STUDY SUMMARY: THE APPLICABILITY OF SOLID PROPELLANTS FOR A NOVA-CLASS INJECTION VEHICLE
AND A COMPARISON WITH A LIQUID VEHICLE OF COMPARABLE CAPABILITY
JPL: California Institute of Technology, Pasadena, California
NASA CR 136573; N74-72160; 9 May 1962

FOREWORD

Early in 1961, JPL began studying injection vehicle aspects of direct ascent approach for the manned lunar landing mission. At the invitation of NASA HQ, preliminary results of this all-solid NOVA study were presented to the joint NASA/AF Large Launch Vehicle Program Group (generally known as the Golvin Committee) on 3 August 1961. This committee, recognizing the unconventional features of the concept, contracted with Boeing and Space Technology Labs, Inc. to make independent evaluations of the solid NOVA studies reported in JPL's Technical Memorandum 33-52 and Addendum A thereto. It was planned that at the end of the one-month evaluations they would submit critiques on 1) the technical feasibility of the concept, 2) the realism of the JPL estimate of schedule, 3) the accuracy of the cost estimate. Within a week the work statements had been expanded to include 1) an evaluation of the concept if payload to escape were increased from originally studied 130,000 lbs to 156,000 lbs, 2) a supplementary evaluation of the adaptability of a solid NOVA or its components as a backup for a Saturn-class vehicle, and 3) quantititave comparisons of the several reliabilities of liquid and all solid NOVA injection vehicles. On a submission of the critiques by Boeing and Space Technology Labs, the committee decided against company briefings or discussions because of other pressing matters. Approximately one month later, Mr. E Mitchell, then assistant director for Propulsion at NASA HQ, exspressed interest in results of the studies; thus on 6 October 1961 Boeing, STL, and JPL representatives provided briefings to NASA Propulsion personnel on the original solid NOVA concept and the results of two industrial critiques. Since that time, no studies have been performed. In recent months, NASA has decided, after careful deliberation, to perform the manned lunar landing mission using a lunar orbit rendezvous mode basedon a liquid propellant Saturn C-5 vehicle for the manned operations and a lunar logistics vehicle for cargo and manned support. However, the need for vehicles with a payload capability of about 500,000 lb in LEO, in the NOVA class, for the more difficult missions beyond the manned lunar landing, has also been recognized by NASA, and the new studies of all liquid and hybrid staged NOVA vehicles have been initiated by several groups. Because the lab's studies of all-solid propellant vehicle systems indicate that it is unusually promising and of general interest, this report, which gathers together all significant results of the studies, is being released as an appropriate addition to current knowledge. Only that phase dealing with the way in which the solid NOVA would fit into the manned lunar landing mission and NASA's long range plans has been omitted because it is no longer applicable. An appendix that discusses background information and the status of applicable solid rocket work has been incorporated as a convenient reference; in addition, several developments from JPL programs are discussed because they have strong bearing on the solid NOVA program under study. The opinions expressed in this report do not necessarily reflect the views of NASA.

ABSTRACT

Studies show that very large, all solid propellant injection vehicle systems in the NOVA class are technically feasible. The vehicle studied is a four-stage, solid propellant rocket having a gross weight in the 30 million pound class, a payload in LEO of 500,000 lbs, and a payload through escape of 130,000 lbs. The first three steps would inject the fourth stage into parking orbit from which the fourth stage would inject the spacecraft into the transfer orbit to the Moon or planets. Designs examined were conceptual and do not represent optimized configurations. Additional studies must be made in depth before the final system and industrial complex requirements can be specified. These new studies must include an examination of the applicability of noncryogenic liquid propellant systems for third and or fourth stages because of the potentially greater vehicle flexibility for a range of missions. When compared to liquid vehicles of equivalent performance capability, it is concluded that the risk associated with development of this vehicle system would be much lower and that the reliability of the resultant system would be predictably higher at a very early point in the flight program- provided that the proposed conservative philosophy were used. As a result of the studies, confidence in the philosophy and program approach advocated has been reinforced. From a technical standpoint, it appears that the vehicle injection system can be made available for the first flight approximately four years after the start of go-ahead, and for operational manned space missions one year after that date. Total costs are estimated to be significantly less than for competitive systems, provided that the philosophy and development approach advocated are implemented. Growth potential for the basic systems seems particularly favorable for the progressively more difficult missions which can be foreseen. There do not appear to be any major technical problem areas. However, early emphasis should be given to thrust vector control (TVC) and meeting guidance and control (GNC) as well as other subsystem reliability requirements. Combustion instability is not expected; and it is believed that even if it is encountered it would not become a serious problem.

INTRODUCTION

Prior to the time that a definitive plan could be formulated by NASA for the manned lunar landing, a number of groups undertook studies of various vehicles and modes of performing the mission. Because of its extensive experience in liquid and solid propellant technology and vehicle system development (Corporal, Sargeant, and the Explorer and Juno upper stages), the JPL voluntarily initiated some studies of large launch vehicles.

Independent of these studies, JPL activities in the Ranger, Surveyor, and Mariner projects had revealed the complexity and difficulty of even the simplest unmanned lunar or planetary missions. Thus, on reviewing the manned mission, it became evident that the total mission reliability was only a part, would be wholly inadequate unless the subsystem reliabilities were substantially better than they had been for most missile and space programs. There appeared to be at least four basic reasons for this. First, man, as a nonexpendable payload, will demand unusually high overall mission reliability-- perhaps 85%-90% probability that the mission will succeed, and 97%-99% probability that, despite a mission abort, the astronaut will be recovered. Second, the larger vehicles required wil be derived by clustering as many as 8-10 propulsion systems per stage and or using more stages. Consequently, if the design philosophy used in the past is preserved, vehicle reliability will drop below the barely acceptable values of current vehicles- particularly if they must use newly developed propulsion systems. Third, because of high vehicle cost the launching rate and production rate will be significantly less than in the past- probably 60-120 vehicles over a ten-year period. Therefore, there will be very few flights with which to develop system reliability; some means must be found to provide high reliability at a much earlier point in the flight development program. Fourth, and most important- space mission will become increasingly demanding in the required number of sequential operations that must be performed before the mission can be completed successfully. For example, a manned lunar or planetary landing mission would probably consist of the following major operations (involving perhaps a total of seventy to eighty critical steps performed in sequence): 1. First stage launch, 2. second stage operations into parking orbit, 3. third stage injection through escape, 4. midcourse propulsion and guidance corrections, 5. terminal guidance and retro-propulsion into lunar or planetary orbit, 6. descent toward the Moon or planet, 7. hovering to a soft landing, 8. lunar surface operations, 9. takeoff for return, 10. midcourse propulsion and guidance corrections, 11. Earth landing through a restricted corridor and recovery. Quite possibly there would also be, at appropriate times: 12. vehicle rendezvous (in Earth orbit with multiple launchings of components and/or rendezvous at the destination), and 13. orbital docking of the vehicle components. It must be recalled that vehicle launches into Earth orbit (only the first two of the 13 listed operations) have resulted in approximately 25-50% failures. Many of the later operations in the list of 13 must be performed in new and hostile (perhaps unknown) environment, there will be little opportunity to practice landings and takeoffs at the destination, and success of a number of the operations (ie rendezvous) is strongly time-dependent. The magnitude of the challenge may be illustrated by noting that mission reliability is an inverse exponential function of the sequence of operations.

SOLID PROPELLANT NOVA INJECTION VEHICLE SYSTEM

The injection vehicle system and its industrial support complex include the launch vehicle, its means of production, the associated transportation complex, the offshore launch pad and ground support equipment, and a range support area (Fig 1.) The spacecraft is considered only as it affects the vehicle system. Additional material in support of the study is contained in App. B.

A. INJECTION VEHICLE

The vehicle in its preliminary form was estimated to have a payload capability of 130,000 lbs to escape, or 500,000 lbs in LEO, and a launch weight of 25 million pounds. Subsequent and more sophisticated calculations show that the vehicle launch weight is closer to 30 million pounds if the vehicle has to provide the indicated payload capability. It is believed that this revision will affect size and cost only and in no way bears on the feasibility of the concept. All discussions in Sec. 3 are based on the 25 million pound vehicle.
1. DESCRIPTION AND OPERATION
The injection vehicle (Fig. 2) consists of four stages of clustered solid propellant motors of two designs. Seven type A motors for the first stage, and three identical motors make up the second stage. Six smaller, type B motors, are clustered together to form the third stage, and a single type B motor forms the fourth stage. This vehicle has a gross weight of about 25 million pounds, in Fig 3. it is shown in comparison with a 6 million pound liquid vehicle having the same payload capability. The Washington Monument model gives an indication of scale. The first three stages are designed to achieve parking orbit. After the required coasting period, the fourth stage is ignited and the 130,000 lb. payload is brought to escape velocity at burnout. A thrust vector control system is used on each stage to maintain directional control. However, no thrust termination devices are used on the large primary motors. Each stage burns to completion, and small vernier or secondary motors trim the velocity errors as the vehicle goes into its parking orbit and possibly again at stage four burnout. The verniers required are small rockets containing about 5,000 pounds to 15,000 pounds of either storable liquid or solid propellant. Their thrust termination is within the state of the art. The solid rocket motor casings, tied together in shear, serve as the primary airframe structure. Columns and truss members are used for interstage structure, which contains provision for positive separation. The diameter of the first stage assembly is 77 feet, and the height of the injection vehicle to the separation plane between the fourth stage and payload is 220 feet.
2. PERFORMANCE
Performance parameters are presented in Table 1. Three dimensional point mass trajectories were computed assuming eastward launch from AMR (Atlantic Missile Range). The first two stages were flown gravity turn after a short vertical ascent. The third stage was flown using a constant attitude in intertial space and, alternately, with constant inertial pitch. The two modes of computation gave essentially equivalent results. It should be noted that the maximum acceleration, occurring near the end of the third stage burn, is 5.3g, an acceleration tolerable by man. The mass will be distributed between the third and fourth stages such that the velocity at the end of the third stage burn is slightly short of achieving circular orbital velocity. Part of the propellant weight in the vernier, assigned to the fourth stage, is used to make up this velocity difference and insert the stage and payload into a circular LEO. The sizing was accomplished by assuming values for the specific impulse and for the stage propellant mass fraction. The primary considerations used in the sizing and trajectory design were performance capability, airloads, achievement of a parking orbit at the end of the third stage burn, and metting manned acceleration requirements.

3. PROPULSION
The motors represent a typical design concept to demonstrate the feasibility of making very large units. Detailed characteristics are shown in Table 2. Motor A is used for the first and second stages; Motor B is used for the third and fourth stages. A specific impulse (ISP) of 245 seconds at 1000 psi and sea level optimum expansion was assumed. The performance parameters assumed here are all within the present state of the art. The reproduceability of performance should exceed that of present, relatively small motors because continuous mixing techniques are contemplated and large amounts of propellant will be cast in each motor. It is believed combustion instability is unlikely to be encountered; if it does occur, modern techniques should preclude any serious delay in schedule. In recent years, powdered aluminum has been found to be an excellent suppressant for combustion instability in existing solid propellant boosters which use ammonium perchlorate solid propellants. As aluminum quantities are increased up to 15-17%, propellant performance as well as aluminum's effectiveness as a suppressant increases. None of the relatively large motors that utilize these high aluminum composite propellants (Minuteman, Pershing, Polaris) has shown any sign of combustion instability; the propellants under consideration contain these high aluminum concentrations. Unitized motors were used in this study in order to investigate ground facilities required to support this type of a design. It is not necessarily recommended that they be chosen over segmented motors. However, since new propellant processing facilities would be constructed for this vehicle, they can be designed to accommodate either approach. The final decision between unitized or segmented motors should result from a study of the effects of the choice on the flight vehicle. Experience has shown that it would be undesirable to compromise the flight vehicle because of ground support equipment requirements unless a question of feasibility is involved. Propellant would be processed in a new, continuous mix plant at a site strategically chosen for the raw material supply, its proximity to the launch site, and the practicability of shipping finished rocket motors. Propellant facilities similar to those under consideration are already in production and have demonstrated the quality of product and high production rates necessary for the program. For example, Aerojet-General Corporation recently cast a 100 inch diameter charge, weighing about 200,000 lbs, at the indicated rate, then cured and fired it successfully. It is of interest to note that the static test record, based on high frequency response instrumentation, gave no indication of combustion instability.

4. THRUST VECTOR CONTROL (TVC)
A TVC system is needed in each stage to compensate for the effects of CG displacement, thrust misalignment, and unequal thrust (Particularly at ignition and burnout). It must also counteract aerodynamic forces during first stage burn. Canting of the rocket motor nozzles so the thrust vector is directed near the burnout CG of the stage helps reduce some of these disturbances but cannot be used when the angle of cant is too large. TVC systems that could be used include jet vanes, secondary injection, moveable nozzles, auxiliary rocket motors (verniers), and jet tabs. Each has distinct advantages and disadvantages, and a choice can result only from a detailed system study considering guidance structures and spacecraft. Jet vane systems have demonstrated good flight records and offer a feasible solution to this problem. However, a gas pressurized system may prove to be inherently more reliable and was chosen for more detailed study in order to provide a basis for weight and cost estimates.

5. STRUCTURES
The largest structural weight item in the vehicle is the rocket motor casings. Fortunately, considerable experience in the design of cylindrical pressure vessels is available, and the large size considered here should introduce no problems which differ basically from those already encountered elsewhere. Consistent with the philosophy already presented, a heat-treatable martensitic steel with a yield strength of 165,000 psi was chosen in this study. Toughness and ductility shoudl be high at this strength level. However, this conclusion is tentative until more information is available for the 3/4 to 7/8 inch thick material considered for the type A motor case. The interstage structure, as presently conceived, is either a space frame or a set of three spaced columns. Loads would be transmitted to the motor cases as concentrated loads acting on truss pads attached to the motor domes or as concentrated line shears. This technique was used on the Sargeant motor case and is extensively used in the large water tanks and pressure vessels in the power generating and chemical industries. Two types of rocket nozzles were considered: 1) a graphite/steel heat-sink type, and 2) an ablative plastic type. The use of ablative nozzles appears very attractive for this application because the linear ablation rate will be the same or slightly less than that of smaller motors; therefore the percentage change in throat area will be acceptably small. Either type appears feasible; scaling laws predict less severe conditions than for existing nozzles despite the relatively long burn times, provided that weights are scaled with impulse.

6. STRUCTURAL DYNAMICS
Major structural dynamics effects such as overall dynamic loads and aeroservoelastic stability were examined qualitatively to ascertain whether extrapolation in size or weight would adversely affect the feasibility of the solid propellant NOVA. By using aeroelastic model theory, and by assuming that dynamic magnification factors associated with transit through discrete gusts are independent of vehicle size, it can be shown that a large vehicle should encounter no more severe loading in relation to its strength than a small vehicle is dynamically similar. Dynamic instability of the type characterized by adverse coupling between the autopilot and the elastic airframe can become a serious problem if a conventional flight control sensor installation is employed. However, an adaptive flight control system using "transducer arrays" (a number of rate gyros discretely positioned over the length of the vehicle without outputs electronically summed, can be employed to suppress the coupling of the autopilot with the body bending modes. Thus, low bending mode frequencies, per se, need not lead to dynamic instability problems. The solid propellant vehicle system carrying liquid payload rockets or liquid secondary injection system poses relatively minor liquid sloshing problems because of the low percentage of liquid mass.

7. AERODYNAMICS
The need to aerodynamically shroud the vehicle has been examined from the aspects of heating, drag, and unsteady flow characteristics. Maximum laminar and turbulent heating rates were examined for the velocity-altitude information obtained from a powered flight computer trajectory. A conservative analysis indicates that the heating will be no problem on exposed structural members or motor casings. An estimate of aerodynamic drag for a body of this type is not readily calculated. The examination of some experimental data and calculations based on two simple drag models indicate that the peak drag to thrust ratio is approximately 0.15, an acceptable value. Unsteady flow effects between motor casings may result in high local vibration loads. Local fairing should alleviate this condition. In none of the cases studied could a positive requirement for shrouding be determined. Consequently, only a shroud from the payload to the top of the fourth stage has been indicated. The weight of this shroud was charged to the second stage, since it would be discarded at the end of the second stage burn. The maximum dynamic pressure (max q) expected is approx. 1400 psf, and at first stage separation the dynamic pressure is approx. 400 psf. These pressures are acceptable for the vehicle under consideration. Trajectory shaping can be used to reduce the max pressure to a value of less than 1000 psf if desirable.

8. ASSEMBLY AND ALIGNMENT
With proper attention paid to details, assembly and alignment should provide no difficulty. Provision will have to be made for a temporary framework to suppor the motors of a stage during construction until the stage is structurally tied together. Erection loads on the individual motor cases will have to be considered in their design. Vehicle loads caused by wind and wave action during storm conditions while on the launch pad are expected to be small for the blunt, dense, solid propellant vehicle, and temporary guys or bracing are adequate protection. Techniques are available to provide CG control without measuring absolute weight. If further study indicates that accurate absolute weight measurement is required, an increase in the existint capability for accurately calibrating load cells is required. Load cells of the necessary size are available.

9. GUIDANCE REQUIREMENTS
The guidance of the vehicle has not been examined in detail, but it is felt that no unusual problems will be present. The general philosophy of this program can be applied to the guidance area, and adequate space with a controlled environment and ample weight for the required equipment have been provided. The weight should allow for underrating components and providing redundancy wherever needed. For example, 3,000 lbs and 350 cu. ft. of space were allowed for the guidance and control electronic compartment alone. Since it is considered that midcourse correction capacity will be required in the spacecraft, the precision and performance required of the injection vehicle are rather modest. Indeed, guidance capability equivalent to that required for military weapons is adequate; no advancement in the state of the art appears necessary.

SPECIALIZED FACILITIES
1. PROPELLANT PROCESSING FACILITY
Because of the quantity of propellant needed and the size of the loaded motors, water transport at the propellant processing facility is required. The site should be 1) convenient to Cape Canaveral (the assumed launch site) by barge transport, preferably through an inland waterway to avoid long exposure to the open sea, and 2) provide in-plant barge transport during processing. There are many islands and coastal areas from Texas to South Carolina that could meet the above requirements. A typical site, Skidaway Island, in Georgia, has been chosen only to demonstrate the feasibility of such and approach (see Fig 5 and 6). It is assumed that the island is undeveloped and that all waterways used during processing will be dredged completely. A description of the facilities is given in Table 3. The propellant materials considered are typical components of propellants, now used in large motor programs. New production capability is required for all of the ammonium perchlorate used in this program (8 million pounds per month at a vehicle launch rate of one every two months). Power requirements for this production are not excessive. The other materials required are readily available or could be made available on relatively short notice. Propellants of the type under consideration have never been known to detonate at the operating temperatures to be used. Nevertheless, since no experience at this size is available, all facilities have been sized and sited on the basis of class 9 or 10 propellant.

2. LAUNCH SITE OPERATIONS
The launch site, GSE, and assembly and transportation techniques are governed by the following criteria: 1) the complex must be in operation within 2.5 years of the commencement of the program if it is to be used for motor static firing tests; 2) the complex must be an economical and practical system for accomplishing the assembly and launch operation; and 3) the possibility of loss, because of a launch failure, of costly long lead time GSE must be minimized. The launch site is assumed to be located near existing range facilities and sources of manpower at Cape Canaveral. It is evident that because of the acoustic and explosive safety distances involved, the solid propellant could not be launched directly from the Cape. The choice of an offshore launching pad is indicated when the additional launch complex requirements are considered. In determining support systems for the assembly and launching of a vehicle of this sie, it is useful to observe the experience which exists in other areas of endeavor. Normal operations in the large civil engineering industry and in marine and naval architecture closely parallel the erection and handling techniques required for this vehicle. It is from this existing technology that the optimum solution should be derived. Fixed underwater supporting structures for the depths required are encountered in bridge foundations and dam construction. Shipment of the weights required occurs daily in normal tug and barge operations. The erection and assembly of large pieces of of equipment is within ship building and repair technology. (Indeed, the largest manmade moving objects are ships of one form or another.) Thus the optimum launch complex to meet the stated objectives is a fixed offshore launch pad supported by vessels and barges and utilizing the Cape Canaveral range and support facilities.

3. TRANSPORT AND STORAGE
Loaded solid propellant rocket motors are stored in their respective loading barges at the propellant processing site. Motors are transferred to the transport barges by the floating crane at the processing site. They are then tugged to the launch site at a rate determined by the launching schedule. Guidance, telemetry, and associated equipment is transported to the launch support area at the Cape by conventional means. After checkout and maintenance, the equipment is barged to the launch platform for installation.

4. SITE CONSTRUCTION
Since vehicle assembly and checkout procedures require four months (Fig 7), two launch pads are required to allow launchings on two-month centers. These pads (Fig 8) would be placed approximately 6-12 miles off the coast of Cape Canaveral, with a similar distance separating them for acoustic and quantity-distance safety requirements. The launch pad would then be in 60 to 100 feet of water. This depth is typical of that found in bridge foundations, and this construction is the type desired. For a solid porpellant vehicle, the dimensions of the pad need only be large enough to provide structural support for the vehicle. The large weight of the vehicle as erected would provide considerable stability against overturning due to wind and storm conditions. Hurricane protection would be limited at most to temporary guying and bracing. All auxiliary operations would emanate from the crane or support vessels. A hold-down structure would be neither desirable nor practical for use during the launching of a solid propellant vehicle. A breakwater one mile in overall length is required to shelter each pad area sufficiently to allow shipborne operations in all but gale conditions. This breakwater would be of rock construction and 10 feet above high tide. General support, electronic checkout, engineering personnel, and auxiliary equipment receiving and storage buildings would be located on or near the Cape. Several docks and a crane (YD-4) would be required to transfer this equipment to barges for transport to the launch pads. The distance offshore between the pad and the Cape has been specified as that allowing for inhabited areas, according to the mass of the propellant involved. Therefore, the launch control center would not be a blockhouse in the normal sense but, structurally, an ordinary building located on the Cape in the general support area. This center would contain the launch control instrumentation and equipment and be connected to the pads by underwater cable for phone and electrical connections. Most of the prelaunch instrumentation and monitoring measurements would be made, however, by direct radio link between the vehicle and LCC. An umbilical mast would provide electrical connections to the spacecraft up until launch, as well as emergency disarming or astronaut exit ladders. This mast would pivot at its base and drop to the water at launch.

5. VEHICLE ASSEMBLY EQUIPMENT
The primary assembly problem is the erection and handling of the first and second stage motors. A crane of 1000 ton capacity and 200 ft hook height is required, with a secondary hook of 200 tons and 300 ft height for the upper stages and payload. The construction and operation of this crane and its support structure are considerably simplified when based on waterborne operations. Currently, the largest movable crane in the US is the Navy YD-171, mounted on a self-propelled moving barge. This is one of four constructed in 1941 and has a 450 ton capacity and 160 foot hook height. Several fixed cranes of this capacity, but with lower hooks, also exist. A scaled-up version of the flat bottomed barge and crane may be specially constructed for this purpose and built to contain personnel and service areas. Its construction is considered quite feasible. The use of a floating crane for the handling and careful positioning of very large loads is well established procedure. It provides the most economical and shortest lead time solution to the erection problem and allows for convenient removal from the pad area prior to launch. Personnel service platforms around the vehicle are made in prefab sections and attached directly to the vehicle structure. The loads introduced are small compared to the load carrying capability of a solid propellant vehicle skin. Personnel access is via an elevator tower on the craneship and gangwalks across the vehicle. The platforms are removed prior to flight.

6. STATIC TESTING
The full scale test firings of the individual full size motors could be conveniently accomplished at one of the launch pads. A motor, supported by suitable structure, could be mounted in an inverted position on the pad. Normal launch control instrumentation would be used for this series of tests. The alternate pad would be used for the first flight vehicle. Upon completion of the static test program, the external support structure would be removed and this pad converted to a flight pad for the second flight firing. Alternatively, a static test stand could be constructed at the propellant processing plant and utilized for this purpose. The added area and cost for such a test stand are included in the cost analysis.

7. FIELD OPERATIONS
Individual motors of types A and B would arrive at the launch complex by barge from the propellant processing plant storage area. They would be erected and assembled on the launch pad by the crane barge (YD). Interstate structure, in large prefab sections, would be barged out and installed. The crane-barge would then be transferred to the alternate pad and replaced by the auxiliary support ship (CLS). During the remaining period, the GNC, telemetry, and spacecraft equipment together with the auxiliary gear would be installed and the final checkout operations accomplished prior to launch. The firing of this vehicle from an offshore location at the Cape presents no extraordinary problems with respec tto range procedure. Normal downrange tracking, range safety, and launch monitoring are identical to regular Cape launchings. Firing windows are reasonable and the actual final countdown time for the multistage solid propellant vehicle is relatively short compared to that for existing vehicles. The shorter burning times also result in significantly simpler tracking operations during launch into orbit than for equivalent liquid vehicles.

E. SPACECRAFT CONSIDERATIONS
Spacecraft studies have been limited to crude feasibility determinations of 1) possible configuration constraints, and 2) spacecraft design problems peculiar to the employment of the solid propellant vehicle for a manned mission. The results of the studies can be summarized as follows: 1) there are no spacecraft design problems peculiar to the solid propellant vehicle; 2) the injection capabilities of the solid propellant vehicle appear to be adequate for manned missions if so desired; 3) there are no major spacecraft configuration constraints or limitations due to the injection vehicle. Of the spacecraft environments associated with a large solid vehicle, (vibration, linear acceleration, acoustical, etc.) the one which appears to be the most severe when compared with a liquid vehicle of the same capability is the acoustical environment. Differences in the other environments are minimal. The analysis of the acoustical environment was based on extrapolations from data from smaller motors, However, the results are considered to be conservative. The calculated sound pressure levels are (solid)-=167 db, and (liquid)-=161 db. These levels correspond to a distance of 200 feet from a solid propellant vehicle of 40 million pounds of thrust or a liquid vehicle of 9 million pounds of thrust. Actually, the lower exhaust velocity of the solid propellant motor causes it to have lower acoustic efficiency, so that the pressure level from a liquid propellant vehicle might well be higher than that from a solid propellant vehicle. In any case, it is important to note the high pressure level from either. Factors such as sound absorption in the air (which increases at these higher intensities because of nonlinear damping) and directivity of the sound should decrease the levels by 20 db. Reflection from the pad coudl be minimized by flowing water under the booster at liftoff. These factors suggest taking the pressure level at 150 db. The effectiveness of ear protectors for the astronaut is limited by bone conduction to about 40 db reduction. Thus the pressure level at the ear will be reduced to 110 db, which is below the threshold of pain at 140 db. The maximum total recommended level for speech comprehension is 110 db, so that talking with astronauts during liftoff may be difficult, although additional attenuation by the cabin walls may make it reasonable. It should be emphasized that the attenuation of the sound with distance is equally important.

