Search Marshall


Text Size

Fact sheet number: FS-2000-09-174-MSFC
Release date: 08/00

Linear Aerospike Engine —
Propulsion for the X-33 Vehicle

Photo description: drawing of linear aerospike engine.

One of the key challenges in designing the next generation of launch vehicles is the development of an efficient propulsion system that is lightweight yet powerful enough to allow for flight. The propulsion system must also offer low-cost operations, improved reliability and short turnaround times.

NASA and its industry partner in the X-33 Advanced Technology Demonstrator program, Lockheed Martin Aeronautics Co., of Palmdale, Calif., have taken a 30-year-old idea the linear aerospike engine and updated it for the 21st century by incorporating new technologies and materials. The effort is managed by NASAs Marshall Space Flight Center in Huntsville, Ala., NASAs Lead Center for Space Transportation Systems Development and Center of Excellence in Propulsion.

The aerospike engine is being developed from groundwork laid in the 1960s and 1970s by the Rocketdyne Propulsion & Power unit of The Boeing Company in Canoga Park, Calif. Unlike conventional rocket engines, which feature a bell nozzle that constricts expanding gasses, the basic aerospike shape is that of a bell turned inside out and upside down. When the reconfigured bell is "unwrapped" and laid flat, it is called a linear aerospike.

The linear aerospike features a series of small combustion chambers along the unwrapped bell, also called the ramp, that shoot hot gases along the ramp's outside surface to produce thrust along the length of the ramp, hence the name "linear aerospike."

With the aerospike, the ramp serves as the inner wall of the virtual bell nozzle, while atmospheric pressure serves as the "invisible" outer wall. The combustion gasses race along the inner wall (the ramp) and the outer wall (atmospheric pressure) to produce thrust.

Efficient at all altitudes

The key to a conventional bell nozzle's level of performance is its width. At high pressure -- i.e. sea level -- the gasses are more tightly focused, so a bell nozzle with a narrow interior surface works best. At low pressure -- i.e. higher altitudes -- a wider interior works best as the gasses will expand farther.

For example, the initial stage of the Saturn rocket which carried the Apollo astronauts to the Moon featured a narrow nozzle to produce an ideal straight-edged column of exhaust at sea level. However, the command module which orbited the Moon featured a much wider bell nozzle better suited for controlling the combustion gasses in the vacuum of space.

Since the width of the bell nozzles can't change to match the atmospheric pressure as the rocket climbs, bell nozzles are normally designed to provide optimum performance at one certain altitude or pressure. This is called a "point design," and engineers accept the performance loss the nozzle will encounter at any altitude other than the one it was designed for.

The aerospike eliminates this loss of performance. Since the combustion gasses only are constrained on one side by a fixed surface -- the ramp -- and constrained on the other side by atmospheric pressure, the aerospike's plume can widen with the decreasing atmospheric pressure as the vehicle climbs, thus maintaining more efficient thrust throughout the vehicle's flight.

However, the aerospike engine's ramp is cut short to save weight, since the amount of thrust contributed by the end of the ramp is small compared to its weight. Designers compensate for this by pumping secondary exhaust from the aerospike's gas generator into the base region to add to the atmospheric pressure and elongate the wake.

A key feature of the linear aerospike is the manner in which the engine distributes the thrust along the X-33's long base. This helps distribute the structural loads and makes for a lighter vehicle since heavy structural reinforcement in a confined area isn't required.

The X-33's direction of flight will be controlled by varying the thrust side to side and engine to engine, rather than by adjusting the direction of a bell nozzle. The lack of systems for thrust vectoring -- such as gimbals, hydraulics and flex lines -- also will make the aerospike easier to maintain than conventional engines and help keep the X-33's weight down.

Lockheed Martin's industry partner for the engine, Boeing Rocketdyne, has built components of four liquid hydrogen and liquid oxygen-fueled, fuel-cooled linear aerospike engines with engineering support from NASA's Marshall Space Flight Center in Huntsville, Ala. The engines will be ground tested prior to installing two engines into the X-33 for flight test.

Marshall Center also will work with the industry team to develop the engine technologies required for a full-scale reusable launch vehicle (RLV), if NASA and the industry team decides to develop an RLV after the X-33 program is completed.

The X-33 will feature two aerospike engines, each independently supplying 20 of the 40 combustion chambers running along the base of the X-33 to produce a combined 412,000 pounds of thrust at sea level. The specific impulse is estimated at 339.9 seconds at sea level, 429.8 seconds in a vacuum.


NASA engineers at the Marshall Center have conducted a number of tests for the linear aerospike engine. In the spring of 1997, Marshall Center conducted tests of three hydrogen-cooled thrusters, or thrust cells, mounted side by side and attached to a 4-foot-long copper alloy ramp. The test series collected data on cell-to-cell plume interaction, cell-to-cell feed system interaction and heating. Marshall engineers followed the thrust cell tests with tests the same year on aerospike ignition and gas generator systems.

In 1997, NASA and Lockheed Martin tested a 5-percent scale model of the lifting body/aerospike configuration in the supersonic wind tunnels at the Air Force's Arnold Engineering Development Center near Tullahoma, Tenn. The wind-tunnel tests were aimed at characterizing the interaction between the aerospike engine exhaust and the lifting body's aerodynamic behavior.

Engine flight qualification testing began in 1998 at NASA's Stennis Space Center in Mississippi. By the spring of 2000, the first aerospike engine had been successfully operated at full power and exceeded the expected operating time that it will experience in test flights from California to Utah.

The first aerospike engine for the X-33 program successfully completed 14 planned hot fire tests in the spring of 2000, accumulating more than 1,460 seconds of total operating time. In addition, engineers have successfully demonstrated the aerospike engines ability to vary its thrust from side to side and top to bottom. This capability to vary its thrust -- called differential throttling -- will be used to control the X-33s direction of flight.

The test stand at Stennis was upgraded in the fall of 2000 to accommodate testing of two engines side-by-side, as they will be installed in the X-33. The dual-engine configuration is scheduled to begin testing in 2001.


The X-33 is being developed under a cooperative agreement between NASA and Lockheed Martin, which began July 2, 1996. It is a subscale technology demonstrator prototype of a RLV which Lockheed Martin has named "VentureStar," and which the company hopes to develop. Through demonstration flights and ground research, the X-33 will provide information needed to decide whether to proceed to the development of a full-scale, commercial RLV.

Lockheed Martin plans to conduct up to 15 flight tests of the X-33. These tests will originate from Edwards Air Force Base, Edwards, Calif., and fly to Michael Army Air Field at Dugway Proving Ground, Utah, and later to Malmstrom Air Force Base, Great Falls, Mont.