BLAST FROM THE PAST; a few good ideas may return to the light of day…

Rockwell International Star-Raker proposal

King of the Wings into space concepts.
by Kelly Starks


Figure 1: Rockwell’s Star-Raker in comparison with a Boeing 747

If you’re going to talk about low cost access to space, and winged Horizontal Takeoff-Horizontal Landing (HTHL), Single-Stage to Orbit (SSTO) vehicles, you need to talk about Rockwell’s Star-Raker, proposed to the Department of Energy (DOE) in the late 1970’s, in response to the DOE study of the Space Solar-Powered Satellite (SSPS) concept. Star-Raker offered a massive change in capacity and price from what had been considered, and turned the whole SSPS concept on its ear – which infuriated some of the advocacy groups for the concept.



Figure 2: Space Solar-Powered Satellites (SSPS)

SSPS was a very popular concept among space advocacy groups in the 1970’s, involving building huge solar collector farms in orbit. Aside from the perceived benefits of solar power arrays placed in orbit (no weather or day/night cycles, and more intense sunlight means a daily average of roughly twenty times as much power gathered per collector) and the huge interest in a non-oil-based power supply during the oil crisis era (when people were being assured by President Jimmy Carter that all oil and gas supplies in the world would be exhausted by the 1990’s), the vast construction effort to build them was seen by space advocates (most especially the L-5 Society) as “the key” to founding major industrial colonies in space. Since the calculated margin cost per pound to orbit with Shuttles (the assumed lowest possible launch technology possible in the day) was roughly $200+ (in ‘70’s dollars), and the target 300 solar power platforms would weigh 10,000 tons each, it would be utterly unaffordable to build these platforms from three million tons of components shipped up to orbit, for $1.2 trillion in 70’s dollars. So space advocates assumed you’d need to colonize space and build with resources in space. However, they had completely misjudged the nature of launch costs.

The specific SSPS proposals varied, but a common assumption was a fleet of three hundred, 10,000-ton SSPS platforms in orbit. This three million ton lift requirement was clearly vastly beyond the capability of any launch system available or in development. It would, for example, require 100,000 flights of the space shuttle which, given the existing capacities of the launch pads and best case assumptions of the shuttles, could take centuries to do. Ignoring that, the space shuttles’ margin cost of roughly $200+ per pound to orbit would mean over $1.2 trillion in launch costs in late 1970’s dollars to lift everything. This was clearly infeasible. (Note: the Space Colonization efforts required launch rates and capacities well beyond the capacity of the shuttle systems as well.)

We need a bigger launch capacity

Between 1978 and 1986, the U.S. Congress authorized the Department of Energy (DoE) and NASA to jointly investigate the SSPS concept. All of the resulting designs by these organizations and advocates required masses to orbit far beyond the capacity of any launcher systems that were operational or in development; but the requirements weren’t beyond the capacity of launcher systems that could be developed, or were being researched. Also, launch costs are largely driven by economies of scale, or rather the total lack of them, in launch markets then, and now. The very scale of tonnage the SSPS programs would need to launch into orbit — estimates were at least a thousand tons per day, half the total in human history to date, and roughly that of the entire 30 year shuttle program — would unavoidably drive costs far down. So, clearly, heavy or super heavy lift capacity craft capable of extremely high flight rates were needed.

Launch vehicle manufacturers were invited to submit suitable design concepts. Three of the baseline concepts used for the SSPS studies were:

  • Boeing’s “Reusable Aerodynamic Space Vehicle” (RASV), an all-rocket HTHL SSTO winged craft launched at high speed from a magnetic levitation trackway, boosted to orbit with rockets fueled solely from internal fuel tanks, and glided back to a runway like a shuttle orbiter.
  • Boeing’s rocket-powered VTVL TSTO (400 ton cargo capacity) configuration
  • Rockwell International’s “Star-Raker” Turbo-ramjet/rocket HTHL SSTO (100 ton cargo capacity).


Figure 3:  North American Rockwell Star-Raker in Orbit

The Rockwell Star-Raker was given some preference by the Department of Energy since it seemed better suited for high flight rates. Offering 100 tons of cargo per flight in a 20 x 20 x 141.5 ft cargo bay, a fleet of 22 Star-Rakers was considered quite capable of lifting the target 1,600 tons per day,  with a projected cost to orbit in 1978 dollars of $22-$33 per kilogram, or $10-$15 per pound – ($36 to $55 a pound in 2014 dollars).

