Subject: Space-tech Digest #154 Contents: P2 discussion, peroxides, pressurization methods (16 msgs) ------------------------------------------------------------ From: henry@zoo.toronto.edu Date: Mon, 21 Jun 93 11:56:25 EDT Subject: Re: P2 1/5 Introduction To: space-tech@cs.cmu.edu >This design proposal suggests that efforts like Ariane and even the >Japanese H-2 are geared at something else besides commercial space >launcher development! I must be crazy to think that goverment pork is >something restricted to the US. The scales drop from my eyes... Well, that's not the only reason why launchers like Ariane end up the way they do. The other side of the coin is that they are designed and built by people who are (historically) mostly in the missile business. The same cost-is-no-object performance-at-any-price philosophy carries over. Henry Spencer at U of Toronto Zoology henry@zoo.toronto.edu utzoo!henry ------------------------------ From: henry@zoo.toronto.edu Date: Mon, 21 Jun 93 12:56:33 EDT Cc: space-tech@cs.cmu.edu Subject: Re: P2 To: Bruce Dunn I like the idea of fluid injection for thrust vectoring. However... >... A further series of ports allow tangential injection of >peroxide into the gas flow for roll control. How does this work? Exactly what is the torque exerted against? My impression was that you couldn't get significant roll forces by fluid injection into an axisymmetric nozzle. >... exhaust does not contain solid particles, as does the >exhaust from a solid rocket. Overall, it should not be hard to find suitable >low cost ablative materials for chamber and nozzle construction. A possible complication is that most existing ablative designs use a refractory throat insert to keep the crucial throat geometry and area constant despite erosion of the ablator. The only large ablative nozzle I've ever seen without this was the Apollo LM descent engine, which ran at quite low pressure. Have you looked at this question? >ASRM retains field joints for assembling sections of the motor, but the >individual steel components of the sections are welded together (rather than >being bolted together as in the original SRB design). It's not quite correct to say that the original SRB factory joints are bolted; they're pinned, like the field joints. (In fact, the factory and field joints are identical except that the factory joints are mated before fuel casting.) In the SSTO comparison at the end, you suggest that the primary (cost) advantage of SSTO is reusability. There is actually another one: reliability assurance through testing rather than finger-crossing. :-) Being able to flight-test the exact vehicle that will carry customer cargos is a major asset. It seems to me that there are basically three approaches to ensuring adequate reliability: 1. Test flying. Requires reusable vehicle. 2. Precision manufacturing plus meticulous inspection. Very expensive and not as effective as commonly thought. 3. Iterative refinement of design and manufacturing process to minimize variation and sensitivity to variation. Takes patience. :-) You might want to address this issue. Henry Spencer at U of Toronto Zoology henry@zoo.toronto.edu utzoo!henry ------------------------------ Date: Mon, 21 Jun 93 11:44 PDT To: space-tech@cs.cmu.edu Subject: P2: Reply to Henry From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) > Henry writes: > I like the idea of fluid injection for thrust vectoring. However... > > >... A further series of ports allow tangential injection of > >peroxide into the gas flow for roll control. > > How does this work? Exactly what is the torque exerted against? My > impression was that you couldn't get significant roll forces by fluid > injection into an axisymmetric nozzle. I was thinking of having the roll control injectors on one face of small refractory internal fins running in the gas flow direction in the lower part of the nozzle. The expanding gas would react against the fin. If this doesn't work, the fallback position is catalytic peroxide trusters or very small bipropellant thrusters somewhere on the periphery of the vehicle. I would rather avoid the complications of these however. > >... exhaust does not contain solid particles, as does the > >exhaust from a solid rocket. Overall, it should not be hard to find > suitable > >low cost ablative materials for chamber and nozzle construction. > > A possible complication is that most existing ablative designs use a > refractory throat insert to keep the crucial throat geometry and area > constant despite erosion of the ablator. The only large ablative nozzle > I've ever seen without this was the Apollo LM descent engine, which ran > at quite low pressure. Have you looked at this question? I was thinking that ablation would not matter so much in engines with large throats. For a given amount of ablator burned off, the percentage change in the throat is less as the throat gets larger. The lower stage engine throat is simply huge - burning off several centimeters