Advancing Launch Technology
John McKernan
>In article <1991Apr1.1...@eplrx7.uucp> lei...@eplrx7.uucp (Walt Leipold) writes:
>>For
>>the sake of discussion, let's assume a modest 500g acceleration in a purely
>>horizontal launcher....
>We can actually get much more than this right now in the lab, but
>perhaps this is a good number for a system with low materials
>costs.
I don't think a 500g or greater acceleration EM launching device would replace
chemically powered launchers. It might be useful for launching bulk materials,
but it can't launch large complex systems like space stations, the Earth
observation system, Gallileo, human beings, etc. So even if we had an
EM launcher, we would still need chemically powered launchers.
It does seem possible to replace chemical launchers with a laser powered
launch system, but not for at least 15 or 20 years. The technology to
generate and maintain a beam of sufficient energy for the required length of time, and then focus and aim it simply doesn't exist. There
has been a lot of research on lasers, but there is still a LONG way to
go before we get to the level of technology we need.
Since we're stuck with chemically powered launchers for at least the
next 15 or 20 years, it makes a lot of sense to try and reduce their
cost by an order of magnitude or so. And Nick, you can repeat yourself
until you're blue in the face, but it's just a fact that with such a
small number of complete chemical launcher design iterations, we just
DON'T KNOW what the lowest possible cost of that technology is.
John L. McKernan. jm...@sun.com
Disclaimer: These are my opinions but, shockingly enough, not necessarily Sun's
-------------------------------------------------------------------------------
"It's kind of a macho thing,
programmers are always trying to be weirder than their machines."
Nick Szabo
>What *are* the absolute minimum costs of a a *practical* EML system? ....
[a reasonable analysis]
>For
>the sake of discussion, let's assume a modest 500g acceleration in a purely
>horizontal launcher....
We can actually get much more than this right now in the lab, but
perhaps this is a good number for a system with low materials
costs.
>Energy: [$1/lb.]
Reasonable for EML. Depending on the fuel used, a light gas gun might
lower this by a factor of 5, since electricity is a fairly inefficient form
of energy.
>Hardware:
>
>How much does a linear foot of high-capacity, reliable, maintainable,
>weatherproof railgun cost? Ten thousand dollars? If so, your 26-km
>railgun will cost $850 million dollars.
To avoid wear, a coilgun is probably a better option than railgun.
The above estimate is also pretty good for a first generation
coilgun; Sandia projects $1,0000 million for the whole first-generation
project. At least two possible technical advances, if they pan out, could
greatly reduce this cost for subsequent projects:
* Building practical systems at >500g
* Mass-produced warm superconductors
Given these, we could go down to 10 km and $1,000/foot, or $32 million.
But currently, $850 million is pretty good estimate for the first-generation
system.
>Real estate:
>...I'll assume that this piece of land will cost about $100 million
The best thing to do might be to bridge the gun across several used
oil tankers or platforms, hundreds of miles away from inhabited areas, to
avoid noise pollution. This could also cost up to $100 million.
>Management:
>
>We'll assume a revolution in management styles gives us an administration
>cost of only $4 million/month....
This doesn't take any management revolution, since many smaller airports,
handling much larger amounts of cargo, operate at an overhead of $4
million per _year_. However, $4 million/month sounds reasonable for
the first generation project.
>Liability insurance is a big unknown....
Not necessarily. Insurance costs are based on reliability, which
in guns is typically 2-3 orders of magnitude greater than for rockets.
>[total cost of] $9/lbm doesn't sound like a lot, but remember that this
>kind of rough estimate was used to predict $100/lbm for the Shuttle a
>couple of decades ago....
Well, if you want to doom _any_ attempt to lower launch costs based
on one NASA chem rocket project, there's not much to say about that....
But there are many, many differences between EML and yet another chemical
launcher like the Shuttle:
* Even your projection, which contains several pessimistic assumptions,
is 10 times lower than any chemical rocket launch projection, and is
500 times lower than realistic rocket launch cost projections and
actual rocket launch practice.
* A totally different technology: fuel is not stored on board.
Shuttle is, when all is said and done, another chemical rocket,
which follows the curve of costs for internally chemical powered
vehicles.
