Transterrestrial Musings

Comments

Is anyone putting serious money into low cost RLV research, today? Orbital not suborbital.

NASA is not. How much is Kistler spending on R&D these days?

Musk is not doing RLVs, right? SpaceX seeks a better cheaper expendable, IIRC.

Bigelow is offering $50 million but is that enough money to get anything accomplished?

Posted by Bill White at March 8, 2005 10:27 AM

Zubrin makes a good case for heavy lift here:

http://www.space.com/spacenews/pdf/zubrin.pdf

In short, if we assume that ELVs will continue to have a 2% failure rate, any lunar exploration plan that requires four or more launches for each mission to the lunar surface (and many rendezvous) results in a number of mission failures and the grounding of the program for a significant period of time.

Posted by Bill Mulger at March 8, 2005 11:03 AM

I also found the article to be very frustrating. If there is a consensus on the issue, then I would have preferred to at least seen some of the discussion that was used to form this consensus.

At a minimum, we should follow the lessons learned in the past. We need assured access to space for any program, which roughly doubles the cost of any new development. Once we get over the sticker shock, let's see how ambitious we really are for new development.

Posted by Dan Schmelzer at March 8, 2005 11:36 AM

SpaceX has designed their first stages to be recovered and reused. Subsequent stages are expendable.

If you want (relatively) cheap space lift, you can spend a lot of money developing a reusable system and hope the refurbishing costs are low enough to help keep the total costs low as you amortize the R&D costs over many launches. It didn't work out that way with the Shuttle, but that doesn't mean no one can develop an inexpensive reusable system. It just means that no one has done it yet.

On the other hand, you can go the "Big Dumb Booster" approach with simple systems designed for easy manufacture, then build a bunch of them. Also, don't let the engineers mess with the design constantly. The Russian Soyuz booster is a great example of this (and the "dumb" reference doesn't mean stupid, it means simple). Since it was developed in the 1960s as an outgrowth of the booster used to launch Sputnik, they've launched close to 2000 of them (counting the Molynia variant with an additional upper stage). I've heard the per launch cost is under $20 million, it can carry a substancial payload, has a high reliability rate, and is even man-rated. Simple, reliable, cheap (about $1000 per pound to LEO). What more could anyone want?

Now, if SpaceX succeeds with their Falcon I and V, then it could stand the launch industry on its ear. I hope it happens.

Posted by Larry J at March 8, 2005 12:25 PM

I am with Larry J on this:

". . . Now, if SpaceX succeeds with their Falcon I and V, then it could stand the launch industry on its ear. I hope it happens. . . . "

But can Musk beat the Russians on price?

Posted by Bill White at March 8, 2005 12:37 PM

Heh! I just flashed a Janis Joplin riff:

"Oh Lord won't ya' buy me a Mercedes RLV, our rivals fly Soyuz; we're filled with envy."

China, of course, flies Son of Soyuz.

Posted by Bill White at March 8, 2005 12:41 PM

> Future space based sensors will require very large apertures, and the
> ultra-lightweight systems and materials which many experts hoped to see
> developed by then will just not be there. The power requirements of a
> future military laser communications satellite might also be much larger
> than expected.

Why do people assume that very large rockets are automatically easier to develop than lightweight systems, orbital assembly, etc.?

And if that hypothetical future laser communications satellite turns out to weigh more than expected, does that really justify the development of an entirely new ELV? Should we develop a new ELV every time a satellite doesn't fit on its launcher?

If a mobile home weighs too much for existing trucks, does the manufacturer spend billions to develop a bigger truck -- or split the load into two modules?

Posted by Edward Wright at March 8, 2005 01:24 PM

I have to agree with Rand on this one. Taylor's
argument falls flat as soon as you realize that
about 75-80% of the mass of any lunar transport
system is propellant. Propellant doesn't have to
be shipped in one lump. It's a liquid for crying
out loud (well at least if you're sane/care about
costs)! Why do you need a big HLV when you can
just launch the transport on one MLV, the cargo
on another, and then tank it up with two or three
more? Heck, propellant delivery is likely going
to be one of the big high-flight-rate markets for
RLVs in the future.

As for orbital not being suborbital, I know that
a lot of people like to pat themselves on the back
for being so smart, but they're wrong. I won't
go into why here (as its part of what I'm doing
for my job), but it should be relatively obvious
to anyone who actually runs the numbers.

~Jon

Posted by Jonathan Goff at March 8, 2005 01:26 PM

>> It is a well known feature of launch vehicle economics that larger boosters
>> are more economic than smaller boosters, with costs/kg scaling roughly as the
>> inverse square root of the total payload.

> This is such an oversimplified rule of thumb as to render it useless for
> this analysis. It assumes similar designs, simply scaled up.

I'm not sure it's correct even then. Aircraft fabrication costs scale with dry weight to a power of roughly 0.8 (from memory, I don't have references in front of me). I would expect ELVs to be fairly similar because the techniques are fairly similar. I wonder where that 0.5 figure comes from and if it's based on a large enough sample to account for design differences.

> Two rockets of radically different design, one large, one small, wouldn't
> be expected to follow this rule.

It also assumes both rockets are produced in similar quantities -- which won't be the case. That means you have to account for learning curve effects.

IBM mainframes appeared to be more cost-effective than microcomputers, until you took into account the number of machines than might be build.

Posted by Edward Wright at March 8, 2005 01:51 PM

> Is anyone putting serious money into low cost RLV research, today? Orbital not suborbital.

