The Artemis Project is a commercial venture to put a permanent manned base on the moon. To do this, a Lunar Transfer Vehicle (LTV) will be used to ferry the base modules and crew back and forth between earth and lunar orbits. Current plans call for rocket thrust to provide all of the velocity changes needed for these journeys. However, the potential exists to use an aerobraking technique to slow the LTV into a low earth orbit upon its return from the moon, and thereby lower the amount of propellant the craft must carry.
This study attempts to assess the feasibility of modifying the baseline rocket-only LTV to use aerobraking. Preliminary aerodynamic, trajectory, aeroheating, and thermostructural analyses have been performed for two potential aerobraking configurations. The techniques used and the results from these analyses are discussed in this paper. Comments on the feasibility of an aerobraking LTV and comparisons with the baseline LTV are then made.
The Artemis Society is a non-profit organization dedicated to establishing a permanent manned lunar base within the next decade using private funding. To accomplish this, the society intends to "sell the project as a spectacle" and use the profits from a merchandising campaign to include movies, magazines, toys, coffee mugs, and anything else that could be sold to help the cause1.
A summary of the current reference mission is as follows. The lunar base habitat will be taken into Low Earth Orbit (LEO) on a Space Shuttle flight and "parked" at the International Space Station or at an independent orbital staging base. The habitat itself will consist of a triple-length SPACEHAB module, similar to the SPACEHAB module flown on the Space Shuttle. A subsequent shuttle flight will transport a Lunar Transfer Vehicle (LTV) and astronauts to the station, where the LTV and the habitat will be mated as shown in Figure 1. The mated stack will then leave the station, and the LTV will fire its liquid oxygen/liquid hydrogen (LOX/LH2) rocket engine to take the stack into a free-return lunar trajectory similar to those used by the Apollo program. Once in orbit around the moon (after another engine firing), the two craft will separate and the Lunar Habitat and astronauts will descend to the lunar surface. After operations on the surface are completed, the astronauts will ride back up into lunar orbit on a small ascent stage and rendezvous with the orbiting LTV. The LTV, with astronauts on board, will then fire its engine to return to earth. Upon arrival at earth, the LTV is to fire its engine once more to slow the craft and place it into LEO, where it will eventually rendezvous with the shuttle and the space station1.
An alternative to having the LTV fire its engine to slow into LEO upon returning from the moon is to have the craft enter the earth's atmosphere briefly. The LTV would emerge from the atmosphere after it had shed enough velocity to reach the space station at the apogee of its exit orbit. Once at the apogee of its new orbit, a small firing of the LTV's engines would be necessary to circularize its orbit. The potential fuel savings from this technique, known as aerobraking, are substantial. The velocity change, or delta-v, required to slow the LTV into LEO is about 10,300 ft/s. If a rocket-only approach is used, all of that delta-v must come from expending fuel. But if an aerobraking technique is used, the earth's atmosphere can provide almost all of the delta-v needed, and the result is that only several hundred ft/s of delta-v must come from expending fuel. However, the fuel weight savings are tempered by the need for increased structural weight to support flight in the atmosphere, as well as by the need for a vastly improved (and therefore heavier) thermal protection system (TPS).
The purposes of this project were to study the feasibility of using an aerobraking technique to slow the LTV into a low earth orbit upon its return from the moon and to see if aerobraking offers any performance advantages. In so doing, it was desired to keep the LTV as close to its current configuration as possible to keep costs down. Please note that this study was not an attempt to design an optimal aerobraking lunar transfer vehicle. It should also be noted that the use of aerobraking was not constrained to occur on the first flight of the LTV. It was envisioned that subsequent flights could provide an enhanced TPS from earth, or that a permanent lunar settlement might be capable of making disposable TPS materials from lunar resources.
The LTV model used in this study is shown in the bottom right corner of Figure 1. The main pressurized section of the LTV consists of a single SPACEHAB module and is a cylinder 13.5 ft in diameter and 10 ft long, sliced off on one side with a chord that spans 94°. The flat side of the LTV contains windows, so any aerobraking configuration needs to make that the leeward side of the craft. On the front of the LTV there is a docking port. At the rear there is a rocket motor that extends about 4 ft. Distributed around the rear of the craft are fuel tanks of an as yet unspecified size. The weight of the craft, without an engine, tanks, or a heat shield is estimated to be 4,000 lb. For both concepts considered in this study, a faring extending 2 ft from the rear of the craft was added to help protect the engine and tankage2.
