Subject: Space-tech Digest #152 Contents: Bruce Dunn's P2 booster design: P2 3/5 Design Choices P2 4/5 Calculations P2 5/5 Economics ------------------------------------------------------------ To: space-tech@cs.cmu.edu Subject: P2 3/5 Design Choices From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) OVERALL DESIGN: The overall design of the P2 vehicle was driven by the following considerations: 1) A two stage vehicle as a basic launcher - one stage is too much of a technical challenge for an expendable vehicle, and 3 stages are not needed for low earth orbit 2) A single engine per stage - this gives simplified engine installation (no propellant manifolds or multi-engine thrust structures) and reduces potential failure points 3) Pressure fed engines using liquid propellants - this gives low cost and reliability, along with low development costs 4) A payload exceeding existing launchers, with better cost and reliability than them - this requirement is driven by a need for initial market penetration in the comsat delivery field 5) Ready development of heavy lift launchers from the basic vehicle - the minimal vehicle must provide building blocks for larger vehicles in a way that SSTO designs can't 6) Storable, non-toxic propellants - for ease and safety of launch operations 7) Construction techniques and components from the shipbuilding and chemical processing industries rather from the aerospace industry - for low cost and multi-sourcing PROPELLANT SELECTION: The standard pump-fed rocket propellant combination of RP-1 and LOX, although having a higher Isp than RP-1/peroxide, is not easy to employ in pressure fed booster designs. The lower bulk density of the RP-1/LOX combination leads to larger tanks for a given mass of propellant. The LOX tank must be made of aluminum, as normal steels are too brittle at cryogenic temperatures and cryogenic compatible stainless steels do not have the tensile strengths required for lightweight tanks. The walls of an aluminum tank for the sizes of boosters used would be up to 36 mm thick, leading to expensive fabrication. The fuel and oxidizer tanks are at markedly different temperatures, leading to mechanical difficulties at attachment points due to differential expansion of materials, and requiring insulation where the fuel line passes through the LOX tank. Cooling of pressurization gas by the cryogenic propellant increases the amount of helium needed for pressurization. Finally, the use of a cryogenic oxidizer complicates launching procedures. A rough calculation gives a LOX/RP-1 pressure fed lower stage with a mass fraction of 0.888, compared with 0.913 for peroxide/RP-1. This loss of mass fraction essentially eliminates the advantage of higher specific impulse for LOX. For a given size of vehicle, whose mass is limited by the take-off thrust available, the performance of pressure-fed RP-1/LOX appears to be no better than that of RP-1/peroxide, and the design appears to be much harder to execute. All stages burn RP-1/peroxide at a mixture ratio of 7 to 1. Hydrazine derivatives are rejected as fuels because of their toxicity and the fact that they are considerably more expensive than hydrocarbons. Alternative fuels would be propane (slightly decreased performance because of lower density) or JP-5 (denser, and possibly slightly better performing, depending on Isp). Hydrogen peroxide is used as an oxidizer as it gives good performance, is not cryogenic, is dense, and gives propellant combinations with a high oxidizer to fuel mixture ratio and thus high bulk densities (important for pressure fed rockets). N2O4 is rejected as an oxidizer because of its toxicity (when spilled it reverts to highly toxic NO2 gas). Nitric acid is rejected as an oxidizer because of low performance, and because in its common form of IRFNA (Inhibited Red Fuming Nitric Acid) it contains a high proportion of NO2. The dangers of oxidizers containing NO2 must be emphasized. In the event of a launch pad or in-flight accident, several hundred tons of oxidizer can be released. This could be potentially catastrophic if winds were to blow the toxic cloud over inhabited areas. Hydrogen peroxide is relatively non-toxic - spilled peroxide does not give off fumes, and skin contact results in a bleached patch of skin without permanent damage. Any accidental spills in the pad area can be flushed away with water (3% hydrogen peroxide in water is benign enough to be sold in drug stores for cleaning open wounds). 