Subject: Space-tech Digest #145 Contents: Pressurizing tanks for SSTO (10 msgs) ------------------------------------------------------------ Date: Thu, 11 Feb 93 16:34 PST To: Space-tech@cs.cmu.edu Subject: Tank blow-down in a pressure fed SSTO From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) In a previous posting, I agreed with a suggestion by Paul Dietz that for a pressure fed SSTO it would not be necessary to pressurize at full pressure for even as much as 50% of the ascent, as the later part of the flight needs little thrust due to the huge propellant burnoff. I indicated that a spreadsheet model suggested that full pressurization for 20% of the flight was sufficient. Paul asked me by E-mail whether I was assuming adiabatic expansion of the pressurization gas (gas expanding in thermal isolation, and cooling as it expands) or isothermal expansion (assuming the gas is in good contact with materials which will keep its temperature from dropping as it expands). I had, for simplicity, assumed isothermal expansion. I have now modeled adiabatic expansion, which may better represent the helium in a large tank. Assuming adiabatic expansion results in a severe sag in acceleration when full pressurization is only carried to the point that 20% of the propellant is burned - when half the propellant is burned, the acceleration of the Burnside Clapp vehicle would drop to 0.6 G. If full pressurization is maintained until 30% of the propellant is burned, then the minimum mid-flight acceleration is raised to 1.1 G. With full pressurization to 40% of propellant burnt, the minimum acceleration in mid-flight becomes 1.75 G. From this, it would look like the helium requirements are increased from the approximately 0.8 tons listed in a previous posting. The alternative to the extra helium is to try to add heat to the gas as it expands, to prevent it from cooling. Paul has some ideas on this - perhaps he will post them. Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ Date: Thu, 11 Feb 93 20:14 PST To: space-tech@cs.cmu.edu Subject: Methods for delivering high pressure helium From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) Pressure-fed vehicles need a source of high pressure gas. The lightest inert gas is helium. However, to pressurize a large tank with high pressure helium requires a source for the helium. In the P2-E concept, I used liquid helium from a non-pressurized tank, pumped to high pressure with a turbopump, and vaporized and warmed in a gas generator. This posting compares this scheme with three other schemes which don't involve a high powered turbopump - comments would be appreciated. Consider the problem of providing helium pressurization for a propellant tank which initially requires gas at 15 MPa (roughly 2000 psi, which is approximately the tank pressure of the Burnside Clapp hydrocarbon/peroxide SSTO, and 2 times the pressure of the Dunn P2-E expendible booster). Assume that a propellant tank requires 1 ton of ambient temperature helium inside the tank to provide the 15 MPa of pressure at the stage just before blowdown begins. Thus the system that provides helium must generate 1 ton of helium at 15 MPa and approximately 300 K. Below are the results of some rough calculations which explore the mass of different systems for generating the helium at the correct pressure and temperature. All calculations below use the experimentally determined density of helium gas at particular temperatures and pressures, as specified in tables from a refrigeration reference book. The ideal gas law is not obeyed very well at high pressures, and is wildly incorrect at predicting helium behaviour in some of the regimes used below. Method 1: Put the helium in a high strength tank at ambient temperature. This tank must have a pressure of several times 15 MPa, as it must through a regulator provide 1 ton of gas at this pressure. Note that this means that more than 1 ton of gas must be in the supply tank, as even after blowdown, there is residual gas in the tank itself. This "unusable" gas is determined by the ratio of the storage pressure to the 15 MPa tank pressure. To a first approximation, the higher the storage pressure, the smaller the unusable gas and the lighter the tank. At a storage pressure of 30 MPa, a system delivering 1 ton of helium at 15 MPa has 1 ton of unusable helium