Propellant depots: the fiscally responsible and feasible alternative to SLS
by Andrew Gasser
|We must ask ourselves, “is there a better way to perform HSF in America?” We contend there is a better way that is not only fiscally responsible, but also relies heavily upon the free market.|
Many inside NASA have quietly pushed for HSF exploration utilizing an architecture of our current roster of Evolved Expendable Launch Vehicles (EELVs) and propellant depots. In fact, former NASA administrator Mike Griffin called propellant depots “the holy grail of deep-space exploration” in testimony before the House Science, Space, and Technology Committee last month during a hearing about future human space flight plans. While many NASA insiders and pundits have claimed that there were studies that showed propellant depots were scientifically possible and a better choice economically, NASA leadership has claimed that they have not seen such evidence. If NASA leadership has not seen the studies on propellant depots, then who is burying the studies like the one that was leaked to NASA Watch on October 12th? We have known about this study for some time and believe that propellant depots are not only scientifically feasible but also that such an approach is the fiscally responsible option for America to regain its lead in HSF exploration.
Griffin, though, is skeptical about using propellant depots, based on last month’s testimony. On heavy-lift, he stated, “Because of economies of scale inherent to the design of launch vehicles, the cost-per-pound of payload to orbit nearly always improves with increasing launch vehicle size. Thus, a heavy-lift vehicle should be designed to be as large as possible within the constraints of the facilities and infrastructure available to build and transport it. This provides the greatest marginal capability at the lowest marginal cost.” While his statement is true based on price trends for Atlas V, Delta IV, and Falcon launch vehicles, the largest vehicles do not provide the lowest cost-per-pound to orbit. Two launch vehicles defined by NASA’s Human Exploration Framework Team (HEFT) had an estimated cost-per-pound to orbit between $8,000 and $11,000 for payloads of 154,000 and 220,000 pounds (70,000 and 100,000 kilograms); the Atlas V and Delta IV families’ price-per-pound to orbit ranges between $5,000 and $9,000 for launch capacity between 22,000 and 55,000 pounds (10,000 and 25,000 kilograms); and posted launch prices Space Exploration Technologies’ (SpaceX) Falcon launch vehicles are $1,000 and $3,000 per pound for 23,000 and 117,000 pounds (10,400 and 53,000 kilograms) payload capacity. Atlas V, Delta IV, and Falcon 9 vehicles are currently in operation while the Falcon Heavy is in development; the two HEFT vehicles, by contrast, are purely conceptual. Estimates for Ares 5 launch cost ranged from $540 million to $1.5 billion to place 280,000 pounds (127,000 kilograms) in LEO, or $2,000 to $5,400 per pound.
When discussing fuel depots and launch cost, Mike Griffin goes on to say, “Ares 5 offers the lowest cost-per-pound for payload to orbit of any presently known launch vehicle design.” Clearly the values above do not support his claim as existing launch vehicles have costs-per-pound between $3,000 and $9,000 while the conceptual Ares 5 and HEFT launch vehicles have estimated costs between $2,000 and $11,000 per pound to orbit.
In 2005, shortly after the Exploration Systems Architecture Study was released, Griffin, speaking at an American Astronautical Society meeting in Houston, said that on-orbit propellant was worth $10,000 per kilogram ($4,550 per pound) to NASA. This value is close to the Delta IV Heavy cost per pound to orbit and, one can infer, the expected cost per pound to orbit for Ares 5. At the same time posted launch prices for SpaceX Falcon vehicles were approximately $4,000 per kilogram ($1,814 per pound), significantly less than the stated value of on-orbit fuel.
|Expending billions of dollars and human resources for infrastructure will limit the exploration and settlement capabilities of the United States.|
Estimated launch costs for NASA’s Ares 5 and HEFT vehicles are based on two to four missions to the Moon per year. Atlas V, Delta IV, and Falcon 9 costs reflect launch rates between three (Delta IV) and eight (Falcon 9) per year. Using existing rockets for exploration missions with the two vehicle masses each less than 25,000 kilograms and with over 150,000 kilograms of fuel required per mission would add at least 16 flights to the current manifest for two missions per year. Increasing the launch rate of existing systems can significantly reduce launch cost, especially as rates grow from 1 to 50 per year as shown by Gstattenbauer, Franke, and Livingston in their paper, “Cost Comparison of Expendable, Hybrid, and Reusable Launch Vehicles,” AIAA-2006-7211-801, presented at the AIAA Space 2006 Conference in San Jose, California.
Griffin’s testimony also included the following statement: “Further, a fuel depot requires a presently non-existent technology – the ability to maintain cryogenic fuels in the necessary thermodynamic state for very long periods in space.” While it is true the required technologies have not yet been flown in space, they have been developed, ground-tested, and are ready for space flight tests. In addition, their performance characteristics have been incorporated into upper stage and depot design concepts indicating cryogenic propellants can be maintained in the appropriate thermodynamic state for over a year with zero oxygen boil-off and less than 0.05% per day boil-off of the initial hydrogen mass. By incorporating cryocoolers (cryogenic refrigeration units), it may be possible to eliminate hydrogen boil-off as well. To mature these technologies and make them available for exploration missions, NASA is currently funding four contractors to define an appropriate cryogenic propellant storage and transfer technology demonstration mission with a target launch date in 2016.