100% hydrogen peroxide is a relatively stable material, with decomposition rates of about 1% per year. Although hydrogen peroxide decomposes energetically to oxygen and water, it is impossible to get bulk hydrogen peroxide to detonate unless a strong booster charge is used. Runaway decomposition can only happen at elevated temperatures or with grossly impure material. Anecdotal reports of difficulties with peroxide instability probably date from wartime German experience with highly impure peroxide in improperly prepared tanks. Peroxide decomposition is a surface catalyzed phenomenon, and is less troublesome when volumes are large. In practice, aside from slow decomposition, hydrogen peroxide is as least as stable as the various hydrazine fuels which are already in use in the aerospace industry. 50 to 90% peroxide solutions are readily obtainable commercial chemicals, and refining peroxide to 100% can be done distillation or freezing methods. For use in a booster, peroxide could be bought through competitive bids on the open market, shipped to the launch site at moderate strengths, then refined there to 100%. The RP-1/peroxide combination gives good performance, is dense, and uses relatively non-toxic propellants. While hydrocarbon/peroxide engines have not been widely used in space launchers to date, there is a considerable information on combustion with this combination. Extensive development work was done in the 1950s using peroxide and jet fuel for "super-performance" rockets for short term bursts of thrust in military aircraft. A useful reference is "Advantages of Hydrogen Peroxide as a Rocket Oxidant" by David Andrews (J. Brit. Interplanetary Society, 43:319-328, 1990). ENGINE DESIGN: The engines assume ablative cooling. This makes for a simple engine design with low manufacturing costs. Avoiding regenerative cooling also eliminates the pressure drops involved in cooling passages, a factor important in a pressure fed engine where pressure drops can only be offset with higher tank pressures and a substantial cost in terms of tank mass. Heat fluxes per unit surface area in the throat and nozzle are less than are commonly encountered in ablatively cooled solid propellant engines because of the relatively low chamber pressure and combustion temperature (3 MPa and 3000 K, vs. for example approximately 6 MPa and approximately 3500 K in a shuttle SRB). The 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. The thrust vector control system is similar to that used for the solid rocket boosters of the Titan launcher. On these boosters, pressurized nitrogen tetroxide is supplied from a cylindrical tank strapped somewhat awkwardly to the side of the solid rocket motor. The liquid is injected through a series of 24 ports arranged in a ring around the nozzle. The flow is controlled by 6 valves. In the P2, high pressure peroxide for thrust vector control purposes can be tapped off the main engine propellant feed, allowing the pressurized liquid for thrust vector control to be obtained "free". The alternative to peroxide injection for thrust vector control would be to mount the engine conventionally so that the entire engine is steerable. This system would require hydraulic actuators and control valves, a source of hydraulic power (pressurized RP-1 might be usable), engine pivots and flexible high pressure propellant lines for the engine. These would make the design more complicated and introduce more failure points than the simple peroxide injection system, whose only moving parts are several valves. The engines assume the use of conventional propellant injection, with the injector design optimized for low manufacturing cost and low pressure drop. Many previously designed peroxide/hydrocarbon engines have utilized catalytic decomposition of peroxide followed by injection of fuel. This gives an easily restartable engine which has good combustion efficiency in a small chamber. The added complications, mass, and pressure drop of catalytic peroxide decomposition however are probably not worthwhile in large engines which do not require restartability. Since the nozzle of the P2 engines does not move during thrust vector control, the option exists to support a nozzle from the outside using a space-frame structure fixed to the lower propellant tank. This may allow mass savings relative to a nozzle which is self supporting, and which must withstand rapid steering motions. TANK DESIGN: Good mass fractions are obtained for the upper and lower stages, in spite of the fact that the boosters are pressure fed. This is partly due to the high bulk density of the propellants chosen, partly due the use of very high strength tank material, and partly due to the fact that all pressure vessels with the exception of the lower stage oxidizer tank are spherical. The volumetric ratio between peroxide and fuel is 4 to 1 at a mass ratio of 7 to 1. Thus, when a booster is built with a lower stage containing four times as much propellant as the upper stage, the diameter of the upper stage oxidizer tank matches the diameter of the lower stage fuel tank and allows a simple cylindrical interstage structure and