Jerry Pournelle RT JERRYP [Chaos Master] Fri Jan 29, 1993 This was posted in the SPACE RT. Shortly thereafter I was told to stay out of the SPACE RT. My apologies if it seems the wrong tone: it was in answer to persistent questions about 'engineering data'. If anyone in SPACE RT wonders where I am, i have been given instructions to stay out of it on pain of being unable to access it. Fascinating. DATA AND NUMBERS Jerry E. Pournelle, Ph.D. Copyright 1993 by Jerry E. Pournelle Let's make one final attempt to answer a very persistant and what seems to me a very odd demand for engineering data. I have been informed that it was insulting to give the rocket equation and derivations as a reply to the demand for "numbers" and 'engineering rigor.' If so, I apologize: I didn't think I was being insulting, and indeed that little essay on staging is apparently going to be published in journals that don't consider it insulting. I thought I had furnished considerable information. I still think so, but apparently the moderator here disagrees. If what is demanded is some kind of operational data on single stage to orbit, then let me say at once, of course there isn't any. How could there be operations data on something that has never been built? That is a bit like demanding rigorous engineering data on a 707 before we will contemplate building a DC-3. It might make sense to have that, but you won't ever get it. We can give some analogies which seem meaningful. (1) Airlines typically operate at a small multiple of fuel costs. There is no obvious reason why this should not apply to space operations. Perhaps instead of 3 times fuel costs a 'spaceline' might have to operate at 9 time fuel costs (I generally use '5' in my analyses); but our present costs run in the order of 200 times fuel costs, and that makes no sense. Rockets are not less efficient than jet engines, and are much more efficient than piston engines. (2) We have never built a fully recoverable ship, so there isn't any way to generate engine flight test data: but we have static tested RL-10 engines, throttled them, test fired them, and in general had a lot of experience with them. The RL-10 was, as I said in that paper, designed by aircraft engine people, and that may provide a clue to its reliability. It also has a low chamber pressure, and that also contributes to reliability. Finally, as an expander engine, it doesn't fail in a spectacular explosion. All this argues that it will not be difficult to develop something like a 'super RL-10': something with chamber pressures under 1400 psi (avoiding critical pressure regimes); and having altitude compensation either by being mounted in an aerospike, or with bells retracted for lower altitude operations and extended at altitude. None of this seems to frighten engine design people. So what 'hard numbers' are wanted? A full engine design with all the specs? That's not beyond our competence to generate, but it is expensive, and probably foolish to lock things up that tight just now. The Lunar Society believes that a good 1400 PSI or so expander engine of the RL-10 persuasion is probably the most critical item for getting to space, and we are hoping to raise the money to do some development; and we will certainly have 'numbers' to show potential donors. Whether those 'numbers' would be much use to most of the readers here is another story. There is an amazing amount of data, and much more of the engineering work has been done than we thought. (3) Airlines typically operate with 120 or so employees per airplane, with about half of those selling tickets. The SR-71 program had about 48 employees per airplane. We see no reason why an SSTO should need more than 100. What 'hard data' do we have? For this ship, none. Since the ship hasn't been built, HOW COULD WE HAVE??? We do know Thor, a rather complex system, was designed for a crew of 8 commanded by a sergeant. We know that Delta Clipper looks to be operated by well under 100 people, with only 3 in the 'block house', and THAT WILL BE HARD DATA after we fly DC/X. But without DC/X we would have precious little source material. (4) Similarly with ground handling, maintainability, and the rest. We have none for the not yet constructed SSTO ship. How could we have? We have lots of data about airplanes and airlines. The Delta Clipper design was intended to use existing scissors platform loaders and stuff like that. Presumably any sane full size design will also: after all, airplane companies have been designing ships for maintainability and reliability for a long time. Why should it be that much harder for spacecraft? (5) The 2 stage vs SSTO: this is an argument about which is 'better' when we have not built either yet. It is my view that anyone who believes that 2 recoverable stages are simpler than one will believe anything, but