Um, yeah, and the beach at Kitty Hawk is not the Atlantic, so clearly the airplane is of no value for intercontinental travel. And microcomputers are useless for anything but games because they can't even run COBOL.

Ocean liners, mainframes, and ELVs forever?

> "Oh Lord won't ya' buy me a Mercedes RLV, our rivals fly Soyuz; we're filled with envy."

Except that the "Mercedes" would, in fact, be cheaper than the Soyuz, making your analogy completely meaningless.

The reasons for building RLVs have nothing to do with your rivalry with China. (Not "our" rivalry -- international rivalry is your gig, Bill. Please don't impute your motives to all of us.)

> China, of course, flies Son of Soyuz.

No, Bill, China does not fly on the Son of Soyuz. Only one Chinese flew on it.

If you define human spaceflight as flying only one human, there might be merit to your approach.

It's funny that, while you keep saying the US needs to copy China, you still choose to live in the US. Why not move to China and enjoy the fruits of that system right now? Assuming, of course, those fruits are as bountiful as you imply.

Posted by Edward Wright at March 8, 2005 02:13 PM

China only flew one man on their first flight, but they're planning on carrying two in the same vehicle this year. If I recall correctly, their space capsule can carry up to three people. As for their booster, it was home grown over a period of many years. It really doesn't have any commonality with Soyuz.

A fully reusable space booster would be terrific, if it could compete with expendables for total cost. For example, suppose it took $10 billion to complete the R&D for such a booster (much less than what the Shuttle cost), that each one ended up costing a billion dollars to build (much less than the Shuttle cost), and each was capable of flying 100 missions. Further, suppose they built 5 of the boosters, the per mission marginal cost was $20 million (1/10th of the Shuttle cost), and the booster could put 20,000 pounds into LEO. (OK, the Shuttle can launch about 50,000 pounds).

R&D = $10 billion
5 boosters = 5 billion
Total = $15 billion
Fleet flight capability = 500 missions
Total cost of 500 missions = $25 billion

500 missions * 20,000 lbs per mission = 1 million pounds to LEO

Cost per pound = $2,500

Suppose the payload per mission was 50,000 pounds. The cost per pound would still be $1000, or about the cost of the Soyuz.

Of course, you can change the assumptions by saying the payload would be higher, that the per mission cost would be lower, etc. However, higher lift capability means more expensive R&D costs. If the booster is operated by NASA, the personnel costs per mission would probably be a lot higher (marginal Shuttle mission costs are reportedly over $200 million). Fully reusable systems tend to have lower payload fractions than expendables. In short, reusability isn't an assured path to lower launch costs.

Now, if you really needed heavy lift, it seems the most cost effective method would be to revive the Energyia. The R&D cost would be minimal, the engines are still in production, so you'd need to build new central tanks (we already know how to do that), launch facilities, updated software and avionics, etc. In the end, you'd be able to put something like 200,000 pounds into orbit for less cost than reviving the Saturn V or creating a HLV from Shuttle components.

Posted by Larry J at March 8, 2005 02:38 PM

All I ask for is that the market be allowed to sort these types of questions out. In a healthy space launch market, eventually someone will go heavy just because it's an unexplored avenue/market. Then if it flops, so be it.

Sadly, we're still in the very early stages of such a market, if it emerges at all.

Posted by Jon Acheson at March 8, 2005 03:55 PM

Assuming for a minute, you are stuck with EELV or shuttle I posit the following:

If you must go ELV or Shuttle derived, the Delta IV with the 6 SRB's and propellant xfeed that gets you 100,000lbs seems to be the cheapest option. You could proabally push that to 65 tons with propellant densification and an RL-60 upper stage. IIRC, The RL-60 and the SRB's already exist and are fully or nearly fully developed.

The Delta could be incrementally improved as need developed.

Would not a multi-core Delta very heavy still contribute to economy of scale as this is still utilizing the same production line and the same launch facilities and the same engineering and support personel as the smaller Delta IV variants?

Isn't the goal here to reduce the cost per core?

Does it make that much of a difference if the cores fly as part of clusters last stand v2.0 or fly as individual Delta IV mediums?

Taylor seems to infer a Shuttle Derived EELV would be the cheaper option but I feel for the above reasons, the Delta IV (or possibly a similar Atlas V variant) would be cheaper as many costs would be shared amongst the entire launch vehicle family as opposed to the SDHLV which would be a limited niche vehicle.

Posted by Mike Puckett at March 8, 2005 04:05 PM

Jon Goff writes:

". . .Heck, propellant delivery is likely going
to be one of the big high-flight-rate markets for
RLVs in the future.. . ."

I agree 100% with this assesment of future demand for Earth to LEO capability. Slogging fuel will be a significant source of demand.

But I also believe it will be far less expensive to loft tanks of methane or LH2 in a flimsy disposable with a very high fuel to dry weight ratio.

Put your crew in a Cadillac RLV (nix the Mercedes) with fancy heat shields and redundant avionics and landing gear, etc. . . .

Fair enough.

But for tanks of methane or LH2 give me mass produced engines for the first and second stage with the fuel needed to attain LEO contained in tanks made from composites and other lightweight utterly disposable materials and seek to launch with over 90-95% of non-payload liftoff weight being fuel.

Use gold plated re-fillable pens for the crew and Bic stick disposables for the fuel.

Posted by Bill White at March 8, 2005 04:16 PM

> China only flew one man on their first flight, but they're planning on carrying
> two in the same vehicle this year. If I recall correctly, their space capsule can
> carry up to three people.

