It’s impossible to tell the story of the Surveyor lunar missions without relating it to the history of the development of the Centaur upper stage. Their developments proceeded in parallel and they both had their share of difficulties. Five of the first seven Centaur test flights failed and some quarters related this to delays in the Surveyor program. Fortunately operational capability was achieved at nearly the same point for both Centaur and Surveyor in early 1966 in time for the first Surveyor launch on May 30. For the complete history of Centaur see NASA’s “Taming Liquid Hydrogen, The Centaur Upper Stage Rocket.”
The Centaur was conceived in the mid-1950s by Kraft Ehricke at General Dynamics as a second stage for the Atlas ballistic missile with the mission of boosting payloads to synchronous orbit or to escape velocity. The design was based upon a pressure-stabilized structure, like the Atlas, and the use of liquid hydrogen and liquid oxygen as high-energy propellants. Two RL-10 engines, designed and built by Pratt & Whitney, provided 16500 pounds of thrust each. General Dynamics wrote a proposal to the Air Force for the design and development of the Centaur in 1957. A further GD proposal was accepted by ARPA in August of 1958 with $36 million for GD and $23 million to Pratt & Whitney. This shortly became a joint ARPA and Air Force program.
After the formation of NASA in 1958, NASA’s Marshall Space Flight Center took over the Air Force’s role. At this time plans called for the use of Centaur to launch the Mariner missions to Venus and Mars, the Surveyor lunar mission, and the DoD’s Advent geosynchronous communications satellite with the first launch planned for January 1961. The Advent mission placed a requirement for several Centaur engine restarts as a parking orbit was required to position the satellite for injection into a transfer orbit following a 5-hour coast prior to injection into a geosynchronous orbit.
In the spring of 1962, Marshall informed JPL that Centaur could not inject the required 2500 pounds into a translunar trajectory and asked for a reduction in payload capability to 2100 pounds. This change required that the Surveyor be redesigned to accommodate the lesser payload. During 1962 there was Congressional pressure to cancel both the Centaur and Surveyor programs.
The Centaur development was in trouble quickly with several engine explosions on test stands at P&W. On May 8, 1962 the first flight was lost when an insulation shield failed and the vehicle exploded 55 seconds after liftoff. Further problems arose due to managerial neglect at MSFC with their focus on development of the Saturn launch vehicle. In August 1962 Werner von Braun recommended cancellation of the Centaur program and use of the Saturn C-1 with an Agena third stage for the Mariner and Surveyor missions. NASA headquarters vetoed this proposal and in September transferred the Centaur program to the Lewis Research Center in Cleveland. Lewis had considerable previous experience in the handling of cryogenic propellants and therefore was deemed to be capable of managing the Centaur project for NASA.
Abe Silverstein, the newly named director of LRC, took over personal management of the Centaur program. The contract with GD was renegotiated at $321M for 14 flight vehicles plus a number of test articles with a fixed fee of $31M. NASA insisted upon a number of changes at GD including adoption of PERT and a move from matrix to project management as well as a strengthening of systems engineering. Silverstein also decided that the requirement for a two-burn Surveyor mission be abandoned until the single-burn or direct ascent mission had been successfully achieved. This placed additional constraints on the Surveyor mission, but fortunately the mission was still possible in the mid-1960s.
My personal knowledge of Centaur expanded greatly at this time. In May 1963 I attended a Centaur familiarization course held at GD in San Diego. I was also the Hughes representative to the Centaur coordination meetings that began when LRC took over the project. These meetings were held at JPL, GD in San Diego, and LRC in Cleveland with representatives from GD, NASA Lewis, JPL and Hughes.
Some of the issues I remember were the Centaur payload capability estimates that concerned us all. LRC introduced the idea of a 3-sigma propellant reserve that disturbed JPL and Hughes as the Surveyor weight was uncertain at this time and every bit of performance was crucial. As I recall the propellant reserve amounted to about 160 pounds. Since this propellant reserve had to be carried to injection into the translunar trajectory it was equivalent to payload weight. In retrospect, it just seems like good engineering practice to have this kind of reserve. Another issue that concerned this group was the possibility of the Surveyor Canopus sensor detecting the spent Centaur while trying to lock on the star. This possibility was avoided by a retrothrust maneuver following an attitude change that also prevented the Centaur from impacting the moon.
Centaur Test Flights
May 8, 1962 Centaur F-1, after spending 15 months on the launch pad the vehicle exploded when Centaur insulation panels failed shortly after launch.
November 27, 1963 AC-2, first successful launch of Atlas-Centaur with a Centaur single burn of 380 seconds. No payload was carried.
June 30, 1964 AC-3, Centaur engines shut down prematurely after a hydraulic pump failure.
December 11, 1964 AC-4, attempted to demonstrate RL-10 engine restart in space failed as the ullage rockets did not provide sufficient thrust to settle propellant. Carried a 2100-pound Surveyor mass model.
March 3, 1965 AC-5 carried a Surveyor SD-1 dynamic mass model. Atlas exploded on Launch Complex 36A severely damaging the complex. NASA had mothballed LC36B at 90% completion to save money so it had to be completed and LC36A had to be repaired.
August 11, 1965 AC-6 flawless single burn mission of 435 seconds with Surveyor SD-2 dynamic model.
April 7, 1966 AC-8, from LC36B failed to restart for second burn after 25 minute coast in a parking orbit Surveyor dynamic model SD-3
October 26, 1966 AC-9, successful Centaur restart after a 25 minute coast with Surveyor dynamic model SD-4.
All seven Surveyor missions were successfully launched by the Atlas Centaur: Surveyors I, 2 and 4 using the direct ascent mode and Surveyors III, V, VI and VIII using the parking orbit mode. The Atlas-Centaur went on to launch a number of other Hughes spacecraft including two ATS missions, eight Intelsat IV spacecraft, six Intelsat IVA spacecraft, two Pioneer Venus missions for NASA, two Comstar spacecraft, and two Galaxy 376 spacecraft.
The following comment is from Steve Dorfman: After Challenger exploded NASA backed off of it’s position that all US satellites be launched on shuttle opening up possibilities for Lockheed Martin’s Titan 3 and General Dynamic’s Atlas Centaur IIA. We had just won the Navy UHF competition to deliver ten satellites in orbit, giving us the responsibility for buying the Launch Services. We won the contract assuming a very good prices for launching without having final firm bids from LM and General Dynamics. A very intense competition between those two followed with both companies seeing this as a possibility to jump start their commercial launch business and Hughes looking for an agressive price to fit our own aggresive bid for the first large order for the HS 601. LM’s bid was based on 5 dual launches on Titan 3 which was an unacceptably high risk wheras GD bid 10 seperate launches. Al Lovelace the ex NASA administrator and GD lead for commercial Atlas flew into LA on the company jet and he and I met at the private jet facilities to hammer out a deal which came close to what we needed and was very aggresive on their part. This started GD and Atlas into the commercial lunch business and they wound up successfully launching 12 UHF HS601s which was a successful and profitable contract for Hughes. The Atlas Centaur remains the leading commercial launch vehicle today. It was win-win for both companies.