Commentary | Reinventing Space: Dramatically Reducing Space Mission Cost — Reducing Launch Cost
This is the 10th in a series of articles on how to go about dramatically reducing space mission cost while maintaining a high level of mission utility. Central to the process of reducing mission cost is finding ways to dramatically reduce the cost of launch, particularly for small satellites. While launch is typically not the highest cost element of a space mission, it drives the other costs. So long as it costs on the order of $20,000 per kilogram to put stuff into orbit, the cost per kilogram of spacecraft will remain high. It is difficult to justify building spacecraft for “only” a few million dollars if the minimum cost for a dedicated launch to orbit is $30 million or more.
Alternatives to a Dedicated Launch to Orbit
The single most effective approach to reducing both cost and schedule is to not launch to orbit at all. Depending on the goals of the experiment, test or mission, there are multiple alternatives to a dedicated orbital launch.
Balloon flights can provide hours or days at high altitude at very low cost. If zero gravity is important, drop towers and drop tubes can provide excellent conditions for 5 to 10 seconds if you drop a payload of up to 1,000 kilograms inside a vacuum tube. (Drop towers for component testing are available at NASA Glenn Research Center in the United States and at the Center of Applied Space Technology and Microgravity, or ZARM, in Germany.) The data and payload are available essentially immediately and the experiment can typically be repeated twice per day.
Periods of zero gravity up to about 20 to 25 seconds (and even longer periods of lunar gravity or Mars gravity) are available from aircraft parabolic flights. Up to 40 parabolas a day can be flown, but perhaps the major benefit is that the experimenter and a few others can fly along, watch what happens and make adjustments and corrections in real time or over the course of several days. Originally flown only by NASA, parabolic flights are now commercially available from Zero G Corp. All of the above options are several orders of magnitude faster and lower cost than a dedicated launch to orbit.
The next step up from parabolic flights are suborbital flights on sounding rockets. These can provide up to 12 minutes of excellent zero gravity at an altitude of up to 1,200 kilometers. This means you can get to low Earth orbit (LEO) altitudes and above with vacuum and a full view of the Earth and space, just as you would in LEO. The only thing missing is the orbital velocity and a large chunk of the price tag.
For going all the way to obit at lower cost for small payloads, the principal options are rides as secondary payloads or shared launches. The ASAP (Ariane Structure for Auxiliary Payload) ring on the Ariane 5 provides accommodations for up to six payloads of 100 kilograms each and multiple slots can be used. The ASAP ring has been in use for many years and has provided a ride to orbit for many low-cost satellites. More recently, the ESPA (EELV Secondary Payload Adapter) ring has been developed to provide similar services for the Atlas and Sharing the launch on a variety of vehicles is also possible, but of course it requires coordination among the various payloads and where they want to go.vehicles.
Depending on the specific mission needs, there are quite a few alternatives to a dedicated launch to orbit. Of course, each approach has both strengths and limitations, but all of them can provide potentially large reductions in both cost and schedule.
For larger spacecraft there are fewer options for reducing cost and schedule, although the use of some of the alternatives above for testing elements of the system may be able to find problems early in the program and therefore avoid more expensive fixes later.
Design for Multiple Launch Vehicles
Perhaps the best option for reducing both cost and schedule for a dedicated launch to orbit, or at least for helping to prevent overruns, is to design the spacecraft for multiple launch vehicles. The cost of launch is typically negotiated between whoever is buying the launch and the launch provider. Clearly, there is more potential for negotiation if more than one launch provider is possible. Designing for multiple launch vehicles is usually not hard or expensive because the payload environments of all of the launch vehicles are typically similar, except for the Minotaur, which provides up to 13 g’s of axial acceleration because it is made from decommissioned ICBMs for which the loads were not a principal design consideration.
An equally important reason for designing for multiple launch vehicles is to protect the schedule. Recall that launch systems have approximately a 90 percent success rate. When a launch failure occurs, there is a significant downtime until the next launch of that system. In addition, if your payload was the next in line at the time of the failure, it may get moved further back by higher-priority launches when the launch system resumes operations. For this reason, nearly all of the constellation builders use multiple launch providers; this also provides a continuing negotiating position. Thus, if an organization needs to launch a constellation of 50 satellites, it may choose Launch Provider A for 15, Launch Provider B for 15 and then reserve the last 20, depending on the performance of the first set of launches. Note that constellations may also use launch vehicles of different sizes by launching multiple satellites on a larger launcher. This can work out well or badly.launched its entire constellation without a launch failure. Unfortunately, in 1998, lost 12 satellites on a single Zenit-2 launch failure.
Develop a Small Dedicated Launch Vehicle
A key to reducing both cost and schedule for systems of all sizes is the development of a low-cost, small, responsive launch vehicle. This is needed for both operational smallsats and for rapid testing of both technology and processes applicable to larger systems. It also provides for the rapid introduction of new technology, which is evolving particularly quickly in small spacecraft. Generally, developing a small launcher for a particular satellite system is regarded as far too much risk and cost for space programs. However, for many small launcher concepts the nonrecurring development can be recovered in savings from one or two launches, making this an extremely attractive option if even a small constellation is needed. Several organizations are currently in the process of planning for or building small launchers, so it is possible that small launch will become a competitive market, which should drive costs down even further.
Build to Inventory with Launch on Demand
Building launch vehicles to inventory, as needed for launch on demand, can significantly reduce cost and schedule and increase utility by allowing spacecraft to be launched when they are ready, rather than when the launch vehicle is ready. Today, it is not uncommon for spacecraft to wait one, two or more years for a launch. This increases costs by leaving the spacecraft dollars sitting on a shelf, and also reduces the return on investment by not getting results as rapidly as would be desirable. And of course immediate launch in response to man-made or natural disasters or destruction of on-orbit assets (due to debris collisions or anti-satellite activities, for example) is effectively impossible without launch on demand. This is a capability that the Russians and former Soviet Union have had for over 30 years. During the 73-day Falklands War in 1982, the Soviets launched 29 payloads into orbit, most in direct response to the war. Launch on demand also prevents us from having to cover all of the world, all of the time, with all possible sensors needed to collect the data we need.
For small launchers, that cost less than $5 million, build to inventory and launch on demand are not particularly expensive and are primarily an issue of whether it is worth the interest cost on the money required to build the vehicle for the time period from when it is completed until it is launched. Thus, at 10 percent interest, holding the vehicle in inventory for six months would increase the build cost by only 5 percent and the total launch cost by less than that, say 4 percent, plus an incremental cost for storage and maintenance. In this case, a relatively small investment could lead to very large cost savings for space missions as a whole.
Design of Low-cost Launch Vehicles
The design and development of launch systems is beyond the scope of this summary. However, John London’s 1994 book “LEO on the Cheap: Methods for Achieving Drastic Reductions in Space Launch Cost” is available for free online and, even though somewhat outdated, provides an excellent overview of why launch systems cost as much as they do and ways to reduce launch system cost. A number of summary papers are available on the Reinventing Space website, including a cost model that is intended to compare reusable vs. expendable launch vehicles, which has been updated to model the added cost of launch on demand systems.
The 11th article in this series will look at approaches to reducing the cost of communications, ground systems and operations.
James R. Wertz is president of Microcosm Inc. He is co-author of “Reducing Space Mission Cost,” published in 1996, and has been teaching a graduate course at the University of Southern California on that topic since then. If you have questions, comments or suggestions, or simply want to discuss these issues, he can be reached at firstname.lastname@example.org. Information on the joint Microcosm/USC Reinventing Space Project can be found at www.smad.com/reinventingspace.html.