Design Example: The MCD/SLV

Working in the TITAN program office was excellent preparation for this new assignment. Among the very many aspects of the design and operation of SLVs, I learned of the detail design, history and developmental problems of solid, liquid, and hybrid rocket propulsion systems, particularly their efficiencies, propellant choices, development times, costs and reliabilities. When visiting contractor facilities - and this included manufacturers of other than TITAN SLVs - I was often successful in finding those individuals who were overall system designers from whom I was able to gain some of insight into their design thinking, and learn of the problems encountered in development and fabrication.

During this time period, I ran across several reports prepared by propulsion contractors who were under subcontract to Hughes on the SMICBM program (the subject of the April 10 column). It was my impression that there was a lack of interest in designing anything meaningful. Definitely there were no attempts made to originate engine designs that might appreciably reduce cost and increase simplicity, ruggedness and reliability. No follow-up work was conducted by the contractors; when the SMICBM program was terminated, contractor interest in minimum cost design ceased as well.

I do not plan to describe as much of the MCD/SLV as contained in The Aerospace Corporation report, "Proposed Minimum Cost Space Launch Vehicle System" by A. Schnitt and Col F.W. Kniss, July 1968, copies of which may be obtained by written request addressed to Aerospace. What I do plan to provide is how I applied the results of the parametric analysis, how I proceeded with its design, where I obtained much of my information and help, and my reasoning in reaching certain design decisions.

From rough-cut cost and stage weight analyses, I made an early decision to use two stages to LEO, thereby accepting a slightly higher gross weight than that of a 3-stage vehicle. Since the first stage would be rugged and the major cost segment of the SLV, I planned to recover it at sea. I knew I should pay about $6 a pound for the basic, Stage 1 hardware, such as, tankage, propellant ducting, and combustion chamber shell. The cost of Stage 2 basic hardware was about $30 a pound, but this value was considered flexible because multiple system interactions were influencing its optimum value.

My initial hardware decision was to choose liquid rocket propulsion. It was an easy choice over solids and hybrids. I considered propulsion to be the key subsystem because I felt that its design strongly influenced the selection of most other vehicle components and subsystems. In examining the weights of existing turbopump-fed liquid engine subsystems, I was surprised to learn that the advertised low weight-to-thrust ratios for minimum weight design liquid rocket engines did not exist in reality because of the relatively large amount of propellants trapped in the turbopump, in the propellant ducting, and in the piping, at stage burnout. This revelation prompted me to consider higher weight, simpler, pressure-fed systems as an initial choice.

I planned to use the same TITAN SLVs hypergolic propellants in both stages. It would certainly lower launch costs. Of more importance, the propellants are storable at ambient temperatures and are compatible with steel that I valued as a low-cost construction material. (Several years later, these propellants were classified as environmentally undesirable and became extremely expensive.)

The parametric analyses showed that, in order to realize minimum payload cost as well, a family of MCD/SLVs should be developed that has a wide range of payload weight capability. In order to minimize R&D costs, the vehicles should be scaleable to the fullest extent possible. This meant that the propulsion systems should be scaleable, but did not necessarily imply that upper stages of large vehicles would be used as first stages of small vehicles.

I had set aside the use of strap-on stages as being structurally inefficient and costly. This meant that the first stages of the larger vehicles would be too large to be manufactured in current aerospace industry production facilities. This restriction on production facilities was not considered detrimental because large first stages would have near-commercial construction.

I had learned that "shower-head" type of fuel injection engine configurations usually led to high development costs and time with each attempt at scaling to higher thrusts. The intricate, small, multiple fuel injection orifices were considered a "show-stopper" to designing for minimum cost. What was needed was an easily fabricated engine with large fuel injection orifices that is also scaleable, say, from several thousand to several million pounds of thrust. A small loss in efficiency might well be tolerated.

At the time I shared an office complex with the head of the liquid propulsion department in the TITAN office. I described my propulsion requirements to him. He identified the TRW LEMDE pressure-fed engine that was used in the Apollo program as the most likely candidate. He reasoned that it would probably lend itself to scaling to multi-million pound thrusts since the current 10,000 pound engine was capable of being throttled to 1,000 pounds. We contacted the designer of the LEMDE engine at TRW, and he concurred with our thinking. His extended cooperation and assistance was further stimulated by the fact that he was acquainted with the analytical work leading to MCD and was an ardent adherent. With the promise of a scaleable engine, I concluded that a single engine per stage would result in much simplification and weight savings.

Having established this keystone subsystem, I proceeded to define a conceptual design of the SLV. Because of the time constraint to produce a minimum, or a near-minimum cost design of appreciably lower cost than current vehicles, I did not attempt to conduct elaborate tradeoff analyses. Instead I selected what I judged might be the optimum subsystems, feeling that tradeoff analyses should best be performed as part of the R&D program.

A critical example of consequences of this design approach was the selection of a pressure-fed propellant subsystem. There was a small group of propulsion specialists at Aerospace who had performed sufficient work on the design of a much simplified turbopump (of slightly higher weight) to believe that a pressure-fed propellant system would be cost-optimum. From what I have read of developments under the current Air Force EELV program, a simplified turbopump of this description has been developed. The Aerospace group may have indeed been right 30+ years ago.

Another critical design decision was the propellant tank material. Steel was judged to be least costly provided it had sufficient tensile strength and the weld strength would not be less than that of the basic material. I called upon friends at the Battelle Memorial Institute to advise me. They suggested HY-140, a steel used by the Navy in the fabricating the hulls of advanced submarines. Indeed, abiding by the parametric analysis, a large first stage should more resemble a submarine than an aerospace-typical structure and be more compatible with sea recovery.

The tank thickness required to sustain the propellant pressurization was considered sufficient to permit simplifications that would result in major cost and weight savings. For instances, a common bulkhead was used between tanks, and the single engine was bolted directly to the bottom of a tank.

"Main Tank Injection" was selected as the subsystem for pressurizing the tanks. In this subsystem the fuel was injected into the oxidizer tank while the oxidizer was injected into the fuel tank. The hypergolic property of the propellants caused minor explosions that produced hot, pressurizing gases. Further developmental tests yielded inconsistent results, and was the project was dropped. In the work subsequently done by others in designing an MCD/SLV, other pressurization subsystems were proposed which were considered to be satisfactory, such as, a simplified hot gas generator.

With the contents of the Aerospace report, previously identified, available to industry and NASA during its preparation and certainly after its release, there was a flurry of activity by industry, the Air Force and NASA. Several industry activities were supported with internal funds. Other activities were supported by the Air Force and NASA. Many of these activities will be described in the next column.
 

Do you agree or disagree with the design decisions made in 1965? How do they relate to decisions being made for today's vehicles? [no discussions were submitted for this question]

Next Column: The MCD/SLV Continued

Return to Published Columns