Document created: 27 August 04
Air University Review, November-December 1970

Some Thoughts On
 Reusable Launch Vehicles

William G. Holder
Captain William D. Siuru, Jr.

The Apollo successes have vividly dramatized the magnitude, sophistication, and capability of the U.S. space program. The gigantic Saturn V/Apollo combination represented the culmination of seven years of technological aerospace advances made by literally hundreds of aerospace companies and government agencies. Unfortunately, the Saturn launch vehicle—worth about $100 million—can be used only once, a fact which, when coupled with the increasing domestic demands on the U.S. budget, threatens the continuation of a vigorous space exploration program. The realistic solution to the economic squeeze is to develop a reusable launch vehicle (RLV).

The cost per pound of payload delivered to low earth orbit by today’s launch vehicles runs from $400 to $2500, not including the elaborate prelaunch assembly and checkout procedures. To achieve our ambitious future space objectives with the limited dollars available, the cost and complexity of launch vehicles must more nearly approach the cost and complexity associated with aircraft operations. A payload delivery cost of $40 to $125 per pound delivered to orbit is a realistic and reasonable goal to strive for in a first-generation RLV. The maintenance and launching of such a vehicle should, ideally, require no more than a pilot, a copilot, a crew chief, and a few ground personnel.

Let us now consider how much the U.S. could afford to invest in developing an RLV and some of the benefits other than economic that might result if an RLV were developed.

economic implications and factors

Probably the simplest way to illustrate the economics of reusable systems is to show how many launches of an RLV are required to amortize investment costs for development, initial inventory, and facilities. The number of launches after amortization represents savings over a comparable expendable system. Three payload classes of launch vehicles are shown in Figure 1, the smallest equating in capability and cost to the Titan IIIM and the largest to the Saturn V class. Between the two is a median vehicle with a payload in the 100,000-pound category. All payloads are to orbits of 100 nautical miles. A band of allowable investments is given, representing savings of 75 and 90 percent over the expendable system.

Figure 1. Economics of recovery. Example: For a 35,000-pound-payload system and a projection of 80 launches, approximately $1 billion could be invested in a reusable launch vehicle (RLV) with savings achievable after the 80th launch. Since the RLV costs only 10 to 25 percent of the cost for a comparable expendable system (i.e., $2 to $4 million per launch versus $17.5 million per launch), the RLV could perform the missions of smaller, yet more costly, expendable launch vehicles. Such as 35K-lb-payload RLV could be used in place of all vehicles between the Thor and Saturn IB, and thus the 80 launches could be accumulated in a relatively short time, with significant savings in subsequent years.

Several points should be noted. First, the investment costs for an RLV are not really tremendous compared to the huge investment involved in a system like the Saturn V. Second, the number of launches before savings result is relatively small, considering that there have been at least 50 space launches of all payload sizes per year in the last five years.

Obviously, in justifying an RLV, the planner must maximize the achievable savings and minimize the investment required. To obtain greatest usage and thus savings, any new launch system must be flexible enough to encompass a wide range of payload weights delivered to varied orbital and interplanetary destinations. What is proposed is a whole new concept in launch operations. This concept envisions the use of a “space truck,” as opposed to the “special vehicle for each mission” concept of today’s space operations.

An attractive future launch capability is one of about 50,000 pounds delivered to low earth orbit. Referring to Figure 1, one can see at a usage rate of 50 launches per year and for an investment of about $4 to $6 billion, an RLV could be amortized over a four-year period. The $4 to $6 billion investment is probably more than required for the job, so amortization could occur somewhat earlier. A launch rate of 50 per year is also reasonable, since this vehicle could be used to shuttle men and supplies to our future space stations, perhaps on a monthly or even biweekly basis. In addition, this vehicle could encompass most of the missions now performed by the Saturn, Titan, Atlas, and perhaps even the Thor family of launch systems.

An added feature of low-cost launch vehicle is its ability to create new uses. That is, as launch costs diminish, it becomes more practical to use space for additional endeavors, giving an added base for investment amortization and thus further increasing the savings per launch. Additionally, an RLV which is designed with growth potential in mind provides a longer operating life span and therefore allows a greater number of years of use after the investment costs have been paid off.