Rockwell had been researching the Star-Raker design with Marshall Space Flight Center since the 1960’s, and by the late 1970’s were confident that new materials technologies, and a light pressure-stiffened wet-wing design (similar to the pressure-stiffened Atlas booster used to carry Mercury flights into orbit), would make HTHL SSTO possible. The lower wing loading of the design would make surface temperatures during re-entry several hundred degrees lower than the Space Shuttle. The turbo-ramjet engines would allow the Star-Raker to carry double the payload than Boeing’s all-rocket-based horizontal take off Reusable Aerodynamic Space Vehicle concept (Figure 4), with the same gross liftoff mass for both craft, although the Star-Raker’s dry mass would be 45% higher than the Boeing design and the vehicle would be exposed to a more severe aerodynamic heating environment.


Figure 4: Boeing’s Reusable Aerodynamic Space Vehicle (RASV)

The Star-Raker team was not only confident they could build it, but expected each craft could sustain a rate of up to three flights per day. With this, a reasonably sized fleet could lift the target 1,600 tons per day, from a fairly normal, airport-like facility. The low capital costs of the facility and a small fleet of craft, and low maintenance cost per flight led to the estimate of $10-$15 per pound to orbit. This is perhaps twenty times less than the $220 margin cost per pound projected for the space shuttles, or roughly a thousand times less than the total cost per pound to orbit demonstrated by the shuttle fleet. Launching SSPS from Earth in kit form would thus be more economical than colonizing space to construct them (as outlined in the 1975 study). Star-Raker showed that aircraft-like operations could deliver costs to orbit seven to ten times the air freight cost per pound from the US to Australia, and presumably offer similar cost factors per passenger to orbit, even at the comparatively small flight rate of thousands of flights for the Star-Raker fleet, versus tens of millions of flights for a fleet airliners such as the Boeing 747.


Figure 5: Star-Raker coming in for a landing.  Note ten running jet engines, and three inactive main rockets and two orbital maneuvering rockets at base of tail.
Operational Comparisons


Figure 6: Star-Raker ground ops at commercial airport, with a second Star-Raker taking off above.


Figure 7: Loading and unloading of 3 Star-Rakers.  Star-Rakers were expected to fly to commercial airports on their jet engines alone, to be loaded with their cargo.  They would then fly to a spaceport to be refueled, and loaded with Liquid Oxygen to boost themselves into orbit with the cargo.  Note swing open nose for cargo loading/unloading.

Rockwell studied the operational issues and requirements for launching 1600 tons of payload into low Earth orbit per day to support the construction of the referenced solar power satellite fleet. They specifically compared their Star-Raker turbofan/air-turbo-/exchanger/ramjet HTHL SSTO with 100 ton payload capacity to Boeing’s rocket-powered, vertical-takeoff, vertical-landing (VTVL) two-stage to orbit (TSTO), with 400 ton payload capability. (Note: the payload assumed for the Star-Rakers varies between various studies.)

Boeing’s two-stage HLLV

Figure 8: Boeing’s two-stage HLLV

Boeing’s VTVL TSTO vehicle would require ten launch pads, requiring extensive refurbishment between missions to meet the launch rate requirement of four flights per day from the Kennedy Space Center. Two new, high-bay Vertical Assembly Buildings (VAB) would also be required as opposed to two aircraft maintenance-type hangars for the Star-Raker. The Boeing VTVL TSTO would need 5.5 days to recover from the ocean, where it landed, and to restack two extremely heavy stages in the VAB. This assumes there was no recovery damage, which was considerably more likely for ocean landings than for runway landings. So although the air-breathing Star-Raker concept would require some advanced technologies, it appeared to be better suited for high flights rates (16/day) than the vertically-launched TSTO.

For the Star-Raker, a single-runway air base would support an entire fleet of thirty craft. By comparison, the VTVL TSTO’s launch range would have to be 850 square kilometers in area to accommodate a fleet of 22 vehicles and the launch noise they generated (120 decibels at 13km versus <120 decibels at 1km for the Star-Raker).