a common propellant tank design. Tanks are of welded steel construction. This is similar to the construction method used for the "260 inch" (6.6 meter) diameter solid rockets built and tested in the mid 1960s by Aerojet-General for NASA, and the ASRM (advanced solid rocket motor) currently being built for NASA. The 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). The P2 design assumes the use of Ladish D6aC steel, heat treated to a minimum yield strength of 1340 MPa. This gives tank walls ranging in thickness from approximately 3 mm (upper stage fuel tank) to 10 mm (lower stage oxidizer tank, barrel section). This steel has been widely used at this yield strength for the construction of solid rocket motor casings such as those for the Titan and Shuttle. A number of alternate steels are available which could also be used - some of these have higher strengths while retaining adequate toughness, and their use could provide thinner tank walls and a modest increases in performance. The use of lower strength steel was not found to be a practical alternative to the use of ultra high strength steel. Steel with a yield strength of 689 MPa, (100,000 psi) resulted in substantially higher tank masses, and lower performance. Payload was reduced from 14.6 metric tons to 6.1 tons for a total vehicle mass of 1000 tons. Alternatively, in order to boost the same payload, the lower strength steel required a vehicle mass of approximately 2400 tons to match the performance of a 1000 ton vehicle with D6aC steel. CHOICE OF TANK PRESSURIZATION GAS: Helium (MW 4) is used for tank pressurization. The next lightest gas which would be non-reactive with both the fuel and oxidizer is nitrogen (MW 28). If nitrogen were to be used for the lower stage, while retaining helium for the upper stage, the payload would be reduced by approximately 15%. If nitrogen were to be used with both stages in place of helium, the payload would drop by 60%. This gas does not therefore appear to be attractive for pressurization. Gases which are potentially reactive with one of the two propellants could possibly by utilized if the propellants are separated from the gases by a flexible membrane, or by ensuring that the gas mixture above propellants is so fuel or oxidizer rich that it cannot support combustion. The logical candidate for an alternate pressurization gas is hydrogen (MW 2). Hydrogen would give only a very slight payload increase over helium, but is considerably less expensive and could be heated in a simplified gas generator burning a portion of the hydrogen with hydrogen peroxide. The possibility of using pressurizing gases which can be burnt with one another or with extra fuel or oxidizer at the end of a flight is initially attractive. For example, the peroxide tank is 80% of the pressurized propellant volume, and it could be pressurized with oxygen rather than helium. Gaseous oxygen remaining in the tank when the peroxide runs out would be burned with remaining fuel, or with fuel and a burnable fuel pressurization gas. However, all such schemes require either an additional dedicated engine, or a specialized main engine which can utilize both liquid and gaseous propellants and can transition easily between them. The slight potential advantages of such a system in terms of payload and pressurizing gas costs seem to be outweighed by the engine design problems and the necessity to use two separate gases to pressure the main propellants. CHOICE OF TANK PRESSURIZING SYSTEM: There are several alternatives to the proposed scheme for delivering warm helium (or possibly hydrogen) for tank pressurization. One of these schemes might prove preferable to the baseline system of a spherical pressure vessel with liquid helium. It should be noted that storage of high pressure helium at ambient temperature is not a feasible option for helium supply - the needed helium supply tank would weigh more than the total mass of the main propellant tanks. 1) The helium could be stored in a light weight non-pressurized tank, and pressurized by a small peroxide-powered turbopump. This would save the mass of the large liquid helium pressure vessel and the mass of the gaseous helium pressurization supply, at the expense of adding a turbopump and its reactant supply. An advantage of this scheme is that the helium tank need not be spherical, and could be fitted into the interstage area if desired. 