3. COST CONCLUSIONS
The cost study of the solid propellant NOVA injection vehicle system has revealed some rather unusual, albeit tentative, conclusions. The conclusions reached are a logical consequence of the basic program philosophies.
a. The development costs for this vehicle are considerably less than for a liquid propellant vehicle of the same capability
b. Production costs per vehicle are roughly comparable to a similar liquid vehicle, depending somewhat on the injection mission and total number of flights.
c. Injection costs in dollars per pound of payload for a 300 mile LEO for a 20 vehicle program are $169 per pound total cost and $91 per pound production cost. For these computations a reliability of 95% and 515,000 lbs of injected weight were used.
d. The development costs are low because the usual full scale development support programs are not required-- no full scale vehicle structural testing, no captive firings of clusters, no battleship propulsion program.
e. Production costs per pound for the structure are low because of the inherent simplicity of the solid propellant rocket. Size was simply exchanged for complexity at a nearly constant total production cost.

H. GROWTH POTENTIAL
Once feasibility of a vehicle is established, it becomes a major interest to examine the system flexibility for quick growth to a vehicle capable of performing much more difficult missions in the foreseeable future. A cargo version of the large spacecraft under consideration can place approx. 35,000 lbs gross weight on the lunar surface. Such devices as Moon-mobiles, prefab structures, and life support systems could be delivered directly from Earth, intact and ready to operate with no assembly or disassembly. The 30 million pound solid propellant NOVA would appear to have considerable growth potential beyond this. Substitution of liquid hydrogen stages based on engines now under development, for the third and fourth stages, such that the gross weight is unchanged, results in a vehicle that could place approx 110,000 lbs of men and equipment on the Moon. The first two, very large solid propellant stages woudl be used as developed for the original vehicle. The first three stages of this solid propellant or hybrid vehicle should be capable of placing approx. 930,000 lbs in LEO. If one were to use this weight as an electric powered spacecraft, the payload on the lunar surface using LOX-LH2 for landing on the Moon, would rise from 110,000 lbs to about 215,000 to 440,000 lbs. depending on the allowable transit time. Alternately, this three stage hybrid vehicle with its electric spacecraft should be capable of performing a three-man Mars landing and return in approx 590 days; this performance estimate is based on the radiation shielding of 50,000 lb, 15 lb/man/day sustenance, and a Mars Orbit Rendezvous of the spacecraft with the Mars Excursion Vehicle.

IV. NOVA DESIGN VERIFICATION AND COMPARISONS

The Large Launch Vehicle Program Group (the Golovin Committee), noting the unconventional aspects of the all-solid NOVA approach, requests the Boeing Company and Space Technology Laboratories to make independent and objective evaluations of the solid NOVA concept for 1) technical feasibility of the vehicle concept, 2) accuracy of the cost estimates, 3) realism of the schedule, 4) effect on the vehicle concept if the payload were increased from 130,000 lbs to a more recent estimate of 156,000 lbs, 5) flexibility of the vehicle as a backup for Saturn, and 6) a comparison of the relative reliability of liquid and solid NOVA vehicles. A review of the pertinent conclusions from the critiques (Ref 1 and 2 respectively) should be valuable at this point in confirming or rejecting the concept.

A. CRITIQUES OF THE SOLID NOVA CONCEPT BY STL AND BOEING

The salient features from the JPL, Boeing, and STL studies are compared in Table 5. In this specific comparison a vehicle with a payload capability to escape of 130,000 lbs is used because the original studies were carried out and evaluated on that basis. The comparison with liquid vehicles in section IVB, will use vehicles having an escape payload capability of 156,000 to 160,000 lbs. The critiques differ in details from the JPL designs (this was not unexpected, for they were conceptual design studies) but on the critical questions were in relatively good agreement.

1. TECHNICAL FEASIBILITY

Although vehicle weight estimates varied somewhat among the three studies, all concluded that weight per se was of minor importance and that there were no technical barriers, such as nozzle erosion, propellant physical properties, burn time, combustion instability, guidance, or vehicle dynamics that would prevent the accomplishment of the mission when using an all solid NOVA vehicle. The conclusion was found to be equally valid for a solid Nova with a 156,000 lb payload.

2. GROWTH POTENTIAL

Both Boeing and STL were quick to recognize the unusually large growth potential of the solid Nova. Each commented favorably on it as a distinct advantage.

3. FLEXIBILITY AS A SATURN BACKUP

Boeing also noted that the upper three stages of the Nova, with only a 5% penalty in the original Nova weight, would be capable of delivering 266,000 lbs into LEO, adequate for a Saturn C-5 rendezvous type boost. STL and Boeing also found that other vehicles with escape payloads of 21,000, 30,000, 40,000, and 45,000 lbs were feasible when using the A and B type motors as modules in the first three stages and an Earth storable liquid system in the fourth stage, only performance capability was checked in these cases. Boeing and STL appeared to differ regarding costs and schedules, their evaluations will be discussed separately.
A. Boeing costs and schedules: Boeing agreed closely with the cost estimates of the original JPL report for the 25 million pound vehicle discussed, ie $1.7 vs. 1.65 billion, reporting costs as realistic. In a final analysis, Boeing believed a heavier vehicle was advisable, its program costs would be about $2.2 billion dollars. A schedule of five years was considered realistic by Boeing. Some concern and possibly misunderstanding of the Golovin Committee may have arisen over apparent discrepancies between the Boeing studies on the solid NOVA and separate Boeing solid booster studies being carried out at the time for MSFC. If so, it is unlikely that the discrepancies were ever explained to the committee, STL and Boeing were not asked for briefings on their critiques when explanations might have been made. Later on 6 October, when JPL, Boeing, and STL personnel presented their results of their NOVA studies to the propulsion personnel at NASA HQ, Boeing did explain, when questions arose, that 1) the two studies were not made by different Boeing study groups, and 2) the schedule differences reported in the two studies were not inconsistent. In the Marshall case they were making a parametric study, whereas the JPL cas they had been asked to examine a specific vehicle system.
B. STL COSTS AND SCHEDULE: The STL report concluded that hte cost of vehicle system development, facilities, and 20 vehicle flights would be $4.2 billion, or approximately twice that estimated by Boeing and JPL. Their schedule too, was longer, requiring 6.5-7 years compared to Boeing and JPL estimating 5 years. During the 6 October briefing at NASA HQ, personnel from STL explained their higher costs and longer schedule on the basis of their experience with the Air Force in the ICBM program, such as Minuteman. It should be noted that the JPL method of implementing a program differs significantly from the Minuteman approach. In the briefing, however, STL did conclude that their cost estimates would be much lower than reported and their schedule would agree with the schedules of Boeing and JPL if the solid NOVA philosophy and approach could be implemented. Boeing believed the philosophy and approach could and should be implemented. Because of the different frames of reference or assumptions made in the Boeing and STL studies, one can draw a very important conclusion: Approximately $1 to 2 billion can be saved in a 20 vehicle program and 1.5 to 2 years can be cut off the 6.5 to 7 year schedule if the solid NOVA philosophy and program approach can indeed be implemented.

B. COMPARISON OF SOLID AND LIQUID NOVA CLASS VEHICLES

The comparisons will be confined to 1) vehicle and propulsion reliabilities, and 2) vehicle schedules, and 3) program costs. Most of the data gathered during the latter part of 1961, when the liquid vehicle configuration given the most serious consideration for the NOVA mission, now designated the C-8, consisting of 8 F-1 engines in the first stage, 8 J-2 LO2/LH2 engines in the second stage, and 2 J-2 engines in the third stage. Its payload capability for the escape mission assuming no engine out was approximately 160,000 lbs. Therefore, the characteristics of the solid NOVA with the 156,000 lbs escape payload, derived from the Golovin Committee supplementary work statement, will be used for comparing schedules and costs in section B2 and B3.

1. FLIGHT RELIABILITY OF LIQUID AND SOLID SYSTEMS

Although vehicle reliability as a function of liquid or solid NOVA time schedules will be of major interest in comparing the two, a review of absolute reliabilities for some recent liquid and solid vehicles is also valuable in itself. Unlike costs and schedules which represent best estimates of future values, reliabilities represent concrete evidence of what has actually been demonstrated in the past, they give some indication of the likelihood of mission success. Fig 10 shows composite plots of overall vehicle reliability for 1) four solid propellant vehicles and 2) six liquid propellant vehicles as a function of the number of vehicle launches. The former group included Polaris, Minuteman, Pershing, and BOMARC B booster; the latter group included Atlas, Titan, Jupiter, Thor, Redstone, and BOMARC A booster. Individual reliability records of these vehicles, taken from Ref. 1, are shown in appendix C. The composite plots of Fig. 10 reveal that solid propellant vehicles have demonstrated significantly higher reliability than liquid propellant vehicles at a given early date. Early reliability is especially significant because the number of total launchings of a NOVA sized vehicle, even for a 10 year period, will probably only be about 60-120; there will only be a limited number which can improve flight reliability. The outstanding exception is Polaris. Development of Polaris probably constituted the gravest challenge to engineering rocket technology the country has seen. The relatively low reliability in early flights, as well as in the static test record (see section IVB2), reflects the very significant advancements in the state of the art that were required in many technical areas: 1) design to severe volume constraints for vehicle stowage, 2) chambers based on notch-sensitive materials of unusually high strength, 3) high performance propellants with much higher flame temperatures and more erosive exhaust gases, 4) multiple nozzles which incorporated unusual materials and design features, and 5) new technique for TVC, jetevators. Failures in the early stages of the static test and flight program were to be expected.

2. SCHEDULES

In the further comparison of the liquid C-8 and solid NOVA programs, the schedule, or time required to launch astronauts, for the solid NOVA has been estimated at 5 years by JPL and 5.25 years by Boeing (Ref 1 part II) The liquid C-8 schedule has been estimated at 6 years by the Fleming Committee, and in separate studies, at 5.75 years by Rocketdyne. It is important to note that Dr. Dergarabedian, director of STL's Systems Research Laboratory, has permitted a quote of STL's comparative evaluation based on their solid Nova critique and their liquid class Nova vehicles for MSFC. "the solid Nova will be available at least a year sooner than the liquid system" STL's estimates of 6.5-7 years for the solid Nova, therefore, means at least 7.5-8 years for the liquid C-8 and must not be compared with the 5.75-6 year estimates of liquid Novas by other agencies. The ground rules were different. "STL believes virtually all estimates by other agencies are idealized (based on success) rather than realistic schedules and are therefore too short."

3. PROGRAM COSTS
Dr. Dergarabedian, Director of STL's System Research Laboratory, again has permitted a quote of their comparative evaluation, "they estimate the liquid C-8 will cost approximately twice that of the solid NOVA or about $8.7 billion." STL cost estimates should not be compared to other cost estimates shown because the assumptions for the studies differ as mentioned in the section on schedules. RAND estimated the costs of the liquid C-8 at $4.7 billion, if the launch rate is one vehicle every two months, while Rocketdyne has estimated $4.5 billion (Ref. 3). RAND reports that MSFC costs for the liquid C-8 are in excellent agreement with their estimates. RAND has also estimated the cost of a liquid C-8 program that included 100 vehicles (Table 10, to take into account the decreasing recurring cost of the liquid vehicle, at $10.5 billion. The solid Nova program for 100 vehicles has been estimated by JPL, for a constant vehicle cost, at $8.3 billion; in practice, production costs of the solid Nova vehicle would, of course, decrease. It is of interest to note the cumulative vehicle program costs to perform a specified practical mission when 1) liquid C-8 vehicles and 2) solid NOVA vehicles are used. Fig. 13 indicates the cumulative injection vehicle system costs to supply a lunar base with 150,000 lbs of useful payload per year. The Nova, C-8, Saturn C-5, and Saturn C-3 were estimated to have, respectively, 30,000, 15,000, and 7,500 lbs of useful payload (payload weight after weight of the weight of the landed spacecraft vehicle itself). The values at zero time represent the costs for R&D, facilities, and 10 vehicle flights. Based on JPL estimates, the total solid-Nova vehicle program cost, after 100 flights and 18.5 years, would $8.3 billion. The solid Nova data in this figure, unlike the liquid system, do not include decreasing production costs for the vehicles and probably penalize the solid vehicle. Both solid and liquid vehicle estimates assume reliabilities of 100%, and therefore, penalize the solid Nova again in a relative sense. If STL cost estimates had been used, both curves would have had higher values but would undoubtedly resemble those shown, unfortunately, no cost breakdown for the liquid vehicle was available from STL for preparing such curves. The Saturn C-3 and C-5 costs, estimated by RAND, have been included as a matter of interest. Although they start with lower development costs than the liquid C-8, they rise rapidly to values appreciably higher. After 10 years their program costs would be $9.0 billion and $8.6 billion, respectively. Figure 14 indicates program costs to supply the lunar base using 1) solid Novas and 2) liquid C-8's, when vehicle reliabilities estimated by Boeing (see section B2) are used and when the recurring costs of the vehicle are decreased with time again. Again, RAND's costs are used for the liquid C-8, with the initial vehicle launch cost dropping from $135 million per vehicle to $70 million for the hundredth launch; JPL cost estimates for the solid Nova decrease from $75 million initially to $65 million for the hundredth launch. Thus, if the liquid C-8 were used to supply the lunar base for 16 years, the overall program would cost about $11.7 billion; if the solid Nova were used, the overall program would cost about $8.25 billion, or 70% of the former. It may be concluded that the solid Nova of conservative design would have a significantly lower development cost and lower recurring costs to perform the specified mission than would the liquid C-8, the Saturn C-5, or the Saturn C-3.

4. GROWTH POTENTIAL

No attempt will be made to compare the growth potential of the solid Nova with the liquid C-8 when advanced developments are incorporated; too many possible combinations exist to make a meaningful study at this time. However, a comment pertaining to the growth is worth noting. The very large solid Nova first stage provides a much greater lift capability than other ground launched systems under consideration. With any given advanced propulsion system as upper stage, the solid Nova will provide better overall performance capability than the other first stages. The reason for this is simply that the solid Nova first stage is much larger and has a higher thrust (about 50 million pounds). The large launch weight that was considered a drawback of the solid Nova by many at the outset of the studies proves to be an asset in this respect.

V. CONCLUSIONS

An examination of manned lunar and planetary missions for the foreseeable future reveals that, despite the obvious challenges to our booster and performance capability, system reliability for the overall mission profile constitutes the greatest potential obstacle to successful mission accomplishment. The latter obstacle arises because of major constraints from 1) non-expendable payloads, ie men, 2) the need for large, complex boosters, 3) very high unit cost, and 4) a long and complex sequence of critical operations to complete a mission. One important aspect of the mission profile, the injection vehicle, has been examined conceptually, with reliability, rather than performance or minimum weight, as the ranking criterion and the following conclusions are reached:
1) Vehicle system reliability will probably prove to be unacceptably low for complex manned space missions if performance (weight reduction) and reliability carry the same relative importance in the new development programs that they have in past missile and space programs.
2) A design and development philosophy which incorporates the important principles 1) unusual conservatism and 2) design within the state of the art is advocated as a mechanism whereby substantially higher reliability can be realized.
3) By using this philosophy and capitalizing on unique characteristics of an all solid rocket vehicle, a basically new approach to high early vehicle reliability in flight becomes possible; program implementation would consist of 1) analyses and evaluation of the motor designs, vehicle structure, and vehicle flight dynamics in scale model tests during the preliminary design and large scale tooling period, followed by 2) assembly of scaled up, full size components as a complete vehicle for immediate use.
4. An all solid propellant injection vehicle in the 25 million to 35 million pound launch weight class, studied by JPL and critically evaluated by STL and Boeing, has been found feasible.
5. When compared to liquid propellant or hybrid stage vehicles of comparable performance capabilities, it is believed that the development risk of the all solid Nova would be very low and the resultant vehicle reliability at an early date predictably high- provided the suggested philosophy and program approach were utilized.
6. From a technical viewpoint, the schedule for the all solid Nova injection vehicle studied is significantly shorter and the program costs markedly lower than hybrid stage and liquid propellant vehicles of comparable payload capability despite the much greater launch weight of the solid vehicle.
7. By replacing solid upper stages with liquid hydrogen/liquid oxygen stages or nuclear stages as these become available, the growth potential of the all solid vehicle is especially favorable for space missions which require larger payloads and greater sophistication. Such hybrid stage vehicles show great promise for future planetary missions using electric or nuclear propulsion.

I. INJECTION VEHICLE

A. DESCRIPTION AND OPERATION

The injection vehicle in its primary form was estimated to have a payload capability of 130,000 lbs to escape or 500,000 lbs to LEO and a launch weight of 25 million lbs. Subsequent and more sophisticated calculations show that the vehicle launch weight is closer to 30 million lbs if the vehicle is to provide the indicated payload capability. It is believed that this revision will affect size and cost only and in no way bears on the feasibility of the concept. All discussions in Appendix B are based on the 25 million pound vehicle.

1. VEHICLE

The injection vehicle (fig. B-1) consists of four stages of clustered solid propellant motors. The motors in a stage are joined by intrastage structure; successive stages are joined by interstage structures which have provision for positive separation. Each stage has its own thrust vector control (TVC) system. The fourth stage carries the 130,000 lb escape payload, guidance, control, and telemetry for the injection vehicle, and a vernier propulsion system. For this study, the injection vehicle was sized to inject its fourth stage into a parking orbit. The fourth stage then injects the 130,000 lb payload into an escape trajectory. The typical sequence of operations is:
1. stage one ignition
2. stage one burnout and separation
3. stage two ignition
4. stage two burnout and separation
5. stage three ignition
6. stage three burnout and separation
7. vernier velocity correction into parking orbit
8. orbital coast
9. stage four ignition
10. stage four burnout and payload separation.
11. vernier velocity correction into final escape trajectory

The desireability of maintaining flexibility with respect to the altitude of the parking orbit is discussed in Section I-Q. To accomplish this would require a resizing of the first three stages of the vehicle and of the vernier velocity system on the fourth stage. Flexibility and other mission requirements might require the use of a storable, restartable liquid fourth stage. Detailed consideration of these requirements ws beyond the scope of this study, but in further studies of the concept to greater depth the applicability of such liquid upper stages should be examined.

2. SPACECRAFT

The spacecraft is composed of the following elements:
1. Three man Apollo Command Module
2. Mission Module (Apollo) *note- at the time, NASA was strongly considering a Soyuz-like three-component CSM consisting of a Service Module, much smaller reentry Command Module, and a disposable Mission Module
3. Midcourse, lunar retro, and takeoff propulsion systems (direct ascent lunar mission mode, which was preferred then)
4. Associated structure, guidance and control, and communications systems.
5. Abort propulsion and structures.

The operational sequence for a manned or unmanned lunar landing and return is:
1. Launch to injection
2. Midcourse correction
3. Transit
4. Site acquisition and lunar retro firing
5. Lunar landing
6. Lunar operations
7. Takeoff
8. Transit
9. Midcourse correction
10. Earth reentry and recovery

B. CONFIGURATION

The injection vehicle configuration is shown in Fig B-1; a weight summary is shown in Table B-1. The following factors entered into the decisions which resulted in this configuration:
1. Only two different solid propellant rocket motors would be developed. The Type A motor would be used in stages 1 and 2, and the Type B motor would be used in stages 3 and 4. This obviously requires that the number of motors in stages 1 and 2 be integral multiples of the Type A motor and that the same be true for the Type B motor in stages 3 and 4.
2. The logistics of motor manufacture, transportation, handling, and field erection dictate a unit weight that can be easily handled.
3. The mission constraints require that: 1) the fourth stage and payload be injected into parking orbit at the end of third stage burn, and 2) the fourth stage be able to inject a 130,000 lb payload into an escape trajectory from parking orbit.
4. The requirement that present state of the art be used led to the assumption of specific impulse and stage mass weight ratio used in sizing the vehicle.

The geometry of clustered cylinders and the desire to provide reasonable load paths for interstage and intrastage structure played a role in determining some of the geometrical details of this particular configuration. The result shown is based on the constraints and the considerations described above and is considered to be representative and adequate for the purpose of this study. It is certainly not intended to be put forth as an optimum configuration. The physical size of the vehicle, 77 foot base diameter and 218.5 feet to the top of the fourth stage, is similar to vehicles which have already been considered in various other studies. The gross weight at takeoff, 25 million pounds, is larger than previously considered. It can be compared to some water heaters used in steam electric plants that are 150 feet high and weigh 20 million pounds or to an aircraft carrier weighing 120 million pounds. Interstage or intrastage, and TVC systems structure is not shown on Fig B-1 becuase of the small scale. These items will be discussed in the following sections. The absence of a complete aerodynamic shroud is intentional since the need is not defined at the present time. Each of the solid propellant motors utilizes a single fixed nozzle and has a thrust program designed to produce acceptable accelerations at stage burnout. In the design shown, it was necessary to assume step mass fraction (ratio of propellant to total stage weight) in order to estimate the performance capability of the vehicle. A stage mass fraction of 0.87 was assumed for, the first and second stages, and 0.91 was assumed for the third and fourth stages. Preliminary weight estimates of case and nozzle design, propellant charge configuration, interstage structure, cluster configuration, and TVC requirements have been made and are discussed in Section I-R. Aerojet-General has stated that the first stage of a 7 million pound gross weight vehicle can achieve a stage mass fraction of 0.889, utilizing 7 motors of 140 inch diameter as propulsion elements. By the utilization of a single unit of 288 inches in diameter (roughly the same size as the Type A motor) they can achieve a value of 0.893. Grand Central Rocket Company has stated that the first stage of a 10 million pound gross weight vehicle, consisting of 16 ten foot diameter motors, would have a stage mass fraction of 0.886. It is expected that these numbers may be optimistic and attainable values can be established only by thorough preliminary design, integrating mission system requirements, engineering mechanics, propulsion, guidance, control, vehicle telemetry, power, and last but not least, payload vehicle interactions.

C. ROCKET MOTORS

The Type A motor is 300 inches in diameter and has a length to diameter ratio of 3.26:1. It contains a little less than 2 million pounds of propellant. The motor has, basically, a star shaped grain perforation. Motor volumetric loading is approximately 82%. The average vacuum thrust is 6.4 million pounds and the burn time is 85 seconds. The Type B motor is 220 inches in diameter and has a length to diameter ratio of 2.33:1. It contains about 350,000 pounds of propellant and also has a star shaped grain perforation. Motor volumetric loading is approximately 88%. The average vacuum thrust is 740,000 lbs and the burn time is 138 seconds. These motors represent a typical design concept to demonstrate the feasibility of very large units withing the current state of the art. The size of the motor and grain configuration can be varied to satisfy particular motor requirements. Both motors are designed to utilize existing available propellants, with solid loading of approximately 82 and 88%, and containing powdered aluminum in the 15-20% range. The Type A motors were designed for a nominal motor chamber pressure of 800 PSI with a burn rate of 0.64 inches per second. The Type B motors were designed for a nominal motor chamber pressure of 350 PSI with a 0.46 inch per second burn rate. These burning rates are available in current propellants both from Aerojet and Thiokol. A specific impulse of 245 seconds at 1000 PSI and sea level optimum expansion was assumed. This performance level has been verified many times with both polyurethane and PBAA propellants in current motors. The difference between the measure ISP value and the calculated ISP value far exceeds the correction for heat loss and nozzle divergence angle. It is believed that the bulk of these excess losses is due to velocity and thermal nonequilibrium of the aluminum oxide particles (products of combustion of aluminized propellants) with the gas. Calculations performed at JPL show that aluminized propellants which deliver an ISP of 245 (at 1000 PSI and sea-level optimum expansion) in ordinary nozzles should deliver substantially more in very large nozzles such as those considered for the solid NOVA. The vacuum specific impulse used for A and B motors represents conservative estimates of this ISP improvement in very large nozzles based on measured aluminum oxide particle sizes and one dimesional, 2 phase flow calculations performed on the IBM 7090. The reproduceability of performance for the type of very large units under consideration is expected to exceed that of present, relatively small motors. For example, deviation in burn rate will be less because of amuch larger sample of propellant is exposed to burning.

D. MOTOR CASE DESIGN

The design of the motor cases of the size considered in this study presents no problems basically different from those that would be encountered in the design of smaller size cases.

E. NOZZLE DESIGN

Single fixed nozzles will be used for both type motors. They represent configurations scaled up from known practical steel, graphite throat nozzles. They have been canted in the first and second stages so that the thrust of each motor passes through the vehicle CG at burnout. Their length is established through the use of standard design procedures for the bell shaped exit nozzles. Nozzle materials and fabrication methods already developed seem entirely adequate for building nozzles for these large motors. Since time was not available during this study for detailed nozzle design, weights were estimated from existing information. Two nozzle designs are used in the vehicle. The average thrust of Type A motors on the first and second stages is 6.4 million pounds; the average thrust on the Type B motor on the third and fourth stages is 740,000 lbs. A survey of available reports containing nozzle designs was made and a plot of thrust vs. nozzle weight was prepared (Fig B-6). The weight of the two nozzles was estimated from this plot. The nozzles considered in preparation of Fig B-6 have expansion ratios of about 10:1 and a chamber pressure range of 700-1000 PSI. The nozzle for the type B motor has an expansion ratio of 33:1 but the chamber pressure is only 360 PSI. Since the nozzle weight can be expected to increase with an increase in expansion ratio and decrease slightly with a decrease in chamber pressure, an estimate of 3500 lbs for the nozzle weight seems reasonable. The estimate of 34,000 lbs for the weight of the large type A motor nozzle appears to be sufficiently conservative, since as can be seen from Fig B-6, the weight ranges from 23,000 to 39,000 lbs for a thrust of 6.4 million pounds. A plot of total motor impulse vs weight was also prepared; it indicates that the large type A nozzle is reasonably conservative. The weight of the type B smaller nozzle appears to be less conservative in this plot, but the value quoted above was used since the weight range for the examples considered here, at a slightly lower total impulse, is rather large. Since the diameter of the nozzle throats considered for this vehicle is large, ablative type nozzles may be more desireable and easier to design in this size than in the current advanced developments being pursued. Although not considered here, they represent a design which promises a weight savings, at no decrease in reliability, over that estimated in the weight analysis.

F. COMBUSTION INSTABILITY

Combustion instability is not expected to be a problem in large solid propellant rocket motors. Although many motor programs have had combustion instability problems, a solution has always been found and the programs have been successfully concluded. There are a number of major reasons for not expecting combustion instability in a large composite ammonium perchlorate aluminum solid propellant motor. The last three large composite solid propellant rocket motors, Minuteman, Polaris, and Pershing first stages, have not had combustion instability problems using the type of propellants under consideration. Aluminized propellants have shown instability have been in large L/D motors at relatively high pressures, around 1500 PSI. Recent experience at JPL with a nonaluminized polyurethane has indicated a stable region below 750 PSI and logitudinal oscillations at higher pressures. Small amounts of aluminum gave stable burning at 1000 PSI. Unfortunately, there are no data in the low frequency range of large motors.