In comparison to Boeing’s winged horizontal take off, all-rocket based RASV (Reusable Aerodynamic Space Vehicle) HTHL SSTO concept mentioned above (Figure 4 above).  The RASV would require a special, very long, magnetic levitation runway to take off from, and another normal runway for landing.  There was no way it could fly in and out of normal airports to pick up their cargo, and then fly to the space port to boost into space, like the Star-Rakers.  Again, the turbo-ramjet engines would allow the Star-Rakers to carry double the payload of Boeing’s, with the same gross liftoff mass for both ships, and would eliminate the need for the magnetic levitation launcher track the RASV needed. Even assuming the RASVs could fly as often per day, it would require a fleet twice as large, with twice the capital and maintenance costs to keep the fleet running.

The Star-Raker was to be compatible with C-5A Galaxy cargo handling facilities and airports, with 2440-4270 meter runways. Indeed, one operational concept had the Star-Rakers flying airports near the cargo suppliers to be loaded. After the cargo was loaded at traditional airports, the ship would then fly to the launch port to be lifted onto its take off cradle, fully fueled with liquid oxygen and liquid hydrogen, and would boost to orbit with no further cargo handling. This makes Star-Raker far more flexible than either Boeing’s VTVL two-stage craft or its magnetic levitation-launched HTHL RASV, at lower cost.

Star-Raker design features

Gross mass: 2,278,800 kg 5,023,800 lb
Payload: 100,000 kg 220,000 lb
Length: 94.50 m 310.00 ft
Span: 110.00 m 360.00 ft
Thrust: 20,480.00 kN 4,604,080 lbf
Apogee: 556 km 345 mi

Star-Raker system design features

Figure 9: Details of Star-Raker wing exterior and interior structure, engine details, and general specifications.

Star-Raker Design Features underside

Figure 10: Star-Raker Design Features[1]

[1] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1 -Space Transportation”. November 1981. Page 40 (1-17).

Star_Raker inboard Profile

Star-Raker Vehicle Section results

Figure 11: Vehicle profiles and sectional cutaways[2]

[2] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1 -Space Transportation.” November 1981. Page 41 (1-18).

The Star-Raker design used ten, 140,000 pound-force turbo-ramjet jet engines to power the craft to Mach 7.2, with a takeoff speed of 225 knots up from a 14,000-foot runway. This eliminated over half the weight of fuel and LOx a pure rocket craft would need to reach the same speed. Three 1.06 million lbf LOx/LH shuttle SSME-type engine rockets kick in at Mach 6, take over completely by Mach 7.3, and continue from there up to 300 mile high orbits.

Upon reaching orbit, the whole nose of the Star-Raker would swing to the side to remove cargo. Reentering, the low wing-loading on the now lightly loaded craft would mean the surface temperatures of the skin would be manageable. Increased ascent temperatures while transporting cargo, would be absorbed by the cryogenic fuel.

Star-Raker Isotherms

Figure 12: Figure showing assent temperature load on the underside of the Star-Raker and NASA shuttle. [3]

[3] Independent Research and Development Data Sheet – Earth-to-LEO Transportation System for SPS. Project Number 243, Fy 1979.


 Figure 13:  SSTO Launch Trajectory[4]

[4] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1 -Space Transportation.” November 1981. Page 45 (1-22).

The most unusual feature of the Star-Raker design was the pressure-stiffened wings. Normally wings of this size, stiff enough to lift such a weight off a runway at 225 mph, would make the craft too heavy to reach orbit. But by allowing the boil-off gas to build up pressure in the wings, the wing pressure-induced tension stiffened them like the hull of the early Atlas missiles. Without the load of cryogenic fuel in the wings, the wings would be several times weaker – but the craft would be several times lighter, and could fly as a normal (if fast) jet aircraft.

Another unusual feature of the Star-Raker was a parachute-dropped takeoff cradle: effectively a heavy landing gear cradle capable of supporting the Star-Raker fully loaded with fuel and LOx for a flight to orbit. In this way, the lighter landing gear would be fully capable of handling the Star-Raker after its fuel/Oxygen load was consumed, and for normal flight operations. The heavy landing gear would be dropped after take-off, so its weight wouldn’t need to be lifted to orbit.