left in the storage tank when 1 ton has been delivered. Thus, the tank must be sized to hold 2 tons of helium. Such a tank, if spherical, has a mass of 15.08 tons if built of steel (1100 MPa working stress, 300 K), plus the 1 ton of unusable helium, for a total of 16.08 tons. Going to a pressure of 60 MPa results in lower waste helium, and a total tank plus waste helium mass of 11.53 tons. Higher pressure do not help very much, and tank walls get very thick. Aluminum tanks would weigh about the same, while Kevlar composite tanks would be somewhat lighter. In addition to the tank mass, a gas warming system must be provided to warm the helium as it adiabatically expands from the tanks - estimate 0.1 ton plus 0.1 ton of reactants, for a total of 11.7 tons. In E-mail to me, Paul Dietz suggests doing the warming by putting a small percentage of hydrogen and oxygen in the helium, and running the gas through a catalyst bed on its way to the propellant tank. Method 2: Store the helium as a high pressure gas at a reduced temperature. This necessitates the use of aluminum tanks for reasonable toughness at cryogenic temperatures. There is an immediate problem of how to generate a tank of cold high pressure gas (this is tricky, as pumping gas into a tank heats it, as anyone who has filled a SCUBA tank knows). By far the simplest way is to partially or completely fill a tank with liquid helium, then to seal the tank and warm it to evaporate the helium and bring it up to the desired temperature. This sets a limit on how much gas can be put into a tank - the density of liquid helium is 0.122. If a tank is filled with liquid helium (temperature approx. 4 K) and sealed and warmed, the helium rapidly turns into a cold gas. Since the tank is sealed, the density of the gas is, by definition, also 0.122. By looking up in tables the temperature and pressure combinations which correspond to a density of 0.122, one can follow what will happen to the pressure of the tank as it is warmed. If a tank of liquid helium for example is sealed and warmed to 100 K, the resulting pressure is 40 MPa. This temperature and pressure are actually near optimum for delivery of gas. Lower temperatures (and pressures) suffer from too much undeliverable waste gas, while higher temperature (and pressures) suffer from the higher mass of the required tanks. At 100 K and 40 MPa, the tank plus unusable gas for delivering 1 ton of helium at 15 MPa have a mass of 6.18 tons. To this must be added the mass of a gas generator system to heat the helium to the desired temperature, plus the reactants. As a guesstimate, assume a gas generator of 0.1 tons, plus 0.2 tons of reactants. Total mass for delivering 1 ton of ambient temperature helium is thus 6.18 + 0.1 + 0.2 = 6.48 tons. Method 3: Store the helium as a liquid in an insulated light weight tank, ***inside*** one of the main propellant tanks. Use a small low pressure pump to run liquid helium through a gas generator (total mass 0.3 tons with reactants) which generates ambient temperature helium for tank pressurization. This generates a "bootstrap" process in which the liquid helium is pressurized by its own product gas, requiring only a small pump to move the helium to the gas generator. To a first approximation, ignore the volume taken up by the helium tank itself, and assume that in order to supply 1 ton of helium, the tank needed only 1 ton of helium ( ie for simplicity assume no unusable helium). Upper stage experience has shown that the tankage for liquid hydrogen is typically about 15 % of the total hydrogen mass. For helium which is denser (0.122 vs. 0.07) assume that tankage is 10%. This means that it only requires 0.1 ton of tank to hold the required ton of helium. However, it has been necessary to increase the size of one of the main propellant tanks by the volume of liquid helium. At a density of 0.122 and assuming that main tank is steel, this requires adding approximately 1.3 tons of additional mass. Added mass would be lower for the Burnside Clapp design, due to the use of composite tanks. The mass requirement for providing 1 ton of helium is thus 0.1 ton (helium tank) plus 0.3 tons (gas generation) plus 1.3 tons (increase in main tank) plus a low pressure pump (assume 0.1 ton) for a total of 1.8 tons. Method 4: Store the helium in a light weight tank ***outside*** the