This study looked at the use of two aerobraking vehicle concepts flown with three different profiles. For both concepts, an aerobraking configuration was created in and aerodynamically analyzed by the Aerodynamic Preliminary Analysis System (APAS) software. The aerodynamic data were then used in a planar trajectory simulation. Output from the trajectory simulation was fed into the MINIVER aerothermal analysis code to determine the aerodynamic heating rates experienced by each concept. Each of these analyses will be discussed further in turn.
It should be noted at this point that the weights of both aerobraking LTV concepts initially were not known with much certainty. The weight of the LTV itself, the engine and fuel tanks, lunar return cargo, and any additional weight imposed by a heat shield and/or structural modifications necessary to support aerobraking were all subject to a great deal of uncertainty. As a result, each of the configurations were analyzed over a range of entry weights: 7500, 10000, 12500, and 15000 lb. (Also, please note that throughout this paper masses will be referred to as weights measured on the surface of the earth.)
The two different aerobraking LTV configurations are shown in Figures 2 and 3. Both are based loosely on concepts for aeroassisted orbital transfer vehicles found in references 3 through 5. The first configuration, shown in Fig. 2, has a mostly hemispherical nose cap and aft engine/tank skirt added to the basic LTV geometry. The hemispherical nose cap serves to take the worst part of the aerodynamic and aero-thermodynamic loads. The aft skirt serves to protect the fuel tanks and engine from the same loads. The nose cap would probably have to be disposable, and in that case it would be jettisoned after a single pass through the atmosphere. This configuration will hereafter be referred to as the hLTV.
The second configuration, shown in Fig. 3, has a sphere-cone (65° cone angle) shaped heat shield which extends to a 9 ft. radius from the centerline. (The LTV itself at most only extends 6.75 ft from the centerline.). This heat shield, like the hLTV's nose cap, will also probably be disposable. Though this shield is designed to take almost all of the (aero/thermo)dynamic loads, a skirt for the tanks and engine is still required. This configuration will henceforth be referred to as the scLTV.
The Aerodynamic Preliminary Analysis System (APAS) software was used to perform aerodynamic analyses of the configurations. APAS is a FORTRAN code developed by Rockwell and used by NASA Langley, among others. It allows the user to input a wide range of vehicle geometries, and can then analyze those geometries over conditions ranging from low subsonic to hypersonic flight. Given a set of test conditions (such as Mach number, altitude, center of gravity location, surface roughness, rotation rates, and angle of attack or sideslip sweeps), APAS can provide the user with all six aerodynamic force and moment coefficients. APAS has been shown to be a effective tool for aerodynamic analysis in engineering-level design studies. For the hypersonic regimes encountered in this study, APAS uses hypersonic impact methods coupled with empirical skin friction methods to perform its analysis6.
The aerodynamic data derived from APAS are displayed in Tables 1-3. Each of the tables shows the trimmed values of lift and drag coefficients, as well as the trim angle of attack over a range of Mach numbers that encompasses the atmospheric portion of the flight. The data were derived using a reference area of 135 ft2, a reference length of 10 ft, and a center of gravity location 6 ft from the front of the original LTV, along the axis of the cylinder.
Table 1 shows the data for the hLTV. This configuration aerodynamically trims at a small (<1°) positive angle of attack throughout the Mach number range. This is unfortunate for two reasons: 1) it means the craft can only produce a small amount of lift to help control its flight; and 2) such a small trim angle will offer very little protection from the aerothermal environment to the leeward side of the craft. The vehicle center of gravity location prevented trimming at a higher angle.
Tables 2 and 3 show the aerodynamic data for the scLTV at zero and nonzero trim angles of attack, respectively. The zero angle of attack data show that this craft produces substantially more drag than the hLTV because of its larger frontal area, and also produces no lift. The nonzero angle of attack data show that the vehicle's trim angle varies with Mach number over a range of positive angles. These positive trim angles will help to protect the leeward side of the craft, but also produce a substantial negative lift. This negative lift is not detrimental, since the craft can be banked to fly with its lift vector pointing in any direction (up, down, or sideways) perpendicular to the drag as a means of controlling the its flight path.
Since the methods used by APAS are approximate, there is some uncertainty associated with the values obtained for the aerodynamic coefficients. Viscous interaction effects often erode the lift of hypervelocity flight vehicles at very low densities and high velocities (during the transition between free molecular and continuum flows), and no attempt was made to model these effects in this study. However, the effect of viscous interactions are somewhat minimized here because the flow near the flight's perigee, where the aerodynamic forces are the largest, is likely to be purely continuum flow. (Continuum flow is where the methods used by APAS are valid.) Also, the drag coefficients are seen to decrease as the Mach number increases, probably because some of APAS' methods do not assume a hypersonic Mach number independence. The large number of significant figures used in presenting the data is not meant to imply a great certainty. It is meant to provide the actual numbers used in the simulation discussed below.