Three people, maybe three or four times a year?

Yawn. Wake me when something interesting comes on. :-)

> A fully reusable space booster would be terrific, if it could compete
> with expendables for total cost.

Why would anyone want to compete with expendables? Microcomputers didn't compete with mainframes, they developed new markets, and mainframes at least made money. Expendables don't.

> suppose it took $10 billion to complete the R&D for such a booster

Why should we suppose that?

> and each was capable of flying 100 missions.

Why would anyone design a vehicle for such a trivial number of flights?

> Further, suppose they built 5 of the boosters, the per mission marginal cost
> was $20 million (1/10th of the Shuttle cost),

Again, why should we suppose this?

A good transportation system distributes costs roughly evenly between hardware (capital costs), fuel (propellant), and labor. Rocket propellant is cheap. $20 million worth of labor is 200 man-years (at $100,000 per man-year, including overhead) or 10,000 man-weeks. That's a lot of labor. What are they all going to be doing? If you fly once a week, you can't have 10,000 people doing touch labor on the vehicle between flights. Even if you wanted to, it's physically impossible. They'd get in each other's way.

> Of course, you can change the assumptions by saying the payload would be higher,
> that the per mission cost would be lower, etc. However, higher lift capability
> means more expensive R&D costs.

The DC-3 hauled much more cargo than the Spruce Goose. It did not have higher R&D costs. You err in assuming that the number of payloads/missions equals the number of flighs. It doesn't. When the US Air Force had the mission to resupply Berlin, they didn't do it with one giant airplane.

> Fully reusable systems tend to have lower payload fractions than
> expendables.

So? The proper figure of merit is cost, not payload fraction.

> If the booster is operated by NASA, the personnel costs per mission
> would probably be a lot higher

Why should it be operated by NASA?

> In short, reusability isn't an assured path to lower
> launch costs.

Not if you make poor design assumptions. One of your assumptions is that you *don't* have good reusability -- you throw the hardware away after only 100 flights.

> Now, if you really needed heavy lift, it seems the most cost effective method
> would be to revive the Energyia. The R&D cost would be minimal

Based on what? The engineers who built it say it would cost $10 billion. That's hardly minimal.

If you needed a propeller cargo plane, would it be more cost-effective to revive the Spruce Goose, or to use some DC-3s that are readily available? R&D costs *count*.

> In the end, you'd be able to put something like 200,000 pounds into orbit
> for less cost than reviving the Saturn V or creating a HLV from Shuttle
> components.

A broken foot is less expensive than a broken leg. That's not a good reason to want a broken foot.

Posted by Edward Wright at March 8, 2005 04:34 PM

> But I also believe it will be far less expensive to loft tanks of methane
> or LH2 in a flimsy disposable with a very high fuel to dry weight ratio.

Belief does not count for much in the world of numbers.

Optimizing for fuel ratio makes little sense when fuel is the cheapest part of your operating costs.

> Put your crew in a Cadillac RLV (nix the Mercedes) with fancy heat
> shields and redundant avionics and landing gear, etc. . . .

So, Federal Express should ship all their packages in expendable capsules? No point in building airplanes with redundant avionics and landing gear, etc., right? Piloted airplanes have to be expensive because they're "fancy"?

Posted by at March 8, 2005 04:52 PM

I have serious issues with several pieces of Taylor Dinerman's defense of heavy lift. His most ridiculous argument is probably this:

A heavy-lift launch system could be used to replace large numbers of military satellites that had been lost to a “Space Pearl Harbor” type attack. The fact that it could do so in a single launch, rather than after a long drawn out series of launch campaigns, should make this option attractive to US strategic planners trying to cope with future worst-case scenarios.

Hell no. If you want survivability, you do not put all your eggs in a single huge basket that is easier to target and you certainly do not use a rocket that requires a fixed launch pad.

Preferably you use mobile launch platforms and launch vehicles. Not to mention hardened, storable hardware and cold launch facilities. Curiously all the traits of... an ICBM.

You would use submarines or road-mobile launch platforms. Such as the system SpaceX is interested in. Their only fault is that you cannot store a missile or do a cold launch. They use LOX/Kerosene.

Rand still seems enamourated with RLVs. Personally I could care less. But yes, if you really want to use an RLV, it is easier to use a small vehicle. Because retrieving it is easier. If you use a landing strip, you need one big enough. if you can land on the rough ground, you still need to go back to base for refit, using a chopper is useless if it weights too much, the best alternative is a descent onto the sea, where you can relatively easily be towed back into port by a boat. But there is still the issue of how much damage you get on the landing, etc. Hence why some insisted on using wings, so they could land on airstrips, etc. Of course a vehicle which lands vertically has no such issues.

Personally, I think ELVs are perfectly fine. I also know that what makes a machine expensive is the cost of labour, for design, production and maintenance. The more parts something has, the more expensive that thing is. By their very nature an RLV has more parts. You need higher-performance that is usually got from complexity, not to mention that you need extra recovery systems not present in an ELV.

Reusing the hardware saves on production costs and some say design costs, because you can do non-destructive vehicle testing. You do however get extra maintenance costs, an ELV needs zero maintenance, maintenance costs which in the case of Shuttle, proved prohibitive. No, having to disassemble a complex engine part by part to inspect and fix before each launch or 10 launches is not a good idea. Neither is it to manually inspect each tile from several thousands in the orbiter.

Posted by Gojira at March 8, 2005 07:42 PM

"And microcomputers are useless for anything but games because they can't even run COBOL."