G. SEGMENTED VS. UNITIZED ROCKET MOTORS

The choice of a unitized (monolithic) vs. segmented motor construction does not appear to be very simple. From the standpoint of reliability each segment of a segmented unit contains most of the probable causes for failure of a solid propellant motor. For example, each segment will have associated with it one joint; each segment will have associated with it two ends which tend to be the regions of pull-away (separation) of the propellant from the case, and each segment will be processed and trimmed separately, inviting other failures if proper controls are not exercised. A unitized motor has a similar joint attaching the nozzle to the case and also similar modes of failure. From the reliability standpoint it can be argued that the number of segments or unitized motors should be minimized. The minimizing of the number of units implies maximizing the size of each unit to do a specific job. A number of arguments have been proposed for minimizing the size of a segment from the standpoint of ground support equipment, processibility, and transportability. From the long experience that is available with missile projects, it can be stated that it would be an error to grossly compromise the injection vehicle on the basis of ground support equipment. Existing processing facilities cannot handle and overload of the size being considered. Therefore, separate process facilities which would have to be established could just as easily be designed to handle large unitized motors as single smaller segments. The concepts proposed in this report are based on what appears to be the more difficult problem, that of handling large single unitized motors, although it is not necessarily recommended that this be the final solution. It is believed that there are problems in the ares of hardware fabrication and vehicle assembly which are minimized with the use of smaller segments and, if the flight vehicle can be designed better by the use of segmented motors, the latter should be used. The final decision between segmented or unitized motors should result from a study of the effects of the choice on the flight vehicle.

H. PROPELLANT PHYSICAL PROPERTIES

The structural integrity of the large propellant grain has been examined and found satisfactory. It is anticipated that there will be a scaling effect in extrapolating propellant physical properties of small grains to large grains, and this effect should be studied during the preliminary design. However,on the basis of experience with state of the art large motors, this effect is not thought to be limiting.

J. INTRASTAGE STRUCTURE

The intrastage structure for all stages has two main purposes: 1) to transmit shear forces between motor cases, and 2) to tie motors together radially. The shear forces between motors arise from thrust variation, a time lag between motor startup and burnout of individual motors, and body bending. The shear forces in the first stages can be transmitted by means of short, thin walled cylinders filling the interspace between the clustered motors. These cylinders may be attached to the motor cases with high strength bolts, through lugs welded onto the cases during fabrication. The third stage motor cluster could develop shear transfer between motors by means of half-cylinders, because of the nesting of the third and fourth stages. Essentially, the shear will be transferred in the same manner as in the first two stages. The motors of the first three stages will be tied together to resist relative radial motion. This will be accomplished by means of a system of cradels and ties at the top and bottom of each motor case. The tie arrangement will be placed at the junction of the case dome closure so that a minimum amount of restraint will be offered to the growth of the cases resulting from pressurization.

K. INTERSTAGE STRUCTURE

The following study goals have been considered when investigating the interstage structure: 1) feasibility considerations for several structural concepts. 2) development of one or more workable conceptual designs. 3) approximate analysis and weight estimate for at least one design. The bulk of the current work has been oriented toward the first to second stage interstage structure. Separation has been considered only to the the extent that it affects structural configuration. It has been assumed that full ring bulkheads at the ends of each stage are not required, that concentrated loads can be applied to the motor cases, and that interstage structure is capable of transferring thrust differences between motors. The largest loading on the structure is of course axial thrust. Side loads caused by wind shears, thrust misalignments, and the control forces resulting from them will vary along the length of the vehicle because of rigid body and elastic dynamic considerations. Vibration in both the axial and transverse directions will be added to the above loadings. Conserative and realistic load estimates were made and both were used. It is believed that they bracket the actual loads. In each case it was conservatively assumed that the maximum axial and transverse loads occur simultaneously. Four types of structures considered are: 1) Truss, 2) Longeron-tie rod, 3) Monocoque, 4) Nozzle utilization. It is considered that the interstage structure design should be predicated on the required strength rather than rigidity. However, the effect of rigidity on dynamic response must be carefully considered in a preliminary design. Preliminary consideration of truss, longeron-tie rod, and moncoque structures showed each to be feasible. The light weight of an unstiffened nozzle structure rules it out as a practical solution. Sketches of a truss and longeron structure design are in Fig B-7 and B-8, respectively. Analyses have been made of a monocoque and modified monocoque structure. The material assumed for the interstage structures is steel with an ultimate strength of 180ksi and 100 ksi at weld joints connecting major structures, with a safety factor of 1.25. The truss design is shown in Fig B-7. If a tubular cross-section is used, the lower members could have a diameter of 24 inches with a thickness of 1.33 inches for number 1 loading, or a thickness of 0.85 inches for the number 2 loading. Longeron design is shown in Fig B-8. A shell-type structure would require rings at the bottom and top, and the separation plane. It may be possible to use partial, scalloped rings for the top and bottom. The critical condition for this design appears to be shell instability. For a solid shell, the required thicknesses of approximately 1.7 and 1.35 inches.

L. THRUST VECTOR CONTROL

A method of controlling the thrust vector of each stage must be provided to compensate for the effects of center of gravity displacement, thrust misalignment, and unequal thrusts (especially at ignition and burnout). For the first stage it is needed to counteract aerodynamic loads. It must also provide the necessary maneuvering forces. Because of expected motor-to-motor variances in thrust level, it appears very desirable to cant the nozzles so the main thrust vectors for the for the individual units will pass through the vehicle center of gravity. It is expected that the greatest difference in thrust level between units will occur near the beginning of the tailoff of the thrust-time curve. As a result, it would be desirable to point the nominal thrust vector through the stage CG at burnout. The effect of aerodynamic loads might modify this conclusion for the first stage; this can be determined only by more detailed study. Based upon opinions of most of the contractors who have studied the subject, the choice of thrust vectoring means for a big solid booster system lies between jet vanes and secondary injection of of liquid into the expansion cone of the nozzle. It is believed that at least one other system should be considered: that of auxiliary rocket motors which have been developed by Allison and Vickers for NASA. A brief discussion ofthe features of all these systems follows.

1) JET VANES-- Roll, pitch, and yaw moments can be obtained by a single nozzle installation from four (or three) wedge-shaped aerodynamic surfaces positioned 90 (or 120) degrees apart in the expansion cone of the nozzle. Drag force on the jet vane during burning, decreasing the effective motor impulse, is one disadvantage of this system. Another is the materials problem that results from immersing a substance in the high temperature exhaust stream. Jet vanes are used on Sergeant and Pershing.

2) AUXILIARY ROCKET MOTORS-- This method of TVC uses rocket motors or gas supplied nozzles at a location chosen to give optimum pitch and yaw moments for a given stage, usually near the interstage structures. Individual, self-contained motors can be used or, in a system that uses gas supplied nozzles, a single common gas generator can be used. In both cases, the exhausting of these secondary gases occurs throughout main stage burning, changing the missile velocity vector only through nozzle orientation. One advantage of this system is that it is possible to keep the auxiliary nozzles exhausting for a specific time after main stage burnout, providing control forces until the subsequent stage ignites. The major disadvantage of this system, compared to the others considered, is the unreliability added to the added to the vehicle systems through the inclusion of an additional propellant device, whether used for direct TVC or for generating gas for several nozzles. Both of these methods are presently under development.

3) SECONDARY FLUID INJECTION-- TVC in pitch and yaw can be obtained for single-nozzle configurations through injection of fluid into the expansion cone of the nozzle. Decomposition gases can also be used. To date, secondary injection has not been flight tested; however, systems are in advanced development stage for use on Polaris and Minuteman. A special feature of this system is that the injected material affect both the effective mass ratio of the unit and specific impulse, since material is being expelled during operation of the main engine. For this analysis, a secondary injection system which may prove to be inherently more reliable was chosen for further detailed study in order to provide TVC weight and cost estimates. Such a system provides jet deflection as a result of introducing high velocity fluid stream into one side of the divergent cone of the nozzle. An oblique shock is created to cause an effective deflection of the exhaust gases linearly proportional to the mass flow of injectant. It has the advantage over other typical systems of being light in weight while requiring few moving parts. Thiokol Chemical Corp. (ref 24) found that such a system would weigh less than half an equivalent system employing jet vanes. It was also shown to be lighter than a simple auxiliary jet system. However, this conclusion isn't universally applicable. In order to estimate the TVC system, the following assumptions were made: 1) CG offset per stage, 6 inches. 2) Net thrust misalignment per stage, 0.25 degrees (this includes thrust differences of 3.7% 3) Nozzles are canted nominally through stage burnout CG. 4) Ignition delay of 0.3 second 5) Burning time variation of =/- 3.7%. 6) All motors on one side of the CG ignite before all motors on the other side. 7) Effects of motor to motor burn time variations are minimized by long thrust tailoff (approximately 10% of burn duration). This can be further minimized in the first stage by giving the center motor a slightly longer burn time. 8) Maximimum wind shear equivalent to 0.3 g side acceleration with a moment arm of 25 feet (the force is believed conservative by a factor of 2.5 and the moment arm by a factor of 4). 9) CG travel as shown in Fig B-9 was used. It was found that side forces applied at the nozzle exits and total side impulses as indicated below would be required (expressed as a fraction of stage axial thrust or total impulse). Based on these requirements, two systems for the first stage were briefly investigated: a turbopump system and a pressurized system. The turbopump was visualized as powered by a monopropellant liquid or solid propellant gas generator. To minimize turbopump control problems a bypass would provide return flow to the injectant tank as jet deflection demands drop off. Ullage pressure for the injectant tank would be provided by either bleeding turbopump exhaust gas or by heated injectant expanded from the high pressure side of the pump. The pressurized system would utilize helium gas stored at 5000 psi and regulated to deliver injectant at the required pressure. Both systems would utilize multiple valves for each cluster arranged to provide proper pitch, yaw, and roll control moments. (Ideally each motor should be self-sufficient at burnout). Programming would be provided to ensure that all injectant is expelled prior to burnout in order that full utilization can be made of the energy available. The turbopump hydraulic horsepowr requirements roughly approximate that of the Rocketdyne F-1 engine turbopump. This unit requires 128 lb/sec of gas for the turbine and weighs almost 3000 lbs.

M. AERODYNAMICS

1. AERO-FAIRING-- The necessity for aerodynamically fairing the vehicle has been examined for the clustered-type vehicle under consideration. Such a fairing may be necessary owing to three aerodynamic factors-- 1) heating, 2) drag, and 3) unsteady flow effects. Aerodynamic stability is not considered in this section. A) heating-- both maximum laminar and turbulent heating rates were examined. The results of this plot clearly show that aerodynamic heating would not be a serious matter to structural members. Even the use of turbulent heating (which is improbable) in this example would not indicate a greater temperature rise because of the small dimensions involved. Correspondingly, heating of the exposed engine tanks would be even less serious because the heating rates are reduced as a result of the effects of large nose radii and large dimensions from the nose to sonic point. B) AERODYNAMIC DRAG-- An open type configuration such as considered in this study, does not lend itself to simple analysis for the estimation of drag. Nevertheless, it is essential to make an estimate in order to determine what effects may occur to the flight parameters, which are presently based on a drag curve for a fully faired vehicle. Fig B-13 shows the drag coefficient vs Mach number used in the flight calculations. Also shown on the figure is the dynamic pressure, which is observed to peak at a Mach number of about 1.7. For estimating the drag of the large vehicle considered here, pure turbulent friction drag is not an important factor because of the high Reynolds numbers involved. Thus, the major problem is pressure drag at supersonic speeds, although major contributions can arise from the interaction of shock waves and boundary layers and from flow separation effects. Vehicle size will affect only these latter sources. In addition to examining a small amount of existing experimental data, two simple drag models were calculated. These are: 1) no fairings assumed: all motor cases independently exposed to the free stream airflow (stage four and payload considered as one body). 2) partial fairing assumed. Case 1 is not a very realistic condition to consider, since adjacent motors would cause mutual interference, so that the effective drag would be greater than the sum of the individual bodies. However, motors of the first and second stages would be exposed to flow at much less than free stream dynamic pressure, thus lowering the drag. The net results of these somewhat compensating effects is not predictable in short period of study. However, it may be reasonable to assume that by properly streamlining all structural members, placing conical noses on top of stage 3 motors, and judiciously using local fairings throughout the structure, it may be possible to keep the drag coefficient down to twice the basic streamlined case, or a drag coefficient of about 1.0 at Mach 1.7. Considerable wind tunnel development would be required on such an open configuration. Theoretical work on determining proper motor lateral spacing may be fruitful in reducing interference drag. For the externally faired case, the major drag is derived from the nose cone and the transition region between stages 1 and 2 as indicated in Fig B-13. For this case, it is observed that the drag coefficient is about 60% higher than the basic curve. Additional computer effort would be required to ascertain the effect of these potential drag coefficient increases on flight parameters. However, taking the thrust as about 40 million pounds, a twofold drag increase would result in a peak drag-to-thrust ratio of only about 0.15, which is acceptable. C) UNSTEADY FLOW EFFECTS-- This effect may be the one which requires a faired configuration. Fluctuating flows through the partially open vehicle may induce vibrational modes in the structure at levels which are not acceptable. However, as in the case of drag reduction, local fairing and streamlining would be beneficial.
2) BOOST PHASE LATERAL AIR LOADS-- The following trajectory assumptions were made when estimating the boost phase lateral air loads: 1) vehicle CG follows intended no wind flight path (zero-alpha trajectory), crabbing into the wind as necessary to avoid drift, 2) TVC gimble angle varied as necessary to produce the small angle of attack (alpha) required to cancel driftand, at the same time, to counterbalance air load pitching moments. Inasmuch as the above mentioned crabbing alleviates the air loads on the vehicle about 20%, a slight more conservative (pessimistic) assumption would be assume the constant attitude vertical flight. However, the associated drift velocities make this representation unrealistic.
3) BASE HEATING-- The heating of items stored around the nozzles has not been examined. Experience gained by tests and flights of the Saturn first stage should be applicable. The large nozzle exit area and the close spacing of the nozzles woudl make it relatively easy to insert a lightweight fiberglass shield to prevent the flow of hot gases up between the nozzles if it should prove necessary. Radiation heating of the outer surfaces of the motor cases by exhaust flame appears relatively insignificant.

IV. SPACECRAFT CONSIDERATIONS

A. PHILOSOPHY
Spacecraft studies have been limited to a crude feasibility determinations based on 1) 130,000 lbs injected weight, 2) configuration constraints, and 3) spacecraft design problems peculiar to the employment of the solid NOVA for a manned lunar landing and return mission. The results of the studies to date can be summarized as follows: 1) there are no spacecraft design problems peculiar to the solid NOVA when compared to the liquid NOVA; 2) the injection capabilities of the solid NOVA appear to be adequate for a manne lunar mission, 3) there are no major spacecraft configuration constraints or limitations due to the injection vehicle. For the purposes of this study, Apollo 3 man mission and command modules of the Convair M-1 type have been assumed. Mission abort capabilities are assumed to be consistent with the guidelines established by the Space Task Group. It is recommended that the spacecraft be capable of accomplishing the entire mission automatically. Man would perform monitoring functions associated with the control loop and he would help implement scientific measurements and observations. Manned override control capabilities would be provided for emergencies. A possible feature of the manned mission would be to provide an alternate return vehicle on the surface of the Moon as a contingency for possible failures during landing. In this event, man would be equipped to transport himself over the lunar surface from one vehicle to the other.

B. DESIGN: SPACECRAFT CONFIGURATION STUDY--
The Apollo command and mission module concept in the Convair configuration M-1 was arbitrarily selected as the basic vehicle. Two propulsion configurations were considered; an all liquid configuration of the storable propellant type, and a hybrid, using solid rockets for the retro to the Moon and takeoff and three storable liquid vernier engines for mid course correction and vernier descent to the surface of the Moon.

1. WEIGHT BREAKDOWN-
The Apollo studies indicate weights for the command and mission module which are compatible with the permissible weight figures as determined in this study. The approximate gross spacecraft weight breakdown is as follows: Command module- 5,650 lbs; Mission module- 3,500 lbs; Propulsion system- 117,500 lbs; Guidance, control, interstage, propulsion support structure, etc.- 3,350 lbs.

2. CONFIGURATION: LIQUID PROPULSION SYSTEM
Figure B-25 shows the spacecraft with the all liquid propulsion system. A Titan II second stage thrust chamber and pumping system is used for the retro and lunar liftoff maneuvers. Liquid verniers with a 10:1 throttling capability operating from the same pumping system as the main engine are used for midcourse correction and for limited hovering capability at lunar landing. Attitude control can be accomplished by twelve 7 lb thrust engines arranged in pairs on the vehicle. The retro tanks which are to be left on the Moon are used as landing support structure. Energy absorbing material is attached to the tanks. Three stabilizers are deployed during the hovering maneuver. Part of the interstage structure could be deployed to serve as the landing gear and stabilizer rather than using the tanks for this purpose.

3) CONFIGURATION: HYBRID SYSTEM
Figure B-26 shows the vehicle with the hybrid propulsion system in place of the all liquid system. Here, three solid rockets are used for the main retro system and one for the lunar liftoff. The same storable propellant liquid engines, now with their own packaging and pumping systems, are used for the midcourse and vernier maneuvers.

4) ABORT SYSTEM CONFIGURATION
Eight solid motors mounted on a tower on the front of the command module supply the necessary velocity to perform the abort function during third or fourth stage firing, at which time the complete vehicle is separated from the booster. Firing four of these motors performs the abort mission from the pad when only the command module is separated.

5) VELOCITY INCREMENT: WEIGHT CALCULATION
Tables B-6 and B-7 show weight calculations for the normal mission employing liquids and solids, respectively. Should propellant not be used in midcourse or vernier touchdown, a hovering time greater than 60 seconds could be utilized. The desireability of and capability for this maneuver can be determined by the men aboard the spacecraft.

6) COMPARISON OF DESIGN LOADINGS
The maximum acceleration of the injection system is about 5 g. The spacecraft propulsion system of the liquid type imposes 6 g maximum, the hybrid configuration imposes 9 g maximum acceleration on the spacecraft. These levels compare favorably with the design levels adopted in the Apollo studies and justify using structural weights given therein.

STUDY SUMMARY: THE APPLICABILITY OF SOLID PROPELLANTS FOR A NOVA-CLASS INJECTION VEHICLE
AND A COMPARISON WITH A LIQUID VEHICLE OF COMPARABLE CAPABILITY
JPL: California Institute of Technology, Pasadena, California
NASA CR 136573; N74-72160; 9 May 1962

FOREWORD

Early in 1961, JPL began studying injection vehicle aspects of direct ascent approach for the manned lunar landing mission. At the invitation of NASA HQ, preliminary results of this all-solid NOVA study were presented to the joint NASA/AF Large Launch Vehicle Program Group (generally known as the Golvin Committee) on 3 August 1961. This committee, recognizing the unconventional features of the concept, contracted with Boeing and Space Technology Labs, Inc. to make independent evaluations of the solid NOVA studies reported in JPL's Technical Memorandum 33-52 and Addendum A thereto. It was planned that at the end of the one-month evaluations they would submit critiques on 1) the technical feasibility of the concept, 2) the realism of the JPL estimate of schedule, 3) the accuracy of the cost estimate. Within a week the work statements had been expanded to include 1) an evaluation of the concept if payload to escape were increased from originally studied 130,000 lbs to 156,000 lbs, 2) a supplementary evaluation of the adaptability of a solid NOVA or its components as a backup for a Saturn-class vehicle, and 3) quantititave comparisons of the several reliabilities of liquid and all solid NOVA injection vehicles. On a submission of the critiques by Boeing and Space Technology Labs, the committee decided against company briefings or discussions because of other pressing matters. Approximately one month later, Mr. E Mitchell, then assistant director for Propulsion at NASA HQ, exspressed interest in results of the studies; thus on 6 October 1961 Boeing, STL, and JPL representatives provided briefings to NASA Propulsion personnel on the original solid NOVA concept and the results of two industrial critiques. Since that time, no studies have been performed. In recent months, NASA has decided, after careful deliberation, to perform the manned lunar landing mission using a lunar orbit rendezvous mode basedon a liquid propellant Saturn C-5 vehicle for the manned operations and a lunar logistics vehicle for cargo and manned support. However, the need for vehicles with a payload capability of about 500,000 lb in LEO, in the NOVA class, for the more difficult missions beyond the manned lunar landing, has also been recognized by NASA, and the new studies of all liquid and hybrid staged NOVA vehicles have been initiated by several groups. Because the lab's studies of all-solid propellant vehicle systems indicate that it is unusually promising and of general interest, this report, which gathers together all significant results of the studies, is being released as an appropriate addition to current knowledge. Only that phase dealing with the way in which the solid NOVA would fit into the manned lunar landing mission and NASA's long range plans has been omitted because it is no longer applicable. An appendix that discusses background information and the status of applicable solid rocket work has been incorporated as a convenient reference; in addition, several developments from JPL programs are discussed because they have strong bearing on the solid NOVA program under study. The opinions expressed in this report do not necessarily reflect the views of NASA.

ABSTRACT

Studies show that very large, all solid propellant injection vehicle systems in the NOVA class are technically feasible. The vehicle studied is a four-stage, solid propellant rocket having a gross weight in the 30 million pound class, a payload in LEO of 500,000 lbs, and a payload through escape of 130,000 lbs. The first three steps would inject the fourth stage into parking orbit from which the fourth stage would inject the spacecraft into the transfer orbit to the Moon or planets. Designs examined were conceptual and do not represent optimized configurations. Additional studies must be made in depth before the final system and industrial complex requirements can be specified. These new studies must include an examination of the applicability of noncryogenic liquid propellant systems for third and or fourth stages because of the potentially greater vehicle flexibility for a range of missions. When compared to liquid vehicles of equivalent performance capability, it is concluded that the risk associated with development of this vehicle system would be much lower and that the reliability of the resultant system would be predictably higher at a very early point in the flight program- provided that the proposed conservative philosophy were used. As a result of the studies, confidence in the philosophy and program approach advocated has been reinforced. From a technical standpoint, it appears that the vehicle injection system can be made available for the first flight approximately four years after the start of go-ahead, and for operational manned space missions one year after that date. Total costs are estimated to be significantly less than for competitive systems, provided that the philosophy and development approach advocated are implemented. Growth potential for the basic systems seems particularly favorable for the progressively more difficult missions which can be foreseen. There do not appear to be any major technical problem areas. However, early emphasis should be given to thrust vector control (TVC) and meeting guidance and control (GNC) as well as other subsystem reliability requirements. Combustion instability is not expected; and it is believed that even if it is encountered it would not become a serious problem.

INTRODUCTION

Prior to the time that a definitive plan could be formulated by NASA for the manned lunar landing, a number of groups undertook studies of various vehicles and modes of performing the mission. Because of its extensive experience in liquid and solid propellant technology and vehicle system development (Corporal, Sargeant, and the Explorer and Juno upper stages), the JPL voluntarily initiated some studies of large launch vehicles.

Independent of these studies, JPL activities in the Ranger, Surveyor, and Mariner projects had revealed the complexity and difficulty of even the simplest unmanned lunar or planetary missions. Thus, on reviewing the manned mission, it became evident that the total mission reliability was only a part, would be wholly inadequate unless the subsystem reliabilities were substantially better than they had been for most missile and space programs. There appeared to be at least four basic reasons for this. First, man, as a nonexpendable payload, will demand unusually high overall mission reliability-- perhaps 85%-90% probability that the mission will succeed, and 97%-99% probability that, despite a mission abort, the astronaut will be recovered. Second, the larger vehicles required wil be derived by clustering as many as 8-10 propulsion systems per stage and or using more stages. Consequently, if the design philosophy used in the past is preserved, vehicle reliability will drop below the barely acceptable values of current vehicles- particularly if they must use newly developed propulsion systems. Third, because of high vehicle cost the launching rate and production rate will be significantly less than in the past- probably 60-120 vehicles over a ten-year period. Therefore, there will be very few flights with which to develop system reliability; some means must be found to provide high reliability at a much earlier point in the flight development program. Fourth, and most important- space mission will become increasingly demanding in the required number of sequential operations that must be performed before the mission can be completed successfully. For example, a manned lunar or planetary landing mission would probably consist of the following major operations (involving perhaps a total of seventy to eighty critical steps performed in sequence): 1. First stage launch, 2. second stage operations into parking orbit, 3. third stage injection through escape, 4. midcourse propulsion and guidance corrections, 5. terminal guidance and retro-propulsion into lunar or planetary orbit, 6. descent toward the Moon or planet, 7. hovering to a soft landing, 8. lunar surface operations, 9. takeoff for return, 10. midcourse propulsion and guidance corrections, 11. Earth landing through a restricted corridor and recovery. Quite possibly there would also be, at appropriate times: 12. vehicle rendezvous (in Earth orbit with multiple launchings of components and/or rendezvous at the destination), and 13. orbital docking of the vehicle components. It must be recalled that vehicle launches into Earth orbit (only the first two of the 13 listed operations) have resulted in approximately 25-50% failures. Many of the later operations in the list of 13 must be performed in new and hostile (perhaps unknown) environment, there will be little opportunity to practice landings and takeoffs at the destination, and success of a number of the operations (ie rendezvous) is strongly time-dependent. The magnitude of the challenge may be illustrated by noting that mission reliability is an inverse exponential function of the sequence of operations.

SOLID PROPELLANT NOVA INJECTION VEHICLE SYSTEM

The injection vehicle system and its industrial support complex include the launch vehicle, its means of production, the associated transportation complex, the offshore launch pad and ground support equipment, and a range support area (Fig 1.) The spacecraft is considered only as it affects the vehicle system. Additional material in support of the study is contained in App. B.

A. INJECTION VEHICLE

The vehicle in its preliminary form was estimated to have a payload capability of 130,000 lbs to escape, or 500,000 lbs in LEO, and a launch weight of 25 million pounds. Subsequent and more sophisticated calculations show that the vehicle launch weight is closer to 30 million pounds if the vehicle has to provide the indicated payload capability. It is believed that this revision will affect size and cost only and in no way bears on the feasibility of the concept. All discussions in Sec. 3 are based on the 25 million pound vehicle.
1. DESCRIPTION AND OPERATION
The injection vehicle (Fig. 2) consists of four stages of clustered solid propellant motors of two designs. Seven type A motors for the first stage, and three identical motors make up the second stage. Six smaller, type B motors, are clustered together to form the third stage, and a single type B motor forms the fourth stage. This vehicle has a gross weight of about 25 million pounds, in Fig 3. it is shown in comparison with a 6 million pound liquid vehicle having the same payload capability. The Washington Monument model gives an indication of scale. The first three stages are designed to achieve parking orbit. After the required coasting period, the fourth stage is ignited and the 130,000 lb. payload is brought to escape velocity at burnout. A thrust vector control system is used on each stage to maintain directional control. However, no thrust termination devices are used on the large primary motors. Each stage burns to completion, and small vernier or secondary motors trim the velocity errors as the vehicle goes into its parking orbit and possibly again at stage four burnout. The verniers required are small rockets containing about 5,000 pounds to 15,000 pounds of either storable liquid or solid propellant. Their thrust termination is within the state of the art. The solid rocket motor casings, tied together in shear, serve as the primary airframe structure. Columns and truss members are used for interstage structure, which contains provision for positive separation. The diameter of the first stage assembly is 77 feet, and the height of the injection vehicle to the separation plane between the fourth stage and payload is 220 feet.
2. PERFORMANCE
Performance parameters are presented in Table 1. Three dimensional point mass trajectories were computed assuming eastward launch from AMR (Atlantic Missile Range). The first two stages were flown gravity turn after a short vertical ascent. The third stage was flown using a constant attitude in intertial space and, alternately, with constant inertial pitch. The two modes of computation gave essentially equivalent results. It should be noted that the maximum acceleration, occurring near the end of the third stage burn, is 5.3g, an acceleration tolerable by man. The mass will be distributed between the third and fourth stages such that the velocity at the end of the third stage burn is slightly short of achieving circular orbital velocity. Part of the propellant weight in the vernier, assigned to the fourth stage, is used to make up this velocity difference and insert the stage and payload into a circular LEO. The sizing was accomplished by assuming values for the specific impulse and for the stage propellant mass fraction. The primary considerations used in the sizing and trajectory design were performance capability, airloads, achievement of a parking orbit at the end of the third stage burn, and metting manned acceleration requirements.