The trajectory analyses for the aerobraking LTV configurations considered in this study were done in three phases. The first phase uses two-body orbital mechanics to analytically compute the trajectory from just after the LTV has left the moon's gravitational influence (assumed to be at the moon's mean orbital radius) until it reaches the earth's atmosphere (assumed to start at 400,000 ft altitude). The second phase uses a numerical integration of the equations of motion for planar flight to simulate the atmospheric portion of the LTV's flight. The final phase once again uses two-body orbital mechanics to calculate the trajectory from atmospheric exit (again at 400,000 ft) to apogee of this transfer orbit, where the burn required to circularize the orbit is computed. The goal is to have the LTV end up in the same orbit as the Space Station, which is assumed to be in a circular orbit at about 220 nautical miles (nmi) altitude. The initial conditions of the orbit are specified by assuming the LTV is at the apogee of its earth orbit with a velocity of 650 ft/s after leaving the moon's sphere of influence, and by varying the (two-body) orbit perigee height until the proper final conditions (a 220 nmi circular orbit) resulted.
The equations of motion governing planar flight over a spherical, non-rotating planet are shown in Figure 4, and were taken from reference 7. These equations formed the basis for the trajectory simulation program, called STAMP, written in FORTRAN to analyze the aerobraking portion of the LTV's flight. STAMP requires the user to specify a set of starting conditions (velocity, altitude, and flight path angle) and then numerically integrates the equations of motion using a fourth order Runge-Kutta technique until a terminal condition (altitude > 400,000 ft) is reached. STAMP also requires the user to specify the aerodynamic lift and drag coefficient, CL and CD, as a function of Mach number, and then linearly interpolates to find the current value. STAMP uses an atmospheric model based on the 1959 standard atmosphere. The mass of each LTV was assumed to be constant throughout its atmospheric flight, though in reality the mass would likely decrease slightly due to heat shield ablation and fuel expended by the reaction control system.
Many different LTV trajectories were computed during the course of this study. Tables 4, 5, and 6 summarize some important statistics for each LTV weight from the trajectories of the hLTV, zero lift scLTV, and lifting scLTV, respectively. Each table shows the two-body perigee height, hp, the atmospheric flight time, the maximum axial acceleration (g-loading) and dynamic pressure encountered, the apogee and perigee altitudes of the transfer orbit, and the delta-v and fuel mass (weight), mf, required to circularize the transfer orbit at its apogee for each of the four LTV weights. (Table 6 also shows the flight path angles at atmospheric entry, which will be discussed shortly.) The fuel masses were derived assuming an engine specific impulse, or Isp, of 420 s.
A few things from Table 4 bear mention. All of the trajectories spend between 6 and 7 minutes in the atmosphere, all have relatively modest peak g-loadings (3.5-3.67 g), and require small (<450 ft/s) circularization burns when compared with the 10,300 ft/s needed for a direction injection into LEO. As one might expect, the heaviest LTV needs to go the deepest in the atmosphere to slow itself into the proper transfer orbit. As a result, the 15,000 lb LTV spends the most time in the atmosphere (400 s), has the highest peak dynamic pressure (meaning the largest aerodynamic forces) (436 lb/ft2, or psf), will likely experience the greatest aerodynamic heating, and because of its weight consumes the most fuel during circularization. Unfortunately, the low amount of lift generated by this configuration leads to both control and heating problems, and caused this configuration to be only partially analyzed.
Table 5 shows the trajectory statistics for the scLTV at zero degree angle of attack. Since this configuration has much more drag than the hemisphere nose LTV, it is reasonable to expect that it does not require as deep a pass through the atmosphere, which is indeed the case. This results in slightly reduced g loadings (3.35 vs. 3.61 for the 10,000 lb, or 10 klb, LTV), slightly increased atmospheric flight times (425 vs. 389 s), and substantially lower dynamic pressures (85 vs. 300 psf). It also results in less elliptic transfer orbits, which in turn lead to a small reduction in the circularization burns (373 vs. 435 ft/s).
These zero lift scLTV trajectories are highly sensitive to the flight path angle at atmospheric entry, ge. Errors in the entry angle of 0.05 degrees will send the craft crashing to earth (too steep) or out to an orbit with a apogee of 20,000 nmi (too shallow). This minuscule margin for error was deemed unacceptable. As a result, work using the zero lift scLTV was discontinued. (A similar problem resulted from the very low lift of the hLTV, and was one of the reasons why work on it was not carried much further.)
