A debate has long raged between those who believe that reusable launch vehicles (RLV) are the only—or at least the best—way to fly to and from Earth orbit and those who emphasize the continuing place of expendable launch vehicles (ELV) in future space access operations. Advocates of RLVs claim that the model for a prosperous future in space is the airline industry, with its thousands of flights per year and its exceptionally safe and reliable operations. Several models exist for future RLVs, however, and all compete for the attention—and the development dollars—of the Federal government.
One especially important model for a next-generation RLV emerged during the Reagan administration when senior government officials began to talk about the “Orient Express,” a hybrid air and spaceplane that would enable ordinary people to travel between New York City and Tokyo in about one hour. Such a concept was quite simple in theory although enormously complex in reality. It required developing a passenger spaceplane with the capability to fly from an ordinary runway like a conventional jet. Flying supersonic it would reach an altitude of about 45,000 feet when the pilot would start scramjet engines, a more efficient, faster jet engine that has the potential to reach hypersonic speeds in the mach 3 realm. These take the vehicle to the edge of space for a flight to the opposite side of the globe, from whence the process is reversed and the vehicle lands like a conventional airplane. It never would reach orbit, but it would still fly in space, and the result is the same as orbital flight for passengers but for less time. It would even be possible, RLV supporters insisted, to build such a spaceplane that could reach orbit.
One of the most significant—and perhaps foolish—efforts to develop this reusable spaceplane was the National Aerospace Plane, a joint NASA/Air Force technology demonstrator begun during the Reagan administration. Like the Space Shuttle of the 1970s, this vehicle was touted as a single-stage-to-orbit (SSTO) fully reusable vehicle—one that could travel to and from Earth orbit without dropping stages once they had expended their fuel—using air breathing engines and wings. After billions spent, NASP never progressed to flight stage. It finally died a merciful death, trapped as it was in bureaucratic politics and seemingly endless technological difficulty, in 1994.
NASA began its own RLV program after the demise of NASP, and the agency’s leadership expressed high hopes for the X-33, a small suborbital vehicle that would demonstrate the technologies required for an operational SSTO launcher. This is the first of a projected set of four stages that NASA contends will lead to a routine space faring capability. The X-33 project, undertaken in partnership with Lockheed Martin, had an ambitious timetable to fly by 2001. But what would happen after its tests were completed remains unclear. Even assuming complete success in meeting its R&D objectives, the time and money necessary to build, test, and certify a full-scale operational follow-on version remains problematic. Who would pay for such an operational vehicle also remains a mystery, especially since the private sector has become less enamored with the joint project over the years and has eased itself away from the venture.
There is also an understanding that the technical hurdles have proven more daunting than anticipated, as was the case thirty years ago with the Space Shuttle and more recently with the NASP. Any SSTO, and X-33 holds true to this pattern, would require breakthroughs in a number of technologies, particularly in propulsion and materials. And when designers begin work on the full-scale SSTO, they may find that available technologies limit payload size so severely that the new vehicle provides little or no cost savings compared to old launchers. If this becomes the case, then everyone must understand that NASA will receive the same barbs from critics as had been seen with the shuttle. They condemned NASA for “selling” the Space Shuttle program as a practical and cost-effective means of routine access to space and then failing to deliver on that promise.
This is not to say that SSTO will never work, or that the X-33 should not have been pursued. It is NASA’s job to take risks and push the technological envelope. But in an effort as important as creating the next generation of launchers, technological risks must be recognized for what they are and accepted. While the goal of the RLV program has been to enable the development of a launch system that is significantly cheaper, more reliable, and more flexible than presently available it is possible to envision a future system that cannot meet those objectives. To accomplish this task, furthermore, the Federal government should be prepared to expend the resources necessary to drive R&D technology. Something approaching a $25 billion investment to achieve significant RLV progress is not out of the question. Such a public investment would go far toward bringing to fruition a next generation launch vehicle to replace the shuttle, supporters believe.
Then there is an alternative position that suggests that the most appropriate approach to space access is through the use of throwaway “big, dumb boosters” that are inexpensive to manufacture and operate. While reusable rockets may seem to be an attractive cost-saving alternative to expendables because they allow repeated use of critical components such as rocket motors and structural elements, ELV advocates claim, actually they offer a false promise of savings. This is because all RLV savings are predicated on maximizing usage of a small number of vehicles over a very long period of time for all types of space launch requirements. Accordingly, cost savings are realized only when an RLV flies many times over many years. That goal is unattainable, they claim, because it assumes that there will be no (or very few) accidents in the reusable fleet throughout its life span.
The reality, ELV advocates warn, is that the probability of all RLV components operating without catastrophic failure throughout the lifetime of the vehicle cannot be assumed to be 100 percent. Indeed, the launch reliability rate of even relatively “simple” ELVs—those without upper stages or spacecraft propulsion modules and with a significant operational experience—peaks at 98 percent with the Delta II and that took thirty years of operations to achieve. To be sure, most ELVs achieve a reliability rate of 90-92 percent, again only after a maturing of the system has taken place. The Space Shuttle, a partially reusable system, has attained a launch reliability rate of 99 percent, but only through extensive and costly redundant systems and safety checks. In the case of a new RLV, or a new ELV for that matter, a higher failure rate has to be assumed because of a lack of experience with the system. Moreover, RLV use doubles the time of exposure of the vehicle to failure, because it must also be recovered and be reusable after refurbishment. To counter this challenge, more and better reliability has to be built into the system and this must exponentially increase R&D and operational costs.
NASA’s goal for the next generation RLV’s operational reliability has been established at 99.9 percent. No launch vehicle has ever achieved that goal. And those who argue for ELV development note that attaining such a goal is virtually impossible. At the very least it will require a much larger fleet than currently envisioned and operational costs will rise to assure reliability through redundancy. Stephen A. Book, an engineer with the Aerospace Corporation, holds that “we can pay a large amount of money to build in very high reliability or we can pay a large amount of money to acquire several more vehicles.” There is no other option, he contends, adding that building more “use-once, throw-away” vehicles make the most sense for most future launches. Certainly for the placement of satellites into Earth orbit, Book concludes, expendable launchers are the most economical vehicle. They may never, however, be desirable for human space flight.
Designing for one-use only, those arguing for ELV development suggest, simplifies the system enormously. One use of a rocket motor, guidance system, and the like means that it only needs to function correctly one time. Acceptance of an operational reliability of 90 percent or even less would further reduce the costs incurred in designing and developing a new ELV. Indeed, many experts believe that reliability rates cannot be advanced more than other 1.5 percent above the 90 percent mark without enormous effort, effort that would be strikingly cost inefficient.
Some provocatively suggest that new ELVs should be designed with the types of payloads to be carried clearly in mind, accepting the risk inherent in any space launch environment where 90 percent reliability would be the norm. For expensive and one-of-a-kind scientific and military satellites, as well as for expensive commercial spacecraft, vehicles with a reliability rate and a higher price tag could be acceptable. But for most payloads, especially logistics supplies and the like going to the International Space Station, reliabilities as low as 65 percent might be acceptable. And, it goes without saying that for human space flights NASA’s goal of 99.9 percent operational reliability is not too high a goal to seek. Picking the most promising concepts to support, however, will be a task not without difficulties.
What path should we pursue into the future?