Not to nitpick (and this does not take away from your argument) but you can get a version of 'COBOL' for your micro.

Posted by Brian Dunbar at March 8, 2005 08:03 PM

An honest-to-God RLV would be waaay cool. I am not against honest-to-God RLVs.

So who will build one first, and when? ;-)

If fuel is your lowest cost then to minimize your dry weight would save the most money. Why pay to carry landing gear and life support (and the weight of an astronaut!) into orbit when fuel tanks and water bladders only need a one way trip?

Why man-rate your cargo carrier at all?

Posted by Bill White at March 8, 2005 09:11 PM

Bill,

> I agree 100% with this assesment of future
> demand for Earth to LEO capability. Slogging
> fuel will be a significant source of demand.

Only if it's affordable. But yeah, if you have
any sort of cislunar economy going on, you're
actually going to need more propellant flights
than passenger flights. If lunar adventure
travel happens, propellant flights for that may
be almost as big of a launch market as passenger
flights to LEO travel destinations.

> But I also believe it will be far less expensive
> to loft tanks of methane or LH2 in a flimsy
> disposable with a very high fuel to dry weight
> ratio.

Why? I've looked at it from both sides, and I
really don't see any benefit to launching fuel
on BDBs as compared to launching it on an RLV
with high flight rates.

I guess if you're stuck in the flying vehicles
only half a dozen times per year mode of thinking,
sure a BDB would be cheaper than an RLV. But
why even bother making an RLV if you don't intend
to fly the thing at least 50-100 times a year?

> Put your crew in a Cadillac RLV (nix the
> Mercedes) with fancy heat shields and redundant
> avionics and landing gear, etc. . . .

But if you have a nice RLV that has low operations
costs, why not also ship fuel and cargo as well as
flying passengers? Sure, with propellant delivery
you can afford to lose a flight every once in a
while, but it isn't an obvious win for either RLVs
or BDBs--both could likely deliver fuel for the
same price in $/lb in orbit. But if you have a
good people deliverer anyway, making a variant
(or a module) that caries propellants or cargo
won't be too expensive, and it just allows you
to amortize your vehicle over more flights (as
well as justifying having a larger fleet of
vehicles flying).

> But for tanks of methane or LH2 give me mass
> produced engines for the first and second stage
> with the fuel needed to attain LEO contained in
> tanks made from composites and other lightweight
> utterly disposable materials and seek to launch
> with over 90-95% of non-payload liftoff weight
> being fuel.

Why? Cost is the key item--total lifecycle cost
for the whole system. It isn't obvious that doing
it your way is superior. It isn't obviously
inferior either, but I don't see why you think its
so great.

~Jon

Posted by at March 8, 2005 10:10 PM

Unless you can design and build a rocket as simple as a Bic (pen, razor, lighter, whatever), disposable rockets cannot make sense economically.

Posted by David Masten at March 8, 2005 10:16 PM

Ugh, I didn't much care for the HLV article either. I didn't feel like going to the effort to refute the assumptions and reasoning througout the piece, but I will take a wack at this one bit...

"A heavy-lift launch system could be used to replace large numbers of military satellites that had been lost to a “Space Pearl Harbor” type attack."

What really stands out here is that an HLV replacement mission would place all the replacement satellites in virtually identical orbits! Yet the very value of a constellation of military communication and recon satellites lies from the fact they are in very different orbits. Different orbits means separate launches for each satellite.

The U.S. military are not fools. They realize the importance of our space assets which is why their project FALCON phase I is aimed at the task of rapid replacement of wartime loses. For this job they don't want some monster HLV, they want a small booster to launch a 1,000 pound satellite into polar orbit as fast and cheaply as possible.

Posted by Brad at March 8, 2005 10:20 PM

Bill,

> An honest-to-God RLV would be waaay cool. I am
> not against honest-to-God RLVs.
>
> So who will build one first, and when? ;-)

I wouldn't be too surprised to see the first
operational commercial suborbital RLV (SS1 was
mostly a flight test vehicle that was also used
to win a prize--I'm referring to vehicles flying
regular space tourism or research flights) within
the next year or two, and I wouldn't even be
surprised to see the first orbital RLV before the
end of the decade. Lots of stuff could go wrong,
but don't make a career out of making snarky
cynical comments--life has a way of making such
people look rather unprophetic at times.

> If fuel is your lowest cost then to minimize
> your dry weight would save the most money.

Well, your highest costs are actually development
cost, operations costs (such as payroll), launch
site infrastructure costs, etc. Hardware costs
are often a distance next place. From some
estimates I've done in the past, I wouldn't be
surprised if the actual hardware cost of a SpaceX
Falcon I was less than $1.5M. Trying to minimize
dry mass usually drive cost up (due to higher
design costs and sometimes higher operational
costs).

> Why pay to carry landing gear and life support
> (and the weight of an astronaut!) into orbit
> when fuel tanks and water bladders only need a
> one way trip?

Because it may very well be cheaper? You must
remember, that you don't just have to get to
orbit--your fuel tanker has to reach some
destination there, and transfer its fuel or at
least the tank to something else to be useful.
That implies a lot of throw-away hardware that
has to be wasted every flight if you don't
reuse it. What's wrong with having a pilot
on board?

You're learning all the wrong lessons from the
Space Shuttle. The Space Shuttle's mistake
wasn't that it had crew and cargon on the same
vehicle. It's that it tried to be space station
with wings that also doubled as a Medium/Heavy
Lift launch vehicle. A real cargo truck that
only had minimal life support capabilities (ie
like an airplane) would likely actually end up
being cheaper to operate than an unpiloted
vehicle.