Star-Raker Multi-cycle Turbofan-turbo-ramjet and inlets

Figure 14: Multi-cycle Turbofan/turbo-ramjet and inlets.  These provide thrust [5]

[1] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1—Space Transportation” November 1981. Page 42 (1-19).   ( )

The Star-Raker’s proposed multi-cycle air-breathing engine system was derived from the General Electric CJ805 aircraft engine, the Pratt and Whitney SWAT-201 supersonic wraparound turbofan/ramjet engine, the Aerojet Air Turbo-rocket, Marquardt’s variable plug-nozzle, ramjet engine technology, and Rocketdyne’s tubular-cooled, rocket engine technology.

Star-Raker mass table.

Figure 15: The weight breakdown of the Star-Raker hull, systems, cargo, and fuel/LOx load, in metric tons.


How does the design look today?

A study done by NASA in late 1981, referencing these three designs, was expecting much lower cost to orbit numbers than folks of the time expected:

“…The workshop decided that, although rather advanced technology and well developed operational management would be required, it was proper to target the average cost of gross cargo payloads into LEO [Low-Earth Orbit] at $30 [1979]/kg for construction of the initial SPS [Solar Power Satellite]. The further cost goal for repetitive construction of 30 to 60 SPS would need to be reduced to $15 [1979]/ kg for all operational payloads for ESLEO [earth surface to low earth orbit] and would require the use of advanced, long-lived vehicles with a sophisticated operational organization”[6] (emphasis added).

“ evolutionary series of heavy-lift and personnel-launch vehicles with chemical rocket propulsion can be targeted realistically to move heavy masses into LEO for $30 [1979]/kg by the year 2000. More advanced propulsion technology and vehicles may make $15 [1979]/kg a goal in the foreseeable future[7] (emphasis added.)

[6] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1—Space Transportation.” November 1981. Page 138.

[7] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1—Space Transportation.” November 1981. Page 248.

So looking back on it now, how realistic were those numbers?  Or more importantly, what could we do now with more advanced modern technology?  If it was proper in 1981 to expect to be able to lift “gross cargo payloads into LEO at $30 1979)/kg for construction of the initial SPS” and down to $15 (1979)/ kg” for a more advanced, mature, and larger scale operation ($96 and $48 per kg or $43-$22 per pound in 2015 dollars.), what could we do now?

In retrospect, hydrocarbon Mach 7 turbo-ramjets, and hydrocarbon<> LOx rocket engines would be much lighter, possibly half the weigh as expected for the Star-Raker, and allow a lower dry weight craft by eliminating most of the bulk and weight of liquid hydrogen tanks. Though the heavier fuel would make the takeoff weight higher (requiring either more efficient wings, or faster takeoff speeds), the dry weight and costs and operational complexity should be less. In general, modern systems are much more reliable and lighter weight than those of almost forty years ago. Similarly, modern materials (metals, composites, fiber-reinforced metals, etc.) would greatly lower the weight of the airframe and hull by perhaps by 30%. Ultra-high toughness ceramic composite (UHTCC) ceramic composite leading edges could not only be sharper and more aerodynamically efficient, they could offload heat that would otherwise spread out over the wings. Similar panels could also greatly lower the weight of other thermal protection system panels, further lowering the dry weight of the craft. At the least, a hydrocarbon-fueled Star-Raker of a similar size could have a 30% lower dry weight than the original design for the same cargo capacity, and the hydrocarbon engines can be built out of off the shelf parts. In short, it would be easier to do now.

I was involved in a project to commercially field a smaller craft than the Star-Raker with similar engines, though fueled with conventional jet fuel and liquid oxygen, and using more modern composites. Rather than costing $36 to $55 per pound to orbit in 2014 dollars, we were calculating more like $15-$20 per pound.

The problem of the need for heavy landing gear for a fully fueled/loaded Star-Raker was looked at by the British company Reaction Engines Limited’s Skylon team, but rather than assuming the need for a heavy takeoff cradle, they developed ways to dramatically lower the weight of the landing gear. The weight of normal landing gears is driven by the very large tires and wheels needed to distribute the weight over normal runways, and the heavy weight of uncooled brake disks. Specifying a super-hard runway for launches to orbit eliminated the heavy wheels and tires. A water-cooled braking system would allow smaller, lighter brakes still capable of handling emergency take-off abort loads. The water would add considerable weight, but could be dumped after the craft has taken off. With a similar landing gear driving the weight of a takeoff- capable landing gear down to 1.5% of the gross takeoff weight, as with Skylon, Star-Raker wouldn’t need a heavy landing gear or drop cradle for flights to orbit, even if assuming bigger, softer tires and with some extra weight allowed due to other weight reductions.