or whatever of ablator won't make much of a difference. For the upper stage engine, the throat will change more, and a refractory insert might be useful. > >ASRM retains field joints for assembling sections of the motor, but the > >individual steel components of the sections are welded together (rather > than > >being bolted together as in the original SRB design). > > It's not quite correct to say that the original SRB factory joints are > bolted; they're pinned, like the field joints. (In fact, the factory and > field joints are identical except that the factory joints are mated before > fuel casting.) Thanks for the clarification. I haven't been able to get much information about the factory joints. One reference stated that they were "bolted" so I used the term. > In the SSTO comparison at the end, you suggest that the primary (cost) > advantage of SSTO is reusability. There is actually another one: > reliability > assurance through testing rather than finger-crossing. :-) Being able to > flight-test the exact vehicle that will carry customer cargos is a major > asset. It seems to me that there are basically three approaches to > ensuring > adequate reliability: > > 1. Test flying. Requires reusable vehicle. > > 2. Precision manufacturing plus meticulous inspection. Very expensive and > not as effective as commonly thought. > > 3. Iterative refinement of design and manufacturing process to minimize > variation and sensitivity to variation. Takes patience. :-) > > You might want to address this issue. Points well taken. I think I would add a fourth and fifth method of ensuring reliability, which are different than what I think you are implying in number 3). 4. Extreme simplification of design, to minimize the number of parts requiring inspection, and minimize the number of failure points. 5. Use of off-the-shelf components and construction techniques with a long history of practical experience. As an example of 5, I mean using valves, fittings etc. from the petrochemical and chemical processing industries, rather than having the aerospace industry design new, light weight components. As an aside, the steel casings for the 6.7 meter diameter solid rockets tested for NASA in the 1960s were built in a shipyard, even though they were made of maraging steel. -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ To: Bruce Dunn Cc: space-tech@cs.cmu.edu, gwh@lurnix.COM Subject: Re: P2: Replies to Paul Dietz Date: Mon, 21 Jun 93 11:34:37 -0700 From: gwh@lurnix.COM > I agree that minimizing cost per unit payload is the primary design >criterion. For some design choices, I am not sure what would result in the >lowest cost. However, the use of lower strength steels looks highly >unattractive, at least for this two stage vehicle (see below for comments on >stage number). The D6aC steel proposed is not an exotic high alloy steel, >as was the 18Ni250 originally used in the P2 design. Even if the steel is >several times the cost of lower strength steels it would pay to use it, as >steel costs are likely to be only a smallish fraction of the total vehicle >costs. If the D6aC steel comes in at $10 per kg (several times what >100,000 psi steel costs according to George), the 38 metric tons of steel in >the P2 vehicle will cost $380,000. The roughly 800 tons of hydrogen >peroxide, at $2 per kg, has a cost of $1,600,000. Using steel at 100,000 psi >strength, as proposed by George, results in a vehicle over double the size >for a given payload. Even if the lower strength steel were given to the >designer for free, the extra propellant alone for the larger vehicle would >cost many times the savings realized. In my current vehicle, using Nitric Acid, the tank and fab costs are roughly equal to the propellant costs, and that seems to have optimized my design in the cost range. Bruce, remember that most of the tankage price will be in fabrication and testing, not just the raw steel... I can buy T-1 steel for under $0.50/lb in quantity ($1.10/kg), but welding it and testing the welds is going to drive tank and structural cost up to closer to $1.25/lb. And D6aC is going to take more exotic welding, forming, and testing methods than T-1 will, so it may have as bad a price multiplier. That having been said, you're right that it looks like for your vehicle the optimal solution is to use very high strength steel. >Mitchell Burnside Clapp thinks that a pressure fed >peroxide /JP-5 vehicle can get to orbit with a single stage. This however >assumes abandoning low cost construction in order to lower vehicle dry mass. Mitchell also appears to have used a composite form of Unobtanium in his design, remember 8-) ... >the vehicle, but also development and operations costs. I think that I am on >one side of the argument (using D6aC steel but only two stages), while George >Herbert is on the other side (conventional steel, but three (or more?) >stages). My current point design has five stages, believe it or not. That turned out to minimize vehicle