV based ballistic capsule would not be useful. After all, Tito & Shuttleworth would never have been able to visit ISS without commercialized off-the-shelf Soviet technology. You have to start somewhere, and I think a combined ISS habitation module / "space hotel" paid for by Robert Bigelow or somebody actually could be immensely useful. Even if only three or four billionaires can afford to visit the facility. It would provide some interesting data as well as useful P.R..

MARCU$

Posted by Marcus Lindroos at September 22, 2003 11:32 AM

If you want hundreds of orbital flights per year, you have to invest billions in new engine & materials R&D whereas as simple capsule will only support maybe half a dozen very expensive flights but DDT&E costs are much less.

That's an opinion, Marcus, but not a fact. Once the suborbital folks get going, we'll see how many "billions" we have to spend on engines and materials. I suspect that the anwer will be less than one.

Posted by Rand Simberg at September 22, 2003 01:34 PM

Post a comment