* The first EML is to be designed be Sandia, not NASA, and subsequent
EML's should be designed, owned and operated privately. Therefore, we
will avoid aerospace/chem rocket world concepts of launch cost and
operation.
The choice is between large ($4-5 billion) chemical rocket
projects, which are proven incapable of significantly lowering launch
costs, or smaller ($100-$500 million) R&D/prototyping projects to
demonstrate a whole suite of new technologies (gas gun, EML, laser
launch, etc.). If EML passes this phase, we would go to a $1 billion
first-generation system, which would dramatically and quickly lower launch
costs. We can introduce a greatly superior technology for less than
the cost of building yet another chemical rocket. This is a whole new
ballgame.
--
Nick Szabo sz...@sequent.com
"If you want oil, drill lots of wells" -- J. Paul Getty
The above opinions are my own and not related to those of any
organization I may be affiliated with.
S Schaper
**************************************************************************
Zeitgeist Busters!
UUCP: {crash tcnet}!orbit!pnet51!schaper
INET: sch...@pnet51.orb.mn.org
Aslan is on the move!
****************************************************************************
Nick Szabo
>...it's just a fact that with such a
>small number of complete chemical launcher design iterations, we just
>DON'T KNOW what the lowest possible cost of that technology is.
You repeat this as if it was an argument in your favor. The fact that
chemical rockets have such long design cycles is one of their problems.
If we want to wait around until 2500 to colonize space (no exageration
-- this is based on the curve of chemical rocket launch costs since 1960),
and assume no other new technology between now and then, what you say might
be reasonable. But then you go on to talk about 10-20 years for newer
technology to come on-line. The long design cycle means we will not reduce
chem rockets an order of magnitude, or anywhere near that, in the 10-20 years
it may take to develop laser launch (much less the 5 years Sandia says it
needs for EML -- already less than half NASA's typical chem rocket design
cycle, for the first implementation of the technology!). Indeed, the
government chem rocket proposals we have seen in this newsgroup where
numbers have been given (ALS, HLLV) do not reduce costs at all, when
examined with realistic operation and market assumptions.
When NASA was contemplating the Shuttle in the late 60's, RAND did a
study that showed NASA would be better off sticking with Saturn V for
manned, heavy lift ability. They reasoned that money going into
Shuttle R&D would reduce the money available to potential Shuttle customers.
The Shuttle would dry up demand for itself, Shuttle would fly far less than
the proposed 55 flights per year, and the R&D costs of Shuttle would
not amortize to a lower cost than Saturn V. All the while, NASA was
declaring that Shuttle would drop prices by your order of magnitude.
A bold try, I must admit, going to reusability, but the bottom line is,
Shuttle still costs $8,000/lb., the same technology in a different
dress.
Private industry, on a smaller scale and with different paradigms, is
making progress on launch costs, in its own fashion (going after response
time and entry level launch costs instead of cost/lb., for example).
While private industry progresses, government is no longer able to
reduce chem rocket costs. Government launch R&D should indeed go
directly to EML, gas gun, tethers, and laser launch. That 10-20 years is
in large part a function of how much basic science and technology effort we
put into it. If we keep pouring most of the money into chem rockets, it
will probably be longer than 20 years; if we change our priorities now,
it could be less than 10. To sum up, there is room for incremental
and paradigm improvement of chemical rockets by private industry.
Government should advance technology or get out of the way.
Doug Mohney
>You repeat this as if it was an argument in your favor. The fact that
>chemical rockets have such long design cycles is one of their problems.
>If we want to wait around until 2500 to colonize space (no exageration
>-- this is based on the curve of chemical rocket launch costs since 1960),
>and assume no other new technology between now and then, what you say might
>be reasonable.
You are asserting this? 2500? Are you smoking something? GAG.
Signature envy: quality of some people to put 24+ lines in their .sigs
-- > SYS...@CADLAB.ENG.UMD.EDU < --
Nick Szabo
[I write -- launch cost curve doesn't give efficient space colonization
until beyond 2500]
>You are asserting this? 2500? Are you smoking something?
I am sorry it violates your wishes. Now sit down, get the data for
launch costs between 1957-1991, and fit the curve for yourself.