main propellant tanks. Use a high pressure pump to run liquid helium through a gas generator (total mass 0.3 tons with reactants) which generates ambient temperature helium for tank pressurization. As noted for method 3, the tank mass for the helium is about 0.1 ton, while the turbopump and reactants for the turbopump might have a mass of something like 0.2 tons. Total mass to deliver 1 ton of helium is therefore 0.5 tons. In summary, to deliver 1 ton of helium at 300 K requires adding the following masses to a vehicle: 11.7 using ambient temperature helium in a pressure vessel 6.5 using 100 K helium in a pressure vessel 1.8 using liquid helium stored in a sub-tank inside a main propellant tank 0.5 using pumped liquid helium from an unpressurized tank The principle lesson seems to be that storage of helium as a gas is prohibitive in mass. Putting liquid helium in a subtank inside a main propellant tank avoids the high pressure turbopump needed for liquid helium stored at low pressure. This method however is not as light weight as a pumped system that avoids putting the liquid helium in a pressure vessel. Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ To: Bruce Dunn Cc: space-tech@CS.CMU.EDU, gwh@soda.berkeley.edu Subject: Re: Methods for delivering high pressure helium Date: Thu, 11 Feb 93 22:39:34 -0800 From: George William Herbert You missed an option there, or a variant of an option: use gaseous Tridyne to pressurize the tank (by volume, 0.8 He, 0.06 O2, 0.14 H2 if I remember right). Tridyne is inert at any pressure and temperature; the fractions of Oxygen and Hydrogen are low enough that combustion can't start. You generate energy (pressure) by running it, by pressure feed usually, past a platinum catylist that initiates a O/H combustion. No moving parts, no fuss, no muss, no bother. I've got some supporting documentation in the various Shuttle Hybrid Booster study reports that are lying on my living room floor, but digging it up right now would be "exceptionally painful", so I'll do it later 8-) [besides which, you should be able to work out the performance details from basic principles and chemical reactions tables ;-) ] -george william herbert Retro Aerospace gwh@soda.berkeley.edu gwh@retro.com ------------------------------ Date: Fri, 12 Feb 93 08:45:21 -0500 From: dietz@cs.rochester.edu To: Bruce_Dunn@mindlink.bc.ca, Space-tech@cs.cmu.edu Subject: Re: Tank blow-down in a pressure fed SSTO Bruce mentioned the problem with adiabatic expansion causing a larger pressure drop. Let me first say that as isothermal expansion is an optimistic assumption, adiabatic expansion is pessimistic. The gas will pick up some heat from the fuel and walls, as well as from condensation of vapors as the gas cools. A gas undergoing adiabatic expansion satisfies P V^k = constant where k is the ratio of specific heats C_p/C_v. For helium, k is nearly 5/3 (helium is almost a perfect monoatomic gas). The polyatomic contaminants in the helium (from the hot gas used to heat the liquid helium) will depress this a bit. C_v for helium is about 3.2 kJ/kg K. If the helium is an ideal gas (which it is not at high pressure, but never mind), then expanding 300 K helium by a factor of two reduces its temperature by about 110 K. We will therefore need roughly 300 MJ/tonne of heat to maintain isothermal expansion. This heat could be added in a number of ways. The helium could be warmed by injection of hot gas. This could lead to decomposition of peroxide. However, the partial pressure of peroxide in the head space will be very small, a few kPa at most. This is << .1% of the partial pressure of the helium. So an explosion there seems unlikely. One would likely design some kind of diffuser to mix the gases before they come in contact with the remaining propellant. Another idea is to vary the mixture ratio of the helium and combustion gases in the gas generator, so that the temperature increases later in the flight. Perhaps a safer approach to reheating the gas is to enhance heat transfer between the propellant and the gas. The specific heat of RP-1 (for example) is about 1.9 kJ/kg K at 300 K. RP-1 is about 40 times denser than helium at 15 MPa, so, naively, it can warm its own volume in helium by 100 K and experience only a cooling of 4 K (more, actually, since fuel is constantly being withdrawn from the tank). One