3. PROPULSION
The motors represent a typical design concept to demonstrate the feasibility of making very large units. Detailed characteristics are shown in Table 2. Motor A is used for the first and second stages; Motor B is used for the third and fourth stages. A specific impulse (ISP) of 245 seconds at 1000 psi and sea level optimum expansion was assumed. The performance parameters assumed here are all within the present state of the art. The reproduceability of performance should exceed that of present, relatively small motors because continuous mixing techniques are contemplated and large amounts of propellant will be cast in each motor. It is believed combustion instability is unlikely to be encountered; if it does occur, modern techniques should preclude any serious delay in schedule. In recent years, powdered aluminum has been found to be an excellent suppressant for combustion instability in existing solid propellant boosters which use ammonium perchlorate solid propellants. As aluminum quantities are increased up to 15-17%, propellant performance as well as aluminum's effectiveness as a suppressant increases. None of the relatively large motors that utilize these high aluminum composite propellants (Minuteman, Pershing, Polaris) has shown any sign of combustion instability; the propellants under consideration contain these high aluminum concentrations. Unitized motors were used in this study in order to investigate ground facilities required to support this type of a design. It is not necessarily recommended that they be chosen over segmented motors. However, since new propellant processing facilities would be constructed for this vehicle, they can be designed to accommodate either approach. The final decision between unitized or segmented motors should result from a study of the effects of the choice on the flight vehicle. Experience has shown that it would be undesirable to compromise the flight vehicle because of ground support equipment requirements unless a question of feasibility is involved. Propellant would be processed in a new, continuous mix plant at a site strategically chosen for the raw material supply, its proximity to the launch site, and the practicability of shipping finished rocket motors. Propellant facilities similar to those under consideration are already in production and have demonstrated the quality of product and high production rates necessary for the program. For example, Aerojet-General Corporation recently cast a 100 inch diameter charge, weighing about 200,000 lbs, at the indicated rate, then cured and fired it successfully. It is of interest to note that the static test record, based on high frequency response instrumentation, gave no indication of combustion instability.

4. THRUST VECTOR CONTROL (TVC)
A TVC system is needed in each stage to compensate for the effects of CG displacement, thrust misalignment, and unequal thrust (Particularly at ignition and burnout). It must also counteract aerodynamic forces during first stage burn. Canting of the rocket motor nozzles so the thrust vector is directed near the burnout CG of the stage helps reduce some of these disturbances but cannot be used when the angle of cant is too large. TVC systems that could be used include jet vanes, secondary injection, moveable nozzles, auxiliary rocket motors (verniers), and jet tabs. Each has distinct advantages and disadvantages, and a choice can result only from a detailed system study considering guidance structures and spacecraft. Jet vane systems have demonstrated good flight records and offer a feasible solution to this problem. However, a gas pressurized system may prove to be inherently more reliable and was chosen for more detailed study in order to provide a basis for weight and cost estimates.

5. STRUCTURES
The largest structural weight item in the vehicle is the rocket motor casings. Fortunately, considerable experience in the design of cylindrical pressure vessels is available, and the large size considered here should introduce no problems which differ basically from those already encountered elsewhere. Consistent with the philosophy already presented, a heat-treatable martensitic steel with a yield strength of 165,000 psi was chosen in this study. Toughness and ductility shoudl be high at this strength level. However, this conclusion is tentative until more information is available for the 3/4 to 7/8 inch thick material considered for the type A motor case. The interstage structure, as presently conceived, is either a space frame or a set of three spaced columns. Loads would be transmitted to the motor cases as concentrated loads acting on truss pads attached to the motor domes or as concentrated line shears. This technique was used on the Sargeant motor case and is extensively used in the large water tanks and pressure vessels in the power generating and chemical industries. Two types of rocket nozzles were considered: 1) a graphite/steel heat-sink type, and 2) an ablative plastic type. The use of ablative nozzles appears very attractive for this application because the linear ablation rate will be the same or slightly less than that of smaller motors; therefore the percentage change in throat area will be acceptably small. Either type appears feasible; scaling laws predict less severe conditions than for existing nozzles despite the relatively long burn times, provided that weights are scaled with impulse.

6. STRUCTURAL DYNAMICS
Major structural dynamics effects such as overall dynamic loads and aeroservoelastic stability were examined qualitatively to ascertain whether extrapolation in size or weight would adversely affect the feasibility of the solid propellant NOVA. By using aeroelastic model theory, and by assuming that dynamic magnification factors associated with transit through discrete gusts are independent of vehicle size, it can be shown that a large vehicle should encounter no more severe loading in relation to its strength than a small vehicle is dynamically similar. Dynamic instability of the type characterized by adverse coupling between the autopilot and the elastic airframe can become a serious problem if a conventional flight control sensor installation is employed. However, an adaptive flight control system using "transducer arrays" (a number of rate gyros discretely positioned over the length of the vehicle without outputs electronically summed, can be employed to suppress the coupling of the autopilot with the body bending modes. Thus, low bending mode frequencies, per se, need not lead to dynamic instability problems. The solid propellant vehicle system carrying liquid payload rockets or liquid secondary injection system poses relatively minor liquid sloshing problems because of the low percentage of liquid mass.

7. AERODYNAMICS
The need to aerodynamically shroud the vehicle has been examined from the aspects of heating, drag, and unsteady flow characteristics. Maximum laminar and turbulent heating rates were examined for the velocity-altitude information obtained from a powered flight computer trajectory. A conservative analysis indicates that the heating will be no problem on exposed structural members or motor casings. An estimate of aerodynamic drag for a body of this type is not readily calculated. The examination of some experimental data and calculations based on two simple drag models indicate that the peak drag to thrust ratio is approximately 0.15, an acceptable value. Unsteady flow effects between motor casings may result in high local vibration loads. Local fairing should alleviate this condition. In none of the cases studied could a positive requirement for shrouding be determined. Consequently, only a shroud from the payload to the top of the fourth stage has been indicated. The weight of this shroud was charged to the second stage, since it would be discarded at the end of the second stage burn. The maximum dynamic pressure (max q) expected is approx. 1400 psf, and at first stage separation the dynamic pressure is approx. 400 psf. These pressures are acceptable for the vehicle under consideration. Trajectory shaping can be used to reduce the max pressure to a value of less than 1000 psf if desirable.

8. ASSEMBLY AND ALIGNMENT
With proper attention paid to details, assembly and alignment should provide no difficulty. Provision will have to be made for a temporary framework to suppor the motors of a stage during construction until the stage is structurally tied together. Erection loads on the individual motor cases will have to be considered in their design. Vehicle loads caused by wind and wave action during storm conditions while on the launch pad are expected to be small for the blunt, dense, solid propellant vehicle, and temporary guys or bracing are adequate protection. Techniques are available to provide CG control without measuring absolute weight. If further study indicates that accurate absolute weight measurement is required, an increase in the existint capability for accurately calibrating load cells is required. Load cells of the necessary size are available.

9. GUIDANCE REQUIREMENTS
The guidance of the vehicle has not been examined in detail, but it is felt that no unusual problems will be present. The general philosophy of this program can be applied to the guidance area, and adequate space with a controlled environment and ample weight for the required equipment have been provided. The weight should allow for underrating components and providing redundancy wherever needed. For example, 3,000 lbs and 350 cu. ft. of space were allowed for the guidance and control electronic compartment alone. Since it is considered that midcourse correction capacity will be required in the spacecraft, the precision and performance required of the injection vehicle are rather modest. Indeed, guidance capability equivalent to that required for military weapons is adequate; no advancement in the state of the art appears necessary.

SPECIALIZED FACILITIES
1. PROPELLANT PROCESSING FACILITY
Because of the quantity of propellant needed and the size of the loaded motors, water transport at the propellant processing facility is required. The site should be 1) convenient to Cape Canaveral (the assumed launch site) by barge transport, preferably through an inland waterway to avoid long exposure to the open sea, and 2) provide in-plant barge transport during processing. There are many islands and coastal areas from Texas to South Carolina that could meet the above requirements. A typical site, Skidaway Island, in Georgia, has been chosen only to demonstrate the feasibility of such and approach (see Fig 5 and 6). It is assumed that the island is undeveloped and that all waterways used during processing will be dredged completely. A description of the facilities is given in Table 3. The propellant materials considered are typical components of propellants, now used in large motor programs. New production capability is required for all of the ammonium perchlorate used in this program (8 million pounds per month at a vehicle launch rate of one every two months). Power requirements for this production are not excessive. The other materials required are readily available or could be made available on relatively short notice. Propellants of the type under consideration have never been known to detonate at the operating temperatures to be used. Nevertheless, since no experience at this size is available, all facilities have been sized and sited on the basis of class 9 or 10 propellant.

2. LAUNCH SITE OPERATIONS
The launch site, GSE, and assembly and transportation techniques are governed by the following criteria: 1) the complex must be in operation within 2.5 years of the commencement of the program if it is to be used for motor static firing tests; 2) the complex must be an economical and practical system for accomplishing the assembly and launch operation; and 3) the possibility of loss, because of a launch failure, of costly long lead time GSE must be minimized. The launch site is assumed to be located near existing range facilities and sources of manpower at Cape Canaveral. It is evident that because of the acoustic and explosive safety distances involved, the solid propellant could not be launched directly from the Cape. The choice of an offshore launching pad is indicated when the additional launch complex requirements are considered. In determining support systems for the assembly and launching of a vehicle of this sie, it is useful to observe the experience which exists in other areas of endeavor. Normal operations in the large civil engineering industry and in marine and naval architecture closely parallel the erection and handling techniques required for this vehicle. It is from this existing technology that the optimum solution should be derived. Fixed underwater supporting structures for the depths required are encountered in bridge foundations and dam construction. Shipment of the weights required occurs daily in normal tug and barge operations. The erection and assembly of large pieces of of equipment is within ship building and repair technology. (Indeed, the largest manmade moving objects are ships of one form or another.) Thus the optimum launch complex to meet the stated objectives is a fixed offshore launch pad supported by vessels and barges and utilizing the Cape Canaveral range and support facilities.

3. TRANSPORT AND STORAGE
Loaded solid propellant rocket motors are stored in their respective loading barges at the propellant processing site. Motors are transferred to the transport barges by the floating crane at the processing site. They are then tugged to the launch site at a rate determined by the launching schedule. Guidance, telemetry, and associated equipment is transported to the launch support area at the Cape by conventional means. After checkout and maintenance, the equipment is barged to the launch platform for installation.

4. SITE CONSTRUCTION
Since vehicle assembly and checkout procedures require four months (Fig 7), two launch pads are required to allow launchings on two-month centers. These pads (Fig 8) would be placed approximately 6-12 miles off the coast of Cape Canaveral, with a similar distance separating them for acoustic and quantity-distance safety requirements. The launch pad would then be in 60 to 100 feet of water. This depth is typical of that found in bridge foundations, and this construction is the type desired. For a solid porpellant vehicle, the dimensions of the pad need only be large enough to provide structural support for the vehicle. The large weight of the vehicle as erected would provide considerable stability against overturning due to wind and storm conditions. Hurricane protection would be limited at most to temporary guying and bracing. All auxiliary operations would emanate from the crane or support vessels. A hold-down structure would be neither desirable nor practical for use during the launching of a solid propellant vehicle. A breakwater one mile in overall length is required to shelter each pad area sufficiently to allow shipborne operations in all but gale conditions. This breakwater would be of rock construction and 10 feet above high tide. General support, electronic checkout, engineering personnel, and auxiliary equipment receiving and storage buildings would be located on or near the Cape. Several docks and a crane (YD-4) would be required to transfer this equipment to barges for transport to the launch pads. The distance offshore between the pad and the Cape has been specified as that allowing for inhabited areas, according to the mass of the propellant involved. Therefore, the launch control center would not be a blockhouse in the normal sense but, structurally, an ordinary building located on the Cape in the general support area. This center would contain the launch control instrumentation and equipment and be connected to the pads by underwater cable for phone and electrical connections. Most of the prelaunch instrumentation and monitoring measurements would be made, however, by direct radio link between the vehicle and LCC. An umbilical mast would provide electrical connections to the spacecraft up until launch, as well as emergency disarming or astronaut exit ladders. This mast would pivot at its base and drop to the water at launch.

5. VEHICLE ASSEMBLY EQUIPMENT
The primary assembly problem is the erection and handling of the first and second stage motors. A crane of 1000 ton capacity and 200 ft hook height is required, with a secondary hook of 200 tons and 300 ft height for the upper stages and payload. The construction and operation of this crane and its support structure are considerably simplified when based on waterborne operations. Currently, the largest movable crane in the US is the Navy YD-171, mounted on a self-propelled moving barge. This is one of four constructed in 1941 and has a 450 ton capacity and 160 foot hook height. Several fixed cranes of this capacity, but with lower hooks, also exist. A scaled-up version of the flat bottomed barge and crane may be specially constructed for this purpose and built to contain personnel and service areas. Its construction is considered quite feasible. The use of a floating crane for the handling and careful positioning of very large loads is well established procedure. It provides the most economical and shortest lead time solution to the erection problem and allows for convenient removal from the pad area prior to launch. Personnel service platforms around the vehicle are made in prefab sections and attached directly to the vehicle structure. The loads introduced are small compared to the load carrying capability of a solid propellant vehicle skin. Personnel access is via an elevator tower on the craneship and gangwalks across the vehicle. The platforms are removed prior to flight.

6. STATIC TESTING
The full scale test firings of the individual full size motors could be conveniently accomplished at one of the launch pads. A motor, supported by suitable structure, could be mounted in an inverted position on the pad. Normal launch control instrumentation would be used for this series of tests. The alternate pad would be used for the first flight vehicle. Upon completion of the static test program, the external support structure would be removed and this pad converted to a flight pad for the second flight firing. Alternatively, a static test stand could be constructed at the propellant processing plant and utilized for this purpose. The added area and cost for such a test stand are included in the cost analysis.

7. FIELD OPERATIONS
Individual motors of types A and B would arrive at the launch complex by barge from the propellant processing plant storage area. They would be erected and assembled on the launch pad by the crane barge (YD). Interstate structure, in large prefab sections, would be barged out and installed. The crane-barge would then be transferred to the alternate pad and replaced by the auxiliary support ship (CLS). During the remaining period, the GNC, telemetry, and spacecraft equipment together with the auxiliary gear would be installed and the final checkout operations accomplished prior to launch. The firing of this vehicle from an offshore location at the Cape presents no extraordinary problems with respec tto range procedure. Normal downrange tracking, range safety, and launch monitoring are identical to regular Cape launchings. Firing windows are reasonable and the actual final countdown time for the multistage solid propellant vehicle is relatively short compared to that for existing vehicles. The shorter burning times also result in significantly simpler tracking operations during launch into orbit than for equivalent liquid vehicles.

E. SPACECRAFT CONSIDERATIONS
Spacecraft studies have been limited to crude feasibility determinations of 1) possible configuration constraints, and 2) spacecraft design problems peculiar to the employment of the solid propellant vehicle for a manned mission. The results of the studies can be summarized as follows: 1) there are no spacecraft design problems peculiar to the solid propellant vehicle; 2) the injection capabilities of the solid propellant vehicle appear to be adequate for manned missions if so desired; 3) there are no major spacecraft configuration constraints or limitations due to the injection vehicle. Of the spacecraft environments associated with a large solid vehicle, (vibration, linear acceleration, acoustical, etc.) the one which appears to be the most severe when compared with a liquid vehicle of the same capability is the acoustical environment. Differences in the other environments are minimal. The analysis of the acoustical environment was based on extrapolations from data from smaller motors, However, the results are considered to be conservative. The calculated sound pressure levels are (solid)-=167 db, and (liquid)-=161 db. These levels correspond to a distance of 200 feet from a solid propellant vehicle of 40 million pounds of thrust or a liquid vehicle of 9 million pounds of thrust. Actually, the lower exhaust velocity of the solid propellant motor causes it to have lower acoustic efficiency, so that the pressure level from a liquid propellant vehicle might well be higher than that from a solid propellant vehicle. In any case, it is important to note the high pressure level from either. Factors such as sound absorption in the air (which increases at these higher intensities because of nonlinear damping) and directivity of the sound should decrease the levels by 20 db. Reflection from the pad coudl be minimized by flowing water under the booster at liftoff. These factors suggest taking the pressure level at 150 db. The effectiveness of ear protectors for the astronaut is limited by bone conduction to about 40 db reduction. Thus the pressure level at the ear will be reduced to 110 db, which is below the threshold of pain at 140 db. The maximum total recommended level for speech comprehension is 110 db, so that talking with astronauts during liftoff may be difficult, although additional attenuation by the cabin walls may make it reasonable. It should be emphasized that the attenuation of the sound with distance is equally important.

3. COST CONCLUSIONS
The cost study of the solid propellant NOVA injection vehicle system has revealed some rather unusual, albeit tentative, conclusions. The conclusions reached are a logical consequence of the basic program philosophies.
a. The development costs for this vehicle are considerably less than for a liquid propellant vehicle of the same capability
b. Production costs per vehicle are roughly comparable to a similar liquid vehicle, depending somewhat on the injection mission and total number of flights.
c. Injection costs in dollars per pound of payload for a 300 mile LEO for a 20 vehicle program are $169 per pound total cost and $91 per pound production cost. For these computations a reliability of 95% and 515,000 lbs of injected weight were used.
d. The development costs are low because the usual full scale development support programs are not required-- no full scale vehicle structural testing, no captive firings of clusters, no battleship propulsion program.
e. Production costs per pound for the structure are low because of the inherent simplicity of the solid propellant rocket. Size was simply exchanged for complexity at a nearly constant total production cost.

H. GROWTH POTENTIAL
Once feasibility of a vehicle is established, it becomes a major interest to examine the system flexibility for quick growth to a vehicle capable of performing much more difficult missions in the foreseeable future. A cargo version of the large spacecraft under consideration can place approx. 35,000 lbs gross weight on the lunar surface. Such devices as Moon-mobiles, prefab structures, and life support systems could be delivered directly from Earth, intact and ready to operate with no assembly or disassembly. The 30 million pound solid propellant NOVA would appear to have considerable growth potential beyond this. Substitution of liquid hydrogen stages based on engines now under development, for the third and fourth stages, such that the gross weight is unchanged, results in a vehicle that could place approx 110,000 lbs of men and equipment on the Moon. The first two, very large solid propellant stages woudl be used as developed for the original vehicle. The first three stages of this solid propellant or hybrid vehicle should be capable of placing approx. 930,000 lbs in LEO. If one were to use this weight as an electric powered spacecraft, the payload on the lunar surface using LOX-LH2 for landing on the Moon, would rise from 110,000 lbs to about 215,000 to 440,000 lbs. depending on the allowable transit time. Alternately, this three stage hybrid vehicle with its electric spacecraft should be capable of performing a three-man Mars landing and return in approx 590 days; this performance estimate is based on the radiation shielding of 50,000 lb, 15 lb/man/day sustenance, and a Mars Orbit Rendezvous of the spacecraft with the Mars Excursion Vehicle.

IV. NOVA DESIGN VERIFICATION AND COMPARISONS

The Large Launch Vehicle Program Group (the Golovin Committee), noting the unconventional aspects of the all-solid NOVA approach, requests the Boeing Company and Space Technology Laboratories to make independent and objective evaluations of the solid NOVA concept for 1) technical feasibility of the vehicle concept, 2) accuracy of the cost estimates, 3) realism of the schedule, 4) effect on the vehicle concept if the payload were increased from 130,000 lbs to a more recent estimate of 156,000 lbs, 5) flexibility of the vehicle as a backup for Saturn, and 6) a comparison of the relative reliability of liquid and solid NOVA vehicles. A review of the pertinent conclusions from the critiques (Ref 1 and 2 respectively) should be valuable at this point in confirming or rejecting the concept.

A. CRITIQUES OF THE SOLID NOVA CONCEPT BY STL AND BOEING

The salient features from the JPL, Boeing, and STL studies are compared in Table 5. In this specific comparison a vehicle with a payload capability to escape of 130,000 lbs is used because the original studies were carried out and evaluated on that basis. The comparison with liquid vehicles in section IVB, will use vehicles having an escape payload capability of 156,000 to 160,000 lbs. The critiques differ in details from the JPL designs (this was not unexpected, for they were conceptual design studies) but on the critical questions were in relatively good agreement.

1. TECHNICAL FEASIBILITY

Although vehicle weight estimates varied somewhat among the three studies, all concluded that weight per se was of minor importance and that there were no technical barriers, such as nozzle erosion, propellant physical properties, burn time, combustion instability, guidance, or vehicle dynamics that would prevent the accomplishment of the mission when using an all solid NOVA vehicle. The conclusion was found to be equally valid for a solid Nova with a 156,000 lb payload.

2. GROWTH POTENTIAL

Both Boeing and STL were quick to recognize the unusually large growth potential of the solid Nova. Each commented favorably on it as a distinct advantage.

3. FLEXIBILITY AS A SATURN BACKUP

Boeing also noted that the upper three stages of the Nova, with only a 5% penalty in the original Nova weight, would be capable of delivering 266,000 lbs into LEO, adequate for a Saturn C-5 rendezvous type boost. STL and Boeing also found that other vehicles with escape payloads of 21,000, 30,000, 40,000, and 45,000 lbs were feasible when using the A and B type motors as modules in the first three stages and an Earth storable liquid system in the fourth stage, only performance capability was checked in these cases. Boeing and STL appeared to differ regarding costs and schedules, their evaluations will be discussed separately.
A. Boeing costs and schedules: Boeing agreed closely with the cost estimates of the original JPL report for the 25 million pound vehicle discussed, ie $1.7 vs. 1.65 billion, reporting costs as realistic. In a final analysis, Boeing believed a heavier vehicle was advisable, its program costs would be about $2.2 billion dollars. A schedule of five years was considered realistic by Boeing. Some concern and possibly misunderstanding of the Golovin Committee may have arisen over apparent discrepancies between the Boeing studies on the solid NOVA and separate Boeing solid booster studies being carried out at the time for MSFC. If so, it is unlikely that the discrepancies were ever explained to the committee, STL and Boeing were not asked for briefings on their critiques when explanations might have been made. Later on 6 October, when JPL, Boeing, and STL personnel presented their results of their NOVA studies to the propulsion personnel at NASA HQ, Boeing did explain, when questions arose, that 1) the two studies were not made by different Boeing study groups, and 2) the schedule differences reported in the two studies were not inconsistent. In the Marshall case they were making a parametric study, whereas the JPL cas they had been asked to examine a specific vehicle system.
B. STL COSTS AND SCHEDULE: The STL report concluded that hte cost of vehicle system development, facilities, and 20 vehicle flights would be $4.2 billion, or approximately twice that estimated by Boeing and JPL. Their schedule too, was longer, requiring 6.5-7 years compared to Boeing and JPL estimating 5 years. During the 6 October briefing at NASA HQ, personnel from STL explained their higher costs and longer schedule on the basis of their experience with the Air Force in the ICBM program, such as Minuteman. It should be noted that the JPL method of implementing a program differs significantly from the Minuteman approach. In the briefing, however, STL did conclude that their cost estimates would be much lower than reported and their schedule would agree with the schedules of Boeing and JPL if the solid NOVA philosophy and approach could be implemented. Boeing believed the philosophy and approach could and should be implemented. Because of the different frames of reference or assumptions made in the Boeing and STL studies, one can draw a very important conclusion: Approximately $1 to 2 billion can be saved in a 20 vehicle program and 1.5 to 2 years can be cut off the 6.5 to 7 year schedule if the solid NOVA philosophy and program approach can indeed be implemented.

B. COMPARISON OF SOLID AND LIQUID NOVA CLASS VEHICLES

The comparisons will be confined to 1) vehicle and propulsion reliabilities, and 2) vehicle schedules, and 3) program costs. Most of the data gathered during the latter part of 1961, when the liquid vehicle configuration given the most serious consideration for the NOVA mission, now designated the C-8, consisting of 8 F-1 engines in the first stage, 8 J-2 LO2/LH2 engines in the second stage, and 2 J-2 engines in the third stage. Its payload capability for the escape mission assuming no engine out was approximately 160,000 lbs. Therefore, the characteristics of the solid NOVA with the 156,000 lbs escape payload, derived from the Golovin Committee supplementary work statement, will be used for comparing schedules and costs in section B2 and B3.

1. FLIGHT RELIABILITY OF LIQUID AND SOLID SYSTEMS

Although vehicle reliability as a function of liquid or solid NOVA time schedules will be of major interest in comparing the two, a review of absolute reliabilities for some recent liquid and solid vehicles is also valuable in itself. Unlike costs and schedules which represent best estimates of future values, reliabilities represent concrete evidence of what has actually been demonstrated in the past, they give some indication of the likelihood of mission success. Fig 10 shows composite plots of overall vehicle reliability for 1) four solid propellant vehicles and 2) six liquid propellant vehicles as a function of the number of vehicle launches. The former group included Polaris, Minuteman, Pershing, and BOMARC B booster; the latter group included Atlas, Titan, Jupiter, Thor, Redstone, and BOMARC A booster. Individual reliability records of these vehicles, taken from Ref. 1, are shown in appendix C. The composite plots of Fig. 10 reveal that solid propellant vehicles have demonstrated significantly higher reliability than liquid propellant vehicles at a given early date. Early reliability is especially significant because the number of total launchings of a NOVA sized vehicle, even for a 10 year period, will probably only be about 60-120; there will only be a limited number which can improve flight reliability. The outstanding exception is Polaris. Development of Polaris probably constituted the gravest challenge to engineering rocket technology the country has seen. The relatively low reliability in early flights, as well as in the static test record (see section IVB2), reflects the very significant advancements in the state of the art that were required in many technical areas: 1) design to severe volume constraints for vehicle stowage, 2) chambers based on notch-sensitive materials of unusually high strength, 3) high performance propellants with much higher flame temperatures and more erosive exhaust gases, 4) multiple nozzles which incorporated unusual materials and design features, and 5) new technique for TVC, jetevators. Failures in the early stages of the static test and flight program were to be expected.

