In 1974, during the design and development of the Space Shuttle Orbiter, NASA faced the problem of how to get the Orbiter from remote landing sites to launch locations. Costs and technical considerations had led NASA to abandon initial plans to use air-breathing jet engines. Another plan for a large aircraft called VIRTUS was also scrapped. VIRTUS would have carried the Orbiter under the center wing, flanked by twin fuselages and a twin boom tail with power from four Pratt & Whitney JT9D turbofans. While design work on VIRTUS had proceeded as far as wind tunnel tests with a 1/34 scale model, the sheer size, long development time, and cost involved to develop an aircraft that would be built in very small numbers eventually resulted in the end of the VIRTUS project.
Similarly, Jack Conroy, the developer of the Super Guppy concept that NASA was using to transport rocket stages, suggested using a jumbo-class aircraft to carry the Orbiter on its back. Proposals were issued to the industry and Lockheed offered up a twin-fuselage aircraft based on the C-5 Galaxy with the Orbiter suspended underneath a new center wing section. But, like the earlier VIRTUS program, this idea was eliminated from consideration due to cost, development time and the fact that the design was so wide, no runway available could accommodate the aircraft.
Boeing offered a modified version of the 747 that could carry the Orbiter on its back, a much lower risk approach. Boeing even suggested that the large, external tanks could be carried on the back of a 747, but wind tunnel studies showed the idea to be less practical than initially thought. Lockheed subsequently reworked its design to a simple modification of a C-5 Galaxy that could carry the Orbiter on its back, much like Boeing’s proposal.
By mid-1974, Boeing’s 747-based proposal and Lockheed’s simpler C-5 Galaxy-based proposal were the only serious contenders to become the new Shuttle Carrier Aircraft (SCA). On 24 April 1974, NASA selected the C-5 Galaxy proposal from Lockheed because it required the least acquisition expense and would need less structural modification than a Boeing 747. Accordingly, NASA approached the United States Air Force with the proposal and a request to make three to five C-5s available. The USAF was very receptive to the idea. The Lockheed proposal only added 400 to 600 pounds of modifications to the Galaxy without adversely affecting its cargo carrying capacity when not being used to transport the Orbiter.
An arrangement was set up whereby NASA would pay for the modifications and then lease the modified C-5s as needed from the Military Airlift Command. One Galaxy was agreed to be bailed to NASA full-time for development into the SCA and for use in the approach and landing flight tests (ALT) with the planned first Orbiter (which would become the Shuttle Enterprise). Despite some lingering concerns about the effects of the Orbiter’s wake on the C-5’s T-tail, NASA and the USAF had worked out an acceptable arrangement for both parties.
The downturn in the American economy in the early 1970s led numerous US airlines to release their 747 aircraft which were too large for the market at the time. As a result, the acquisition cost of a Boeing 747 to be the SCA was much lower than what Lockheed was proposing. With the ready availability of low-time 747s on the market, NASA abandoned plans for using the Galaxy and decided it would much easier to have complete control of the SCA than to have to compromise with military priorities for use of the C-5 Galaxy.
On 18 July 1974, NASA purchased a used Boeing 747-123 (N9668, msn 20107) from American Airlines. It was the 86th 747 off the production line at Everett and was had been delivered to American on 29 October 1970. By the time of the NASA purchase, it had only logged 8,999 flight hours and 2,985 cycles flying primarily transcontinental services between New York JFK and LAX. NASA re-registered the aircraft as N905NA. Before modification into the SCA configuration, N905NA was used for in-house studies with NASA Ames on wake vortices. Following the conclusion of the wake vortex research program, Boeing initiated the $30 million conversion program on N905NA on 2 August 1976.
Boeing installed new bulkheads to strengthen the fuselage with skin reinforcement at critical stress areas. The horizontal stabilizer structure was also beefed up, along with the addition of telemetry and transponder test equipment, fittings for the Orbiter support struts, and the installation of a 747-200 rudder actuator system. Boeing also developed a set of removable modifications for the SCA – the first was a telescopic forward support assembly that was used only during the atmospheric flight tests with the Shuttle Enterprise. This support would hold the Enterprise at a six-degree angle of attack to facilitate release during the flight tests. A fixed assembly was also developed for use during SCA ferry missions that held the Orbiter at a three-degree angle of attack, which induced less drag during the ferry flights. The aft support assemblies (there were two) were common to both the atmospheric flight tests and ferry flights. Finally, 10 foot by 20 foot vertical endplates were added to the end of the horizontal stabilizer to provide additional stability when carrying the Orbiter. In practice, though, NASA never removed the endplates.
The 747’s trim system was also modified to allow a greater range of trim in pitch to counteract the downwash off the Orbiter’s wing on to the horizontal stabilizer. Most of the main deck interior was stripped out, but some seats were retained for support personnel during the ferry flights. The JT9D engines were converted to allow a thrust increase from 43,500 pounds to 46,950 pounds of thrust. The JT9D engines on the current incarnation of the 747 SCA are rated at 50,000 pounds of thrust. Since each Orbiter has a different empty weight, an adjustable ballast system using standard cargo containers in the forward under-fuselage cargo compartment had to be developed to maintain the center of gravity. On 14 January 1977, Boeing finished the modification work and after a period of flight testing, the SCA was delivered to NASA.
Concerns about flight crew safety during the approach and landing tests with the Shuttle Enterprise led NASA to incorporate an escape system on N905NA, since the flight crew of the Enterprise had ejection seats. The escape system was based on what was used on the 747 prototype during Boeing’s 1969 flight tests; in the event of an emergency, a handle was pulled that blew out thirty fuselage windows to facilitate rapid decompression of the aircraft. Three seconds later, an emergency egress hatch on the lower forward fuselage was blown out with extendable spoiler being deployed. The crew would make their way back from the flight deck to the middle of the upper deck lounge area where a hole leading to a 16 foot escape slide would lead them out the blown hatch and clear of the aircraft. Testing showed the flight crew could bail out of the 747 within 11 seconds.
In 1988, NASA acquired a second 747 to act as a back up to N905NA. This was driven in part by the recommendations following the Challenger accident that a significant number of Shuttle flights should land at Edwards Air Force Base. On behalf of NASA, Boeing purchased a 747-100SR from Japan Air Lines where it had flown as JA8117, msn 20781 and conducted the necessary modification work to bring it up to SCA standards with the new tail number N911NA. On 20 November 1990, it was delivered to NASA and in 1995-1996 both of the 747 SCAs were repainted in NASA’s new colors.
During a ferry mission, the SCA’s maximum speed was 250 KIAS (Mach 0.6) at an altitude of 13,000-15,000 feet with a range of approximately 1,150 miles. Without the Orbiter, the SCA cruises at 24,000-26,000 feet with a range of 6,300 miles. During ferry flights, the usual crew is two pilots and two flight engineers, but only one flight engineer is needed on non-ferry flights. At one point, NASA looked at inflight refueling of the SCA. The equipment was readily available as it was installed on a handful of USAF’s 747s: the E-4 airborne command posts and the two VC-25A presidential transport aircraft. Proximity flight tests were even carried out with N905NA and a KC-135 tanker minus the Orbiter, but the discovery of cracks at the base of N905NA led to the termination of the studies.