way to do this would be to spray fuel into the head space of the tank through some kind of showerhead arrangement. This would require the spraying of on the order of tens of kilograms of fuel per second for the SSTO concept's kerosene tank (and some for the peroxide tank as well). This could be done with pumps, or by using two tanks (one at higher pressure), or by means of tanks equiped with perforated horizontal barriers. Paul F. Dietz dietz@cs.rochester.edu ------------------------------ Date: Fri, 12 Feb 93 07:50 PST To: space-tech@cs.cmu.edu Subject: Re: Methods for delivering high pressure helium From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) > George William Herbert writes: > > > You missed an option there, or a variant of an option: > use gaseous Tridyne to pressurize the tank (by volume, 0.8 He, > 0.06 O2, 0.14 H2 if I remember right). Tridyne is inert at > any pressure and temperature; the fractions of Oxygen and > Hydrogen are low enough that combustion can't start. > > You generate energy (pressure) by running it, by pressure > feed usually, past a platinum catylist that initiates > a O/H combustion. No moving parts, no fuss, no muss, > no bother. Actually, it was mentioned in passing in the posting but I did not make a big deal of it as is it is apparent that with most schemes for delivering high pressure helium, the mass penalties are in the system for generating high pressure, rather than in the system for warming the gas to the desired operating temperature. Putting hydrogen and oxygen in the helium and using a catalyst bed is a simple and convenient way of generating heat. However, it is not necessarily lighter than using a gas generator burning other propellants. The hydrogen and oxygen for energy generation are in effect being stored as gases at low density and high pressure (the helium tank must be made larger to account for their contribution). Nearly the same energy output per mass of reactant can be had from peroxide and a hydrocarbon burned in a gas generator. These have the advantage of high density, and consequently low tank mass for their storage. It should also be noted that a helium/oxygen/hydrogen mix is only usable when the primary storage system for the helium involves gases. With liquid helium, the hydrogen and oxygen would freeze out. (Now there's an explosive for you - mixed crystals of frozen hydrogen and oxygen :-) ). Since the lightest schemes for delivering helium seem to start with liquid helium, this leaves out the possibility of incorporating hydrogen and oxygen into the helium. Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ To: Bruce Dunn Cc: space-tech@cs.cmu.edu, gwh@lurnix.COM Subject: Re: Methods for delivering high pressure helium Date: Fri, 12 Feb 93 12:05:10 -0800 From: gwh@lurnix.COM Bruce correctly points out that any gaseous Helium system will mass more than liquid helium or propellant based pressurizations systems. One advantage Tridyne has, though, is its amazing simplicity and low cost 8-) This is not insignificant in vehicle design... though on a SSTO, the system's optimal point may well be with the absolute lightest system possible, nearly regardless of expense. This is one reason I don't like SSTO all that much: it encourages pushing the edges of vehicle's mass budgets, which is risky and expensive to develop. By the way, Bruce, have you compared the composites structural characteristics I posted to what Burnside Clapp (?) was assuming yet, and if so how far off was he? 8-) Inquiring minds want to know... Also: >(Now there's an >explosive for you - mixed crystals of frozen hydrogen and oxygen :-) ). At this point I will categorically deny any knowledge of that or similar explosive combinations, but agree that it's a problem for spacecraft 8-) -george william herbert Retro Aerospace gwh@lurnix.com gwh@soda.berkeley.edu gwh@retro.com ------------------------------ Date: Fri, 12 Feb 93 19:03 PST To: space-tech@cs.cmu.edu Subject: Burnside Clapp SSTO tanks From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) George Herbert writes: > > By the way, Bruce, have you compared the composites structural > characteristics I posted to what Burnside Clapp (?) was assuming > yet, and if so how far off was he? 8-) Inquiring minds want to > know... > Sorry for the delay. Mitch Burnside Clapp provided me with a