2) The helium could be stored as cold pressurized gas at several times the delivery pressure in an insulated pressure vessel, with the gas being delivered under its own pressure. This would eliminated the gaseous helium blowdown components, but would require a much heavier pressure vessel (2 to 3 times the total mass of the baseline system). This scheme also wastes helium (some helium always remains in the pressure vessel since the output pressure never goes to zero) and it is difficult to charge the tank with cold gas and to keep the tank from warming up after charging (the latter problem is solved in boosters using cryogenic propellants by placing the helium tanks inside the LOX or LH2 tank). Whatever the helium supply method, a system must be provided to warm the gas before use (even if the helium were stored at ambient temperature as a pressurized gas, it would cool upon expansion from its tank). 1) The baseline vehicle has a flow through gas generator, in which the products of stochiometric combustion of peroxide and fuel are intimately mixed with the pressurization gas and delivered to the propellant tanks. The CO2 and H2O from the combustion are only a small portion of the gas flow, and are assumed not to result in problems when they enter the tank (the H2O will condense at the temperatures involved, resulting in a water mist in the helium gas stream). If this is considered objectionable, a heat exchanger type gas generator could be used, with the reactant products dumped overboard. Such a gas generator would be heavier, more complex and less fuel-efficient than a flow-through generator. 2) Engine heat from the combustion chamber could be used to evaporate and warm the helium. Adding this requirement to the engine however complicates the design of the combustion chamber and involves considerable piping to connect the combustion chamber to a source of liquid helium and to conduct the vaporized helium back to the tops of the propellant tanks. If the helium vaporization system were integrated with the engine, it could not be tested independently of full scale engine tests. Finally, pressurization gas would not be available until the engine is running, complicating engine startup. 3) Helium could be warmed by contact with tank propellants, using a heat exchanger in a flow chamber at the tank outlet. Such a scheme however is more difficult to test than a stand-alone system, can never warm the helium completely to ambient temperature, and is likely to require a large heavy heat exchanger because of the limited temperature differential between partly warmed gas and the propellant. RECOVERABILITY: While the basic P2 vehicle is designed to be produced as a low cost expendable, there is a potential to recover and reuse some of the hardware if it were to prove economic. The booster stages of the boosted P2 have roughly the terminal velocity as the burned-out Shuttle solid rocket boosters, while being somewhat lighter. The technology thus exists to recover them by SRB-derived parachutes and water landing, and reuse them or components from them in order to lower launch costs. If tanks valves are closed after propellants are used, helium used for pressurization could be recovered at the same time. Any attempt to salvage and reuse the components would have to take into account economic considerations, including the cost of the recovery and refurbishment operations, the reduction in economy of scale of the production of the expendable stages, and the reduction in payload mass caused by the lower mass fraction of the boosters equipped with parachutes and recovery aids. -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ Date: Mon, 14 Jun 93 22:56 PDT To: space-tech@cs.cmu.edu Subject: P2 4/5 Calculations From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) PERFORMANCE CALCULATIONS: Performance calculations were made using a spread-sheet which models the acceleration of a payload to a velocity of 9300 m/sec (low earth orbit velocity, taking into account gravity and air resistance losses during ascent and assuming a launch at 28.5 degrees latitude). For all configurations, total liftoff mass is governed by the sea level thrust of the P2 engines in operation at takeoff. An initial acceleration of 1.35 g was required of the vehicle. Payloads given are gross, without any allowance for shrouds - addition of allowances for shrouds carried partway to orbit might lower payloads by something like 10%. The spread sheet used to determine payloads to LEO does not actually "fly" an ascent, and gravity and air resistance losses are only estimated. Nevertheless, the spreadsheet gives reasonably accurate results with test cases. The spread sheet estimates a gross payload to low earth orbit of 7.8 metric tons for the Soyuz launch vehicle. When carrying a shroud and escape tower for the first 160 seconds of flight, the literature value for the nominal payload for this vehicle is 6.8 metric tons. For the Ariane 44L launcher, the spreadsheet predicts a gross payload of 10.9 metric tons, against a literature value of 9.6 metric tons net. The optimum chamber pressure and expansion ratio for both the lower and upper stage engine was investigated in a series of vehicles in