Table 6 is somewhat different from the previous two tables. It contains data for three different trajectories for each weight. The first entry for each weight assumes that the LTV is banked so that there is no in-plane (longitudinal) lift. The second assumes that the LTV is oriented to produce a maximum positive (upward) in-plane lift, and the third assumes the LTV produces a maximum negative (downward) lift. Table 6 also contains a column displaying the entry flight path angle for each trajectory.
The reason so many different trajectories were performed for the lifting scLTV was to establish some idea of the margin for error that could be tolerated in the entry flight path angle. As discussed earlier, shapes with very little or no lift have extremely small error tolerances. The lifting scLTV however, can be banked to compensate for entry errors, atmospheric variations, aerodynamic estimation errors, or to change inclination if desired. If the craft enters too steeply, it needs to produce some positive in-plane lift. If it enters too shallow, it needs some negative in-plane lift. In all cases, a guidance system will need to direct bank angle changes during atmospheric flight to ensure that the craft attains the proper inclination upon atmospheric exit.
The zero in-plane lift trajectories in Table 6 are similar to those found in Table 5 for the zero angle of attack scLTV. They are not identical because flying at a nonzero angle of attack actually reduces the drag slightly. The positive lift trajectories are substantially more stressful than the no lift trajectories. They go much deeper into the atmosphere in a shorter time, and as a result incur higher axial g-loadings (5.74 vs. 3.34 for the 10 klb case), have higher dynamic pressures (158 vs. 93 psf), and yield more elliptic transfer orbits that take stronger burns to circularize (576 vs. 378 ft/s). The positive lift trajectories also have nontrivial transverse accelerations. As one would expect, the negative lift trajectories are longer and less stressful than the no lift trajectories, with lower g-loadings (1.87 for 10 klb), lower dynamic pressures (54 psf), and less elliptic transfer orbits taking weaker burns (317 ft/s) to circularize).
The point worthy of most note about these trajectories, though, is the difference in entry flight path angles. In every case, the difference between the extremum values is about 1.1o. That means, assuming no other anomalies occur, that the guidance system only needs to be accurate to within 1.1o (in entry angle) for the LTV to exit in a nominal transfer orbit, which is certainly much more feasible than aiming for a target much less than 0.1o wide.
A plot of the three lifting 10 klb scLTV and hLTV trajectories has been prepared and is shown in Figure 5. This plot serves to provide a visual representation of the differences between them.
The program MINIVER was used to analyze the effects of aerodynamic heating that the LTV would experience. MINIVER is a FORTRAN code developed for NASA Langley and consists of an aeroheating code, LANMIN, and a thermostructural code, EXITS. LANMIN can perform a series of point analyses of the convective heating a vehicle experiences, provided the user specifies the shape of the region being analyzed, the trajectory followed, and the atmospheric model to be used. It uses a variety of largely empirical methods that have been validated extensively using data from Space Shuttle and other atmospheric reentry flights8.9.
In hypervelocity flight, such as is encountered upon returning from the moon, one often finds that equilibrium radiative heating of the vehicle by the atmosphere often dominates the convective heating. Several texts were consulted, including reference 10 (Figure 7-1). Afterward, it was concluded that, based on the flight profiles and nose radii used in this study, equilibrium radiative heating would likely account for no more than 10-15% of the total heat load experienced by the vehicles. It was also recognized that nonequilibrium effects due to the extreme high altitudes of the flights could also contribute to the heating. Therefore, it was decided to account for the radiative heating and nonequilibrium effects encountered by the vehicles through the use of a safety factor.
The results from the aeroheating portion of MINIVER (LANMIN) are shown in Table 7. This table shows the maximum values of radiation equilibrium temperature, heat rate, and pressure, as well as the total heat load at the stagnation point of the scLTV for each of the lifting trajectories discussed above. All of these values are derived using the 1962 U.S. standard atmosphere and assuming a radiative (gray body) emissivity of 0.8. In order to provide an indication of the severity of the aerothermal environment, the values in Table 7 also assume that no heat shield ablation takes place. (The high peak temperatures largely rule out the use of reusable materials in the heat shield, though.) As expected, in general the heavier the craft, the higher the heating parameters are. The one exception to this rule is the 12.5 klb and 10 klb shallow (max - lift) trajectories. Here the 10 klb craft has a slightly higher peak heating rate (48.96 vs. 47.61 Btu/ft2/s) and peak temperature (2907 vs. 2884 °F). This anomaly is probably due to the large time steps needed between outputs in MINIVER for these long trajectories. (MINIVER can only handle a total of 50 time steps.) It is likely that the 12.5 klb output missed the peak of the heating by a few seconds, while the 10 klb output was very close to the peak.