You will find no more than a two order of magnitude drop by 2500,
still well above the cost needed for affordable manned travel
to GEO, L-5 and beyond (3-4 orders of magnitude, depending whose
argument you use).
The way we break out of the curve is by working towards more advanced
launch and upper stage technology, not by wishful thinking and rehashing
of the same old ideas and technologies.
--
Nick Szabo sz...@sequent.com
"The biscuits and the syrup never come out even" -- Robert A. Heinlein
Allen W. Sherzer
>I am sorry it violates your wishes. Now sit down, get the data for
>launch costs between 1957-1991, and fit the curve for yourself.
Perhaps you could give us the figures you are using and the sources? Ever
since you incorrectly claimed that Titan was more expensive per pound than
Delta I have been wondering where you get your numbers.
Allen
--
+-----------------------------------------------------------------------------+
|Allen W. Sherzer | If you love something, let it go. If it doesn't come back |
| a...@iti.org | to you, hunt it down and kill it. |
+-----------------------------------------------------------------------------+
Doug Mohney
>
>[I write -- launch cost curve doesn't give efficient space colonization
>until beyond 2500]
>
>>You are asserting this? 2500? Are you smoking something?
Funny, soon after I posted my message, someone else also wondered if you had
recently participated in drug testing.
>I am sorry it violates your wishes. Now sit down, get the data for
>launch costs between 1957-1991, and fit the curve for yourself.
>You will find no more than a two order of magnitude drop by 2500,
>still well above the cost needed for affordable manned travel
>to GEO, L-5 and beyond (3-4 orders of magnitude, depending whose
>argument you use).
>
>The way we break out of the curve is by working towards more advanced
>launch and upper stage technology, not by wishful thinking and rehashing
>of the same old ideas and technologies.
Nick, if you are so blind as to use a concocted example to support your
arguements, I can't see why you just don't shoot everyone at NASA above the
level of technical manager and hope that they are randomly replaced with
brilliant people who are willing to espouse your ideas. I'm quite sure the data
is correct. However, it is laughable to try to project out the curve for 500
years (can I round off 1990 to 2000 for the sake of arguement?).
It is the same sort of GARBAGE which was used to by the Club of Rome clique in
the late '70s to say the world is going to hell in a handbasket in 20 years,
due to increases in population and limited resource allocation, with NO
ACCOUNTING WHATSOEVER for A) Incremental technological advances B) New
technological advances C) New Technologies applied to old problems and D)
Technological advances which we have no way of predicting, but which come
together due to circumstances.
You can smugly point at the "fitted curve" all you want. As others have pointed
out, there ARE efforts to look at laser launching, tethers, and other New And
Improved technologies aside from chemicals. They all have drawbacks. You want
to throw money at them to make them work NOW. TODAY.
In the Long View, our technological approaches of today are going to be
different from those in 2100. Fundmental advances in physics, chemistry,
computer science, materials, and biology (will we find a way to grow fuel? :-)
will make your "curve" irrelevant.
However, patience here is required. Evolution, not revolution, will get us
into orbit, and it is the things which we cannot see, or project, which will
provide the most radical revolutions in space exploration.
Gary Coffman
>
>[I write -- launch cost curve doesn't give efficient space colonization
>until beyond 2500]
>
>>You are asserting this? 2500? Are you smoking something?
>
>I am sorry it violates your wishes. Now sit down, get the data for
>launch costs between 1957-1991, and fit the curve for yourself.
>You will find no more than a two order of magnitude drop by 2500,
>still well above the cost needed for affordable manned travel
>to GEO, L-5 and beyond (3-4 orders of magnitude, depending whose
>argument you use).
>The way we break out of the curve is by working towards more advanced
>launch and upper stage technology, not by wishful thinking and rehashing
>of the same old ideas and technologies.
Trying to extrapolate from the 1957-1991 curve is like trying to
extrapolate automobile costs from the 1885 Daimler and Benz to
the 1901 Olds curve. It ignores the 1908 to 1914 period that followed
where Henry Ford took basically the same technology and brought costs
down from that of a rich man's toy to everyman's car. The T wasn't
high tech compared to the 1901 Olds, just big, dumb, reliable, cheap,
and made in huge quanities.