> Why man-rate your cargo carrier at all?

Man-rating is a stupid concept. Why do you want
to design a vehicle that will be flying valuable
payloads that isn't reliable? Having a reliable
vehicle is critical for keeping RLV costs low.
It also happens to be a benefit of truly well
thought out RLV development paths (hint Shuttle
doesn't fall in that category).

~Jon

Posted by Jonathan Goff at March 8, 2005 10:21 PM

RLV, ELV, costs and SpaceX

The SpaceX Falcon rockets are intended for reuse, at least the first stage is. So although not completly reusuable the Falcon rockets should not get lumped into the expendable category.

On the other hand the launch costs quoted for the Falcon rockets assume no successful reuse at all. Elon Musk has said that if the reusability works out prices could be even lower. So how might the Falcon compare to Soyuz?

The late model Falcon V, with a P&W RL-10 engined second stage, is supposed to place 20,000 pounds into LEO for a price of $20 million. So with no reuse at all that's 1,000 dollars per pound, just as cheap as the Russian Soyuz.

Posted by Brad at March 8, 2005 10:32 PM

propellent to orbit dirt cheap

Most of the mass moved to orbit for missions beyond LEO will be rocket engine propellent or fuel. Heck even the LEO bound International Space Station burns up 7 tons of fuel per year just for station keeping. It makes sense to split the mission of cargo to orbit into three different jobs: crew, equipment and propellent. So I have a favorite candidate (and one that sadly is not very likely to succeed ) for the job of delivering propellent to LEO.

The Jules Verne Gun company wants (wanted?) to use a cannon to launch 5 ton rockets that would deliver 3.7 ton payloads to LEO. The claimed cost was only 250 dollars per pound! And propellent being insensate materiel would be the ideal payload to withstand the high G launch forces of the orbital cannon.

Posted by Brad at March 8, 2005 11:02 PM

> An honest-to-God RLV would be waaay cool. I am not against honest-to-God RLVs.

> So who will build one first, and when? ;-)

No one could tell you which company would invent the iPod and when, yet iPods were invented. You use products every day that no central planner commanded to come into existance.

The lack of a central planner who can tell you who will do something and when is no indication that it won't happen.

Central planning is not the solution; it's the problem.

> If fuel is your lowest cost then to minimize your dry weight would save the most money. Why pay to carry
> landing gear and life support (and the weight of an astronaut!) into orbit when fuel tanks and water bladders
> only need a one way trip?

Because if you leave off the landing gear, it costs too much to repair your craft after every crash. Did you miss the part about labor costs?

> Why man-rate your cargo carrier at all?

Two reasons.

The first is economics. Unpiloted vehicles cost more to develop and are orders of magnitude less reliable.

The second is more basic: If a vehicle does not allow *us* to go into space, who cares?

The anti-humans-in-space crowd misses the point completely.

Posted by Edward Wright at March 8, 2005 11:21 PM

An honest-to-God RLV requires reliable low (no) maintenance heat shields that can be re-used dozens? hundreds? thousands? of times in order to achieve the flight rates needed to be economical.

Hopefully, we will develop such heat shields.

To my knowledge no such heat shield has ever actually flown. Not even once.

Yes, I covet RLVs - - but to paraphrase Don Rumsfeld, "Let's find a way to make money using the rockets we have. . ."

IF we are stuck with expendables (even with refurbishing which is a good thing) then larger will be cheaper per pound than smaller.

SpaceX is awesome and deserves our total support yet we must also remember that Musk will in essence be "catching up" with the Russians. Boeing abandoned the commercial satellite business for a simple reason. They lost the pricing wars to Moscow and Kiev.

Posted by Bill White at March 9, 2005 05:29 AM

Clark Lindsey has further thoughts:

...there is no better way to kill the VSE program than to start it off with a costly expendable rocket program. . .

= = =

Like Delta IV? ;-)

One of Rand's original points stands unchallenged and is agreed with by me.

. . . NASA and its major contractors continue to pay lip service to involving smaller companies and smaller launchers, but clearly the mainstream industry is firmly back into Apollo mode. And if they don't get out of it soon, they won't even be as successful as that program was . . .

I agree with this 100%

And, if using SDV will cause the VSE to fail, so will using Delta IV.

Atlas? Those guys use Russian engines, right? If we are free to buy Russian, well, let's put Zenit and Proton on the table as well for the civilian aspects of the VSE. Build an Angora plant in Alabama?

Posted by Bill White at March 9, 2005 07:50 AM

Like Delta IV?

No. Sorry.

Clark and I are both referring (well, at least I am) to a new heavy-lift vehicle, much larger than either Delta IV or Atlas V.

That's what I mean by the term "heavy-lift" being so nebulous.

Posted by Rand Simberg at March 9, 2005 08:03 AM

Edward Wright said:
Why would anyone want to compete with expendables? Microcomputers didn't compete with mainframes, they developed new markets, and mainframes at least made money. Expendables don't.

Indeed. Yet even today mainframes are still used. PCs did not kill off mainframes, their market just grew bigger. Mainframes still run your bank account.
When IBM made the S/360 family, it bet the company on it. They made a huge investment and they literally took decades to pay it off.

Not if you make poor design assumptions. One of your assumptions is that you *don't* have good reusability -- you throw the hardware away after only 100 flights.