Table 7 also shows the same data for the 10 klb hLTV. Due to its smaller nose and lower drag, this craft experiences more severe heating than any of the scLTV trajectories. When it became clear that the hLTV was not a viable configuration, work on it was halted. No further discussion of the hLTV will be contained in this paper.
It is worth noting the differences between the 3 trajectories for a given weight. The deep entry (max + lift) trajectory experiences the highest peak heating rate (68.2 Btu/ft2/s for the 10 klb case) and temperature (3197 °F), but it also experiences the lowest total heat load (6,500 Btu/ft2). The shallow entry has the lowest peak heating rate (49 Btu/ft2/s) and peak temperature (2907 °F), but has the highest total heat load (11,090 Btu/ft2). These variations are probably due to differences in length of time spent inside the atmosphere. The deep trajectory accomplishes its slowdown very quickly and absorbs its heat quickly, while the shallow trajectory slows down much more gradually, but picks up more total heat since it spends almost three times as long (729 vs. 263 s) in the atmosphere. This increased total heat load combined with the use of an ablative material made the shallow entries the limiting cases for the design of the Thermal Protection System (TPS).
The thermostructural part of MINIVER (called EXITS) uses the output from LANMIN and finite difference one-dimensional unsteady heat transfer techniques to compute the temperature distribution through the thickness of the TPS material. In the case of an ablative material as was used here, EXITS tracks the amount of material lost to ablation/sublimation. The material chosen for use was SLA-561. It is an elastomeric silicone and was chosen because it was one of the few materials in EXITS' material database that seemed possible to make on the moon. This material has an emissivity of 0.9 (see ref. 12 for property data), which is slightly higher than the 0.8 assumed above and helps reduce the heating slightly. For all of the cases run, the objective was to keep the inner wall of the TPS below 250 °F, which was taken to be the maximum operating temperature of aluminum. (The support shell underneath the TPS was assumed to be aluminum, and was assumed to have no heat capacitance of its own to provide an additional margin of safety.)
The results from EXITS are summarized in Table 8. This table shows the stagnation point TPS thickness, average TPS thickness over the heat shield, and the estimated weight of the heat shield TPS (which includes a safety factor of 20%) for each of the shallow entry (max - lift) trajectories discussed above. EXITS was used to obtain the required stagnation point TPS thicknesses for each of the 4 cases, and for the thickness near the leeside edge of the shield for the 10,000 lb case. For that case, the shield turned out to be 0.35 in. thinner on the edge than at the stagnation point. The average of these two values was used to computed the average thickness, and the average thickness used to compute the weight. Each of the other average thicknesses and weights were obtained assuming that all of the shields would turn out to be 0.35 in. thinner at the edge than at the stagnation point. It is interesting to note that the difference in heat shield TPS weights between the heaviest and lightest craft is only 150 lb. This is because each craft must dissipate a similar total amount of heat.
In addition to the weight of the heat shield TPS, there will be other weight penalties an aerobraking LTV must pay. There must be a support structure for the heat shield, some additional reusable TPS on the rest of the vehicle, fuel for orbit circularization, and possibly increased structural strength in the basic vehicle. Each of these items will now be discussed in more detail for the 10 klb LTV.
The support structure for the heat shield TPS will likely consist of some type of stiffened shell to support its shape and a truss frame that attaches to the LTV. The shell is assumed to be made from an aluminum honeycomb skin with reinforcing stiffeners. The truss will attach to some of those stiffeners on one end, and on the other end it will attach to the region around the LTV docking port already reinforced to handle docking maneuvers. The pressures provided by MINIVER, while not exceedingly large, are mostly concentrated on the heat shield. As such, the truss must be strong enough to transmit almost all of the axial and bending loads the craft experiences. Based on data found in reference 13, the shell and stiffeners were estimated to weigh 1 lb/ft2 (of heat shield area) for a total weight of 275 lb, and the truss was estimated to weigh 125 lb. This gives a combined weight for the 10 klb scLTV of 400 lb.