which the tank and chamber pressure was varied, keeping a constant 1 MPa of pressure available for injector pressure drop. The maximum exit cone diameter for the lower stage engine was set at 5 meters (for easy support of the 6.1 meter diameter boosters on the launch pad). A sea-level thrust of 13,500 kN was required of the lower stage engine, which uniquely defined the throat diameter and expansion ratio for each chamber pressure. Higher chamber pressures were associated with higher Isp (both at sea level and vacuum), but had the penalty of higher tank mass to contain the required higher tank pressures. Between 3 and 6 MPa chamber pressure there was very little change in the payload to low earth orbit - a pressure of 3 MPa was selected as it was associated with the lowest dry mass and helium mass. The maximum exit cone diameter for the upper stage engine was set at 3 meters, with a required thrust of 1000 kN. Delivered payload was at an optimum for a chamber pressure of 3 MPa, supplied by propellants pressurized to 4 MPa. The baseline P2 design relies on a gas generator to vaporize and warm the helium used for tank pressurization. The gas generator requires approximately 200 kg of fuel+peroxide for every ton of helium vaporized - most energy is used for warming, not vaporization. The mass of the reactants required for vaporization is explicitly calculated and included in the liftoff mass of the vehicle. The burnout mass includes the mass of helium pressurization gas, the combustion products of the gas generator, and 0.5% of the total original propellant load as residual unburned propellant. The performance of the P2 booster to a given delta V depends on the mass fraction and specific impulse of the stages. Specific impulse for peroxide and RP-1 was directly calculated for different chamber pressures and expansion ratios using a US Air Force program developed by Curtis Selph (thanks to Mitchell Burnside Clapp for providing a copy of the program). Thrusts for various chamber pressures, throats, expansion ratios and ambient pressures were calculated by a spreadsheet incorporating standard formulas from reference works. In practice, the maximum exit diameter of the engine nozzles was predetermined from design considerations, and the throat needed to provide the desired thrust from a given chamber pressure was determined, thus determining the expansion ratio. The Isp program was then used to determine the theoretical Isp for the particular engine configuration. The results of these theoretical Isp calculations were derated by 5% to allow for losses and inefficiencies in a real world engine. The mass fraction of the booster stages is determined by the burnout mass associated with a given mass of usable propellant. As long as helium is used as the pressurizing gas, most of the burnout mass is the "dry" mass of the vehicle. This in turn is dominated by two items: the engine and the high pressure propellant and helium tanks. ENGINE MASS: The lower stage engine is taken to have a sea level specific thrust of 900 N/kg. This is similar to that of the pump fed F-1, RS-27 and H-1 expendable LOX/RP-1 engines built by Rocketdyne (which have specific thrusts of 800, 900 and 1000 respectively). In practice, since the P2 engine has no gas generators, turbines, pumps, or nozzle steering equipment, it could possibly be built to have a higher specific thrust than that assumed. However for the purposes of the design concept being investigated here, the mass has been estimated conservatively. The complete lower stage P2 engine has a thrust of 13,500 kN, and a mass estimated to be 15 metric tons. As a reality check, the Advanced Solid Rocket Motor currently under development has a sea level thrust of 15,500 kN. The ablative nozzle for this motor has a mass of 8.6 metric tons even though it incorporates a heavy movable joint which would not apply to the P2 nozzle, has extra safety margins for man-rating, and can be refurbished for reuse. The mass of the P2 engine thus looks eminently achievable. The upper stage engine is assumed to have a vacuum specific thrust of 800 N/kg, similar to that of the high expansion ratio Atlas MA-5 sustainer engine. TANK MASS: Tank mass is dominated by the requirements of pressurization. Tank mass is minimized by using spherical tanks for all but the lower stage peroxide tank. Tank masses were calculated by standard pressure vessel formulas for spherical tanks or cylindrical tanks with spherical end caps, with an allowance of an extra 5% mass for local strengthening in critical areas and for attachments etc. The tank mass estimates use a safety factor of 1.2 for an expendable vehicle (yield strength divided by maximum operating stress). This may be compared with a safety factor of 1.25 used for the pressure induced loads in the Shuttle