2. SCHEDULES

In the further comparison of the liquid C-8 and solid NOVA programs, the schedule, or time required to launch astronauts, for the solid NOVA has been estimated at 5 years by JPL and 5.25 years by Boeing (Ref 1 part II) The liquid C-8 schedule has been estimated at 6 years by the Fleming Committee, and in separate studies, at 5.75 years by Rocketdyne. It is important to note that Dr. Dergarabedian, director of STL's Systems Research Laboratory, has permitted a quote of STL's comparative evaluation based on their solid Nova critique and their liquid class Nova vehicles for MSFC. "the solid Nova will be available at least a year sooner than the liquid system" STL's estimates of 6.5-7 years for the solid Nova, therefore, means at least 7.5-8 years for the liquid C-8 and must not be compared with the 5.75-6 year estimates of liquid Novas by other agencies. The ground rules were different. "STL believes virtually all estimates by other agencies are idealized (based on success) rather than realistic schedules and are therefore too short."

3. PROGRAM COSTS
Dr. Dergarabedian, Director of STL's System Research Laboratory, again has permitted a quote of their comparative evaluation, "they estimate the liquid C-8 will cost approximately twice that of the solid NOVA or about $8.7 billion." STL cost estimates should not be compared to other cost estimates shown because the assumptions for the studies differ as mentioned in the section on schedules. RAND estimated the costs of the liquid C-8 at $4.7 billion, if the launch rate is one vehicle every two months, while Rocketdyne has estimated $4.5 billion (Ref. 3). RAND reports that MSFC costs for the liquid C-8 are in excellent agreement with their estimates. RAND has also estimated the cost of a liquid C-8 program that included 100 vehicles (Table 10, to take into account the decreasing recurring cost of the liquid vehicle, at $10.5 billion. The solid Nova program for 100 vehicles has been estimated by JPL, for a constant vehicle cost, at $8.3 billion; in practice, production costs of the solid Nova vehicle would, of course, decrease. It is of interest to note the cumulative vehicle program costs to perform a specified practical mission when 1) liquid C-8 vehicles and 2) solid NOVA vehicles are used. Fig. 13 indicates the cumulative injection vehicle system costs to supply a lunar base with 150,000 lbs of useful payload per year. The Nova, C-8, Saturn C-5, and Saturn C-3 were estimated to have, respectively, 30,000, 15,000, and 7,500 lbs of useful payload (payload weight after weight of the weight of the landed spacecraft vehicle itself). The values at zero time represent the costs for R&D, facilities, and 10 vehicle flights. Based on JPL estimates, the total solid-Nova vehicle program cost, after 100 flights and 18.5 years, would $8.3 billion. The solid Nova data in this figure, unlike the liquid system, do not include decreasing production costs for the vehicles and probably penalize the solid vehicle. Both solid and liquid vehicle estimates assume reliabilities of 100%, and therefore, penalize the solid Nova again in a relative sense. If STL cost estimates had been used, both curves would have had higher values but would undoubtedly resemble those shown, unfortunately, no cost breakdown for the liquid vehicle was available from STL for preparing such curves. The Saturn C-3 and C-5 costs, estimated by RAND, have been included as a matter of interest. Although they start with lower development costs than the liquid C-8, they rise rapidly to values appreciably higher. After 10 years their program costs would be $9.0 billion and $8.6 billion, respectively. Figure 14 indicates program costs to supply the lunar base using 1) solid Novas and 2) liquid C-8's, when vehicle reliabilities estimated by Boeing (see section B2) are used and when the recurring costs of the vehicle are decreased with time again. Again, RAND's costs are used for the liquid C-8, with the initial vehicle launch cost dropping from $135 million per vehicle to $70 million for the hundredth launch; JPL cost estimates for the solid Nova decrease from $75 million initially to $65 million for the hundredth launch. Thus, if the liquid C-8 were used to supply the lunar base for 16 years, the overall program would cost about $11.7 billion; if the solid Nova were used, the overall program would cost about $8.25 billion, or 70% of the former. It may be concluded that the solid Nova of conservative design would have a significantly lower development cost and lower recurring costs to perform the specified mission than would the liquid C-8, the Saturn C-5, or the Saturn C-3.

4. GROWTH POTENTIAL

No attempt will be made to compare the growth potential of the solid Nova with the liquid C-8 when advanced developments are incorporated; too many possible combinations exist to make a meaningful study at this time. However, a comment pertaining to the growth is worth noting. The very large solid Nova first stage provides a much greater lift capability than other ground launched systems under consideration. With any given advanced propulsion system as upper stage, the solid Nova will provide better overall performance capability than the other first stages. The reason for this is simply that the solid Nova first stage is much larger and has a higher thrust (about 50 million pounds). The large launch weight that was considered a drawback of the solid Nova by many at the outset of the studies proves to be an asset in this respect.

V. CONCLUSIONS

An examination of manned lunar and planetary missions for the foreseeable future reveals that, despite the obvious challenges to our booster and performance capability, system reliability for the overall mission profile constitutes the greatest potential obstacle to successful mission accomplishment. The latter obstacle arises because of major constraints from 1) non-expendable payloads, ie men, 2) the need for large, complex boosters, 3) very high unit cost, and 4) a long and complex sequence of critical operations to complete a mission. One important aspect of the mission profile, the injection vehicle, has been examined conceptually, with reliability, rather than performance or minimum weight, as the ranking criterion and the following conclusions are reached:
1) Vehicle system reliability will probably prove to be unacceptably low for complex manned space missions if performance (weight reduction) and reliability carry the same relative importance in the new development programs that they have in past missile and space programs.
2) A design and development philosophy which incorporates the important principles 1) unusual conservatism and 2) design within the state of the art is advocated as a mechanism whereby substantially higher reliability can be realized.
3) By using this philosophy and capitalizing on unique characteristics of an all solid rocket vehicle, a basically new approach to high early vehicle reliability in flight becomes possible; program implementation would consist of 1) analyses and evaluation of the motor designs, vehicle structure, and vehicle flight dynamics in scale model tests during the preliminary design and large scale tooling period, followed by 2) assembly of scaled up, full size components as a complete vehicle for immediate use.
4. An all solid propellant injection vehicle in the 25 million to 35 million pound launch weight class, studied by JPL and critically evaluated by STL and Boeing, has been found feasible.
5. When compared to liquid propellant or hybrid stage vehicles of comparable performance capabilities, it is believed that the development risk of the all solid Nova would be very low and the resultant vehicle reliability at an early date predictably high- provided the suggested philosophy and program approach were utilized.
6. From a technical viewpoint, the schedule for the all solid Nova injection vehicle studied is significantly shorter and the program costs markedly lower than hybrid stage and liquid propellant vehicles of comparable payload capability despite the much greater launch weight of the solid vehicle.
7. By replacing solid upper stages with liquid hydrogen/liquid oxygen stages or nuclear stages as these become available, the growth potential of the all solid vehicle is especially favorable for space missions which require larger payloads and greater sophistication. Such hybrid stage vehicles show great promise for future planetary missions using electric or nuclear propulsion.

I. INJECTION VEHICLE

A. DESCRIPTION AND OPERATION

The injection vehicle in its primary form was estimated to have a payload capability of 130,000 lbs to escape or 500,000 lbs to LEO and a launch weight of 25 million lbs. Subsequent and more sophisticated calculations show that the vehicle launch weight is closer to 30 million lbs if the vehicle is to provide the indicated payload capability. It is believed that this revision will affect size and cost only and in no way bears on the feasibility of the concept. All discussions in Appendix B are based on the 25 million pound vehicle.

1. VEHICLE

The injection vehicle (fig. B-1) consists of four stages of clustered solid propellant motors. The motors in a stage are joined by intrastage structure; successive stages are joined by interstage structures which have provision for positive separation. Each stage has its own thrust vector control (TVC) system. The fourth stage carries the 130,000 lb escape payload, guidance, control, and telemetry for the injection vehicle, and a vernier propulsion system. For this study, the injection vehicle was sized to inject its fourth stage into a parking orbit. The fourth stage then injects the 130,000 lb payload into an escape trajectory. The typical sequence of operations is:
1. stage one ignition
2. stage one burnout and separation
3. stage two ignition
4. stage two burnout and separation
5. stage three ignition
6. stage three burnout and separation
7. vernier velocity correction into parking orbit
8. orbital coast
9. stage four ignition
10. stage four burnout and payload separation.
11. vernier velocity correction into final escape trajectory

The desireability of maintaining flexibility with respect to the altitude of the parking orbit is discussed in Section I-Q. To accomplish this would require a resizing of the first three stages of the vehicle and of the vernier velocity system on the fourth stage. Flexibility and other mission requirements might require the use of a storable, restartable liquid fourth stage. Detailed consideration of these requirements ws beyond the scope of this study, but in further studies of the concept to greater depth the applicability of such liquid upper stages should be examined.

2. SPACECRAFT

The spacecraft is composed of the following elements:
1. Three man Apollo Command Module
2. Mission Module (Apollo) *note- at the time, NASA was strongly considering a Soyuz-like three-component CSM consisting of a Service Module, much smaller reentry Command Module, and a disposable Mission Module
3. Midcourse, lunar retro, and takeoff propulsion systems (direct ascent lunar mission mode, which was preferred then)
4. Associated structure, guidance and control, and communications systems.
5. Abort propulsion and structures.

The operational sequence for a manned or unmanned lunar landing and return is:
1. Launch to injection
2. Midcourse correction
3. Transit
4. Site acquisition and lunar retro firing
5. Lunar landing
6. Lunar operations
7. Takeoff
8. Transit
9. Midcourse correction
10. Earth reentry and recovery

B. CONFIGURATION

The injection vehicle configuration is shown in Fig B-1; a weight summary is shown in Table B-1. The following factors entered into the decisions which resulted in this configuration:
1. Only two different solid propellant rocket motors would be developed. The Type A motor would be used in stages 1 and 2, and the Type B motor would be used in stages 3 and 4. This obviously requires that the number of motors in stages 1 and 2 be integral multiples of the Type A motor and that the same be true for the Type B motor in stages 3 and 4.
2. The logistics of motor manufacture, transportation, handling, and field erection dictate a unit weight that can be easily handled.
3. The mission constraints require that: 1) the fourth stage and payload be injected into parking orbit at the end of third stage burn, and 2) the fourth stage be able to inject a 130,000 lb payload into an escape trajectory from parking orbit.
4. The requirement that present state of the art be used led to the assumption of specific impulse and stage mass weight ratio used in sizing the vehicle.

The geometry of clustered cylinders and the desire to provide reasonable load paths for interstage and intrastage structure played a role in determining some of the geometrical details of this particular configuration. The result shown is based on the constraints and the considerations described above and is considered to be representative and adequate for the purpose of this study. It is certainly not intended to be put forth as an optimum configuration. The physical size of the vehicle, 77 foot base diameter and 218.5 feet to the top of the fourth stage, is similar to vehicles which have already been considered in various other studies. The gross weight at takeoff, 25 million pounds, is larger than previously considered. It can be compared to some water heaters used in steam electric plants that are 150 feet high and weigh 20 million pounds or to an aircraft carrier weighing 120 million pounds. Interstage or intrastage, and TVC systems structure is not shown on Fig B-1 becuase of the small scale. These items will be discussed in the following sections. The absence of a complete aerodynamic shroud is intentional since the need is not defined at the present time. Each of the solid propellant motors utilizes a single fixed nozzle and has a thrust program designed to produce acceptable accelerations at stage burnout. In the design shown, it was necessary to assume step mass fraction (ratio of propellant to total stage weight) in order to estimate the performance capability of the vehicle. A stage mass fraction of 0.87 was assumed for, the first and second stages, and 0.91 was assumed for the third and fourth stages. Preliminary weight estimates of case and nozzle design, propellant charge configuration, interstage structure, cluster configuration, and TVC requirements have been made and are discussed in Section I-R. Aerojet-General has stated that the first stage of a 7 million pound gross weight vehicle can achieve a stage mass fraction of 0.889, utilizing 7 motors of 140 inch diameter as propulsion elements. By the utilization of a single unit of 288 inches in diameter (roughly the same size as the Type A motor) they can achieve a value of 0.893. Grand Central Rocket Company has stated that the first stage of a 10 million pound gross weight vehicle, consisting of 16 ten foot diameter motors, would have a stage mass fraction of 0.886. It is expected that these numbers may be optimistic and attainable values can be established only by thorough preliminary design, integrating mission system requirements, engineering mechanics, propulsion, guidance, control, vehicle telemetry, power, and last but not least, payload vehicle interactions.

C. ROCKET MOTORS

The Type A motor is 300 inches in diameter and has a length to diameter ratio of 3.26:1. It contains a little less than 2 million pounds of propellant. The motor has, basically, a star shaped grain perforation. Motor volumetric loading is approximately 82%. The average vacuum thrust is 6.4 million pounds and the burn time is 85 seconds. The Type B motor is 220 inches in diameter and has a length to diameter ratio of 2.33:1. It contains about 350,000 pounds of propellant and also has a star shaped grain perforation. Motor volumetric loading is approximately 88%. The average vacuum thrust is 740,000 lbs and the burn time is 138 seconds. These motors represent a typical design concept to demonstrate the feasibility of very large units withing the current state of the art. The size of the motor and grain configuration can be varied to satisfy particular motor requirements. Both motors are designed to utilize existing available propellants, with solid loading of approximately 82 and 88%, and containing powdered aluminum in the 15-20% range. The Type A motors were designed for a nominal motor chamber pressure of 800 PSI with a burn rate of 0.64 inches per second. The Type B motors were designed for a nominal motor chamber pressure of 350 PSI with a 0.46 inch per second burn rate. These burning rates are available in current propellants both from Aerojet and Thiokol. A specific impulse of 245 seconds at 1000 PSI and sea level optimum expansion was assumed. This performance level has been verified many times with both polyurethane and PBAA propellants in current motors. The difference between the measure ISP value and the calculated ISP value far exceeds the correction for heat loss and nozzle divergence angle. It is believed that the bulk of these excess losses is due to velocity and thermal nonequilibrium of the aluminum oxide particles (products of combustion of aluminized propellants) with the gas. Calculations performed at JPL show that aluminized propellants which deliver an ISP of 245 (at 1000 PSI and sea-level optimum expansion) in ordinary nozzles should deliver substantially more in very large nozzles such as those considered for the solid NOVA. The vacuum specific impulse used for A and B motors represents conservative estimates of this ISP improvement in very large nozzles based on measured aluminum oxide particle sizes and one dimesional, 2 phase flow calculations performed on the IBM 7090. The reproduceability of performance for the type of very large units under consideration is expected to exceed that of present, relatively small motors. For example, deviation in burn rate will be less because of amuch larger sample of propellant is exposed to burning.

D. MOTOR CASE DESIGN

The design of the motor cases of the size considered in this study presents no problems basically different from those that would be encountered in the design of smaller size cases.

E. NOZZLE DESIGN

Single fixed nozzles will be used for both type motors. They represent configurations scaled up from known practical steel, graphite throat nozzles. They have been canted in the first and second stages so that the thrust of each motor passes through the vehicle CG at burnout. Their length is established through the use of standard design procedures for the bell shaped exit nozzles. Nozzle materials and fabrication methods already developed seem entirely adequate for building nozzles for these large motors. Since time was not available during this study for detailed nozzle design, weights were estimated from existing information. Two nozzle designs are used in the vehicle. The average thrust of Type A motors on the first and second stages is 6.4 million pounds; the average thrust on the Type B motor on the third and fourth stages is 740,000 lbs. A survey of available reports containing nozzle designs was made and a plot of thrust vs. nozzle weight was prepared (Fig B-6). The weight of the two nozzles was estimated from this plot. The nozzles considered in preparation of Fig B-6 have expansion ratios of about 10:1 and a chamber pressure range of 700-1000 PSI. The nozzle for the type B motor has an expansion ratio of 33:1 but the chamber pressure is only 360 PSI. Since the nozzle weight can be expected to increase with an increase in expansion ratio and decrease slightly with a decrease in chamber pressure, an estimate of 3500 lbs for the nozzle weight seems reasonable. The estimate of 34,000 lbs for the weight of the large type A motor nozzle appears to be sufficiently conservative, since as can be seen from Fig B-6, the weight ranges from 23,000 to 39,000 lbs for a thrust of 6.4 million pounds. A plot of total motor impulse vs weight was also prepared; it indicates that the large type A nozzle is reasonably conservative. The weight of the type B smaller nozzle appears to be less conservative in this plot, but the value quoted above was used since the weight range for the examples considered here, at a slightly lower total impulse, is rather large. Since the diameter of the nozzle throats considered for this vehicle is large, ablative type nozzles may be more desireable and easier to design in this size than in the current advanced developments being pursued. Although not considered here, they represent a design which promises a weight savings, at no decrease in reliability, over that estimated in the weight analysis.

F. COMBUSTION INSTABILITY

Combustion instability is not expected to be a problem in large solid propellant rocket motors. Although many motor programs have had combustion instability problems, a solution has always been found and the programs have been successfully concluded. There are a number of major reasons for not expecting combustion instability in a large composite ammonium perchlorate aluminum solid propellant motor. The last three large composite solid propellant rocket motors, Minuteman, Polaris, and Pershing first stages, have not had combustion instability problems using the type of propellants under consideration. Aluminized propellants have shown instability have been in large L/D motors at relatively high pressures, around 1500 PSI. Recent experience at JPL with a nonaluminized polyurethane has indicated a stable region below 750 PSI and logitudinal oscillations at higher pressures. Small amounts of aluminum gave stable burning at 1000 PSI. Unfortunately, there are no data in the low frequency range of large motors.

G. SEGMENTED VS. UNITIZED ROCKET MOTORS

The choice of a unitized (monolithic) vs. segmented motor construction does not appear to be very simple. From the standpoint of reliability each segment of a segmented unit contains most of the probable causes for failure of a solid propellant motor. For example, each segment will have associated with it one joint; each segment will have associated with it two ends which tend to be the regions of pull-away (separation) of the propellant from the case, and each segment will be processed and trimmed separately, inviting other failures if proper controls are not exercised. A unitized motor has a similar joint attaching the nozzle to the case and also similar modes of failure. From the reliability standpoint it can be argued that the number of segments or unitized motors should be minimized. The minimizing of the number of units implies maximizing the size of each unit to do a specific job. A number of arguments have been proposed for minimizing the size of a segment from the standpoint of ground support equipment, processibility, and transportability. From the long experience that is available with missile projects, it can be stated that it would be an error to grossly compromise the injection vehicle on the basis of ground support equipment. Existing processing facilities cannot handle and overload of the size being considered. Therefore, separate process facilities which would have to be established could just as easily be designed to handle large unitized motors as single smaller segments. The concepts proposed in this report are based on what appears to be the more difficult problem, that of handling large single unitized motors, although it is not necessarily recommended that this be the final solution. It is believed that there are problems in the ares of hardware fabrication and vehicle assembly which are minimized with the use of smaller segments and, if the flight vehicle can be designed better by the use of segmented motors, the latter should be used. The final decision between segmented or unitized motors should result from a study of the effects of the choice on the flight vehicle.

H. PROPELLANT PHYSICAL PROPERTIES

The structural integrity of the large propellant grain has been examined and found satisfactory. It is anticipated that there will be a scaling effect in extrapolating propellant physical properties of small grains to large grains, and this effect should be studied during the preliminary design. However,on the basis of experience with state of the art large motors, this effect is not thought to be limiting.

J. INTRASTAGE STRUCTURE

The intrastage structure for all stages has two main purposes: 1) to transmit shear forces between motor cases, and 2) to tie motors together radially. The shear forces between motors arise from thrust variation, a time lag between motor startup and burnout of individual motors, and body bending. The shear forces in the first stages can be transmitted by means of short, thin walled cylinders filling the interspace between the clustered motors. These cylinders may be attached to the motor cases with high strength bolts, through lugs welded onto the cases during fabrication. The third stage motor cluster could develop shear transfer between motors by means of half-cylinders, because of the nesting of the third and fourth stages. Essentially, the shear will be transferred in the same manner as in the first two stages. The motors of the first three stages will be tied together to resist relative radial motion. This will be accomplished by means of a system of cradels and ties at the top and bottom of each motor case. The tie arrangement will be placed at the junction of the case dome closure so that a minimum amount of restraint will be offered to the growth of the cases resulting from pressurization.

K. INTERSTAGE STRUCTURE

The following study goals have been considered when investigating the interstage structure: 1) feasibility considerations for several structural concepts. 2) development of one or more workable conceptual designs. 3) approximate analysis and weight estimate for at least one design. The bulk of the current work has been oriented toward the first to second stage interstage structure. Separation has been considered only to the the extent that it affects structural configuration. It has been assumed that full ring bulkheads at the ends of each stage are not required, that concentrated loads can be applied to the motor cases, and that interstage structure is capable of transferring thrust differences between motors. The largest loading on the structure is of course axial thrust. Side loads caused by wind shears, thrust misalignments, and the control forces resulting from them will vary along the length of the vehicle because of rigid body and elastic dynamic considerations. Vibration in both the axial and transverse directions will be added to the above loadings. Conserative and realistic load estimates were made and both were used. It is believed that they bracket the actual loads. In each case it was conservatively assumed that the maximum axial and transverse loads occur simultaneously. Four types of structures considered are: 1) Truss, 2) Longeron-tie rod, 3) Monocoque, 4) Nozzle utilization. It is considered that the interstage structure design should be predicated on the required strength rather than rigidity. However, the effect of rigidity on dynamic response must be carefully considered in a preliminary design. Preliminary consideration of truss, longeron-tie rod, and moncoque structures showed each to be feasible. The light weight of an unstiffened nozzle structure rules it out as a practical solution. Sketches of a truss and longeron structure design are in Fig B-7 and B-8, respectively. Analyses have been made of a monocoque and modified monocoque structure. The material assumed for the interstage structures is steel with an ultimate strength of 180ksi and 100 ksi at weld joints connecting major structures, with a safety factor of 1.25. The truss design is shown in Fig B-7. If a tubular cross-section is used, the lower members could have a diameter of 24 inches with a thickness of 1.33 inches for number 1 loading, or a thickness of 0.85 inches for the number 2 loading. Longeron design is shown in Fig B-8. A shell-type structure would require rings at the bottom and top, and the separation plane. It may be possible to use partial, scalloped rings for the top and bottom. The critical condition for this design appears to be shell instability. For a solid shell, the required thicknesses of approximately 1.7 and 1.35 inches.

L. THRUST VECTOR CONTROL

A method of controlling the thrust vector of each stage must be provided to compensate for the effects of center of gravity displacement, thrust misalignment, and unequal thrusts (especially at ignition and burnout). For the first stage it is needed to counteract aerodynamic loads. It must also provide the necessary maneuvering forces. Because of expected motor-to-motor variances in thrust level, it appears very desirable to cant the nozzles so the main thrust vectors for the for the individual units will pass through the vehicle center of gravity. It is expected that the greatest difference in thrust level between units will occur near the beginning of the tailoff of the thrust-time curve. As a result, it would be desirable to point the nominal thrust vector through the stage CG at burnout. The effect of aerodynamic loads might modify this conclusion for the first stage; this can be determined only by more detailed study. Based upon opinions of most of the contractors who have studied the subject, the choice of thrust vectoring means for a big solid booster system lies between jet vanes and secondary injection of of liquid into the expansion cone of the nozzle. It is believed that at least one other system should be considered: that of auxiliary rocket motors which have been developed by Allison and Vickers for NASA. A brief discussion ofthe features of all these systems follows.

1) JET VANES-- Roll, pitch, and yaw moments can be obtained by a single nozzle installation from four (or three) wedge-shaped aerodynamic surfaces positioned 90 (or 120) degrees apart in the expansion cone of the nozzle. Drag force on the jet vane during burning, decreasing the effective motor impulse, is one disadvantage of this system. Another is the materials problem that results from immersing a substance in the high temperature exhaust stream. Jet vanes are used on Sergeant and Pershing.

2) AUXILIARY ROCKET MOTORS-- This method of TVC uses rocket motors or gas supplied nozzles at a location chosen to give optimum pitch and yaw moments for a given stage, usually near the interstage structures. Individual, self-contained motors can be used or, in a system that uses gas supplied nozzles, a single common gas generator can be used. In both cases, the exhausting of these secondary gases occurs throughout main stage burning, changing the missile velocity vector only through nozzle orientation. One advantage of this system is that it is possible to keep the auxiliary nozzles exhausting for a specific time after main stage burnout, providing control forces until the subsequent stage ignites. The major disadvantage of this system, compared to the others considered, is the unreliability added to the added to the vehicle systems through the inclusion of an additional propellant device, whether used for direct TVC or for generating gas for several nozzles. Both of these methods are presently under development.

3) SECONDARY FLUID INJECTION-- TVC in pitch and yaw can be obtained for single-nozzle configurations through injection of fluid into the expansion cone of the nozzle. Decomposition gases can also be used. To date, secondary injection has not been flight tested; however, systems are in advanced development stage for use on Polaris and Minuteman. A special feature of this system is that the injected material affect both the effective mass ratio of the unit and specific impulse, since material is being expelled during operation of the main engine. For this analysis, a secondary injection system which may prove to be inherently more reliable was chosen for further detailed study in order to provide TVC weight and cost estimates. Such a system provides jet deflection as a result of introducing high velocity fluid stream into one side of the divergent cone of the nozzle. An oblique shock is created to cause an effective deflection of the exhaust gases linearly proportional to the mass flow of injectant. It has the advantage over other typical systems of being light in weight while requiring few moving parts. Thiokol Chemical Corp. (ref 24) found that such a system would weigh less than half an equivalent system employing jet vanes. It was also shown to be lighter than a simple auxiliary jet system. However, this conclusion isn't universally applicable. In order to estimate the TVC system, the following assumptions were made: 1) CG offset per stage, 6 inches. 2) Net thrust misalignment per stage, 0.25 degrees (this includes thrust differences of 3.7% 3) Nozzles are canted nominally through stage burnout CG. 4) Ignition delay of 0.3 second 5) Burning time variation of =/- 3.7%. 6) All motors on one side of the CG ignite before all motors on the other side. 7) Effects of motor to motor burn time variations are minimized by long thrust tailoff (approximately 10% of burn duration). This can be further minimized in the first stage by giving the center motor a slightly longer burn time. 8) Maximimum wind shear equivalent to 0.3 g side acceleration with a moment arm of 25 feet (the force is believed conservative by a factor of 2.5 and the moment arm by a factor of 4). 9) CG travel as shown in Fig B-9 was used. It was found that side forces applied at the nozzle exits and total side impulses as indicated below would be required (expressed as a fraction of stage axial thrust or total impulse). Based on these requirements, two systems for the first stage were briefly investigated: a turbopump system and a pressurized system. The turbopump was visualized as powered by a monopropellant liquid or solid propellant gas generator. To minimize turbopump control problems a bypass would provide return flow to the injectant tank as jet deflection demands drop off. Ullage pressure for the injectant tank would be provided by either bleeding turbopump exhaust gas or by heated injectant expanded from the high pressure side of the pump. The pressurized system would utilize helium gas stored at 5000 psi and regulated to deliver injectant at the required pressure. Both systems would utilize multiple valves for each cluster arranged to provide proper pitch, yaw, and roll control moments. (Ideally each motor should be self-sufficient at burnout). Programming would be provided to ensure that all injectant is expelled prior to burnout in order that full utilization can be made of the energy available. The turbopump hydraulic horsepowr requirements roughly approximate that of the Rocketdyne F-1 engine turbopump. This unit requires 128 lb/sec of gas for the turbine and weighs almost 3000 lbs.