While not addressed by Griffin in his testimony, one typically hears that including fuel depots in exploration missions significantly decreases the probability of mission success since, as the logic goes, the more launches one needs to conduct a mission the less likely it is they will all be successful. Typical launch reliability is around 98%. Therefore, if a mission requires one launch, the probability of having a successful launch is 0.98. If a mission requires 5 launches, then the probability that all five launches will be successful is 0.90 (0.98 x 0.98 x 0.98 x 0.98 x 0.98); for 12 launches, the probability of successfully launching all payloads would be 0.785. But, if fuel transfer is used to conduct that same mission using two hardware launches and 10 fuel launches, the probability of all launches being successful would be 0.71, if and only if, the hardware is launched first followed by 10 sequential launches transferring fuel directly from the fuel tanker into the departure stage. This assumes all launches have a probability of success of 0.98 and the probability of successful fuel transfer is 0.99.
On the other hand, with a depot, fuel can be launched first followed by mission hardware launch, propellant transfer, and departure. This changes the probability of mission success to the probability of successful hardware launch times the probability of having sufficient fuel on hand times the probability of successfully transferring fuel from the depot to the departure stage. If the probability of successful transfer (docking, transferring, undocking) is 0.99 and only 10 launches are scheduled to provide the required 10 fuel loads, the probability of mission success is 0.701 (0.98 x 0.98 x (0.98 x 0.99)10 x 0.99). However, the probability getting all mission hardware loaded with required propellant can approach 0.95 (0.98 x 0.98 x 0.99 x 0.99996) where 0.99996 is the probability of having 10 or more successful propellant launches and transfers into the depot out of 14 scheduled.
|Depots allow NASA to use existing EELV integration and launch infrastructure that is considerably less expensive. This is a win/win/win scenario that everyone should get behind.|
The NASA Ares 5 and HEFT 70 t and 100 t launch vehicle concepts have a higher estimated cost per pound to LEO than existing launch vehicles. Adding additional capability to existing launch vehicles may provide lower marginal cost per increased capability but it does necessarily mean large launch vehicles offer the lowest cost per pound to orbit. Technologies for storing and transferring cryogenic propellants in a low Earth orbit fuel depot exist, have been developed and tested in the lab, and are ready for a planned demonstration space mission in 2016. Conducting exploration missions with fuel depots and smaller launch vehicles offer similar probabilities of launch success as using two or three large launch vehicles and can drive down launch costs as rates increase as well as offering increased flexibility and lower cost per launch failure.
NASA cannot accurately predict the future flight rate of the SLS or its infrastructure cost. Currently there are only two planned flights for SLS, one in December 2017 and one in 2021. SLS will use many of the now-inefficient system processes used for shuttle. This is not a personal attack on the brilliant men and women who supported shuttle. This is just simply pointing out the bureaucracy that has not allowed NASA to innovate processes and systems in operations in the same fashion as the private sector.
It is fair to say that SLS will have similar costs to shuttle for infrastructure. If we believe Griffin’s testimony, the shuttle infrastructure costs $2.3 billion a year for the first launch and $300 million for each subsequent launch. This economic model is a fiscally irresponsible position to take in our current political climate. Expending billions of dollars and human resources for infrastructure will limit the exploration and settlement capabilities of the United States.
Depots allow NASA to use existing EELV integration and launch infrastructure that is considerably less expensive. This savings could then be passed on to do other things, like develop a lunar lander or a nuclear propulsion module, both of which are unfunded and off the radar with the SLS option. This is a win/win/win scenario that everyone should get behind.
NASA wins: They get a modular architecture that allows customization. However, NASA should not fall into the paradigm that depots can only be used for HSF. Depots can be used for science missions as well. NASA can also divert more money to programs that are subjected to SLS pressures. At the September hearing everyone acknowledged that our commercial crew partners would be the first to return Americans to space. At that same hearing former astronaut Gene Cernan stated that America would need nuclear propulsion to go to Mars. Propellant depots allow for both the commercial crew and technology budgets to be bumped up.
The free market wins: By using depots, the free market will also invest their own resources to help develop the technology. Through mechanisms like Space Act Agreements with milestone payments, NASA can have that limited government oversight that everyone would agree we need. Different companies will have different approaches to achieving the requirements set by NASA. Corporations will put their “own skin in the game”. This will drive costs down with multiple options available. Free market competition will spur further innovation that will only benefit the American space program.
The American taxpayer wins: They are wary of government failure. NASA simply cannot afford another Constellation or James Webb Space Telescope. The reasons for program costs overruns and schedule slips is inconsequential. They have happened and we need to recognize and learn from these mistakes. Everyone, from the administrator on down to the engineers in the trenches, needs to understand that we will not get another chance at this. The taxpayers are watching their government more than ever. As a bonus, the many launches required for propellant depots would not only help drive down EELV costs but also provide a reason for vacationers to visit the Space Coast.
The information presented here proves that the propellant depot architecture is a viable alternative to the Space Launch System. Just as importantly, the propellant depot strategy fits within the country’s need for programs that are in sound monetary policy. NASA needs a strategy that NASA leaders and employees can back in private, as well as in public.