You wish the Shuttle was as reusable as that! The engines need to be overhauled every 11 missions, eventually getting dumped.

The DC-3 hauled much more cargo than the Spruce Goose. It did not have higher R&D costs. You err in assuming that the number of payloads/missions equals the number of flighs. It doesn't. When the US Air Force had the mission to resupply Berlin, they didn't do it with one giant airplane.

The Spruce Goose used flimsy materials and was not rugged enough to be properly reusable. Some could say the Shuttle has the same problem.

Because if you leave off the landing gear, it costs too much to repair your craft after every crash. Did you miss the part about labor costs?

Helicopters do not have a landing gear. They have skids. Yet they manage to be reusable... So do boats.
Please try thinking outside the airplane box. A space launch vehicle is *not* an airplane. For an airplane, the atmosphere is useful because you get lift and oxidizer from it. For a space launch vehicle the atmosphere is a nuisance to be avoided. A good reusable space launch vehicle will look nothing like an airplane. In fact it may be different from all the vehicles we have now. The same way a hovercraft is quite different from a regular boat.
The KISS principle inspires on me the following thinking for a reusable: I want to go up. I want a low mass fraction so I can launch more payload, or even reach orbit. Spherical tanks are more mass eficient, not to mention more robust. Vertically built vehicles can better tolerate the loads on launch. Ideally the vehicle should land with the same orientation as the one used on launch in order to be less complicated (as do aircraft and helictopters). So the answer is... VTVL and small pads using skids or landing legs like a lunar lander.

David Masten said:Unless you can design and build a rocket as simple as a Bic (pen, razor, lighter, whatever), disposable rockets cannot make sense economically.

Read about Big Dumb Boosters and Truax's Sea Dragon, Excalibur, etc.

I agree with Jonathan Goff. Man-rating is stupid. You need high-reliability for all payloads currently launched. Your customer is not interested in losing his expensive satellite.

Posted by Gojira at March 9, 2005 08:09 AM

I am saying. ..

If the VSE relies primarily on Delta IV it will fail, since Delta IV is "a costly expendable rocket" to use Clark Lindsay's phrase.

What can Delta IV do that Proton or Zenit cannot, for 1/3 the cost? Atlas might work better but has Russian engines, so why not just buy Russian and save even more money.

And if the CEV specifications come in too large for a Falcon V but "just right" for Delta IV where does that leave Musk?

Posted by Bill White at March 9, 2005 08:16 AM

If the VSE relies primarily on Delta IV it will fail, since Delta IV is "a costly expendable rocket" to use Clark Lindsay's phrase.

We know you're saying that, Bill. Obviously you'll continue to say that, because you'll continue to miss the point, which is that while expendables in general are a bad idea, whether home-grown or Russian, a new heavy-lift expendable is a killer.

...if the CEV specifications come in too large for a Falcon V but "just right" for Delta IV where does that leave Musk.

With a huge market for propellant delivery, and if he wants to build a passenger vehicle, a large space tourism market. CEV become irrelevant, or used only for trips to the moon, not to LEO.

Posted by Rand Simberg at March 9, 2005 08:28 AM

I think we are missing the main point here which is that the lifter is the main cost piece and the main design piece of the VSE. For NASA to be a customer, it should have its bidders articulate its lift options and it should have them liquidate damages for risky options like a new lifter. That is, if a vendor wants to bid using a new HLV, it should develop the HLV with private money and only be paid on delivery. If a vendor wants to use existing, it should bid using new modular inflatable connectable components or whatever and only be paid when the components snap together. Private insurers can insure fails if the aerospace primes don't want to take on that risk. NASA becomes a customer and gets to keep its money in the event of program failure. It should expect to pay more in event of success for the privilege, but the bidders' financiers will be much better at handicapping the relative success rates to underwrite or invest in the program than NASA has been.

Posted by Sam Dinkin at March 9, 2005 09:01 AM

Bill,

> An honest-to-God RLV requires reliable low (no)
> maintenance heat shields that can be re-used
> dozens? hundreds? thousands? of times in order
> to achieve the flight rates needed to be
> economical.

At least a hundred times. The engines, tanks,
avionics, RCS, and other systems also need to
be able to fly at least a couple dozen times
without much more maintenance than is done between
airplane flights.

> Hopefully, we will develop such heat shields.
>
> To my knowledge no such heat shield has ever
> actually flown. Not even once.

You need to be careful Bill. Just because a lot
of things have been done one way in the past does
not mean that that is the only way to do it. One
of the biggest TPS issues with the shuttle and
with all the old capsules is that all of those
designs were relatively high ballistic coefficient
designs--ie they had a high mass-to-frontal-area
ratio. Lower stage TPS is no sweat--the reentry
environment could be kept to not much worse than
suborbital reentry conditions. For the orbital
stage, you have essentially a fairly "fluffy"
vehicle that is mostly empty tanks when you go
for reentry. Making good reusable TPS for such
a vehicle will be much easier than the TPS for
the shuttle.

But yeah, some practical engineering research
will probably be needed. So what? There are
ways to do that very inexpensively and fast.

> Yes, I covet RLVs - - but to paraphrase Don
> Rumsfeld, "Let's find a way to make money using
> the rockets we have. . ."

Oh, I agree that we could make money with the
"rockets we have" (or at least with the rockets
we'll be soon to have--Falcon I and Falcon V).
I'm just saying that making a good RLV to go
along with that may also make a lot of sense.