spreadsheet printout detailing some of the calculations for the proposed SSTO. Under the section "Tank Details" is listed: Tank pressure: 13,790,000 Pa (13.7 MPa, or 2000 psi) Tank wall density: 942 kg/m^3 Tank wall maximum stres: 2,344 MPa From the size and mass of his tanks (reported elsewhere in the spreadsheet) it appears that the 2,344 MPa is the ultimate tensile strength of the tank wall material. His working stress is lower than this by a factor of 1.3. The paper in which the SSTO is described says "The tanks are composed of the highest strength to weight material available, Kevlar 49", and quotes "Schwartz, M.M. Composite Materials Handbook, McGraw Hill, New York" as a reference for this fact. I think that these values don't withstand scrutiny. Kevlar by itself has a density of 1.45, and the resin used has a density of about 1.2. Assuming 50 % by volume fibers in a composite material gives a density for the composite of 1.32, very close to the 1.33 quoted below by George. I can't see how this can be reconciled with the density of 0.942 used by Mitch. The only thing that I can think of is that 0.942 might be the bulk density of a wound Kevlar tank reinforcement over an aluminum tank, done without resin. The strength quoted by Mitch, 2,344 MPa, is reasonably close to the 2,650 quoted by George for Kevlar fibers with a unidirectional pull and without resin. Another reference I have gives a strength of 2,700 MPa. However, this figure can't be used for the strength of a "ball of yarn" Kevlar winding on a spherical tank, where only a portion of the fibers are available to take stress in any one direction. For fully composite tanks, the Kevlar must in addition be combined with resin, which adds mass but almost no tensile strength. For a fiber wound aluminum tank, I don't know whether a tank wound with Kevlar without resin would be robust enough for a reusable vehicle. George Herbert's values for composite materials are below. Unfortunately, there is no value for 45 degree Kevlar composite. The strength of this may perhaps be estimated from the other data. 90 degree Kevlar has 83% of the strength of 90 degree carbon fiber. If we assume that 45 degree Kevlar has 83% of the strength of 45 degree carbon fiber, we end up with a tensile strength of 200 MPa, and a density of 1.33. This is surprisingly low - it hardly outperforms aluminum at a yield stress of 400 MPa, and a density of 2.7. I suspect that the real number should be better - after all, fiber wound pressure vessels are common enough to make one think that there are mass advantages. > > Kevlar 49, 90 degree woven bidirectional layup (has max strength along > x, y axies, much lower along intermediate angles) in Epoxy matrix: > Ultimate Tensile Strength: 517 MPa > Density: 1.33 > > +-45 degree woven XAS Carbon fiber, (closer to even strength in all axies) > in Epoxy: > UTS: 240 MPa > Density: 1.53 > > +-90 degree woven XAS, Epoxy (again, strong only along x, y axies) > UTS: 625 MPa > Density: 1.53 > > Unidirectional XAS, Epoxy (strength only along one axis) > UTS: 2040 MPa > Density: 1.57 > > Unidirectional Kevlar > UTS: 1379 MPa > Density: 1.38 > > > The fibers involved, without a matrix, have UTS of roughly > 2650 MPa (Kevlar) and 3850 MPa (Carbon XAS). There are slightly stronger > fibers out there, but their performance in an actual composite layup is > similar in bi- and multi-directional usage. Their primary usages are in > uni-directional layups. > > As you can see, there is a factor of 7 reduction in strength > from unidirectional fiber to even bidirectional composite layup; > for a real spherical tank, I would guess that a limiting stress would > be on the order of 165 MPa. > > [For the metric-impaired (Including Me, most of the time): > 1 MPa = 145 PSI > in other words, Carbon XAS has a fiber strength of about > 560,000 PSI but in a non-directional composite it's closer to > 35,000 PSI (and you have to have a safety margin on top of > that...)]