mass... enough that the additional thrust vector control hardware wasn't as expensive as the savings in overall mass. -george william herbert Retro Aerospace KD6WUQ ------------------------------ Date: Mon, 21 Jun 93 18:53:34 -0400 From: dietz@cs.rochester.edu To: space-tech@cs.cmu.edu Subject: Re: P2: Reply to Henry Henry wrote: > A possible complication is that most existing ablative designs use a > refractory throat insert to keep the crucial throat geometry and area > constant despite erosion of the ablator. The only large ablative nozzle > I've ever seen without this was the Apollo LM descent engine, which ran > at quite low pressure. Have you looked at this question? The LM DE's nozzle throat diameter was about 1/9th that of Bruce's P2 first stage engine. The LM DE's chamber pressure was about 1/4 that of Bruce's engine. Since, all else being equal, heat transfer increases slightly sublinearly with pressure, Bruce may be ok (of course, not all else is equal.) Paul ------------------------------ Date: Mon, 21 Jun 93 23:01:47 -0400 From: dietz@cs.rochester.edu To: space-tech@cs.cmu.edu Subject: peroxide compatibility with plastics Bruce suggested coating the inside of the peroxide tank with a plastic, or perhaps with parafin. This brings up some interesting safety problems. The coating had better be rather pure. Monomers, plasticizers, or contaminants could dissolve in the peroxide, to be converted to organic peroxides or other unstable compounds, which could then precipitate or accumulate somehow and explode. I suspect epoxy or other such organic compounds formed by in-place reactions would be right out. Even teflon may not be completely safe in the presence of peroxide. The reaction (CF2)n + n H2O2 --> n CO2 + 2n HF is highly exothermic, liberating around 5 MJ per kilogram of reactants. (Telfon is pretty dense [2.15 g/cc] so maybe this would be a good hybrid propellant combination?) Paul ------------------------------ From: henry@zoo.toronto.edu Date: Tue, 22 Jun 93 00:05:03 EDT To: space-tech@cs.cmu.edu Subject: Re: peroxide compatibility with plastics >Even teflon may not be completely safe in the presence of peroxide. >The reaction > (CF2)n + n H2O2 --> n CO2 + 2n HF >is highly exothermic... This is fairly irrelevant in practice, though. The question is not whether the reaction would occur if it got the chance, but whether it will get the chance under normal conditions. The normal material for LOX tanks in rocketry is aluminum, and aluminum+LOX is *spectacularly* exothermic... but getting it started is sufficiently difficult that it's not a problem in practice. Assessing hazards requires examination of secondary issues like surface films and activation energies, not just the net energy release of a hypothetical reaction. Henry Spencer at U of Toronto Zoology henry@zoo.toronto.edu utzoo!henry ------------------------------ Date: Tue, 22 Jun 93 09:02:21 -0400 From: dietz@cs.rochester.edu To: henry@zoo.toronto.edu Subject: Re: peroxide compatibility with plastics Cc: space-tech@cs.cmu.edu Henry reminds us that a reaction being exothermic is not in and of itself sufficient reason for it to occur. This is of course correct. Still, it does give one pause. With peroxide, I imagine the following scenario. A flaw developes in the coating, which becomes delaminated and allows some peroxide to flow into contact with the metal wall. The peroxide begins to decompose, causing a buildup of hot fluids behind the plastic. At some point, the plastic there ignites. Bruce: how much do the tank walls stretch at full pressure? Paul ------------------------------ Date: Tue, 22 Jun 93 11:39:57 -0400 From: dietz@cs.rochester.edu To: space-tech@cs.cmu.edu Subject: passivation of peroxide tanks re plated coatings to passivate against peroxide decomposition... I read that tin does not catalyze peroxide decomposition (tin salts are used as peroxide "stabilizers", in fact). The application of tin coatings to steel is an old technology; it can be done by dipping or electroplating. Paul ------------------------------ Date: Tue, 22 Jun 93 08:50 PDT To: space-tech@cs.cmu.edu Subject: P2: Wall strain and Isp program From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) > Paul writes: > Bruce: how much do the tank walls stretch at full pressure? Simple question. Just look up Young's modulus for steel, and do a couple of simple divisions. What! The Handbook of Chemistry and Physics does not contain any tables of Young's modulus (except for rare earth elements, which are hardly of concern to structural engineers). I finally found a table in Gordon's "Structures" - steel is 210,000 MPa. Yield strength of the steel I am using is 1340 MPa, and I am using a 1.2 safety factor, giving a working stress of 1117 MPa. Strain (ie. stretch) is thus 1117/210,000, or 0.53%. I will make the Isp program from the US Air Force available to anyone who wants it. It is for MSDOS, and is not particularly user friendly, so I will write a small read-me file for it before