A few additional MINIVER runs were performed to try to estimate the TPS needs for the rest of the LTV. It is likely that all of the leeward side of the craft, as well as the front of the windward side, will be shielded from most of the direct heating by the flowfield. For these regions, calculations from MINIVER suggest that a blanket of TG-15000 should be sufficient protection. TG-15000 is a newly developed lightweight composite blanket of glass fibers in a silicone resin. (See ref. 12 for TG-15000 property data.) It is also likely that the aft end of the windward side of the LTV will be in the flowfield since the craft flies at an angle of attack. For these regions MINIVER results suggest that a combination of AETB-8 ceramic tiles and Toughened Unipiece Fibrous Insulation (TUFI), developed as replacements to the Space Shuttle's underside tiles could be used for protection. (See ref. 14 for property data.) Unfortunately, it is very difficult how much of the craft will be in the flowfield without more sophisticated analysis techniques. For this reason, it has been decided that 2/3 of the LTV will be covered with a 0.4 in. thick TG-15000 blanket and 1/3 will be covered with .75 in. thick AETB-8/TUFI combination. Both are more protection than the shuttle would require. The resulting additional weight is estimated to be 105 lb for the 10 klb case.
The fuel required for circularizing the LTV's transfer orbit was estimated to be about 418 lb for the worst case (deep entry) 10 klb trajectory. This number contains some built-in margin for error, since the Isp assumed was 420 s, a low number for a LOX/LH2 engine. There will also be additional fuel required to operate the reaction control system during atmospheric flight, which is estimated to be 50 lb.
The additional structural weight imposed on the LTV to perform aerobraking is perhaps the most difficult quantity to estimate. Very little data was available on what went into the initial LTV weight estimate. It is known that if the LTV is to be launched on the Space Shuttle, then it will have to be able to withstand at least a 3-g axial acceleration load. If an expendable launch vehicle is used instead, the LTV would likely have to withstand at least 4 or 5 g's. The maximum axial load encountered in the 10 klb LTV trajectories was 5.74 g in the deep entry trajectory. Therefore, it is likely that some axial structural reinforcement will be necessary. Some of the outer skin of the LTV will likely also have to be reinforced to withstand atmospheric pressures, though results from MINIVER indicate that the pressures on the LTV itself should be rather low. The estimate for necessary structural reinforcements to the LTV is 1,350 lb about 33% of the total initial LTV weight. This number is admittedly little better than an educated guess.
A summary of the additional weights due to aerobraking is included as Table 9. The total weight is just under 3,350 lb.
Now that all the data have been presented, some comments are in order. First of all, the scLTV configuration developed here is by no means an optimum design. There are a number of things that come to mind that could be done to decrease its weight further. Foremost among these is to decrease the margin for error in the entry angle. If it is possible, reducing the amount the entry angle can be too shallow will decrease the heat shield TPS weight, since the total heat load will decrease. Likewise, scaling back the amount the angle can be too deep will reduce the structural loading on the vehicle, and decrease the structural reinforcement weights. Another idea to reduce the overall system weight is to change the size of the heat shield. The choice for this heat shield's size was rather arbitrary, and there is likely a better size which balances the TPS and structural reinforcement weights in a more optimal way. There are also other ideas for aerobraking configurations, such as inflatable aerobrakes (ballutes), which could potentially reduce an aerobraking LTV configuration's weight.
Now, a comparison of an aerobraking vs. a rocket-only LTV will be made. If it is assumed that the unmodified LTV weighs 6,650 lb (to correspond roughly with the 10 klb case) with the proper amount of fuel to complete a station rendezvous after LEO insertion and that a direct LEO insertion burn is 10,300 ft/s, then the LTV will require about 7,600 lb of propellant (using the rocket equation with an Isp of 460) for a direct rocket insertion. The 10 klb aerobraking scLTV would then result in a reduction of about 4,250 lb (10,000 lb vs. 14,650 lb unmodified) that must be given a trans earth injection (TEI) burn. If this result is cascaded backwards one more step, using aerobraking results in a savings of almost 5,400 lb over rocket braking, assuming the TEI burn is 3200 ft/s. This means that 5,400 lb more payload could be delivered to a lunar orbit if aerobraking were used.
A payload increase of 5,400 lb sounds very attractive. (The original payload is about 31,200 lb, so this would be an increase of about 17%.) Even if the structural weight estimates are greatly in error (which they admittedly could be), there is likely still going to be some payload increase. There is, however, literally a price to be paid to gain this payload increase. It is relatively easy and cheap to make larger fuel tanks and carry more fuel, but there will be a whole new realm of development costs to be paid for an aerobraking vehicle. Extensive aerodynamic, aero-thermodynamic, and structural testing will have to be performed on any aerobraking configuration, which could markedly increase the total development costs. In addition, a more complex (and therefore costly) control system will be required to control the atmospheric phase of flight.