> SpaceX is awesome and deserves our total support
> yet we must also remember that Musk will in
> essence be "catching up" with the Russians.
> Boeing abandoned the commercial satellite
> business for a simple reason. They lost the
> pricing wars to Moscow and Kiev.

Well, he's trying to offer a domestic booster,
with higher reliability and similar cost. Why
is he playing catch-up? The only Russian boosters
with prices below what he's asking for Falcon V
flights are ones with atrocious reliability
ratings and lousy launch environments.

If SpaceX does succesfully launch their rocket,
I think they'll end up doing *very* well.
Particularly with America's Space Prize now
starting up. They may not have the winning
spacecraft, but they will probably be launching
it (regardless of who build it). A contract for
what was it....two dozen flights? would be a very
good deal for them even if someone else was
getting money for running the operation.

~Jon

Posted by Jonathan Goff at March 9, 2005 09:31 AM

> An honest-to-God RLV requires reliable low (no) maintenance heat shields that can be re-used dozens?
> hundreds? thousands? of times in order to achieve the flight rates needed to be economical.

SpaceShip One had one. It isn't that hard for suborbital vehicles. Orbital vehicles are a bit harder, but fortunately, orbital is not the only market.

> Yes, I covet RLVs - - but to paraphrase Don Rumsfeld, "Let's find a way to make money
> using the rockets we have. . ."

"We"??? Are you a stockholder in Boeing and Lockheed, or are you having pronoun trouble again.

Neither Boeing nor Lockheed are able to make money on Delta and Atlas -- that's why the new national space transportation policy calls for indefinite subsidies. The Chi Comms, despite your admiration, are not making money on their Soyuz clone, either.

Why is it better to try to make money on something you have, that can't make money, than to build something new that can make money?

And by what definition is a Soyuz clone something you already "have"?

> IF we are stuck with expendables (even with refurbishing which is a good thing) then larger will be
> cheaper per pound than smaller.

You haven't proven that -- and if "we" are stuck with ELVs, the obvious solution is to get unstuck.

> SpaceX is awesome and deserves our total support yet we must also remember that Musk will in
> essence be "catching up" with the Russians. Boeing abandoned the commercial satellite business for
> a simple reason. They lost the pricing wars to Moscow and Kiev.

What does that have to do with your desire for the US to build Soyuz clones like the Chinese?

SpaceX isn't interested in communication satellites, except as an interim step. The name of the company is Space *Exploration* Technologies.

Posted by Edward Wright at March 9, 2005 10:21 AM

> When IBM made the S/360 family, it bet the company on it. They made a huge investment
> and they literally took decades to pay it off.

While microcomputers took far less investment and far less time to pay off. Investors care about ROI.

> You wish the Shuttle was as reusable as that! The engines need to be overhauled every 11 missions,
> eventually getting dumped.

Who said anything about the Shuttle?

> Helicopters do not have a landing gear. They have skids. Yet they manage to be reusable...

Skids are landing gear. Get your terminology straight.

> So do boats.

Non sequitar. Boats don't have to fly or land. Spacecraft do.

> For an airplane, the atmosphere is useful because you get lift and oxidizer from it. For a space launch vehicle
> the atmosphere is a nuisance to be avoided.

There's no reason why a spacecraft can't get lift from the atmosphere. Even Soyuz gets lift, except when they screw up the reentry sequence.

> Read about Big Dumb Boosters and Truax's Sea Dragon, Excalibur, etc.

Of course, Truax contradicted himself later in life when he tried to build *little* dumb boosters.

Posted by Edward Wright at March 9, 2005 10:36 AM

Ed,

> SpaceShip One had one.

Actually, my understanding was that the TPS for
SS1 required some work after each flight. It
was a spray on ablative, no?

> It isn't that hard for suborbital vehicles.
> Orbital vehicles are a bit harder, but
> fortunately, orbital is not the only market.

To be fair Ed, we were talking specifically
about orbital RLVs, not suborbital ones. We were
talking about whether orbital RLVs would be
available at some point in the future, or if we
needed to use expendables for orbital launches.

> SpaceX isn't interested in communication
> satellites, except as an interim step. The
> name of the company is Space *Exploration*
> Technologies.

Yup, their goal is to open up private manned
spaceflight. They're being pragmatic about it,
and trying to go after currently existing markets
while the market for private manned spaceflight
ramps up, but that has always been their stated
goal.

As it is, even with a small orbital RLV in use,
I think there would still be a market for people
like SpaceX launching larger pieces on their
Falcon vehicles.

~Jon

Posted by Jonathan Goff at March 9, 2005 11:08 AM

Okay, I'm hardly and expert and I'm sure I'll be picked apart but...

Why not strap together a dozen Solid-Rocket boosters, add something to guide them into orbit, and launch bulk materials like water or kerosine (that could be used as fuel and won't boil off for some time).

The solid-rocket boosters are not as environmentally clean, but I'm not talking about an ongoing program here. Yeah the solid-rocket boosters aren't really man ready either but even if they blow up I'm talking about payloads far less explosive than the shuttles hydrogen.

Seems to me having a large fuel depot in orbit would simplify a lot of design issues. Reentry can now use fuel instead of worrying about heat for example. Any vehicle designed to go to the lunar surface doesn't need to pack its own fuel, it just needs to be able to transfer fuel in orbit. That sort of thing.

Posted by rjschwarz at March 9, 2005 11:41 AM

Interesting (and long) thread - but please note that rockets will not always be the only way to attain orbit. Tethers, space elevators are not practical now but god willing and cross your fingers they will be soon.