sending it out. The file is too big to UUENCODE without a lot of hassle, so I will send copies by snail mail to anyone who gives me an address. I have checked with Mitchell Burnside Clapp - he says there is no problem in distributing the program, even though it originated in the US Air Force (after all, the basic methodology has been around for decades). From there, perhaps someone could put it on a site from which it could be FTP'd for those with that capability. Warning: the program requires a math co-processor, although I have gotten it to work with a shareware software emulator for a co-processor. -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ Date: Tue, 22 Jun 93 09:26 PDT To: space-tech@cs.cmu.edu Subject: Peroxide and Tin From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) > Paul writes: > I read that tin does not catalyze peroxide decomposition (tin > salts are used as peroxide "stabilizers", in fact). The application > of tin coatings to steel is an old technology; it can be done > by dipping or electroplating. > I love it. Maybe the name of the launcher should be changed to the "Tin can 2". Its three stage cousin would be the "Tin can 3". :-) -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ Date: Tue, 22 Jun 93 20:45:14 MET DST From: Magnus Redin To: Bruce Dunn Cc: space-tech@cs.cmu.edu Subject: Re: P2: Replies to Paul Dietz This is a reply to the suggestion to use peroxide injected on either sides of baffels in the booster nozzle to get roll control. (I was to quick to delete the original. ) I have no personal experience but it gives me an impression of being hard to test and difficult to estimate the roll force. You will have to run the booster to test the roll system and you cannot test how it performs in vacuum without an launch. Using catalysit decomposed peroxide in separate engines must be much easier to design and test. I have also a suggestion since the forces are low and you want to minimize the number of critical systems. Use only one set of roll control engines and mount it on the upper stage. You will waste some peroxide but the penalty is perhaps a ton (?) wich should be tolerable. -- Magnus Redin Lysator Academic Computer Society redin@lysator.liu.se Mail: Magnus redin, Rydsv{gen 240C26, 582 ------------------------------ Date: Tue, 22 Jun 93 13:37 PDT To: space-tech@cs.cmu.edu Subject: P2 Design Simplification From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) The currently posted design for the P2 vehicle has the liquid helium for pressurization purposes in a large spherical tank. The liquid is pressurized by hot helium gas, originating from a high pressure tank of helium gas, and heated by a heat exchanger in the gas generator. The gas generator currently burns peroxide and fuel from auxilary propellant tanks, and uses the heat to vaporize and warm the liquid helium. It has occurred to me that maybe this system can be simplified by using solid propellant gas generator cartridges. These can be produced to generate moderately hot gases (perhap 1500 K; details are in the Huzel and Huang book on rocket design, which I don't have handy right at the moment). The liquid helium tank would remain, but would be pressurized by hot gas from a solid propellant cartridge. Hot gas would flow from the cartridge, past a pop-off valve set to limit the pressure, and into the liquid helium tank. The hot gas would hit the surface of the helium, evaporating and warming it to give a tank pressurization gas consisting partly of helium, and partly of the cartridge products. The expelled liquid helium would then be heated in a gas generator which uses a second solid propellant cartridge to provide the heat. Making the proposed changes would eliminate three pressure vessels (the gaseous helium supply and the two auxiliary propellant tanks) and considerable plumbing and valving. The system for pressurizing the liquid helium would be lighter than the current system (no need for the gaseous helium tank), while the gas generator would be heavier (solid cartridge hot gas generator vs. liquid reactant generator). Overall, the system might have about the same mass, but would be simpler. One potential disadvantage is that once the cartridges are touched off, they probably can't be shut off. In the event of a launch pad abort after engine ignition and shutdown, the generated helium gas could be simply dumped as it is generated, rather than being applied to the propellant tanks. Comments? -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ Date: Tue, 22 Jun 93 22:45 PDT To: space-tech@cs.cmu.edu Subject: Solid propellant Gas Generators From: Nick_Janow@mindlink.bc.ca (Nick Janow) Bruce writes: > The hot gas would hit the surface of the helium, evaporating and warming it > to give a tank pressurization gas consisting partly of helium, and partly > of the cartridge products. Make sure the solid particles don't catalyze peroxide decomposition. :) -- Nick_Janow@mindlink.bc.ca ------------------------------ Date: Tue, 22 Jun 93 22:46 PDT To: space-tech@cs.cmu.edu Subject: Droppable Helium Tank From: Nick_Janow@mindlink.bc.ca (Nick Janow) Bruce writes: > An even bigger problem is where to put the large spherical helium tank so > that it can be dropped easily. I'm not sure of the relative size of the nozzle. Would it be possible to place a toroidal He tank at the perimeter of the nozzle (drop straight down)? Even fancier would be to have the tank form the lower section of the nozzle; this would take care of heating the helium. :) I think I remember hearing about a two-piece nozzle on an existing rocket. Of course, this does add to complexity and reduces the modularity of the system (can't test the gas generator separately), so I'm not sure if it would be worthwhile. -- Nick_Janow@mindlink.bc.ca ------------------------------ To: uunet!cs.rochester.edu!dietz@uunet.UU.NET Cc: gwh@lurnix.COM, space-tech@cs.cmu.edu Subject: Re: P2: Replies to Paul Dietz Date: Wed, 23 Jun 93 10:25:22 -0700 From: gwh@lurnix.COM Paul asked me in private mail: > Could you tell us more about what your design looks like now? >In particular, what have you decided about fuels? Last I remember, >you were looking at something involving frozen nitric acid that >scared me half to death. At this point I reccomend you stay scared 8-) To recap for those who may have forgotten, I'm examining in detail a fuel combination that's somewhere between "hybrid" and "solid"; my working description is "bipropellant solid". Basically, coiled up metal wire or mesh (most likely combinations of both, for structural and burn rate tradeoff reasons) is tightly packed inside what's essentially a solid rocket casing. Then you pour in inhibited nitric acid (98% HNO3 + 1+-% HF) and freeze it. The resulting "composite" fuel has relatively good handling characteristics, no moving parts, etc. In reality it`s a bit more complex than that because you want a hole down the middle of the resulting fuel grain, etc, but that's the basic story. The combination is designed to be essentially inert until its on the pad and the nitric acid is added, to reduce handling difficulties. In this way it's following in the footsteps of existing hybrids. Once it's fueled it is essentially a solid, which means that it's about as simple as you can get (except for thrust vectoring). Potential problems include runaway reactions between the nitric acid and the fuel. Since INA is supposed to be relatively good about producing flouride coatings on the contact metals instead of oxidizing them, and my preliminary reaction rate checks and literature checks support that, I'm guessing that's not a problem. I WILL have some info to report on testing this relatively soon, though the first test series has been pushed back to end of summer completion. The point design I'm working on uses Zinc as the fuel metal, because it's pretty cheap and very dense. Stoichiometric mixture densities of frozen HNO3 and solid zinc are about 3.1, which is about twice as dense as the best alternative propellant combinations. My preliminary calculations are that Isp will be somewhere between 260 and 295 at a 600 PSI chamber pressure (yes, this is a wonderfully tight envelope isn't it...). You get the low end if the zinc oxide coming out the nozzle ends up with too much of the energy. The point design assumed the worst and a reasonable drop from ideal to actual Isp, giving a design value of 250 (vac.). I have a 1.1 ton (2400 lb) vehicle on the drawing pad, with a launch mass of 150 tons, five stages, diameter of 2.3 meters and length of about 27 meters. It's sized to fit in standard sized cargo containers for easy transport. Preliminary pricing estimates are between $3.75 million and $4.5 million per flight, including paying off development, support costs, accountants, lawers, profits, and taxes. And it still uses T-1 steel 8-) Two upgraded versions will offer about 2.3 and 5.3 tons payload each, at about $6 million and $10 million respectively. The 5.3 ton vehicle can also put about 1.5 tons into GTO with a price of about $11 million. It's worthwhile to note that the prices per pound on these are $860/lb for the big one, $1185 for the midsized one, and $1650ish/lb for the small one. I've been chasing around pushing on omtimizing rockets for the last three years, and barring problems with this fuel mixture and concept I _think_ I've found a significant price per pound minimum with this configuration. In addition to the little vehicles which I can see affordably developing, I have preliminary plans for launchers in the 10 ton class, the 25 ton class, the 100 ton class, and I have what appears to be a workable 1200 ton launcher. I can't figure out what to DO with the 1200 ton one, (mars mission, maybe? 8-) but I've got some numbers and sketches. Maybe someone here can help figure out useful things to do with that 8-) -george william herbert Retro Aerospace ------------------------------ End of Space-tech Digest #154 *******************