external tank, and the identical factor of 1.25 used for the casing of the reusable ASRM. Before each use, tanks are tested hydrostatically to 10% over their maximum operating pressure. Tank masses are likely to be reasonably accurate, as they are directly calculated by simple formulae and follow existing practice in the design of large steel casings for solid rocket motors. The two determinants of tank mass are the safety factor used, and the yield strength of the material. The D6aC steel proposed has previously been used for Titan and Shuttle solid rocket boosters. The tensile strength of this steel depends on how it is heat treated. Treatment which gives very high tensile strengths also however gives a steel which does not have adequate toughness, and which is subject to brittle fracture. The calculations assume steel treated to a yield strength of 1340 MPa - a treatment used for the steel in Titan and Shuttle boosters and thus known to have adequate toughness. CONSERVATIVE CALCULATIONS: It is the purpose of the P2 conceptual design to present a reasonable prediction of the minimum performance of the vehicle. Where possible therefore, calculations have been made on the basis of conservative design choices or methods of calculation. Some of these are as follows: 1) When the propellants are exhausted, propellant tanks still contain a substantial mass of high pressure helium. If exhausted through the nozzle after engine burnout, this could provide additional thrust for a few seconds. This potential thrust from propulsive gas venting has been ignored in performance calculations. 2) The engine is ablatively cooled. As the ablator burns off the inside surfaces of the nozzle, the burned material contributes to the mass flow and thrust of the engine, and the engine gets lighter second by second, increasing vehicle performance. This effect has been ignored, and the engine is assumed to have an unchanged mass at the end of the flight. 3) The initial thrust/weight of the vehicle has been set to 1.35. In practice, some liquid propelled launch vehicles have used lower initial thrust to weight ratios. If the main engine is kept constant and the lower stage is increased in size to give an initial thrust/weight ratio of 1.2, the payload of the vehicle increases 15%. 4) The calculations used here assumed a fixed mass 800 ton lower stage and a fixed mass 200 ton upper stage without optimization of the relative masses of the stages. Optimization of the relative sizes of the stages would likely increase performance slightly from that calculated here for arbitrary stage sizes, but would disturb the 4 to 1 ratio of stage sizes which allows both the lower stage fuel tank and the upper stage oxidizer tank to be identical spheres. -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ Date: Mon, 14 Jun 93 22:59 PDT To: space-tech@cs.cmu.edu Subject: P2 5/5 Economics From: Bruce_Dunn@mindlink.bc.ca (Bruce Dunn) COMPARISONS WITH EXISTING LAUNCHERS: The P2 minimum launcher may be compared with the following competitors: Vehicle Cost, $M LEO payload Vehicle Mass metric tons metric tons P2 with 10% shroud allowance ??? 13.1 1000 Ariane 44L/LH2-LO2 110 9.6 469 Atlas2/Centaur 80 7.1 163 Delta2/PAM-D 52 5.0 204 It is premature to perform detailed cost estimates for the P2 vehicle. However, as a very rough rule, costs are more associated with complexity and the number of parts in a vehicle, rather than sheer size. Complexity drives up inspection and quality control costs and gets in the way of reliability - this is one of the basic problems of the Shuttle. On the basis of complexity, the P2 compares very well with the commercially successful Ariane 44L vehicle, which has less payload capability to low earth orbit. Vehicle Tank sets Separation Planes Engines Turbopump sets Ariane 44L 7 6 10 10 P2 2 1 2 0 The P2 vehicle avoids the use of solid rocket boosters, eliminating the pollution problems associated with their exhaust (Titan, Delta, Shuttle, some Ariane configurations). It avoids the use of toxic propellants such as nitrogen tetroxide and hydrazine derivatives (Titan, Ariane), and avoids the use of a hydrogen/oxygen stage (Titan IV, Atlas, Ariane, Shuttle). The major development costs for the vehicle will be for two new engines. The engines should have low development and productions costs as they have no gas generator system, no turbines, no pumps and no regenerative cooling. They burn dense, storable, and relatively non-toxic propellants which are delivered to the engine at ambient temperature. Currently, the major hope of many people for reducing the cost of access to low earth orbit is the development of SSTO vehicles such as the DC-1. Relative to the DC-1 conceptual design, the P2 can be evaluated as follows: Item P2 SSTO Development cost moderate high Production