Since projected launch rates now appear to be lower than those projected by studies in the early 1960s, the potential amount of savings achievable with reusable launch systems has decreased. Early studies assumed launch rates in the hundreds per year, which would make even the most “way out” RLV system appear attractive. Today’s launch rates account for fewer than 50 launches of medium- and large-sized vehicles per year. This reduction in launch rate requires a decrease in the allowable investment, and the actual recoverable systems proposed have to be more realistic in their design and technology.

advantages—other than economic

In addition to the economic benefits of recoverable boosters, there are several fringe benefits to be derived. Many recoverable concepts allow a mission abort and recovery of the payload, an especially attractive feature for manned or other particularly high-value payloads.

The flight-test program for an RLV is similar to that for an aircraft in that it can be conducted in a stepwise manner with a minimum of test hardware. A vast amount of operating experience and “shakedown” testing can be accumulated before the first payload is launched. Unlike the totally expendable launch vehicles, the RLV can be flight-tested without incurring more than the cost of the expended propellants and the flight-to-flight maintenance.

One of the constraints that has limited the U.S. to only two space bases capable of launching major payloads is the vehicle overflight problem. The U.S. cannot launch rockets over populated areas for fear of debris jettisoned during the normal flight profile or fear of failure. A reusable vehicle with aerodynamic return capability could overcome this problem, since the launch vehicle would be little different from an airliner flying over a city, provided a flight trajectory was chosen so as to minimize ground noise from the propulsion system. An RLV with man aboard could alleviate many of the situations where the launch vehicle has to be destroyed because of minor malfunction. This alleviation of overflight restrictions could allow more launch sites to be developed or take advantage of heretofore unusable facilities, a feature highly desirable for military missions.

Another desirable feature of most RLV concepts would eliminate some of the transportation problems from manufacturing site to launch site. The RLV with aerodynamic qualities could be ferried from the manufacturing site to the launch site and thus eliminate the special carriers in use today, such as Guppies and special booster barges.

The large booster concepts (boosters larger than the Saturn V) have been mentioned as possible contenders for a considerable portion of future space research money. However, a relatively small RLV could accomplish the missions projected for these large boosters through the assembly-in-orbit concept. While a very large launch system would have very limited application, perhaps less than one launch per year (witness experience with the large Saturn vehicles), a moderately sized RLV could take on the job of the large booster as well as many other smaller missions. An inherent feature of the assembly-in-orbit concept is that a single launch failure does not destroy the entire payload, whereas with the large single-vehicle launch a failure could cause complete destruction of the payload and possibly the whole program. Most of the large-payload concepts already consider some type of resupply and crew rotation capability. An RLV could provide both of these functions in addition to establishing the space station in the first place. A very large booster would certainly require advances in technology and of course new and quite expensive launch facilities, while an RLV could be built with currently available technology and launched from existing facilities usually with only minor modifications.

Today the payload planner is constrained by the maximum payload capability of the largest launch system he can afford. If a single RLV were developed with a constant cost, regardless of payload, he might see payload growth towards the maximum capability of the RLV. If this cost was constant at a value less than for equivalent expendable systems, he would be able to use the maximum payload weight consistent with his particular mission. This in itself could lead to lower costs, since the spacecraft would not have to use such expensive techniques as miniaturization of components and could benefit from such concepts as increased redundancy and longer life components. Of course, it goes without saying that this additional capability could also be used to broaden the mission capabilities of the spacecraft itself.

concepts

Now that the benefits to be derived from the RLV have been established, some of the various concepts will be discussed. A surprising number of recovery concepts have been suggested by both the government and industry, employing every conceivable technology. These concepts range from the minor modification of existing hardware to those based on the greatest advances in technology. In line with today’s realistic outlook, the discussion will be limited to those systems which represent an evolution from the systems of today, rather than the more revolutionary concepts.

The simplest RLV techniques involve the recovery of currently used expendable stages. These schemes usually fall into the class of systems known as vertical takeoff and vertical landing (VTOVL) and are primarily designed to recover the first stage. (Figure 2) This stage is the easiest to recover because its impact point will probably not be farther than 300 miles downrange and it will be subjected to the least severe environment. Recovery devices most frequently proposed are parachutes, paragliders, and deployable rotors.

Figure 2. Vertical takeoff and vertical landing recovery (VTOVL)

While these methods represent the first step in the evolution of recovery technology, they are neither operationally nor economically attractive. First of all, if an expendable stage were adapted for reuse, the number of reuses would be very limited and the refurbishment costs high, since the systems and subsystems were originally designed for only one flight. This type of recovery usually involves landing the stage in water or, in the case of smaller stages, using an aerial snatch technique. As everyone has seen in the Apollo program, the operational problems associated with water retrieval preclude its being an everyday operation. Additionally, landing a piece of aerospace hardware in the ocean can quickly cause corrosion and make refurbishment expensive.