Posted by Brian Dunbar at March 9, 2005 11:44 AM

While I don't agree with the fact that using currently available boosters is in any way cost effective and that scaling them up would only be worse, I do not see why such boosters could not be used in their current form for VSE, assuming a good launch escape system is available. The Delta and Atlas EELV's are severly under-utilized and are operating about 20% of their theoretical maximum flight rates.

If these vehicles flew at their maximum rates with 15 Delta IV Heavy's and 12 Atlas V 552's (baseline) per year, and 20 of those were devoted to VSE, then there would be somewhat less than 900,000 lbs. of payload per year placed in orbit for VSE. This is enough for two sets of Mars Direct style hardware. The marginal costs for these 20 vehicles is about 2.5 billion dollars, which is still less than the space shuttle's 3.4 billion that typically only puts 200,000 pounds (and often zero pounds) per year into orbit.

While there are cheaper rockets on the drawing boards (pick a company) and even on the pads (SpaceX, and Chinese and Russian vehicles), the EELV's are already in production, require no new infrastructure, can be launched from American soil (to appease congress), and provide a large enough payload to make modular construction of spacecraft manageable. They also circumvent the necessity of relying on either the Space Shuttle or Soyuz.

While this is not an optimal solution (I am a big proponent of small, private RLV's and also see some benefit to Saturn V class vehicles in terms of long range space exploration), I see this as being *a* straightforward way of jumpstarting the program. When the RLV's or bigger HLV's come along, which they surely will, modify the space based hardware and use the new rockets. But, if government controlled space exploration is your style, then use what you have and get going as soon as possible, because programs are subject to cancellation at any time.

Posted by Donovan Chipman at March 9, 2005 02:41 PM

I did some back of envelope calculations last night.

A Bigelow Nautilus Module has slightly more volume than Skylab (including the docking module on Skylab).

Thats one module people, one that will weigh in at 45 to 50 thousand pounds.

Not all of the hardware of the VSE will weigh anywhere near what its Apollo counterparts did.

Just one Nautilus module + a docking and station keeping module would make a very nice EML-1 station for a pittance of the cost of the ISS.

Posted by Mike Puckett at March 9, 2005 02:48 PM

Mike,

I think the 40-50klb weight they keep mentioning
for Nautilus includes the docking and power systems.
I could be wrong though.

But yeah, you could make a decent L1 facility using
a Nautalus for say ~$500M even using existing (or
near-term) boosters like Proton and Falcon V.

~Jon

Posted by Jonathan Goff at March 9, 2005 05:13 PM

> To be fair Ed, we were talking specifically about orbital RLVs, not suborbital ones.

Sometimes one needs to change the conversation. The idea that we must use ELVs forever because RLVs have not yet reached orbit, the Moon, or Alpha Centauri is seriously flawed (as is the idea that we need to wait for orbital elevators, scramjets, warp drive, or whatever star treknology is popular de jure).

Posted by Edward Wright at March 9, 2005 06:05 PM

That last unnamed message was by me. Forgot to add my nickname. Sorry about that.

Posted by Gojira at March 9, 2005 06:51 PM

Edward Wright,
While microcomputers took far less investment and far less time to pay off. Investors care about ROI.

Most investors are sheep baahing and hoping they are not being led to slaughter. I have some quotes for you from someone who knew better:

I know of nothing more despicable and pathetic than a man who devotes all the hours of the waking day to the making of money for money's sake.
John D. Rockefeller

If you want to succeed you should strike out on new paths, rather than travel the worn paths of accepted success.
John D. Rockefeller

If your only goal is to become rich, you will never achieve it.
John D. Rockefeller

IBM is still around and their mainframe business is profitable. Compaq, one of the companies which started the PC clone business, is dead and IBM just sold off their own PC division to the Chinese.

Skids are landing gear. Get your terminology straight.

Ah, got me there. :-)

Non sequitar. Boats don't have to fly or land. Spacecraft do.

It is spelled 'sequitur'. Some 'boats' do have to 'land' in a sense. Like a hovercraft.

There's no reason why a spacecraft can't get lift from the atmosphere. Even Soyuz gets lift, except when they screw up the reentry sequence.

On ascent it means you loiter around, wasting more fuel in the process. Descent is almost irrelevant for a Moon launch system. Most of the time you want to send things on a one-way trip to the Moon. Fuel, cargo, food, whatever. Precious little needs to come back. Only people and small items. Launch costs are prohibitive to make mining worthwhile, not to mention that the Moon has not that different a composition from our own Earth. Helium-3 is a pie in the sky vaporware dream.
It is inneficient to make a non-essential aspect the main focus.

Of course, Truax contradicted himself later in life when he tried to build *little* dumb boosters.

As proof of concept. A large power plant is more efficient to run than a small one, does not mean that it is not useful to make a cheap one off scaled prototype to prove it works.

PS: Ignore my last message, it seems it didn't get posted in the first place. Doh.

Posted by Gojira at March 9, 2005 06:52 PM

>> There's no reason why a spacecraft can't get lift from the atmosphere.

> On ascent it means you loiter around, wasting more fuel in the process.

Not for an any known definition of "loiter."

> Descent is almost irrelevant for a Moon launch system.

If you want more than another Apollo program, which spends tens of billions to do almost nothing, perhaps. If you want to do anything useful, realistic two-way transportation is required.

> It is inneficient to make a non-essential aspect the main focus.

In the real world, what matters is cost not theoretical "efficiency."

Posted by at March 9, 2005 07:37 PM

Post a comment