Gary
Fraering Philip
>...down from that of a rich man's toy to everyman's car. The T wasn't
>high tech compared to the 1901 Olds, just big, dumb, reliable, cheap,
>and made in huge quanities.
It was larger than earlier cars, but the fundamental advance
the model T made was in being mass produced. This would not have
happened if Henry Ford had tried to achieve economy of scale by
building cars large enough to hold several hundred people.
--
Phil Fraering
dlbr...@pc.usl.edu
''It's a Flash Gordon/E.E. Smith war, with superior Tnuctip technology
battling tools and weapons worked up on the spot by a billion Dr.
Zarkovs.`` - Larry Niven, describing the end to _Down in Flames_.
Gary Coffman
>In article <27...@ke4zv.UUCP> ga...@ke4zv.UUCP (Gary Coffman) writes:
>
>>...down from that of a rich man's toy to everyman's car. The T wasn't
>>high tech compared to the 1901 Olds, just big, dumb, reliable, cheap,
>>and made in huge quanities.
>
>It was larger than earlier cars, but the fundamental advance
>the model T made was in being mass produced. This would not have
>happened if Henry Ford had tried to achieve economy of scale by
>building cars large enough to hold several hundred people.
Actually the T was smaller than many of it's predecessors though it
could often carry more. What made it interesting was that the design
was simplified for mass production. This had several interesting side
effects. Such a simple vehicle was easy to maintain. Fewer parts meant
less things to go wrong. Making large quanities of a standard design
lowered the cost. The "any color as long as it's black" philosopy
succeeded not because black was necessarily a cheaper color, but
because the custom, one off, painting of special colors was avoided.
This concept would work just as well for small launchers as large
launchers, but I believe that the small launcher field can successfully
be developed by private firms designing to meet current commercial demand.
Heavy lift is going to need an initial government funded pump prime to
overcome the chicken and egg effect. I'm confident that once routine
relatively inexpensive heavy lift is available, payloads will be designed
to use the capacity. Note that I don't even believe that an HLV has to
be cheaper per pound to orbit than smaller launchers. I believe that
there will be payloads that will require heavy lift because it will
be cheaper to launch them assembled than to do assembly in space.
Gary
Fraering Philip
>Note that I don't even believe that an HLV has to
>be cheaper per pound to orbit than smaller launchers. I believe that
>there will be payloads that will require heavy lift because it will
>be cheaper to launch them assembled than to do assembly in space.
1. Assembly in space may be a lot cheaper and easier than you think.
Since the Russians can do it with their rather less developed automation,
shouldn't it be easier with the more developed technology here?
2. Currently it looks like it is the small booster which can take
business away from the larger one in spite of a cost per pound
imbalance in favor of the larger booster. My example, again, is
Pegasus...
--
Phil Fraering || Usenet (?):dlbr...@pc.usl.edu || YellNet: 318/365-5418
''It hardly mattered now; it was, in fact, a fine and enviable
madness, this delusion that all questions have answers, and nothing is
beyond the reach of a strong left arm.`` - Larry Niven and Jerry
Pournelle, _The Mote in God's Eye_
Allen W. Sherzer
>2. Currently it looks like it is the small booster which can take
>business away from the larger one in spite of a cost per pound
>imbalance in favor of the larger booster. My example, again, is
>Pegasus...
This is not a flame but I think you missed the boat here. No payloads
have gone from the larger ones to the smaller ones. It therefore
cannot be said that they are loosing buisness.
What IS happening is that we are seeing a new market develop. This is
not taking buisness away but rather adding more buisness. In other
words, making the pie bigger.
Allen
--
+---------------------------------------------------------------------------+
|Allen W. Sherzer | Allen's tactics are too tricky to deal with |
| a...@iti.org | -- Harel Barzilai |
+---------------------------------------------------------------------------+
Matthew DeLuca
>1. Assembly in space may be a lot cheaper and easier than you think.
>Since the Russians can do it with their rather less developed automation,
>shouldn't it be easier with the more developed technology here?
The Russians haven't actually done any assembly in space of the nature that
is being talked about. What they *have* done is dock lots of capsules
together, which isn't particularly more sophisticated than what we and they
have been doing for the last twenty-five years.
>2. Currently it looks like it is the small booster which can take
>business away from the larger one in spite of a cost per pound
>imbalance in favor of the larger booster. My example, again, is
>Pegasus...