cost per flight low low (reusable) Time to availability moderate long (?) Technical risk low high (?) Propellants storable cryogenic Reliability via simplicity redundancy Repair/refurbishment cost none low(?) Complexity of vehicle low high Need for landing facilities no yes Need for booster drop area yes no All-azimuth capability no (booster drop) yes Sensitivity to mass growth low high Sensitivity to Isp shortfalls low high Growth path to heavy launcher yes no On orbit propellant scavenging yes no Abort sites needed no yes Payload return no yes Ease of upgrading capability easy difficult Potential for Aerospace pork low high Relative to an SSTO such as a DC-1, the P2 is relatively insensitive to failure to achieve the design Isp and stage mass. This is shown in the following table, which indicates the percent reduction in payload induced by a 1% increase in stage mass, or a 1% decrease in Isp. 1% Change in Reduction in payload P2 lower stage mass 0.43% P2 upper stage mass 1.08% P2 both lower and upper stage mass 1.51% P2 lower stage Isp 2.21% P2 upper stage Isp 4.31% P2 both lower and upper stage Isp 6.49% DC-1 stage mass 4.49% DC-1 Isp 12.3% It is apparent that the SSTO has a much higher sensitivity to dry mass and Isp than the P2. This sensitivity is likely to make the design and development of the vehicle very expensive, as it is difficult to make design compromises without impacting the payload. Furthermore, the SSTO is at a much higher level of technology than the P2 and it is much more costly to shave mass or increase Isp for this vehicle than for the P2. The SSTO designer is thus caught in a double bind. Marginal improvements in the vehicle are expensive to make, but are more likely to be required because of the high sensitivity of the design. If the P-E designer has an lower stage which is 1% over his projected mass, he has the option of shrugging his shoulders and taking a 0.43% hit in payload, or making simple changes which reduce the stage mass by 1%. The SSTO designer is less likely to be able to afford his 4.49% hit in payload, and when he tries to shave mass he has a much harder job than the designer of the expendable. Actual flight operations costs are likely to favor the P2. The P-E is an extremely simple vehicle and will require very little checkout before launch. The P2 is not burdened with the costs of recovery, refurbishment or repair between missions (the Achilles heel of the Shuttle). Assuming for the sake of argument that flight operations costs are equal between the two vehicles, any advantage of an SSTO thus must come from the reusability of the hardware. Reusability however comes only at the expense of high development and initial production costs. Allen Sherzer posted some numbers about projected costs for DC-1 flights. These numbers are reworked here to show the competitiveness for the P2. Allen assumed two cases: an optimistic case and a pessimistic case. In both cases, the development cost was amortized over 10 years at 8% interest. Launch costs were assumed to be a constant 6 million (or 12 million for the pessimistic case) for each flight. Allen assumed a flight rate of 200 per year. I have reworked his figures using different flight rates, and assuming that the per flight launch costs are the same for lower flight numbers (not necessarily true, but an assumption which does not affect the relative merits of the two systems). I have assumed that a P2 launch cost (exclusive of hardware) is the same as that of the DC-1. This is probably unnecessarily handicapping the P2, which does not have the recurring maintenance costs of the DC-1. I have assumed that the development cost of the P2 is 20% of that of the DC-1. I have made no allowance for the larger payload of the P2 (several tons more), and have done comparisons on a per flight basis. Optimistic Scenario: DC-1 Yearly Flights Launch Amortize Total per Flight 25 6 29 35 50 6 14.5 20.5 100 6 7.3 13.3 200 6 3.6 9.6 P2 Yearly Flights Launch Amortize Total per Flight Difference 25 6 5.8 11.8 23.3 50 6 2.9 8.9 11.6 100 6 1.5 7.4 5.8 200 6 0.7 6.7 2.9 The "Difference" in the P2 table is the amount by which the launch costs and amortization for the P2 are less than the DC-1. This money is available to pay for expendable hardware. As long as the per vehicle cost of the expendable hardware is less that this difference, it pays to use an expendable vehicle. When similar calculations are done for the pessimistic DC-1 case (10 billion development costs, 12 million launch costs), the money available for expendable hardware doubles, to reach 46.5 million per flight for 25 launches per year, to a low of 5.8 million for 200 flights per year. -- Bruce Dunn Vancouver, Canada Bruce_Dunn@mindlink.bc.ca ------------------------------ End of Space-tech Digest #152 *******************