M. AERODYNAMICS

1. AERO-FAIRING-- The necessity for aerodynamically fairing the vehicle has been examined for the clustered-type vehicle under consideration. Such a fairing may be necessary owing to three aerodynamic factors-- 1) heating, 2) drag, and 3) unsteady flow effects. Aerodynamic stability is not considered in this section. A) heating-- both maximum laminar and turbulent heating rates were examined. The results of this plot clearly show that aerodynamic heating would not be a serious matter to structural members. Even the use of turbulent heating (which is improbable) in this example would not indicate a greater temperature rise because of the small dimensions involved. Correspondingly, heating of the exposed engine tanks would be even less serious because the heating rates are reduced as a result of the effects of large nose radii and large dimensions from the nose to sonic point. B) AERODYNAMIC DRAG-- An open type configuration such as considered in this study, does not lend itself to simple analysis for the estimation of drag. Nevertheless, it is essential to make an estimate in order to determine what effects may occur to the flight parameters, which are presently based on a drag curve for a fully faired vehicle. Fig B-13 shows the drag coefficient vs Mach number used in the flight calculations. Also shown on the figure is the dynamic pressure, which is observed to peak at a Mach number of about 1.7. For estimating the drag of the large vehicle considered here, pure turbulent friction drag is not an important factor because of the high Reynolds numbers involved. Thus, the major problem is pressure drag at supersonic speeds, although major contributions can arise from the interaction of shock waves and boundary layers and from flow separation effects. Vehicle size will affect only these latter sources. In addition to examining a small amount of existing experimental data, two simple drag models were calculated. These are: 1) no fairings assumed: all motor cases independently exposed to the free stream airflow (stage four and payload considered as one body). 2) partial fairing assumed. Case 1 is not a very realistic condition to consider, since adjacent motors would cause mutual interference, so that the effective drag would be greater than the sum of the individual bodies. However, motors of the first and second stages would be exposed to flow at much less than free stream dynamic pressure, thus lowering the drag. The net results of these somewhat compensating effects is not predictable in short period of study. However, it may be reasonable to assume that by properly streamlining all structural members, placing conical noses on top of stage 3 motors, and judiciously using local fairings throughout the structure, it may be possible to keep the drag coefficient down to twice the basic streamlined case, or a drag coefficient of about 1.0 at Mach 1.7. Considerable wind tunnel development would be required on such an open configuration. Theoretical work on determining proper motor lateral spacing may be fruitful in reducing interference drag. For the externally faired case, the major drag is derived from the nose cone and the transition region between stages 1 and 2 as indicated in Fig B-13. For this case, it is observed that the drag coefficient is about 60% higher than the basic curve. Additional computer effort would be required to ascertain the effect of these potential drag coefficient increases on flight parameters. However, taking the thrust as about 40 million pounds, a twofold drag increase would result in a peak drag-to-thrust ratio of only about 0.15, which is acceptable. C) UNSTEADY FLOW EFFECTS-- This effect may be the one which requires a faired configuration. Fluctuating flows through the partially open vehicle may induce vibrational modes in the structure at levels which are not acceptable. However, as in the case of drag reduction, local fairing and streamlining would be beneficial.
2) BOOST PHASE LATERAL AIR LOADS-- The following trajectory assumptions were made when estimating the boost phase lateral air loads: 1) vehicle CG follows intended no wind flight path (zero-alpha trajectory), crabbing into the wind as necessary to avoid drift, 2) TVC gimble angle varied as necessary to produce the small angle of attack (alpha) required to cancel driftand, at the same time, to counterbalance air load pitching moments. Inasmuch as the above mentioned crabbing alleviates the air loads on the vehicle about 20%, a slight more conservative (pessimistic) assumption would be assume the constant attitude vertical flight. However, the associated drift velocities make this representation unrealistic.
3) BASE HEATING-- The heating of items stored around the nozzles has not been examined. Experience gained by tests and flights of the Saturn first stage should be applicable. The large nozzle exit area and the close spacing of the nozzles woudl make it relatively easy to insert a lightweight fiberglass shield to prevent the flow of hot gases up between the nozzles if it should prove necessary. Radiation heating of the outer surfaces of the motor cases by exhaust flame appears relatively insignificant.

IV. SPACECRAFT CONSIDERATIONS

A. PHILOSOPHY
Spacecraft studies have been limited to a crude feasibility determinations based on 1) 130,000 lbs injected weight, 2) configuration constraints, and 3) spacecraft design problems peculiar to the employment of the solid NOVA for a manned lunar landing and return mission. The results of the studies to date can be summarized as follows: 1) there are no spacecraft design problems peculiar to the solid NOVA when compared to the liquid NOVA; 2) the injection capabilities of the solid NOVA appear to be adequate for a manne lunar mission, 3) there are no major spacecraft configuration constraints or limitations due to the injection vehicle. For the purposes of this study, Apollo 3 man mission and command modules of the Convair M-1 type have been assumed. Mission abort capabilities are assumed to be consistent with the guidelines established by the Space Task Group. It is recommended that the spacecraft be capable of accomplishing the entire mission automatically. Man would perform monitoring functions associated with the control loop and he would help implement scientific measurements and observations. Manned override control capabilities would be provided for emergencies. A possible feature of the manned mission would be to provide an alternate return vehicle on the surface of the Moon as a contingency for possible failures during landing. In this event, man would be equipped to transport himself over the lunar surface from one vehicle to the other.

B. DESIGN: SPACECRAFT CONFIGURATION STUDY--
The Apollo command and mission module concept in the Convair configuration M-1 was arbitrarily selected as the basic vehicle. Two propulsion configurations were considered; an all liquid configuration of the storable propellant type, and a hybrid, using solid rockets for the retro to the Moon and takeoff and three storable liquid vernier engines for mid course correction and vernier descent to the surface of the Moon.

1. WEIGHT BREAKDOWN-
The Apollo studies indicate weights for the command and mission module which are compatible with the permissible weight figures as determined in this study. The approximate gross spacecraft weight breakdown is as follows: Command module- 5,650 lbs; Mission module- 3,500 lbs; Propulsion system- 117,500 lbs; Guidance, control, interstage, propulsion support structure, etc.- 3,350 lbs.

2. CONFIGURATION: LIQUID PROPULSION SYSTEM
Figure B-25 shows the spacecraft with the all liquid propulsion system. A Titan II second stage thrust chamber and pumping system is used for the retro and lunar liftoff maneuvers. Liquid verniers with a 10:1 throttling capability operating from the same pumping system as the main engine are used for midcourse correction and for limited hovering capability at lunar landing. Attitude control can be accomplished by twelve 7 lb thrust engines arranged in pairs on the vehicle. The retro tanks which are to be left on the Moon are used as landing support structure. Energy absorbing material is attached to the tanks. Three stabilizers are deployed during the hovering maneuver. Part of the interstage structure could be deployed to serve as the landing gear and stabilizer rather than using the tanks for this purpose.

3) CONFIGURATION: HYBRID SYSTEM
Figure B-26 shows the vehicle with the hybrid propulsion system in place of the all liquid system. Here, three solid rockets are used for the main retro system and one for the lunar liftoff. The same storable propellant liquid engines, now with their own packaging and pumping systems, are used for the midcourse and vernier maneuvers.

4) ABORT SYSTEM CONFIGURATION
Eight solid motors mounted on a tower on the front of the command module supply the necessary velocity to perform the abort function during third or fourth stage firing, at which time the complete vehicle is separated from the booster. Firing four of these motors performs the abort mission from the pad when only the command module is separated.

5) VELOCITY INCREMENT: WEIGHT CALCULATION
Tables B-6 and B-7 show weight calculations for the normal mission employing liquids and solids, respectively. Should propellant not be used in midcourse or vernier touchdown, a hovering time greater than 60 seconds could be utilized. The desireability of and capability for this maneuver can be determined by the men aboard the spacecraft.

6) COMPARISON OF DESIGN LOADINGS
The maximum acceleration of the injection system is about 5 g. The spacecraft propulsion system of the liquid type imposes 6 g maximum, the hybrid configuration imposes 9 g maximum acceleration on the spacecraft. These levels compare favorably with the design levels adopted in the Apollo studies and justify using structural weights given therein.

STUDY SUMMARY: THE APPLICABILITY OF SOLID PROPELLANTS FOR A NOVA-CLASS INJECTION VEHICLE
AND A COMPARISON WITH A LIQUID VEHICLE OF COMPARABLE CAPABILITY
JPL: California Institute of Technology, Pasadena, California
NASA CR 136573; N74-72160; 9 May 1962

FOREWORD

Early in 1961, JPL began studying injection vehicle aspects of direct ascent approach for the manned lunar landing mission. At the invitation of NASA HQ, preliminary results of this all-solid NOVA study were presented to the joint NASA/AF Large Launch Vehicle Program Group (generally known as the Golvin Committee) on 3 August 1961. This committee, recognizing the unconventional features of the concept, contracted with Boeing and Space Technology Labs, Inc. to make independent evaluations of the solid NOVA studies reported in JPL's Technical Memorandum 33-52 and Addendum A thereto. It was planned that at the end of the one-month evaluations they would submit critiques on 1) the technical feasibility of the concept, 2) the realism of the JPL estimate of schedule, 3) the accuracy of the cost estimate. Within a week the work statements had been expanded to include 1) an evaluation of the concept if payload to escape were increased from originally studied 130,000 lbs to 156,000 lbs, 2) a supplementary evaluation of the adaptability of a solid NOVA or its components as a backup for a Saturn-class vehicle, and 3) quantititave comparisons of the several reliabilities of liquid and all solid NOVA injection vehicles. On a submission of the critiques by Boeing and Space Technology Labs, the committee decided against company briefings or discussions because of other pressing matters. Approximately one month later, Mr. E Mitchell, then assistant director for Propulsion at NASA HQ, exspressed interest in results of the studies; thus on 6 October 1961 Boeing, STL, and JPL representatives provided briefings to NASA Propulsion personnel on the original solid NOVA concept and the results of two industrial critiques. Since that time, no studies have been performed. In recent months, NASA has decided, after careful deliberation, to perform the manned lunar landing mission using a lunar orbit rendezvous mode basedon a liquid propellant Saturn C-5 vehicle for the manned operations and a lunar logistics vehicle for cargo and manned support. However, the need for vehicles with a payload capability of about 500,000 lb in LEO, in the NOVA class, for the more difficult missions beyond the manned lunar landing, has also been recognized by NASA, and the new studies of all liquid and hybrid staged NOVA vehicles have been initiated by several groups. Because the lab's studies of all-solid propellant vehicle systems indicate that it is unusually promising and of general interest, this report, which gathers together all significant results of the studies, is being released as an appropriate addition to current knowledge. Only that phase dealing with the way in which the solid NOVA would fit into the manned lunar landing mission and NASA's long range plans has been omitted because it is no longer applicable. An appendix that discusses background information and the status of applicable solid rocket work has been incorporated as a convenient reference; in addition, several developments from JPL programs are discussed because they have strong bearing on the solid NOVA program under study. The opinions expressed in this report do not necessarily reflect the views of NASA.

ABSTRACT

Studies show that very large, all solid propellant injection vehicle systems in the NOVA class are technically feasible. The vehicle studied is a four-stage, solid propellant rocket having a gross weight in the 30 million pound class, a payload in LEO of 500,000 lbs, and a payload through escape of 130,000 lbs. The first three steps would inject the fourth stage into parking orbit from which the fourth stage would inject the spacecraft into the transfer orbit to the Moon or planets. Designs examined were conceptual and do not represent optimized configurations. Additional studies must be made in depth before the final system and industrial complex requirements can be specified. These new studies must include an examination of the applicability of noncryogenic liquid propellant systems for third and or fourth stages because of the potentially greater vehicle flexibility for a range of missions. When compared to liquid vehicles of equivalent performance capability, it is concluded that the risk associated with development of this vehicle system would be much lower and that the reliability of the resultant system would be predictably higher at a very early point in the flight program- provided that the proposed conservative philosophy were used. As a result of the studies, confidence in the philosophy and program approach advocated has been reinforced. From a technical standpoint, it appears that the vehicle injection system can be made available for the first flight approximately four years after the start of go-ahead, and for operational manned space missions one year after that date. Total costs are estimated to be significantly less than for competitive systems, provided that the philosophy and development approach advocated are implemented. Growth potential for the basic systems seems particularly favorable for the progressively more difficult missions which can be foreseen. There do not appear to be any major technical problem areas. However, early emphasis should be given to thrust vector control (TVC) and meeting guidance and control (GNC) as well as other subsystem reliability requirements. Combustion instability is not expected; and it is believed that even if it is encountered it would not become a serious problem.

INTRODUCTION

Prior to the time that a definitive plan could be formulated by NASA for the manned lunar landing, a number of groups undertook studies of various vehicles and modes of performing the mission. Because of its extensive experience in liquid and solid propellant technology and vehicle system development (Corporal, Sargeant, and the Explorer and Juno upper stages), the JPL voluntarily initiated some studies of large launch vehicles.

Independent of these studies, JPL activities in the Ranger, Surveyor, and Mariner projects had revealed the complexity and difficulty of even the simplest unmanned lunar or planetary missions. Thus, on reviewing the manned mission, it became evident that the total mission reliability was only a part, would be wholly inadequate unless the subsystem reliabilities were substantially better than they had been for most missile and space programs. There appeared to be at least four basic reasons for this. First, man, as a nonexpendable payload, will demand unusually high overall mission reliability-- perhaps 85%-90% probability that the mission will succeed, and 97%-99% probability that, despite a mission abort, the astronaut will be recovered. Second, the larger vehicles required wil be derived by clustering as many as 8-10 propulsion systems per stage and or using more stages. Consequently, if the design philosophy used in the past is preserved, vehicle reliability will drop below the barely acceptable values of current vehicles- particularly if they must use newly developed propulsion systems. Third, because of high vehicle cost the launching rate and production rate will be significantly less than in the past- probably 60-120 vehicles over a ten-year period. Therefore, there will be very few flights with which to develop system reliability; some means must be found to provide high reliability at a much earlier point in the flight development program. Fourth, and most important- space mission will become increasingly demanding in the required number of sequential operations that must be performed before the mission can be completed successfully. For example, a manned lunar or planetary landing mission would probably consist of the following major operations (involving perhaps a total of seventy to eighty critical steps performed in sequence): 1. First stage launch, 2. second stage operations into parking orbit, 3. third stage injection through escape, 4. midcourse propulsion and guidance corrections, 5. terminal guidance and retro-propulsion into lunar or planetary orbit, 6. descent toward the Moon or planet, 7. hovering to a soft landing, 8. lunar surface operations, 9. takeoff for return, 10. midcourse propulsion and guidance corrections, 11. Earth landing through a restricted corridor and recovery. Quite possibly there would also be, at appropriate times: 12. vehicle rendezvous (in Earth orbit with multiple launchings of components and/or rendezvous at the destination), and 13. orbital docking of the vehicle components. It must be recalled that vehicle launches into Earth orbit (only the first two of the 13 listed operations) have resulted in approximately 25-50% failures. Many of the later operations in the list of 13 must be performed in new and hostile (perhaps unknown) environment, there will be little opportunity to practice landings and takeoffs at the destination, and success of a number of the operations (ie rendezvous) is strongly time-dependent. The magnitude of the challenge may be illustrated by noting that mission reliability is an inverse exponential function of the sequence of operations.

SOLID PROPELLANT NOVA INJECTION VEHICLE SYSTEM

The injection vehicle system and its industrial support complex include the launch vehicle, its means of production, the associated transportation complex, the offshore launch pad and ground support equipment, and a range support area (Fig 1.) The spacecraft is considered only as it affects the vehicle system. Additional material in support of the study is contained in App. B.

A. INJECTION VEHICLE

The vehicle in its preliminary form was estimated to have a payload capability of 130,000 lbs to escape, or 500,000 lbs in LEO, and a launch weight of 25 million pounds. Subsequent and more sophisticated calculations show that the vehicle launch weight is closer to 30 million pounds if the vehicle has to provide the indicated payload capability. It is believed that this revision will affect size and cost only and in no way bears on the feasibility of the concept. All discussions in Sec. 3 are based on the 25 million pound vehicle.
1. DESCRIPTION AND OPERATION
The injection vehicle (Fig. 2) consists of four stages of clustered solid propellant motors of two designs. Seven type A motors for the first stage, and three identical motors make up the second stage. Six smaller, type B motors, are clustered together to form the third stage, and a single type B motor forms the fourth stage. This vehicle has a gross weight of about 25 million pounds, in Fig 3. it is shown in comparison with a 6 million pound liquid vehicle having the same payload capability. The Washington Monument model gives an indication of scale. The first three stages are designed to achieve parking orbit. After the required coasting period, the fourth stage is ignited and the 130,000 lb. payload is brought to escape velocity at burnout. A thrust vector control system is used on each stage to maintain directional control. However, no thrust termination devices are used on the large primary motors. Each stage burns to completion, and small vernier or secondary motors trim the velocity errors as the vehicle goes into its parking orbit and possibly again at stage four burnout. The verniers required are small rockets containing about 5,000 pounds to 15,000 pounds of either storable liquid or solid propellant. Their thrust termination is within the state of the art. The solid rocket motor casings, tied together in shear, serve as the primary airframe structure. Columns and truss members are used for interstage structure, which contains provision for positive separation. The diameter of the first stage assembly is 77 feet, and the height of the injection vehicle to the separation plane between the fourth stage and payload is 220 feet.
2. PERFORMANCE
Performance parameters are presented in Table 1. Three dimensional point mass trajectories were computed assuming eastward launch from AMR (Atlantic Missile Range). The first two stages were flown gravity turn after a short vertical ascent. The third stage was flown using a constant attitude in intertial space and, alternately, with constant inertial pitch. The two modes of computation gave essentially equivalent results. It should be noted that the maximum acceleration, occurring near the end of the third stage burn, is 5.3g, an acceleration tolerable by man. The mass will be distributed between the third and fourth stages such that the velocity at the end of the third stage burn is slightly short of achieving circular orbital velocity. Part of the propellant weight in the vernier, assigned to the fourth stage, is used to make up this velocity difference and insert the stage and payload into a circular LEO. The sizing was accomplished by assuming values for the specific impulse and for the stage propellant mass fraction. The primary considerations used in the sizing and trajectory design were performance capability, airloads, achievement of a parking orbit at the end of the third stage burn, and metting manned acceleration requirements.

3. PROPULSION
The motors represent a typical design concept to demonstrate the feasibility of making very large units. Detailed characteristics are shown in Table 2. Motor A is used for the first and second stages; Motor B is used for the third and fourth stages. A specific impulse (ISP) of 245 seconds at 1000 psi and sea level optimum expansion was assumed. The performance parameters assumed here are all within the present state of the art. The reproduceability of performance should exceed that of present, relatively small motors because continuous mixing techniques are contemplated and large amounts of propellant will be cast in each motor. It is believed combustion instability is unlikely to be encountered; if it does occur, modern techniques should preclude any serious delay in schedule. In recent years, powdered aluminum has been found to be an excellent suppressant for combustion instability in existing solid propellant boosters which use ammonium perchlorate solid propellants. As aluminum quantities are increased up to 15-17%, propellant performance as well as aluminum's effectiveness as a suppressant increases. None of the relatively large motors that utilize these high aluminum composite propellants (Minuteman, Pershing, Polaris) has shown any sign of combustion instability; the propellants under consideration contain these high aluminum concentrations. Unitized motors were used in this study in order to investigate ground facilities required to support this type of a design. It is not necessarily recommended that they be chosen over segmented motors. However, since new propellant processing facilities would be constructed for this vehicle, they can be designed to accommodate either approach. The final decision between unitized or segmented motors should result from a study of the effects of the choice on the flight vehicle. Experience has shown that it would be undesirable to compromise the flight vehicle because of ground support equipment requirements unless a question of feasibility is involved. Propellant would be processed in a new, continuous mix plant at a site strategically chosen for the raw material supply, its proximity to the launch site, and the practicability of shipping finished rocket motors. Propellant facilities similar to those under consideration are already in production and have demonstrated the quality of product and high production rates necessary for the program. For example, Aerojet-General Corporation recently cast a 100 inch diameter charge, weighing about 200,000 lbs, at the indicated rate, then cured and fired it successfully. It is of interest to note that the static test record, based on high frequency response instrumentation, gave no indication of combustion instability.

4. THRUST VECTOR CONTROL (TVC)
A TVC system is needed in each stage to compensate for the effects of CG displacement, thrust misalignment, and unequal thrust (Particularly at ignition and burnout). It must also counteract aerodynamic forces during first stage burn. Canting of the rocket motor nozzles so the thrust vector is directed near the burnout CG of the stage helps reduce some of these disturbances but cannot be used when the angle of cant is too large. TVC systems that could be used include jet vanes, secondary injection, moveable nozzles, auxiliary rocket motors (verniers), and jet tabs. Each has distinct advantages and disadvantages, and a choice can result only from a detailed system study considering guidance structures and spacecraft. Jet vane systems have demonstrated good flight records and offer a feasible solution to this problem. However, a gas pressurized system may prove to be inherently more reliable and was chosen for more detailed study in order to provide a basis for weight and cost estimates.

5. STRUCTURES
The largest structural weight item in the vehicle is the rocket motor casings. Fortunately, considerable experience in the design of cylindrical pressure vessels is available, and the large size considered here should introduce no problems which differ basically from those already encountered elsewhere. Consistent with the philosophy already presented, a heat-treatable martensitic steel with a yield strength of 165,000 psi was chosen in this study. Toughness and ductility shoudl be high at this strength level. However, this conclusion is tentative until more information is available for the 3/4 to 7/8 inch thick material considered for the type A motor case. The interstage structure, as presently conceived, is either a space frame or a set of three spaced columns. Loads would be transmitted to the motor cases as concentrated loads acting on truss pads attached to the motor domes or as concentrated line shears. This technique was used on the Sargeant motor case and is extensively used in the large water tanks and pressure vessels in the power generating and chemical industries. Two types of rocket nozzles were considered: 1) a graphite/steel heat-sink type, and 2) an ablative plastic type. The use of ablative nozzles appears very attractive for this application because the linear ablation rate will be the same or slightly less than that of smaller motors; therefore the percentage change in throat area will be acceptably small. Either type appears feasible; scaling laws predict less severe conditions than for existing nozzles despite the relatively long burn times, provided that weights are scaled with impulse.

6. STRUCTURAL DYNAMICS
Major structural dynamics effects such as overall dynamic loads and aeroservoelastic stability were examined qualitatively to ascertain whether extrapolation in size or weight would adversely affect the feasibility of the solid propellant NOVA. By using aeroelastic model theory, and by assuming that dynamic magnification factors associated with transit through discrete gusts are independent of vehicle size, it can be shown that a large vehicle should encounter no more severe loading in relation to its strength than a small vehicle is dynamically similar. Dynamic instability of the type characterized by adverse coupling between the autopilot and the elastic airframe can become a serious problem if a conventional flight control sensor installation is employed. However, an adaptive flight control system using "transducer arrays" (a number of rate gyros discretely positioned over the length of the vehicle without outputs electronically summed, can be employed to suppress the coupling of the autopilot with the body bending modes. Thus, low bending mode frequencies, per se, need not lead to dynamic instability problems. The solid propellant vehicle system carrying liquid payload rockets or liquid secondary injection system poses relatively minor liquid sloshing problems because of the low percentage of liquid mass.

7. AERODYNAMICS
The need to aerodynamically shroud the vehicle has been examined from the aspects of heating, drag, and unsteady flow characteristics. Maximum laminar and turbulent heating rates were examined for the velocity-altitude information obtained from a powered flight computer trajectory. A conservative analysis indicates that the heating will be no problem on exposed structural members or motor casings. An estimate of aerodynamic drag for a body of this type is not readily calculated. The examination of some experimental data and calculations based on two simple drag models indicate that the peak drag to thrust ratio is approximately 0.15, an acceptable value. Unsteady flow effects between motor casings may result in high local vibration loads. Local fairing should alleviate this condition. In none of the cases studied could a positive requirement for shrouding be determined. Consequently, only a shroud from the payload to the top of the fourth stage has been indicated. The weight of this shroud was charged to the second stage, since it would be discarded at the end of the second stage burn. The maximum dynamic pressure (max q) expected is approx. 1400 psf, and at first stage separation the dynamic pressure is approx. 400 psf. These pressures are acceptable for the vehicle under consideration. Trajectory shaping can be used to reduce the max pressure to a value of less than 1000 psf if desirable.

8. ASSEMBLY AND ALIGNMENT
With proper attention paid to details, assembly and alignment should provide no difficulty. Provision will have to be made for a temporary framework to suppor the motors of a stage during construction until the stage is structurally tied together. Erection loads on the individual motor cases will have to be considered in their design. Vehicle loads caused by wind and wave action during storm conditions while on the launch pad are expected to be small for the blunt, dense, solid propellant vehicle, and temporary guys or bracing are adequate protection. Techniques are available to provide CG control without measuring absolute weight. If further study indicates that accurate absolute weight measurement is required, an increase in the existint capability for accurately calibrating load cells is required. Load cells of the necessary size are available.

9. GUIDANCE REQUIREMENTS
The guidance of the vehicle has not been examined in detail, but it is felt that no unusual problems will be present. The general philosophy of this program can be applied to the guidance area, and adequate space with a controlled environment and ample weight for the required equipment have been provided. The weight should allow for underrating components and providing redundancy wherever needed. For example, 3,000 lbs and 350 cu. ft. of space were allowed for the guidance and control electronic compartment alone. Since it is considered that midcourse correction capacity will be required in the spacecraft, the precision and performance required of the injection vehicle are rather modest. Indeed, guidance capability equivalent to that required for military weapons is adequate; no advancement in the state of the art appears necessary.