Cost-per-pound is frequently a fallacy. The important number to look at
is how much it costs to get your payload into orbit. If I have a fifty
pound payload to put up, and I put it on a bare-bones $5 million rocket,
I am paying $100,000 per pound. If it's worth the $5 million I paid, though,
then I got a good deal.
In my opinion, this is why there is going to be a solid market for the
small booster; universities and private companies can afford the few
millions of dollars a small rocket must cost to do some basic research,
whereas they cannot begin to afford the fifty-million dollar costs of the
current rockets we are using.
--
Matthew DeLuca
Georgia Institute of Technology "I'd hire the Dorsai, if I knew their
Office of Information Technology P.O. box." - Zebadiah Carter,
Internet: cco...@prism.gatech.edu _The Number of the Beast_
Vincent Cate
>I'm of the opinion that small cheaper launchers will emerge without
>NASA's help. There is already strong commercial interest in Pegasus
>which is not particularly cheap in it's present incarnation. It does
>point the way however by using wings and air breathing engines for
>the initial stage of the journey to orbit.
Speaking of present incarnation etc.
Orbital Sciences is looking into the following:
1) Ramjets to boost after B-52 stage and before rockets fire.
Could more than double the payload or get rid of
first rocket stage.
2) Reusing the first stage.
They will recover the first stage after this coming launch
to study this option (should be within a month).
3) Mass-production to bring costs down.
My guess is they will do more of this when/if they get
all or part of the Motorola Iridium contract.
4) Liquid upper stage
High ISP really pays off in upper stages. Might want
to combine the current last 2 solid stages.
The combination of mass-production and an airbreathing reusable
"first" stage (after B-52 stage) would be fantastic. In any
case, the Pegasus will probably be the first mass-produced rocket,
sort of the Model-T of rockets.
-- Vince
Frank Crary
>Since the Russians can do it with their rather less developed automation,
>shouldn't it be easier with the more developed technology here?
>
assembly work has involved docking pre-fabricated modules together. They
have found this to be reasonably simple. They have also tested tools and
procedures to "assemble" things on orbit. These include unfolding
gird-work structures, welding, bolting things together, etc... They have
found this sort of work to be quite dificult, though not impossible.
On-orbit assembly of, for example, a Mars sample return, might easily
involve a significant amount of construction work, rather than just
docking two modules together.
Frank Crary
UC Berkeley
Gary Coffman
>
>>Note that I don't even believe that an HLV has to
>>be cheaper per pound to orbit than smaller launchers. I believe that
>>there will be payloads that will require heavy lift because it will
>>be cheaper to launch them assembled than to do assembly in space.
>
>1. Assembly in space may be a lot cheaper and easier than you think.
>Since the Russians can do it with their rather less developed automation,
>shouldn't it be easier with the more developed technology here?
For properly modularized payloads that can be simply docked together,
assembly in space is cheap. Given a manned assembly facility in orbit,
complex assembly of big systems *may* be cheap. But complex, big payloads
aren't cheap to assemble in orbit now. Therefore, an HLV, even if it's
costs per pound are higher, could be the cheapest way to get such payloads
delivered in the near future.
>2. Currently it looks like it is the small booster which can take
>business away from the larger one in spite of a cost per pound
>imbalance in favor of the larger booster. My example, again, is
>Pegasus...
Indeed, in cases where scheduling and launching flexibility are of
paramount concern, and the payloads are small, cost per pound is
not a real issue. Pegasus' market looks healthy *without* investing
in a program to make it substantially cheaper per pound of payload.
It's operational costs will probably decrease even more as OSC
progresses up the learning curve. Thus there is no pressing need for a
government funded program to lower the launch costs of small launchers.
For large launchers, there is little demand pull at present. Therefore
the pump priming of a government effort at supply push is needed to
get this system off the ground. Nobody is currently designing big
complex payloads, except Fred, because there is no reasonable way to get
them in space. If it becomes practical to do so, the payloads will likely
come.
Eventually there should be little need for HLVs as space materials
and space manufacturing allow complex structures to be made in space.
But to get from where we are to that utopian future, there must exist
intermediate steps. An HLV is such a step.
Gary