. > > > _Design with Advanced Composite Materials_, Ed. Leslie N Phillips, > Springer-Verlag, New York (1989) pp. 5, 18,21, esp. 73 (table) > Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ To: Bruce Dunn Cc: space-tech@cs.cmu.edu, gwh@soda.berkeley.edu Subject: Re: Burnside Clapp SSTO tanks Date: Fri, 12 Feb 93 22:10:08 -0800 From: George William Herbert >I think that these values don't withstand scrutiny. Kevlar by itself has a >density of 1.45, and the resin used has a density of about 1.2. Assuming 50 >% by volume fibers in a composite material gives a density for the composite >of 1.32, very close to the 1.33 quoted below by George. I can't see how this >can be reconciled with the density of 0.942 used by Mitch. The only thing >that I can think of is that 0.942 might be the bulk density of a wound Kevlar >tank reinforcement over an aluminum tank, done without resin. Aluminum is heavier than kevlar is; it's density is 2.55-2.75 depending on the alloy. Al-Li alloy is lighter by a bit, but still about 2.0 . >The strength quoted by Mitch, 2,344 MPa, is reasonably close to the 2,650 >quoted by George for Kevlar fibers with a unidirectional pull and without >resin. Another reference I have gives a strength of 2,700 MPa. However, >this figure can't be used for the strength of a "ball of yarn" Kevlar winding >on a spherical tank, where only a portion of the fibers are available to take >stress in any one direction. For fully composite tanks, the Kevlar must in >addition be combined with resin, which adds mass but almost no tensile >strength. For a fiber wound aluminum tank, I don't know whether a tank wound >with Kevlar without resin would be robust enough for a reusable vehicle. You can reduce the resin fraction, but you can't eliminate it, and bad things start to happen to manufacturing and lifetime characteristics. >George Herbert's values for composite materials are below. >... >This is surprisingly low - it >hardly outperforms aluminum at a yield stress of 400 MPa, and a density of >2.7. I suspect that the real number should be better - after all, fiber >wound pressure vessels are common enough to make one think that there are >mass advantages. No, trust me, composites do "hardly outperform" aluminum in most applications. They save about 10% of the mass of aluminum for structures because they're a bit stronger per unit mass, and about 10% more if you take advantage of unidirectional composites where stresses are mostly uniaxial. That 10-20% savings is all you usually get, for astronomical cost, but at $7500/kg, any savings is worth it. It's worthwhile noting that in cylindrical tanks with round ends, you gain a bit more because the load is primarily circumfrential (it's twice the stress in the axial direction) and can use unidirectional layups to better advantage there. For instance, you have a layup with layers at +45, -45, 0, -45, +45, 0, ...; this gives adequate strength axially and increased strength where it's needed. -george william herbert ------------------------------ Date: Sun, 14 Feb 93 11:59:37 -0500 From: dietz@cs.rochester.edu To: Bruce_Dunn@mindlink.bc.ca, space-tech@cs.cmu.edu Subject: Re: Methods for delivering high pressure helium Bruce mentioned four methods for delivering helium gas. The third method involved putting an insulated liquid helium tank inside the main pressure vessel. I am not sure this would save much weight: insulation for liquid helium typically includes *vacuum* insulation, so the structure of the tank must support the full pressure of the materials, not just the difference in pressure between the helium and propellant tanks. Can Bruce tell us a bit more about the strength of vacuum insulation materials? Paul ------------------------------ Date: Sun, 14 Feb 93 14:03:44 -0500 From: dietz@cs.rochester.edu To: space-tech@cs.cmu.edu Subject: More on pressurizing Here's another idea for tank pressurization, applicable to both the SSTO and the P2. How about pressurizing the fuel and oxidizer tanks with some gaseous fuels and oxidizers? This way, when the liquid fuels are exhausted, the pressurants can be reacted, so none of the pressurants contribute to the "dry mass". At least in the SSTO, this will tend to happen near the end of the flight when thrust is low anyway, so the lower flow rate of the gases is perhaps not a problem. Obvious pressurants are oxygen (for the oxidizer tank) and hydrogen or methane for the fuel tank. One problem would be that (depending on the relative sizes of the fuel and oxidizer tanks) the fuel/oxidizer ratio could be a bit lean. This could be addressed by letting the liquid oxidizer run out before the liquid fuel. Paul ------------------------------ End of Space-tech Digest #145 *******************