In addition to increased cost, there are a number of other disadvantages to using this aerobraking LTV configuration. One is the bulky size of the heat shield. Something this size would not fit into the shuttle's cargo bay in one piece, and on-orbit assembly could prove problematic. If it was made on the moon, bringing it up from the surface would be difficult with the current design for the ascent stage. There has also been talk of putting some instruments or solar arrays on the outside of the LTV. If aerobraking were used, these instruments or arrays would have to be stored somewhere during atmospheric flight, which may not be possible. In addition, reaction control system thrusters could probably be put only on the leeward side of the craft, since they require some protection from atmospheric heating.
This study has been an attempt to asses the feasibility of aerobraking LTV, and to determine to a first approximation if there are any performance advantages to using aerobraking. It does indeed appear that an aerobraking LTV is at least theoretically feasible and that there are performance advantages. Whether these advantages are worth their price remains to be seen.
Preliminary aerodynamic, trajectory, and aeroheating analyses have been performed for two potential aerobraking LTV configurations. The techniques used and the results from these analyses have been discussed in this paper. Comments on the feasibility of an aerobraking LTV and comparisons with the baseline LTV have been made. The following conclusions are drawn:
1) The hemisphere nose LTV does not appear to be an acceptable aerobraking LTV configuration with its center of gravity in its current location.
2) The lifting sphere-cone nose LTV (scLTV) experiences moderate aerothermal and structural loads, and would appear to be a reasonable aerobraking LTV configuration.
3) The lifting scLTV appears to offer a substantial lunar payload increase over a rocket-only LTV.
4) The lifting scLTV has a number of drawbacks, including increased cost and heat shield transportability concerns, which may overshadow the benefits of increased payload capacity.
The author wishes to acknowledge the following people for their help: Mr. Gregory Bennett of The Lunar Resources Company for his help obtaining the LTV specifications; Ms. Kathyrn Wurster of NASA Langley Research Center for her help in using MINIVER; and Dr. Frederick Lutze of Virginia Tech for his help debugging the trajectory simulation.
2. Bennett, Gregory. Personal correspondence, December 1995-March 1996.
3. Powell, Richard W., Stone, Howard W., and Naftel, J. Chris, Performance Evaluation of the Atmospheric Phase of Two Aeromaneuvering Orbital Transfer Vehicles, AIAA Paper 84-0405, January 1984.
4. Wilhite, Alan W., Arrington, J. P., and McCandless, R. S., Performance Aerodynamics of Aeroassisted Orbital Transfer Vehicles, AIAA Paper 84-0405, January 1984
5. Talay, Theodore A., White, Nancy H., and Naftel, J. Chris, Impact of Atmospheric Uncertainties and Viscous Interaction Effects on the Performance of Aeroassisted Orbital Transfer Vehicles, AIAA Paper 84-0408, January 1984.
6. Engelund, W. C. and Ware, G. M., Aerodynamic Predictions and Experimental Results for an Advanced Manned Launch System Orbiter Configuration, AIAA Paper 93-1020, February 1993.
7. Griffen, Michael D. and French, James R., Space Vehicle Design, AIAA Education Series, 1991.
8. Engel, Carl D., and Praharaj, Sarat C., MINIVER Upgrade for the AVID System, Volume 1: LANMIN User's Manual, NASA Contractor Report 172212, August 1983.
9. Wurster, K. E., Zoby, E. V., and Thompson, R. A., Flowfield and Vehicle Parameter Influence on Results of Engineering Aerothermal Methods. AIAA Paper 89-1769, September 1989.
10. Howe, John T., Hypervelocity Atmospheric Flight: Real Gas Flowfields, NASA Reference Publication 1249, November 1990.
11. Pond, John E., and Schmitz, Craig P., MINIVER Upgrade for the AVID System, Volume 3: EXITS User's and Input Guide, NASA Contractor Report 172214, August 1983.
12. Williams, S. D., and Curry, Donald M., Thermal Protection Materials NASA Reference Publication 1289, December 1992.
13. Scott, C.D., Ried, R. C., Maraia, R. J., Li, Chien-P, and Derry, S. M., An AOTV Aeroheating and Thermal Protection Study, AIAA Paper 84-1710, June 1984.