However, water recovery might have limited application in reducing the cost of reuse of very large boosters with low launch rates, e.g., Saturn V. It might also be profitable to recover a stage for research purposes, to determine how it withstood its portion of powered flight These data could tell the designer if the system was overdesigned and where.

It might be worthwhile to recover only the most expensive elements of a launch vehicle, for example, the propulsion and electronic components. This would require packaging these elements so that they could be separated from the rest of the expendable system and thus be recovered separately.

As for the propulsion system, the rocket engine still appears to be the most practical propulsion unit at least for a first-generation RLV. With this assumption, it is almost mandatory that takeoff occur vertically. Recovery is best accomplished by an aircraft-type configuration, which dictates horizontal landing. This, then, presents a most attractive class of reusable launch systems—vertical takeoff and horizontal landing (VTOHL). Within this class of vehicles can be included partially and totally recoverable systems. In the evolutionary approach to recovery, one may start at the uppermost stage or the lowermost stage.

The first VTOHL vehicle would have a recoverable upper stage/spacecraft launched by expendable lower stages. (Figure 3a) This concept could be readily developed if the U.S. decided to pursue the creation of a maneuverable, reusable spacecraft. The technology from the reusable spacecraft would allow an easy development of a reusable upper stage.

Figure 3. Partially recoverable vertical takeoff and horizontal landing (VTOHL)

The second VTOHL vehicle, consisting of a reusable first stage and expendable upper stages, would be the easiest to obtain if a reusable spacecraft were not developed. (Figure 3b) Such a system could be developed by mating rocket engines with aircraft hardware. Since the severity of the flight environment for such a first stage would be minimal, exotic materials and structural techniques would probably not be required.

The next logical step is a totally reusable system. The choice of staging technique is very important. Both tandem and parallel staging have their strong and weak points. Tandem staging (Figure 4a) allows greater flexibility in the various payloads and upper stages that can be accommodated, since the upper stage/payload can be mounted atop the first stage as with today’s expendable stages. With parallel staging, the payload upper stage must be aerodynamically integrated with the first-stage body, somewhat limiting mission flexibility. (Figure 4b)

Figure 4. Totally reusable vertical takeoff and horizontal landing

A class of launch vehicles using a common first stage but a variety of upper stages could provide a wide range of payload capability, with a commensurate range of payload delivery costs. For example, the first stage could be offloaded, and the second stage could be a small upper stage delivering, say, 15,000 pounds to low earth orbit. The same vehicle when fully loaded could use a large upper stage or stages and provide payloads in the 50,000-pound class.

A tandem-staged vehicle could be initially designed to use expendable upper stages; then at a later date a reusable upper stage/payload could be incorporated. A parallel-staged vehicle requires the whole system to be designed as an integral package because of the aerodynamic interfaces between the various stages. This makes it more difficult to use with a wide range of payloads/upper stages.

Parallel staging does, however, reduce some of the loads and bending moments due to wind load while on the pad and during the early phases of flight because the parallel-staged vehicle is usually squatter than the tandem-staged vehicle. This factor is especially important if the payload/upper stages include wings.

A final VTOHL system would employ expendable outboard propellant tanks that are jettisoned once their load has been consumed by a reusable combination spacecraft/launch vehicle in the center. (Figure 5a) Long engine burntimes and payload envelope constraints present the major problems with this technique. The outstanding advantage is that it allows a reusable system to be developed for a minimum cost, since only one reusable system is involved—the center body.

A derivative of this last technique involves recovering not only the center section but also the outboard tanks. (Figure 5b) This could be accomplished with only one development program, since the center vehicle and the outboard vehicles could be identical in design. This system incorporates an inherent building-block capability. For example, for quite small payloads it might be necessary to launch only the center portion of the vehicle, while for very large payloads a number of the vehicles could be clustered together. This building-block approach already has been used with current expendable systems, e.g., the Titan family.

Figure 5. “Drop-tank” vertical takeoff and horizontal landing

The third type of RLV employs horizontal takeoff and horizontal landing (HTOHL) and is thus most like an aircraft. (Figure 6) This arrangement requires the use of an advanced air-breathing propulsion system in the first stage or, alternatively, a rocket-propelled first stage used in conjunction with some type of sled launch device.