SPECIALIZED FACILITIES
1. PROPELLANT PROCESSING FACILITY
Because of the quantity of propellant needed and the size of the loaded motors, water transport at the propellant processing facility is required. The site should be 1) convenient to Cape Canaveral (the assumed launch site) by barge transport, preferably through an inland waterway to avoid long exposure to the open sea, and 2) provide in-plant barge transport during processing. There are many islands and coastal areas from Texas to South Carolina that could meet the above requirements. A typical site, Skidaway Island, in Georgia, has been chosen only to demonstrate the feasibility of such and approach (see Fig 5 and 6). It is assumed that the island is undeveloped and that all waterways used during processing will be dredged completely. A description of the facilities is given in Table 3. The propellant materials considered are typical components of propellants, now used in large motor programs. New production capability is required for all of the ammonium perchlorate used in this program (8 million pounds per month at a vehicle launch rate of one every two months). Power requirements for this production are not excessive. The other materials required are readily available or could be made available on relatively short notice. Propellants of the type under consideration have never been known to detonate at the operating temperatures to be used. Nevertheless, since no experience at this size is available, all facilities have been sized and sited on the basis of class 9 or 10 propellant.

2. LAUNCH SITE OPERATIONS
The launch site, GSE, and assembly and transportation techniques are governed by the following criteria: 1) the complex must be in operation within 2.5 years of the commencement of the program if it is to be used for motor static firing tests; 2) the complex must be an economical and practical system for accomplishing the assembly and launch operation; and 3) the possibility of loss, because of a launch failure, of costly long lead time GSE must be minimized. The launch site is assumed to be located near existing range facilities and sources of manpower at Cape Canaveral. It is evident that because of the acoustic and explosive safety distances involved, the solid propellant could not be launched directly from the Cape. The choice of an offshore launching pad is indicated when the additional launch complex requirements are considered. In determining support systems for the assembly and launching of a vehicle of this sie, it is useful to observe the experience which exists in other areas of endeavor. Normal operations in the large civil engineering industry and in marine and naval architecture closely parallel the erection and handling techniques required for this vehicle. It is from this existing technology that the optimum solution should be derived. Fixed underwater supporting structures for the depths required are encountered in bridge foundations and dam construction. Shipment of the weights required occurs daily in normal tug and barge operations. The erection and assembly of large pieces of of equipment is within ship building and repair technology. (Indeed, the largest manmade moving objects are ships of one form or another.) Thus the optimum launch complex to meet the stated objectives is a fixed offshore launch pad supported by vessels and barges and utilizing the Cape Canaveral range and support facilities.

3. TRANSPORT AND STORAGE
Loaded solid propellant rocket motors are stored in their respective loading barges at the propellant processing site. Motors are transferred to the transport barges by the floating crane at the processing site. They are then tugged to the launch site at a rate determined by the launching schedule. Guidance, telemetry, and associated equipment is transported to the launch support area at the Cape by conventional means. After checkout and maintenance, the equipment is barged to the launch platform for installation.

4. SITE CONSTRUCTION
Since vehicle assembly and checkout procedures require four months (Fig 7), two launch pads are required to allow launchings on two-month centers. These pads (Fig 8) would be placed approximately 6-12 miles off the coast of Cape Canaveral, with a similar distance separating them for acoustic and quantity-distance safety requirements. The launch pad would then be in 60 to 100 feet of water. This depth is typical of that found in bridge foundations, and this construction is the type desired. For a solid porpellant vehicle, the dimensions of the pad need only be large enough to provide structural support for the vehicle. The large weight of the vehicle as erected would provide considerable stability against overturning due to wind and storm conditions. Hurricane protection would be limited at most to temporary guying and bracing. All auxiliary operations would emanate from the crane or support vessels. A hold-down structure would be neither desirable nor practical for use during the launching of a solid propellant vehicle. A breakwater one mile in overall length is required to shelter each pad area sufficiently to allow shipborne operations in all but gale conditions. This breakwater would be of rock construction and 10 feet above high tide. General support, electronic checkout, engineering personnel, and auxiliary equipment receiving and storage buildings would be located on or near the Cape. Several docks and a crane (YD-4) would be required to transfer this equipment to barges for transport to the launch pads. The distance offshore between the pad and the Cape has been specified as that allowing for inhabited areas, according to the mass of the propellant involved. Therefore, the launch control center would not be a blockhouse in the normal sense but, structurally, an ordinary building located on the Cape in the general support area. This center would contain the launch control instrumentation and equipment and be connected to the pads by underwater cable for phone and electrical connections. Most of the prelaunch instrumentation and monitoring measurements would be made, however, by direct radio link between the vehicle and LCC. An umbilical mast would provide electrical connections to the spacecraft up until launch, as well as emergency disarming or astronaut exit ladders. This mast would pivot at its base and drop to the water at launch.

5. VEHICLE ASSEMBLY EQUIPMENT
The primary assembly problem is the erection and handling of the first and second stage motors. A crane of 1000 ton capacity and 200 ft hook height is required, with a secondary hook of 200 tons and 300 ft height for the upper stages and payload. The construction and operation of this crane and its support structure are considerably simplified when based on waterborne operations. Currently, the largest movable crane in the US is the Navy YD-171, mounted on a self-propelled moving barge. This is one of four constructed in 1941 and has a 450 ton capacity and 160 foot hook height. Several fixed cranes of this capacity, but with lower hooks, also exist. A scaled-up version of the flat bottomed barge and crane may be specially constructed for this purpose and built to contain personnel and service areas. Its construction is considered quite feasible. The use of a floating crane for the handling and careful positioning of very large loads is well established procedure. It provides the most economical and shortest lead time solution to the erection problem and allows for convenient removal from the pad area prior to launch. Personnel service platforms around the vehicle are made in prefab sections and attached directly to the vehicle structure. The loads introduced are small compared to the load carrying capability of a solid propellant vehicle skin. Personnel access is via an elevator tower on the craneship and gangwalks across the vehicle. The platforms are removed prior to flight.

6. STATIC TESTING
The full scale test firings of the individual full size motors could be conveniently accomplished at one of the launch pads. A motor, supported by suitable structure, could be mounted in an inverted position on the pad. Normal launch control instrumentation would be used for this series of tests. The alternate pad would be used for the first flight vehicle. Upon completion of the static test program, the external support structure would be removed and this pad converted to a flight pad for the second flight firing. Alternatively, a static test stand could be constructed at the propellant processing plant and utilized for this purpose. The added area and cost for such a test stand are included in the cost analysis.

7. FIELD OPERATIONS
Individual motors of types A and B would arrive at the launch complex by barge from the propellant processing plant storage area. They would be erected and assembled on the launch pad by the crane barge (YD). Interstate structure, in large prefab sections, would be barged out and installed. The crane-barge would then be transferred to the alternate pad and replaced by the auxiliary support ship (CLS). During the remaining period, the GNC, telemetry, and spacecraft equipment together with the auxiliary gear would be installed and the final checkout operations accomplished prior to launch. The firing of this vehicle from an offshore location at the Cape presents no extraordinary problems with respec tto range procedure. Normal downrange tracking, range safety, and launch monitoring are identical to regular Cape launchings. Firing windows are reasonable and the actual final countdown time for the multistage solid propellant vehicle is relatively short compared to that for existing vehicles. The shorter burning times also result in significantly simpler tracking operations during launch into orbit than for equivalent liquid vehicles.

E. SPACECRAFT CONSIDERATIONS
Spacecraft studies have been limited to crude feasibility determinations of 1) possible configuration constraints, and 2) spacecraft design problems peculiar to the employment of the solid propellant vehicle for a manned mission. The results of the studies can be summarized as follows: 1) there are no spacecraft design problems peculiar to the solid propellant vehicle; 2) the injection capabilities of the solid propellant vehicle appear to be adequate for manned missions if so desired; 3) there are no major spacecraft configuration constraints or limitations due to the injection vehicle. Of the spacecraft environments associated with a large solid vehicle, (vibration, linear acceleration, acoustical, etc.) the one which appears to be the most severe when compared with a liquid vehicle of the same capability is the acoustical environment. Differences in the other environments are minimal. The analysis of the acoustical environment was based on extrapolations from data from smaller motors, However, the results are considered to be conservative. The calculated sound pressure levels are (solid)-=167 db, and (liquid)-=161 db. These levels correspond to a distance of 200 feet from a solid propellant vehicle of 40 million pounds of thrust or a liquid vehicle of 9 million pounds of thrust. Actually, the lower exhaust velocity of the solid propellant motor causes it to have lower acoustic efficiency, so that the pressure level from a liquid propellant vehicle might well be higher than that from a solid propellant vehicle. In any case, it is important to note the high pressure level from either. Factors such as sound absorption in the air (which increases at these higher intensities because of nonlinear damping) and directivity of the sound should decrease the levels by 20 db. Reflection from the pad coudl be minimized by flowing water under the booster at liftoff. These factors suggest taking the pressure level at 150 db. The effectiveness of ear protectors for the astronaut is limited by bone conduction to about 40 db reduction. Thus the pressure level at the ear will be reduced to 110 db, which is below the threshold of pain at 140 db. The maximum total recommended level for speech comprehension is 110 db, so that talking with astronauts during liftoff may be difficult, although additional attenuation by the cabin walls may make it reasonable. It should be emphasized that the attenuation of the sound with distance is equally important.

3. COST CONCLUSIONS
The cost study of the solid propellant NOVA injection vehicle system has revealed some rather unusual, albeit tentative, conclusions. The conclusions reached are a logical consequence of the basic program philosophies.
a. The development costs for this vehicle are considerably less than for a liquid propellant vehicle of the same capability
b. Production costs per vehicle are roughly comparable to a similar liquid vehicle, depending somewhat on the injection mission and total number of flights.
c. Injection costs in dollars per pound of payload for a 300 mile LEO for a 20 vehicle program are $169 per pound total cost and $91 per pound production cost. For these computations a reliability of 95% and 515,000 lbs of injected weight were used.
d. The development costs are low because the usual full scale development support programs are not required-- no full scale vehicle structural testing, no captive firings of clusters, no battleship propulsion program.
e. Production costs per pound for the structure are low because of the inherent simplicity of the solid propellant rocket. Size was simply exchanged for complexity at a nearly constant total production cost.

H. GROWTH POTENTIAL
Once feasibility of a vehicle is established, it becomes a major interest to examine the system flexibility for quick growth to a vehicle capable of performing much more difficult missions in the foreseeable future. A cargo version of the large spacecraft under consideration can place approx. 35,000 lbs gross weight on the lunar surface. Such devices as Moon-mobiles, prefab structures, and life support systems could be delivered directly from Earth, intact and ready to operate with no assembly or disassembly. The 30 million pound solid propellant NOVA would appear to have considerable growth potential beyond this. Substitution of liquid hydrogen stages based on engines now under development, for the third and fourth stages, such that the gross weight is unchanged, results in a vehicle that could place approx 110,000 lbs of men and equipment on the Moon. The first two, very large solid propellant stages woudl be used as developed for the original vehicle. The first three stages of this solid propellant or hybrid vehicle should be capable of placing approx. 930,000 lbs in LEO. If one were to use this weight as an electric powered spacecraft, the payload on the lunar surface using LOX-LH2 for landing on the Moon, would rise from 110,000 lbs to about 215,000 to 440,000 lbs. depending on the allowable transit time. Alternately, this three stage hybrid vehicle with its electric spacecraft should be capable of performing a three-man Mars landing and return in approx 590 days; this performance estimate is based on the radiation shielding of 50,000 lb, 15 lb/man/day sustenance, and a Mars Orbit Rendezvous of the spacecraft with the Mars Excursion Vehicle.

IV. NOVA DESIGN VERIFICATION AND COMPARISONS

The Large Launch Vehicle Program Group (the Golovin Committee), noting the unconventional aspects of the all-solid NOVA approach, requests the Boeing Company and Space Technology Laboratories to make independent and objective evaluations of the solid NOVA concept for 1) technical feasibility of the vehicle concept, 2) accuracy of the cost estimates, 3) realism of the schedule, 4) effect on the vehicle concept if the payload were increased from 130,000 lbs to a more recent estimate of 156,000 lbs, 5) flexibility of the vehicle as a backup for Saturn, and 6) a comparison of the relative reliability of liquid and solid NOVA vehicles. A review of the pertinent conclusions from the critiques (Ref 1 and 2 respectively) should be valuable at this point in confirming or rejecting the concept.

A. CRITIQUES OF THE SOLID NOVA CONCEPT BY STL AND BOEING

The salient features from the JPL, Boeing, and STL studies are compared in Table 5. In this specific comparison a vehicle with a payload capability to escape of 130,000 lbs is used because the original studies were carried out and evaluated on that basis. The comparison with liquid vehicles in section IVB, will use vehicles having an escape payload capability of 156,000 to 160,000 lbs. The critiques differ in details from the JPL designs (this was not unexpected, for they were conceptual design studies) but on the critical questions were in relatively good agreement.

1. TECHNICAL FEASIBILITY

Although vehicle weight estimates varied somewhat among the three studies, all concluded that weight per se was of minor importance and that there were no technical barriers, such as nozzle erosion, propellant physical properties, burn time, combustion instability, guidance, or vehicle dynamics that would prevent the accomplishment of the mission when using an all solid NOVA vehicle. The conclusion was found to be equally valid for a solid Nova with a 156,000 lb payload.

2. GROWTH POTENTIAL

Both Boeing and STL were quick to recognize the unusually large growth potential of the solid Nova. Each commented favorably on it as a distinct advantage.

3. FLEXIBILITY AS A SATURN BACKUP

Boeing also noted that the upper three stages of the Nova, with only a 5% penalty in the original Nova weight, would be capable of delivering 266,000 lbs into LEO, adequate for a Saturn C-5 rendezvous type boost. STL and Boeing also found that other vehicles with escape payloads of 21,000, 30,000, 40,000, and 45,000 lbs were feasible when using the A and B type motors as modules in the first three stages and an Earth storable liquid system in the fourth stage, only performance capability was checked in these cases. Boeing and STL appeared to differ regarding costs and schedules, their evaluations will be discussed separately.
A. Boeing costs and schedules: Boeing agreed closely with the cost estimates of the original JPL report for the 25 million pound vehicle discussed, ie $1.7 vs. 1.65 billion, reporting costs as realistic. In a final analysis, Boeing believed a heavier vehicle was advisable, its program costs would be about $2.2 billion dollars. A schedule of five years was considered realistic by Boeing. Some concern and possibly misunderstanding of the Golovin Committee may have arisen over apparent discrepancies between the Boeing studies on the solid NOVA and separate Boeing solid booster studies being carried out at the time for MSFC. If so, it is unlikely that the discrepancies were ever explained to the committee, STL and Boeing were not asked for briefings on their critiques when explanations might have been made. Later on 6 October, when JPL, Boeing, and STL personnel presented their results of their NOVA studies to the propulsion personnel at NASA HQ, Boeing did explain, when questions arose, that 1) the two studies were not made by different Boeing study groups, and 2) the schedule differences reported in the two studies were not inconsistent. In the Marshall case they were making a parametric study, whereas the JPL cas they had been asked to examine a specific vehicle system.
B. STL COSTS AND SCHEDULE: The STL report concluded that hte cost of vehicle system development, facilities, and 20 vehicle flights would be $4.2 billion, or approximately twice that estimated by Boeing and JPL. Their schedule too, was longer, requiring 6.5-7 years compared to Boeing and JPL estimating 5 years. During the 6 October briefing at NASA HQ, personnel from STL explained their higher costs and longer schedule on the basis of their experience with the Air Force in the ICBM program, such as Minuteman. It should be noted that the JPL method of implementing a program differs significantly from the Minuteman approach. In the briefing, however, STL did conclude that their cost estimates would be much lower than reported and their schedule would agree with the schedules of Boeing and JPL if the solid NOVA philosophy and approach could be implemented. Boeing believed the philosophy and approach could and should be implemented. Because of the different frames of reference or assumptions made in the Boeing and STL studies, one can draw a very important conclusion: Approximately $1 to 2 billion can be saved in a 20 vehicle program and 1.5 to 2 years can be cut off the 6.5 to 7 year schedule if the solid NOVA philosophy and program approach can indeed be implemented.

B. COMPARISON OF SOLID AND LIQUID NOVA CLASS VEHICLES

The comparisons will be confined to 1) vehicle and propulsion reliabilities, and 2) vehicle schedules, and 3) program costs. Most of the data gathered during the latter part of 1961, when the liquid vehicle configuration given the most serious consideration for the NOVA mission, now designated the C-8, consisting of 8 F-1 engines in the first stage, 8 J-2 LO2/LH2 engines in the second stage, and 2 J-2 engines in the third stage. Its payload capability for the escape mission assuming no engine out was approximately 160,000 lbs. Therefore, the characteristics of the solid NOVA with the 156,000 lbs escape payload, derived from the Golovin Committee supplementary work statement, will be used for comparing schedules and costs in section B2 and B3.

1. FLIGHT RELIABILITY OF LIQUID AND SOLID SYSTEMS

Although vehicle reliability as a function of liquid or solid NOVA time schedules will be of major interest in comparing the two, a review of absolute reliabilities for some recent liquid and solid vehicles is also valuable in itself. Unlike costs and schedules which represent best estimates of future values, reliabilities represent concrete evidence of what has actually been demonstrated in the past, they give some indication of the likelihood of mission success. Fig 10 shows composite plots of overall vehicle reliability for 1) four solid propellant vehicles and 2) six liquid propellant vehicles as a function of the number of vehicle launches. The former group included Polaris, Minuteman, Pershing, and BOMARC B booster; the latter group included Atlas, Titan, Jupiter, Thor, Redstone, and BOMARC A booster. Individual reliability records of these vehicles, taken from Ref. 1, are shown in appendix C. The composite plots of Fig. 10 reveal that solid propellant vehicles have demonstrated significantly higher reliability than liquid propellant vehicles at a given early date. Early reliability is especially significant because the number of total launchings of a NOVA sized vehicle, even for a 10 year period, will probably only be about 60-120; there will only be a limited number which can improve flight reliability. The outstanding exception is Polaris. Development of Polaris probably constituted the gravest challenge to engineering rocket technology the country has seen. The relatively low reliability in early flights, as well as in the static test record (see section IVB2), reflects the very significant advancements in the state of the art that were required in many technical areas: 1) design to severe volume constraints for vehicle stowage, 2) chambers based on notch-sensitive materials of unusually high strength, 3) high performance propellants with much higher flame temperatures and more erosive exhaust gases, 4) multiple nozzles which incorporated unusual materials and design features, and 5) new technique for TVC, jetevators. Failures in the early stages of the static test and flight program were to be expected.

2. SCHEDULES

In the further comparison of the liquid C-8 and solid NOVA programs, the schedule, or time required to launch astronauts, for the solid NOVA has been estimated at 5 years by JPL and 5.25 years by Boeing (Ref 1 part II) The liquid C-8 schedule has been estimated at 6 years by the Fleming Committee, and in separate studies, at 5.75 years by Rocketdyne. It is important to note that Dr. Dergarabedian, director of STL's Systems Research Laboratory, has permitted a quote of STL's comparative evaluation based on their solid Nova critique and their liquid class Nova vehicles for MSFC. "the solid Nova will be available at least a year sooner than the liquid system" STL's estimates of 6.5-7 years for the solid Nova, therefore, means at least 7.5-8 years for the liquid C-8 and must not be compared with the 5.75-6 year estimates of liquid Novas by other agencies. The ground rules were different. "STL believes virtually all estimates by other agencies are idealized (based on success) rather than realistic schedules and are therefore too short."

3. PROGRAM COSTS
Dr. Dergarabedian, Director of STL's System Research Laboratory, again has permitted a quote of their comparative evaluation, "they estimate the liquid C-8 will cost approximately twice that of the solid NOVA or about $8.7 billion." STL cost estimates should not be compared to other cost estimates shown because the assumptions for the studies differ as mentioned in the section on schedules. RAND estimated the costs of the liquid C-8 at $4.7 billion, if the launch rate is one vehicle every two months, while Rocketdyne has estimated $4.5 billion (Ref. 3). RAND reports that MSFC costs for the liquid C-8 are in excellent agreement with their estimates. RAND has also estimated the cost of a liquid C-8 program that included 100 vehicles (Table 10, to take into account the decreasing recurring cost of the liquid vehicle, at $10.5 billion. The solid Nova program for 100 vehicles has been estimated by JPL, for a constant vehicle cost, at $8.3 billion; in practice, production costs of the solid Nova vehicle would, of course, decrease. It is of interest to note the cumulative vehicle program costs to perform a specified practical mission when 1) liquid C-8 vehicles and 2) solid NOVA vehicles are used. Fig. 13 indicates the cumulative injection vehicle system costs to supply a lunar base with 150,000 lbs of useful payload per year. The Nova, C-8, Saturn C-5, and Saturn C-3 were estimated to have, respectively, 30,000, 15,000, and 7,500 lbs of useful payload (payload weight after weight of the weight of the landed spacecraft vehicle itself). The values at zero time represent the costs for R&D, facilities, and 10 vehicle flights. Based on JPL estimates, the total solid-Nova vehicle program cost, after 100 flights and 18.5 years, would $8.3 billion. The solid Nova data in this figure, unlike the liquid system, do not include decreasing production costs for the vehicles and probably penalize the solid vehicle. Both solid and liquid vehicle estimates assume reliabilities of 100%, and therefore, penalize the solid Nova again in a relative sense. If STL cost estimates had been used, both curves would have had higher values but would undoubtedly resemble those shown, unfortunately, no cost breakdown for the liquid vehicle was available from STL for preparing such curves. The Saturn C-3 and C-5 costs, estimated by RAND, have been included as a matter of interest. Although they start with lower development costs than the liquid C-8, they rise rapidly to values appreciably higher. After 10 years their program costs would be $9.0 billion and $8.6 billion, respectively. Figure 14 indicates program costs to supply the lunar base using 1) solid Novas and 2) liquid C-8's, when vehicle reliabilities estimated by Boeing (see section B2) are used and when the recurring costs of the vehicle are decreased with time again. Again, RAND's costs are used for the liquid C-8, with the initial vehicle launch cost dropping from $135 million per vehicle to $70 million for the hundredth launch; JPL cost estimates for the solid Nova decrease from $75 million initially to $65 million for the hundredth launch. Thus, if the liquid C-8 were used to supply the lunar base for 16 years, the overall program would cost about $11.7 billion; if the solid Nova were used, the overall program would cost about $8.25 billion, or 70% of the former. It may be concluded that the solid Nova of conservative design would have a significantly lower development cost and lower recurring costs to perform the specified mission than would the liquid C-8, the Saturn C-5, or the Saturn C-3.

4. GROWTH POTENTIAL

No attempt will be made to compare the growth potential of the solid Nova with the liquid C-8 when advanced developments are incorporated; too many possible combinations exist to make a meaningful study at this time. However, a comment pertaining to the growth is worth noting. The very large solid Nova first stage provides a much greater lift capability than other ground launched systems under consideration. With any given advanced propulsion system as upper stage, the solid Nova will provide better overall performance capability than the other first stages. The reason for this is simply that the solid Nova first stage is much larger and has a higher thrust (about 50 million pounds). The large launch weight that was considered a drawback of the solid Nova by many at the outset of the studies proves to be an asset in this respect.

V. CONCLUSIONS

An examination of manned lunar and planetary missions for the foreseeable future reveals that, despite the obvious challenges to our booster and performance capability, system reliability for the overall mission profile constitutes the greatest potential obstacle to successful mission accomplishment. The latter obstacle arises because of major constraints from 1) non-expendable payloads, ie men, 2) the need for large, complex boosters, 3) very high unit cost, and 4) a long and complex sequence of critical operations to complete a mission. One important aspect of the mission profile, the injection vehicle, has been examined conceptually, with reliability, rather than performance or minimum weight, as the ranking criterion and the following conclusions are reached:
1) Vehicle system reliability will probably prove to be unacceptably low for complex manned space missions if performance (weight reduction) and reliability carry the same relative importance in the new development programs that they have in past missile and space programs.
2) A design and development philosophy which incorporates the important principles 1) unusual conservatism and 2) design within the state of the art is advocated as a mechanism whereby substantially higher reliability can be realized.
3) By using this philosophy and capitalizing on unique characteristics of an all solid rocket vehicle, a basically new approach to high early vehicle reliability in flight becomes possible; program implementation would consist of 1) analyses and evaluation of the motor designs, vehicle structure, and vehicle flight dynamics in scale model tests during the preliminary design and large scale tooling period, followed by 2) assembly of scaled up, full size components as a complete vehicle for immediate use.
4. An all solid propellant injection vehicle in the 25 million to 35 million pound launch weight class, studied by JPL and critically evaluated by STL and Boeing, has been found feasible.
5. When compared to liquid propellant or hybrid stage vehicles of comparable performance capabilities, it is believed that the development risk of the all solid Nova would be very low and the resultant vehicle reliability at an early date predictably high- provided the suggested philosophy and program approach were utilized.
6. From a technical viewpoint, the schedule for the all solid Nova injection vehicle studied is significantly shorter and the program costs markedly lower than hybrid stage and liquid propellant vehicles of comparable payload capability despite the much greater launch weight of the solid vehicle.
7. By replacing solid upper stages with liquid hydrogen/liquid oxygen stages or nuclear stages as these become available, the growth potential of the all solid vehicle is especially favorable for space missions which require larger payloads and greater sophistication. Such hybrid stage vehicles show great promise for future planetary missions using electric or nuclear propulsion.

I. INJECTION VEHICLE

A. DESCRIPTION AND OPERATION

The injection vehicle in its primary form was estimated to have a payload capability of 130,000 lbs to escape or 500,000 lbs to LEO and a launch weight of 25 million lbs. Subsequent and more sophisticated calculations show that the vehicle launch weight is closer to 30 million lbs if the vehicle is to provide the indicated payload capability. It is believed that this revision will affect size and cost only and in no way bears on the feasibility of the concept. All discussions in Appendix B are based on the 25 million pound vehicle.

1. VEHICLE

The injection vehicle (fig. B-1) consists of four stages of clustered solid propellant motors. The motors in a stage are joined by intrastage structure; successive stages are joined by interstage structures which have provision for positive separation. Each stage has its own thrust vector control (TVC) system. The fourth stage carries the 130,000 lb escape payload, guidance, control, and telemetry for the injection vehicle, and a vernier propulsion system. For this study, the injection vehicle was sized to inject its fourth stage into a parking orbit. The fourth stage then injects the 130,000 lb payload into an escape trajectory. The typical sequence of operations is:
1. stage one ignition
2. stage one burnout and separation
3. stage two ignition
4. stage two burnout and separation
5. stage three ignition
6. stage three burnout and separation
7. vernier velocity correction into parking orbit
8. orbital coast
9. stage four ignition
10. stage four burnout and payload separation.
11. vernier velocity correction into final escape trajectory

The desireability of maintaining flexibility with respect to the altitude of the parking orbit is discussed in Section I-Q. To accomplish this would require a resizing of the first three stages of the vehicle and of the vernier velocity system on the fourth stage. Flexibility and other mission requirements might require the use of a storable, restartable liquid fourth stage. Detailed consideration of these requirements ws beyond the scope of this study, but in further studies of the concept to greater depth the applicability of such liquid upper stages should be examined.