The Star-Rakers are a tremendous—and all but forgotten—capability that was utterly unnecessary for any program we actually undertook in space. But it shows we have the capability to do more – and do it far more economically than most would assume.

As a passenger craft, a single Star-Raker could have lifted more people into orbit in a day than have so far reached there in all of human history. Theoretically, even if they spent one quarter of each year being serviced (insanely high for most military or commercial aircraft), a fleet of 1,000 Star-Rakers (a moderate-sized production run for airliners), each with a thirty-year service life (average to low for commercial and military aircraft) could lift all of the people of the Earth to orbit in 28 years, for a ticket price, assuming $30/pound operations, of $19,000 each.

In cargo configuration, one Star-Raker could lift as much cargo tonnage per week as has ever been launched in human history. The total estimated ten million ton weight of a 10,000 person L5 colony from the 1975 NASA space settlement study would take a 100-ship fleet of similar cargo craft under fifteen months to lift to orbit, for a total cost of $600 billion. Compare that to the $150 billion dollar Space Station budget, or to the budgets of hundreds of billions of dollars proposed for the return to the moon, or man to Mars programs.

A thousand similar Star-Rakers could lift a billion ton, 30km long Island 3 O’Neill cylinder in twelve years for $60 trillion. By way of comparison, this is less than the $73 trillion dollar global GDP for 2014. Of course you might want to be into space mining by then. Or develop something a little more advanced than a 1970’s era Star-Raker. But until then, they could support about anything anyone has dreamed of doing in space.

Sources Cited

“Earth-to-LEO Transportation System for SPS,” Independent Research and Development Data Sheet, Project Number 243. Rockwell International Space Systems Group, 15 December 1978. Retrieved from < >

“The Final Report of the SPS Space Transportation Workshop, January 29-31, 1980.” The Johnson Environmental and Energy Center, The University of Alabama-Huntsville, October 1980.  Retrieved from <;

Hanley, G.M. “NASA Contractor Report 3321: Satellite Power Systems (SPS) Concept Definition Study – Volume IV: Transportation Analysis.” NASA—Science and Technical Information Branch, 1980. Retrieved from < Star_Raker/NASA-CR-3321_Excerpt.pdf>

Hanley, G.M. and R. Bergeron. ”An Overview of the Satellite Power System Transportation System.” 14th Joint Propulsion Conference, American Institute of Aeronautics and Astronautics (AIAA), July 1978.  Retrieved from <>.

“NASA Technical Memorandum 58238: Satellite Power System: Concept Development and Evaluation Program – Volume VII: Space Transportation.” NASA—Science and Technical Information Branch, 1981. Retrieved from < Star_Raker/NASA-TM-58238_Excerpt.pdf>

Reed, David A., Jr., Hideo Ikawa, and Jonas A. Sadunas. “Star-Raker: An airbreather/Rocket-Powered, Horizontal Takeoff Tridelta Flying Wing, Single-Stage-to-Orbit Transportation System.” Conference on Advanced Technology for Future Space Systems, American Institute of Aeronautics and Astronautics (AIAA). May 1979. Retrieved from  <>

[1] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1 -Space Transportation”. November 1981. Page 40 (1-17).

[2] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1 -Space Transportation.” November 1981. Page 41 (1-18).

[3] Independent Research and Development Data Sheet – Earth-to-LEO Transportation System for SPS. Project Number 243, Fy 1979.

[4] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1 -Space Transportation.” November 1981. Page 45 (1-22).

[5] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1—Space Transportation” November 1981. Page 42 (1-19).   ( )

[6] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1—Space Transportation.” November 1981. Page 138.

[7] NASA Technical Memorandum 58238: “Satellite Power System: Concept Development and Evaluation Program Volume VI1—Space Transportation.” November 1981. Page 248.


One thought on “BLAST FROM THE PAST; a few good ideas may return to the light of day…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s