14. Chiu, S. A., and Pitts, W. C., Reusable Surface Insulations for Reentry Spacecraft, AIAA Paper 91-0695, January 1991.
|Mach Number||Trim Angle of Attack||CLtrim||CDtrim|
|Mach Number||Trim Angle of Attack||CLtrim||CDtrim|
|Mach Number||Trim Angle of Attack||CLtrim||CDtrim|
|Mass (lbm)||hp (ft)||Time (s)||Max axial g's||Max dyn press (psf)||Transfer Orbit (nmi>||DV to Circ. (ft/s)||mf to Circ (lb)|
|15,000||185,500||400.6||-3.50||435.5||221.55 x -25.90||442.7||483.9|
|12,500||190,275||395.2||-3.55||368.6||221.31 x -24.49||439.7||400.5|
|10,000||195,975||388.9||-3.61||299.8||220.13 x -23.04||435.0||317.0|
|7,500||203,125||380.5||-3.67||230.2||221.73 x -20.27||432.5||236.5|
|Mass (lbm)||hp (ft)||Time (s)||Max axial g's||Max dyn press (psf)||Transfer Orbit (nmi>||DV to Circ. (ft/s)||mf to Circ (lb)|
|15,000||22,715||435.3||-3.28||123.7||224.89 x 10.00||381.7||418.1|
|12,500||231,125||430.2||-3.31||104.4||224.80 x 11.14||379.5||346.4|
|10,000||235,850||424.8||-3.35||84.7||221.76 x 12.03||372.7||272.3|
|7,500||241,750||415.8||-3.40||65.3||225.82 x 14.27||375.4||205.7|
|Mass (lbm)||hp (ft)||Max Axial Accel (g's)||Max Dyn Press (psf)||Transfer Orbit (nmi)||Delta-V to Circ. (ft/s)||mf to Circ. (lb)||Entry Angle Gamma-e (deg)|
|15,000 (Zero Lift)||224,925||437.0||-3.26||135.8||223.01 x 8.78||380.8||417.2||-5.155|
|15,000( Max + L=16,750 lb)||170,450||276.3||-5.37||217.9||220.43 x -94.83||571.2||621.4||-5.866|
|15,000 (Max - L=-6,500 lb)||249,044||730.5||-1.87||80.0||226.61 x 40.85||328.0||360.1||-4.786|
|12,500 (Zero Lift)||228,925||431.0||-3.30||114.8||226.75 x 10.51||383.9||350.4||-5.095|
|12,500 (Max + L=14,500 lb)||176,950||269.9||-5.48||189.3||220.77 x -95.80||573.7||520.0||-5.820|
|12,500 (Max - L=-5,475 lb)||252,765||731.3||-1.87||67.0||223.2 x 41.50||321.1||293.8||-4.726|
|10,000 (Zero Lift)||233,675||425.5||-3.34||93.4||224.24 x 11.48||377.9||276.0||-5.024|
|10,000 (Max + L=12,200 lb)||181,100||262.5||-5.74||158.4||220.07 x -97.79||576.3||417.9||-5.766|
|10,000 (Max - L=-4,450 lb)||257,178||729.2||-1.87||53.9||221.69 x 42.27||317.1||232.2||-4.654|
|7,500 (Zero Lift)||239,600||417.5||-3.38||71.9||224.46 x 13.17||375.0||205.5||-4.933|
|7,500 (Max + L=9,600 lb)||186,400||252.8||-6.00||124.3||224.08 x -96.84||581.2||316.0||-5.695|
|7,500 (Max - L=-3,400 lb)||262,654||720.9||-1.88||40.8||224.22 x 43.22||319.7||175.5||-4.564|
|Case||Max Rad. Equil. Temp (deg F)||Max Heat Transfer Rate (Btu/ft2/s)||Total Heat Load||Max Surface Pressure (psi)|
|15,000 (Zero Lift)||3084||60.04||8,514||193.9|
|15,000 (Max + Lift)||3287||75.03||6,507||315.7|
|15,000 (Max - Lift)||2884||47.61||11,910||114.9|
|10,000 (Zero Lift)||2996||54.29||7,786||157.2|
|10,000 (Max + Lift)||3197||68.21||5,848||265.7|
|10,000 (Max - Lift)||2907||48.96||11,090||93.82|
|7,500 (Zero Lift)||2901||48.62||6,880||122.9|
|7,500 (Max + Lift)||3101||61.32||5,086||209.8|
|7,500 (Max - Lift)||2848||45.57||10,270||72.15|
|Case||Stag Pt TPS Thickness (in)||Avg Heat Shield TPS Thickness (in)||Heat Shield TPS Weight (lb)|
|Heat Shield TPS||1,021|
| Heat Shield |