2. SPACECRAFT

The spacecraft is composed of the following elements:
1. Three man Apollo Command Module
2. Mission Module (Apollo) *note- at the time, NASA was strongly considering a Soyuz-like three-component CSM consisting of a Service Module, much smaller reentry Command Module, and a disposable Mission Module
3. Midcourse, lunar retro, and takeoff propulsion systems (direct ascent lunar mission mode, which was preferred then)
4. Associated structure, guidance and control, and communications systems.
5. Abort propulsion and structures.

The operational sequence for a manned or unmanned lunar landing and return is:
1. Launch to injection
2. Midcourse correction
3. Transit
4. Site acquisition and lunar retro firing
5. Lunar landing
6. Lunar operations
7. Takeoff
8. Transit
9. Midcourse correction
10. Earth reentry and recovery

B. CONFIGURATION

The injection vehicle configuration is shown in Fig B-1; a weight summary is shown in Table B-1. The following factors entered into the decisions which resulted in this configuration:
1. Only two different solid propellant rocket motors would be developed. The Type A motor would be used in stages 1 and 2, and the Type B motor would be used in stages 3 and 4. This obviously requires that the number of motors in stages 1 and 2 be integral multiples of the Type A motor and that the same be true for the Type B motor in stages 3 and 4.
2. The logistics of motor manufacture, transportation, handling, and field erection dictate a unit weight that can be easily handled.
3. The mission constraints require that: 1) the fourth stage and payload be injected into parking orbit at the end of third stage burn, and 2) the fourth stage be able to inject a 130,000 lb payload into an escape trajectory from parking orbit.
4. The requirement that present state of the art be used led to the assumption of specific impulse and stage mass weight ratio used in sizing the vehicle.

The geometry of clustered cylinders and the desire to provide reasonable load paths for interstage and intrastage structure played a role in determining some of the geometrical details of this particular configuration. The result shown is based on the constraints and the considerations described above and is considered to be representative and adequate for the purpose of this study. It is certainly not intended to be put forth as an optimum configuration. The physical size of the vehicle, 77 foot base diameter and 218.5 feet to the top of the fourth stage, is similar to vehicles which have already been considered in various other studies. The gross weight at takeoff, 25 million pounds, is larger than previously considered. It can be compared to some water heaters used in steam electric plants that are 150 feet high and weigh 20 million pounds or to an aircraft carrier weighing 120 million pounds. Interstage or intrastage, and TVC systems structure is not shown on Fig B-1 becuase of the small scale. These items will be discussed in the following sections. The absence of a complete aerodynamic shroud is intentional since the need is not defined at the present time. Each of the solid propellant motors utilizes a single fixed nozzle and has a thrust program designed to produce acceptable accelerations at stage burnout. In the design shown, it was necessary to assume step mass fraction (ratio of propellant to total stage weight) in order to estimate the performance capability of the vehicle. A stage mass fraction of 0.87 was assumed for, the first and second stages, and 0.91 was assumed for the third and fourth stages. Preliminary weight estimates of case and nozzle design, propellant charge configuration, interstage structure, cluster configuration, and TVC requirements have been made and are discussed in Section I-R. Aerojet-General has stated that the first stage of a 7 million pound gross weight vehicle can achieve a stage mass fraction of 0.889, utilizing 7 motors of 140 inch diameter as propulsion elements. By the utilization of a single unit of 288 inches in diameter (roughly the same size as the Type A motor) they can achieve a value of 0.893. Grand Central Rocket Company has stated that the first stage of a 10 million pound gross weight vehicle, consisting of 16 ten foot diameter motors, would have a stage mass fraction of 0.886. It is expected that these numbers may be optimistic and attainable values can be established only by thorough preliminary design, integrating mission system requirements, engineering mechanics, propulsion, guidance, control, vehicle telemetry, power, and last but not least, payload vehicle interactions.

C. ROCKET MOTORS

The Type A motor is 300 inches in diameter and has a length to diameter ratio of 3.26:1. It contains a little less than 2 million pounds of propellant. The motor has, basically, a star shaped grain perforation. Motor volumetric loading is approximately 82%. The average vacuum thrust is 6.4 million pounds and the burn time is 85 seconds. The Type B motor is 220 inches in diameter and has a length to diameter ratio of 2.33:1. It contains about 350,000 pounds of propellant and also has a star shaped grain perforation. Motor volumetric loading is approximately 88%. The average vacuum thrust is 740,000 lbs and the burn time is 138 seconds. These motors represent a typical design concept to demonstrate the feasibility of very large units withing the current state of the art. The size of the motor and grain configuration can be varied to satisfy particular motor requirements. Both motors are designed to utilize existing available propellants, with solid loading of approximately 82 and 88%, and containing powdered aluminum in the 15-20% range. The Type A motors were designed for a nominal motor chamber pressure of 800 PSI with a burn rate of 0.64 inches per second. The Type B motors were designed for a nominal motor chamber pressure of 350 PSI with a 0.46 inch per second burn rate. These burning rates are available in current propellants both from Aerojet and Thiokol. A specific impulse of 245 seconds at 1000 PSI and sea level optimum expansion was assumed. This performance level has been verified many times with both polyurethane and PBAA propellants in current motors. The difference between the measure ISP value and the calculated ISP value far exceeds the correction for heat loss and nozzle divergence angle. It is believed that the bulk of these excess losses is due to velocity and thermal nonequilibrium of the aluminum oxide particles (products of combustion of aluminized propellants) with the gas. Calculations performed at JPL show that aluminized propellants which deliver an ISP of 245 (at 1000 PSI and sea-level optimum expansion) in ordinary nozzles should deliver substantially more in very large nozzles such as those considered for the solid NOVA. The vacuum specific impulse used for A and B motors represents conservative estimates of this ISP improvement in very large nozzles based on measured aluminum oxide particle sizes and one dimesional, 2 phase flow calculations performed on the IBM 7090. The reproduceability of performance for the type of very large units under consideration is expected to exceed that of present, relatively small motors. For example, deviation in burn rate will be less because of amuch larger sample of propellant is exposed to burning.

D. MOTOR CASE DESIGN

The design of the motor cases of the size considered in this study presents no problems basically different from those that would be encountered in the design of smaller size cases.

E. NOZZLE DESIGN

Single fixed nozzles will be used for both type motors. They represent configurations scaled up from known practical steel, graphite throat nozzles. They have been canted in the first and second stages so that the thrust of each motor passes through the vehicle CG at burnout. Their length is established through the use of standard design procedures for the bell shaped exit nozzles. Nozzle materials and fabrication methods already developed seem entirely adequate for building nozzles for these large motors. Since time was not available during this study for detailed nozzle design, weights were estimated from existing information. Two nozzle designs are used in the vehicle. The average thrust of Type A motors on the first and second stages is 6.4 million pounds; the average thrust on the Type B motor on the third and fourth stages is 740,000 lbs. A survey of available reports containing nozzle designs was made and a plot of thrust vs. nozzle weight was prepared (Fig B-6). The weight of the two nozzles was estimated from this plot. The nozzles considered in preparation of Fig B-6 have expansion ratios of about 10:1 and a chamber pressure range of 700-1000 PSI. The nozzle for the type B motor has an expansion ratio of 33:1 but the chamber pressure is only 360 PSI. Since the nozzle weight can be expected to increase with an increase in expansion ratio and decrease slightly with a decrease in chamber pressure, an estimate of 3500 lbs for the nozzle weight seems reasonable. The estimate of 34,000 lbs for the weight of the large type A motor nozzle appears to be sufficiently conservative, since as can be seen from Fig B-6, the weight ranges from 23,000 to 39,000 lbs for a thrust of 6.4 million pounds. A plot of total motor impulse vs weight was also prepared; it indicates that the large type A nozzle is reasonably conservative. The weight of the type B smaller nozzle appears to be less conservative in this plot, but the value quoted above was used since the weight range for the examples considered here, at a slightly lower total impulse, is rather large. Since the diameter of the nozzle throats considered for this vehicle is large, ablative type nozzles may be more desireable and easier to design in this size than in the current advanced developments being pursued. Although not considered here, they represent a design which promises a weight savings, at no decrease in reliability, over that estimated in the weight analysis.

F. COMBUSTION INSTABILITY

Combustion instability is not expected to be a problem in large solid propellant rocket motors. Although many motor programs have had combustion instability problems, a solution has always been found and the programs have been successfully concluded. There are a number of major reasons for not expecting combustion instability in a large composite ammonium perchlorate aluminum solid propellant motor. The last three large composite solid propellant rocket motors, Minuteman, Polaris, and Pershing first stages, have not had combustion instability problems using the type of propellants under consideration. Aluminized propellants have shown instability have been in large L/D motors at relatively high pressures, around 1500 PSI. Recent experience at JPL with a nonaluminized polyurethane has indicated a stable region below 750 PSI and logitudinal oscillations at higher pressures. Small amounts of aluminum gave stable burning at 1000 PSI. Unfortunately, there are no data in the low frequency range of large motors.

G. SEGMENTED VS. UNITIZED ROCKET MOTORS

The choice of a unitized (monolithic) vs. segmented motor construction does not appear to be very simple. From the standpoint of reliability each segment of a segmented unit contains most of the probable causes for failure of a solid propellant motor. For example, each segment will have associated with it one joint; each segment will have associated with it two ends which tend to be the regions of pull-away (separation) of the propellant from the case, and each segment will be processed and trimmed separately, inviting other failures if proper controls are not exercised. A unitized motor has a similar joint attaching the nozzle to the case and also similar modes of failure. From the reliability standpoint it can be argued that the number of segments or unitized motors should be minimized. The minimizing of the number of units implies maximizing the size of each unit to do a specific job. A number of arguments have been proposed for minimizing the size of a segment from the standpoint of ground support equipment, processibility, and transportability. From the long experience that is available with missile projects, it can be stated that it would be an error to grossly compromise the injection vehicle on the basis of ground support equipment. Existing processing facilities cannot handle and overload of the size being considered. Therefore, separate process facilities which would have to be established could just as easily be designed to handle large unitized motors as single smaller segments. The concepts proposed in this report are based on what appears to be the more difficult problem, that of handling large single unitized motors, although it is not necessarily recommended that this be the final solution. It is believed that there are problems in the ares of hardware fabrication and vehicle assembly which are minimized with the use of smaller segments and, if the flight vehicle can be designed better by the use of segmented motors, the latter should be used. The final decision between segmented or unitized motors should result from a study of the effects of the choice on the flight vehicle.

H. PROPELLANT PHYSICAL PROPERTIES

The structural integrity of the large propellant grain has been examined and found satisfactory. It is anticipated that there will be a scaling effect in extrapolating propellant physical properties of small grains to large grains, and this effect should be studied during the preliminary design. However,on the basis of experience with state of the art large motors, this effect is not thought to be limiting.

J. INTRASTAGE STRUCTURE

The intrastage structure for all stages has two main purposes: 1) to transmit shear forces between motor cases, and 2) to tie motors together radially. The shear forces between motors arise from thrust variation, a time lag between motor startup and burnout of individual motors, and body bending. The shear forces in the first stages can be transmitted by means of short, thin walled cylinders filling the interspace between the clustered motors. These cylinders may be attached to the motor cases with high strength bolts, through lugs welded onto the cases during fabrication. The third stage motor cluster could develop shear transfer between motors by means of half-cylinders, because of the nesting of the third and fourth stages. Essentially, the shear will be transferred in the same manner as in the first two stages. The motors of the first three stages will be tied together to resist relative radial motion. This will be accomplished by means of a system of cradels and ties at the top and bottom of each motor case. The tie arrangement will be placed at the junction of the case dome closure so that a minimum amount of restraint will be offered to the growth of the cases resulting from pressurization.

K. INTERSTAGE STRUCTURE

The following study goals have been considered when investigating the interstage structure: 1) feasibility considerations for several structural concepts. 2) development of one or more workable conceptual designs. 3) approximate analysis and weight estimate for at least one design. The bulk of the current work has been oriented toward the first to second stage interstage structure. Separation has been considered only to the the extent that it affects structural configuration. It has been assumed that full ring bulkheads at the ends of each stage are not required, that concentrated loads can be applied to the motor cases, and that interstage structure is capable of transferring thrust differences between motors. The largest loading on the structure is of course axial thrust. Side loads caused by wind shears, thrust misalignments, and the control forces resulting from them will vary along the length of the vehicle because of rigid body and elastic dynamic considerations. Vibration in both the axial and transverse directions will be added to the above loadings. Conserative and realistic load estimates were made and both were used. It is believed that they bracket the actual loads. In each case it was conservatively assumed that the maximum axial and transverse loads occur simultaneously. Four types of structures considered are: 1) Truss, 2) Longeron-tie rod, 3) Monocoque, 4) Nozzle utilization. It is considered that the interstage structure design should be predicated on the required strength rather than rigidity. However, the effect of rigidity on dynamic response must be carefully considered in a preliminary design. Preliminary consideration of truss, longeron-tie rod, and moncoque structures showed each to be feasible. The light weight of an unstiffened nozzle structure rules it out as a practical solution. Sketches of a truss and longeron structure design are in Fig B-7 and B-8, respectively. Analyses have been made of a monocoque and modified monocoque structure. The material assumed for the interstage structures is steel with an ultimate strength of 180ksi and 100 ksi at weld joints connecting major structures, with a safety factor of 1.25. The truss design is shown in Fig B-7. If a tubular cross-section is used, the lower members could have a diameter of 24 inches with a thickness of 1.33 inches for number 1 loading, or a thickness of 0.85 inches for the number 2 loading. Longeron design is shown in Fig B-8. A shell-type structure would require rings at the bottom and top, and the separation plane. It may be possible to use partial, scalloped rings for the top and bottom. The critical condition for this design appears to be shell instability. For a solid shell, the required thicknesses of approximately 1.7 and 1.35 inches.

L. THRUST VECTOR CONTROL

A method of controlling the thrust vector of each stage must be provided to compensate for the effects of center of gravity displacement, thrust misalignment, and unequal thrusts (especially at ignition and burnout). For the first stage it is needed to counteract aerodynamic loads. It must also provide the necessary maneuvering forces. Because of expected motor-to-motor variances in thrust level, it appears very desirable to cant the nozzles so the main thrust vectors for the for the individual units will pass through the vehicle center of gravity. It is expected that the greatest difference in thrust level between units will occur near the beginning of the tailoff of the thrust-time curve. As a result, it would be desirable to point the nominal thrust vector through the stage CG at burnout. The effect of aerodynamic loads might modify this conclusion for the first stage; this can be determined only by more detailed study. Based upon opinions of most of the contractors who have studied the subject, the choice of thrust vectoring means for a big solid booster system lies between jet vanes and secondary injection of of liquid into the expansion cone of the nozzle. It is believed that at least one other system should be considered: that of auxiliary rocket motors which have been developed by Allison and Vickers for NASA. A brief discussion ofthe features of all these systems follows.

1) JET VANES-- Roll, pitch, and yaw moments can be obtained by a single nozzle installation from four (or three) wedge-shaped aerodynamic surfaces positioned 90 (or 120) degrees apart in the expansion cone of the nozzle. Drag force on the jet vane during burning, decreasing the effective motor impulse, is one disadvantage of this system. Another is the materials problem that results from immersing a substance in the high temperature exhaust stream. Jet vanes are used on Sergeant and Pershing.

2) AUXILIARY ROCKET MOTORS-- This method of TVC uses rocket motors or gas supplied nozzles at a location chosen to give optimum pitch and yaw moments for a given stage, usually near the interstage structures. Individual, self-contained motors can be used or, in a system that uses gas supplied nozzles, a single common gas generator can be used. In both cases, the exhausting of these secondary gases occurs throughout main stage burning, changing the missile velocity vector only through nozzle orientation. One advantage of this system is that it is possible to keep the auxiliary nozzles exhausting for a specific time after main stage burnout, providing control forces until the subsequent stage ignites. The major disadvantage of this system, compared to the others considered, is the unreliability added to the added to the vehicle systems through the inclusion of an additional propellant device, whether used for direct TVC or for generating gas for several nozzles. Both of these methods are presently under development.

3) SECONDARY FLUID INJECTION-- TVC in pitch and yaw can be obtained for single-nozzle configurations through injection of fluid into the expansion cone of the nozzle. Decomposition gases can also be used. To date, secondary injection has not been flight tested; however, systems are in advanced development stage for use on Polaris and Minuteman. A special feature of this system is that the injected material affect both the effective mass ratio of the unit and specific impulse, since material is being expelled during operation of the main engine. For this analysis, a secondary injection system which may prove to be inherently more reliable was chosen for further detailed study in order to provide TVC weight and cost estimates. Such a system provides jet deflection as a result of introducing high velocity fluid stream into one side of the divergent cone of the nozzle. An oblique shock is created to cause an effective deflection of the exhaust gases linearly proportional to the mass flow of injectant. It has the advantage over other typical systems of being light in weight while requiring few moving parts. Thiokol Chemical Corp. (ref 24) found that such a system would weigh less than half an equivalent system employing jet vanes. It was also shown to be lighter than a simple auxiliary jet system. However, this conclusion isn't universally applicable. In order to estimate the TVC system, the following assumptions were made: 1) CG offset per stage, 6 inches. 2) Net thrust misalignment per stage, 0.25 degrees (this includes thrust differences of 3.7% 3) Nozzles are canted nominally through stage burnout CG. 4) Ignition delay of 0.3 second 5) Burning time variation of =/- 3.7%. 6) All motors on one side of the CG ignite before all motors on the other side. 7) Effects of motor to motor burn time variations are minimized by long thrust tailoff (approximately 10% of burn duration). This can be further minimized in the first stage by giving the center motor a slightly longer burn time. 8) Maximimum wind shear equivalent to 0.3 g side acceleration with a moment arm of 25 feet (the force is believed conservative by a factor of 2.5 and the moment arm by a factor of 4). 9) CG travel as shown in Fig B-9 was used. It was found that side forces applied at the nozzle exits and total side impulses as indicated below would be required (expressed as a fraction of stage axial thrust or total impulse). Based on these requirements, two systems for the first stage were briefly investigated: a turbopump system and a pressurized system. The turbopump was visualized as powered by a monopropellant liquid or solid propellant gas generator. To minimize turbopump control problems a bypass would provide return flow to the injectant tank as jet deflection demands drop off. Ullage pressure for the injectant tank would be provided by either bleeding turbopump exhaust gas or by heated injectant expanded from the high pressure side of the pump. The pressurized system would utilize helium gas stored at 5000 psi and regulated to deliver injectant at the required pressure. Both systems would utilize multiple valves for each cluster arranged to provide proper pitch, yaw, and roll control moments. (Ideally each motor should be self-sufficient at burnout). Programming would be provided to ensure that all injectant is expelled prior to burnout in order that full utilization can be made of the energy available. The turbopump hydraulic horsepowr requirements roughly approximate that of the Rocketdyne F-1 engine turbopump. This unit requires 128 lb/sec of gas for the turbine and weighs almost 3000 lbs.

M. AERODYNAMICS

1. AERO-FAIRING-- The necessity for aerodynamically fairing the vehicle has been examined for the clustered-type vehicle under consideration. Such a fairing may be necessary owing to three aerodynamic factors-- 1) heating, 2) drag, and 3) unsteady flow effects. Aerodynamic stability is not considered in this section. A) heating-- both maximum laminar and turbulent heating rates were examined. The results of this plot clearly show that aerodynamic heating would not be a serious matter to structural members. Even the use of turbulent heating (which is improbable) in this example would not indicate a greater temperature rise because of the small dimensions involved. Correspondingly, heating of the exposed engine tanks would be even less serious because the heating rates are reduced as a result of the effects of large nose radii and large dimensions from the nose to sonic point. B) AERODYNAMIC DRAG-- An open type configuration such as considered in this study, does not lend itself to simple analysis for the estimation of drag. Nevertheless, it is essential to make an estimate in order to determine what effects may occur to the flight parameters, which are presently based on a drag curve for a fully faired vehicle. Fig B-13 shows the drag coefficient vs Mach number used in the flight calculations. Also shown on the figure is the dynamic pressure, which is observed to peak at a Mach number of about 1.7. For estimating the drag of the large vehicle considered here, pure turbulent friction drag is not an important factor because of the high Reynolds numbers involved. Thus, the major problem is pressure drag at supersonic speeds, although major contributions can arise from the interaction of shock waves and boundary layers and from flow separation effects. Vehicle size will affect only these latter sources. In addition to examining a small amount of existing experimental data, two simple drag models were calculated. These are: 1) no fairings assumed: all motor cases independently exposed to the free stream airflow (stage four and payload considered as one body). 2) partial fairing assumed. Case 1 is not a very realistic condition to consider, since adjacent motors would cause mutual interference, so that the effective drag would be greater than the sum of the individual bodies. However, motors of the first and second stages would be exposed to flow at much less than free stream dynamic pressure, thus lowering the drag. The net results of these somewhat compensating effects is not predictable in short period of study. However, it may be reasonable to assume that by properly streamlining all structural members, placing conical noses on top of stage 3 motors, and judiciously using local fairings throughout the structure, it may be possible to keep the drag coefficient down to twice the basic streamlined case, or a drag coefficient of about 1.0 at Mach 1.7. Considerable wind tunnel development would be required on such an open configuration. Theoretical work on determining proper motor lateral spacing may be fruitful in reducing interference drag. For the externally faired case, the major drag is derived from the nose cone and the transition region between stages 1 and 2 as indicated in Fig B-13. For this case, it is observed that the drag coefficient is about 60% higher than the basic curve. Additional computer effort would be required to ascertain the effect of these potential drag coefficient increases on flight parameters. However, taking the thrust as about 40 million pounds, a twofold drag increase would result in a peak drag-to-thrust ratio of only about 0.15, which is acceptable. C) UNSTEADY FLOW EFFECTS-- This effect may be the one which requires a faired configuration. Fluctuating flows through the partially open vehicle may induce vibrational modes in the structure at levels which are not acceptable. However, as in the case of drag reduction, local fairing and streamlining would be beneficial.
2) BOOST PHASE LATERAL AIR LOADS-- The following trajectory assumptions were made when estimating the boost phase lateral air loads: 1) vehicle CG follows intended no wind flight path (zero-alpha trajectory), crabbing into the wind as necessary to avoid drift, 2) TVC gimble angle varied as necessary to produce the small angle of attack (alpha) required to cancel driftand, at the same time, to counterbalance air load pitching moments. Inasmuch as the above mentioned crabbing alleviates the air loads on the vehicle about 20%, a slight more conservative (pessimistic) assumption would be assume the constant attitude vertical flight. However, the associated drift velocities make this representation unrealistic.
3) BASE HEATING-- The heating of items stored around the nozzles has not been examined. Experience gained by tests and flights of the Saturn first stage should be applicable. The large nozzle exit area and the close spacing of the nozzles woudl make it relatively easy to insert a lightweight fiberglass shield to prevent the flow of hot gases up between the nozzles if it should prove necessary. Radiation heating of the outer surfaces of the motor cases by exhaust flame appears relatively insignificant.

IV. SPACECRAFT CONSIDERATIONS

A. PHILOSOPHY
Spacecraft studies have been limited to a crude feasibility determinations based on 1) 130,000 lbs injected weight, 2) configuration constraints, and 3) spacecraft design problems peculiar to the employment of the solid NOVA for a manned lunar landing and return mission. The results of the studies to date can be summarized as follows: 1) there are no spacecraft design problems peculiar to the solid NOVA when compared to the liquid NOVA; 2) the injection capabilities of the solid NOVA appear to be adequate for a manne lunar mission, 3) there are no major spacecraft configuration constraints or limitations due to the injection vehicle. For the purposes of this study, Apollo 3 man mission and command modules of the Convair M-1 type have been assumed. Mission abort capabilities are assumed to be consistent with the guidelines established by the Space Task Group. It is recommended that the spacecraft be capable of accomplishing the entire mission automatically. Man would perform monitoring functions associated with the control loop and he would help implement scientific measurements and observations. Manned override control capabilities would be provided for emergencies. A possible feature of the manned mission would be to provide an alternate return vehicle on the surface of the Moon as a contingency for possible failures during landing. In this event, man would be equipped to transport himself over the lunar surface from one vehicle to the other.

B. DESIGN: SPACECRAFT CONFIGURATION STUDY--
The Apollo command and mission module concept in the Convair configuration M-1 was arbitrarily selected as the basic vehicle. Two propulsion configurations were considered; an all liquid configuration of the storable propellant type, and a hybrid, using solid rockets for the retro to the Moon and takeoff and three storable liquid vernier engines for mid course correction and vernier descent to the surface of the Moon.

1. WEIGHT BREAKDOWN-
The Apollo studies indicate weights for the command and mission module which are compatible with the permissible weight figures as determined in this study. The approximate gross spacecraft weight breakdown is as follows: Command module- 5,650 lbs; Mission module- 3,500 lbs; Propulsion system- 117,500 lbs; Guidance, control, interstage, propulsion support structure, etc.- 3,350 lbs.

2. CONFIGURATION: LIQUID PROPULSION SYSTEM
Figure B-25 shows the spacecraft with the all liquid propulsion system. A Titan II second stage thrust chamber and pumping system is used for the retro and lunar liftoff maneuvers. Liquid verniers with a 10:1 throttling capability operating from the same pumping system as the main engine are used for midcourse correction and for limited hovering capability at lunar landing. Attitude control can be accomplished by twelve 7 lb thrust engines arranged in pairs on the vehicle. The retro tanks which are to be left on the Moon are used as landing support structure. Energy absorbing material is attached to the tanks. Three stabilizers are deployed during the hovering maneuver. Part of the interstage structure could be deployed to serve as the landing gear and stabilizer rather than using the tanks for this purpose.

3) CONFIGURATION: HYBRID SYSTEM
Figure B-26 shows the vehicle with the hybrid propulsion system in place of the all liquid system. Here, three solid rockets are used for the main retro system and one for the lunar liftoff. The same storable propellant liquid engines, now with their own packaging and pumping systems, are used for the midcourse and vernier maneuvers.

4) ABORT SYSTEM CONFIGURATION
Eight solid motors mounted on a tower on the front of the command module supply the necessary velocity to perform the abort function during third or fourth stage firing, at which time the complete vehicle is separated from the booster. Firing four of these motors performs the abort mission from the pad when only the command module is separated.

5) VELOCITY INCREMENT: WEIGHT CALCULATION
Tables B-6 and B-7 show weight calculations for the normal mission employing liquids and solids, respectively. Should propellant not be used in midcourse or vernier touchdown, a hovering time greater than 60 seconds could be utilized. The desireability of and capability for this maneuver can be determined by the men aboard the spacecraft.

6) COMPARISON OF DESIGN LOADINGS
The maximum acceleration of the injection system is about 5 g. The spacecraft propulsion system of the liquid type imposes 6 g maximum, the hybrid configuration imposes 9 g maximum acceleration on the spacecraft. These levels compare favorably with the design levels adopted in the Apollo studies and justify using structural weights given therein.