Figure 6. Current-technology horizotal takeoff and horizotal landing (HTOHL)

There is one important advantage that can be obtained from an HTOHL system: the ability to provide an offset launch capability. (Figure 7a) In other words, the launch vehicle can be flown from the takeoff site to the point on the earth under the desired orbit, alleviating any requirement for plane changes or for orbital phasing. This is a particularly attractive advantage for missions employing an orbital intercept, such as space rescue and satellite inspection. An additional advantage of this kind of system over a rocket- propelled VTOHL is that the same air-breathing engines could be used during both the launch and the flyback phases, thus enabling a lighter overall system. (Figure 7b) In the rocket system the flyback engines just “go along for the ride” during the launch and thus represent a deadweight penalty.

Figure 7. Advanced-technology horizontal takeoff and horizontal landing

The B-70, the supersonic transport (SST), the C-5A, and new configurations based on this technology have all been studied as possible launch platforms. Their flight profiles require the use of untried flight maneuvers such as parallel staging, perhaps a pull-up maneuver involving rocket assist, and ignition of rocket stages after a few seconds of freefall, sometimes in a near-horizontal position. These studies concluded that the high development costs and risks were not warranted in view of the small amount of energy that this type of reusable first stage contributed to the whole mission. More advanced air-breathing propulsion systems such as ramjets and supersonic combustion ramjets (SCRAMJET), in combination with rockets and turbojets, could allow application of this concept with rather dramatic increases in payload over rocket systems for a given takeoff weight.

operational considerations

The development and use of a reusable launch vehicle require adherence to certain principles. For example, to minimize investment costs, the new system should be adaptable to our current launch facilities at the Eastern Test Range (ETR) and Western Test Range (WTR). At both sites there are existing facilities that could be adapted for an RLV: the Saturn pads at the ETR and the Titan facilities at the WTR and ETR.

Another very important consideration concerns standardized payload interfaces. To make an RLV work, the payloads must be designed around the launch system characteristics, rather than the launch vehicles’ being adapted to payload interface requirements, as is often presently done. Included in payload consideration is the requirement that minimum time be devoted to on-pad assembly and checkout because of the high frequency of launches and the limited number of pads. For example, if the 48 launches in 1968 were accomplished by an RLV from only two launch pads, this schedule would allow an average of about two weeks for pad refurbishment, vehicle loading, and preflight between launches.

One constraint for this type of system would be a standard launch profile for all launches. This might require ballasting, off-loading of propellants, and, since the system will be man-rated, throttleable engines in order to provide a launch system with a flexible payload weight capability.

If the U.S. builds a fleet of RLV’s, it should not be overconservative in the actual number of vehicles purchased. Too small a purchase and one or two complete failures, or even normal attrition, could significantly reduce our launch capability. The reinstitution of production lines at a later date to replenish the inventory is always a very expensive proposition.

To achieve our future space objectives within severe budget restrictions, the next launch vehicle developed in the U.S. must be at least partially reusable. Such a reusable system could be used through the 1970s and into the 1980s, just as our Thor, Atlas, and Titan have been used in the 1950s and 1960s. The other features, such as mission abort, offset launch, alleviation of overflight problems, and “man in the loop,” found in some or all of the concepts, are added benefits resulting from a reusable launch vehicle. An additional incentive for initiating development of an RLV is the technical challenge itself. The payoff for pursuing such a challenge will probably be greater than the benefits derived from exploratory trips to Mars and Venus. It would provide a flexible “space truck” that could economically open the vast reaches of space for military, commercial, and scientific purposes.

Foreign Technology Division, AFSC

Contributor

William G. Holder (B.S.A.E., Purdue University) is a space systems analyst with the Foreign Technology Division of Air Force Systems Command. He has worked with the Boeing Company on the Bomarc B and the Saturn V and served three years as a lieutenant in the U.S. Army as an air defense guided missile instructor. Mr. Holder is the author of a number of technical and historical articles and a book, Saturn V—The Moon Rocket (1969) . . . Captain William D. Siuru, Jr., (M.S., AFIT), is an aeronautical engineer in the Foreign Technology Division of AFSC. He was a project engineer in the XB-70 System Program Office at Wright-Patterson AFB and later was assigned to the Development Planning Directorate of Space and Missile Systems Organization (AFSC), where he worked in advanced launch and space systems planning.

Disclaimer

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

Home Page | Feedback? Email the Editor