suppose it to be true? So what? What we have shown is that an X vehicle SSTO can be made to fly, can be incrementally tested, and can probably get to orbit. It is argued that a clever 2 stage might be superior to a stupid single stage. Well, yes, of course, that's true, but it's trivial. What we have here is a demand for optimization: for prototypes rather than X vehicles; for final answers when we don't have enough data. Or so it appears to me. There's a name for that. * * * Now I have said all this for weeks, and the only answer seems to be repeated demands for "numbers" and "engineering rigor"; so many demands that I begin to wonder what the point is in constantly demanding operations data on non-existent systems. That is after all what X PROGRAMS are supposed to do: develop some operations data. So far we have Shuttle data on reusable vehicles: clearly that doesn't justify recoverable vehicles at all, much less single stage to orbit. So? Presumably we have learned SOMETHING since Shuttle was designed and built. Now it's time to take what we have learned and use that to build an X vehicle so we'll know more. Thus: if what is wanted here is an 'admission' that we have no operations data on SSTO ships, then of course I have no choice but to 'admit' it, and ask why anyone would expect us to? If what is wanted is the entire history of the RL-10, then I say it is easily summarized: it's reliable, rugged, and has been tested a lot. That leaves us little doubt that an improved model can be made. Beyond that -- what could we have? I have here many times said WE ARE BEYOND THE LIMITS OF ANALYSIS. The answer to that was a demand for an impartial and 'honest' design analysis, whatever that means, and a claim that an honest year of design work would somehow -- what? Not quite specified. I read it to be a demand for more flogging of the little data we have, and no acquisition of more. I put it do you that there isn't a lot left to say. On the one hand, I have always said: we need an X program to develop more data and operations experience. ON the other we hear continual demands to 'prove it', 'give us numbers', 'it's your job to convince us,' and the rest. Well, perhaps it is our job to convince you, and perhaps I have convinced some of you. It is also very clear that I will never convince others: that they will continue to demand operations data from non-existent programs before allowing those programs to proceed to a stage where they can develop operations data. There's a name for that. -30- ---------- D.HARTSOCK [Dana] Mon Feb 08, 1993 Wm. Gaubatz spoke at the Mid-Continent Space Development Conference. The program director for the Delta Clipper program said the DC-1 would be shipped from Huntington Beach to White Sands sometime in March. It will first be put on a NASA test stand for some static testing. In May plans are to move it to the "spaceport facility," with the first flight test probably in June. The liquid hydrogen tank in the DC-X is aluminum with balsa wood as an insulator I believe. Half the aeroshell is in place and hydraulics are currently being installed. For what it is worth the software for the DC-X is ahead of schedule. I don't know if it has been made clear previously, the DC-X will be flown remotely. Pete Conrad is scheduled to be at the remote piloting controls for the first test. Flight Envelope Expansion Program Test Suite 1: fly to a height of 600 ft. AGL, MaxQ of 10 psi Test Suite 2: fly to a height of 5,000 ft. AGL, MaxQ of 30 psi Test Suite 3: fly to a height of 20,000 ft. AGL, MaxQ of 20-95 psi the rotation manuever should be flown in this test suite During this flight testing the ground support operations are also being tested. I have heard talk now several times of a DC-X Prime. Apparently it would be an upgrade of the DC-X with suborbital capability? Does anyone know if this is an attempt to maintain some funding for the SSRT program without having to ask for the amount necessary to move on to the DC-Y? Dana ---------- T.TELENKO [Trent] Tue Feb 09, 1993 This past weekend, at the Space Frontier Foundation conference, it was disclosed that the Lockeed "Skunk Works" is putting together a team to create a SSTO vehicle. I have since found out that the Lockeed "Skunk Works" has briefed this effort to several congressmen as a part of an effort to redirect the "Starlifter" effort to a fully reuseable SSTO type. This vehicle is winged and based on the technology of the Aurora spy plane. ---------- M.HAYNES9 [Cowbubba] Tue Feb 09, 1993 The only fax number I've found is Secretary Aspin's (703-697-9080), so I sent him a note on DC-X. What I wonder is, where is Aviation Week on this? There was a good article a year ago, but nothing since, just Brazilian crop-dusters and such. This is a perfect cause for such an influential forum. ----------