Elon Musk, chairman and chief executive officer of El Segundo, Calif.-based Space Exploration Technologies (SpaceX), said the cause of the leak was human error, not a design flaw.
“All current analysis shows that the nature of the problem was a pad processing error the day before the launch,” Musk said during an April 5 presentation at the National Space Symposium here.
Musk said the mistake was made by “one of our most experienced pad technicians” but declined to provide further details on the error, saying he did not want to get ahead of an ongoing launch failure investigation SpaceX is conducting with the Pentagon, its customer for the mission.
Sources familiar with the investigation said the pad technician had been working on the rocket’s avionics the night before launch and failed to tighten a tiny fuel pipe fitting that had been loosened in order to perform the avionics work.
In the final minutes before the March 24 liftoff, Falcon 1 flight controllers saw no pressurization drop that would have been an indication of a problem onboard the vehicle.
“If we had been looking at the right data stream at the right time we would have caught it,” Musk said.
Instead, the rocket was launched. About 25 seconds into the flight, a fire broke out around the top of the main engine, damaging the first-stage helium pneumatic system used to pressurize the rocket’s fuel tanks. Four seconds later, once the pressure dropped, the Falcon 1’s Merlin engine shut down.
The rocket subsequently crashed into a reef not 1 kilometer from SpaceX’s Pacific Ocean island launch facility in the Kwajalein Atoll. The rocket’s payload, a small experimental satellite called FalconSat-2 that was built by cadets at the U.S. Air Force Academy, crashed through the roof of an unoccupied storage shed on the island and landed next to the shipping container used to transport it to Kwajalein.
Musk said he views the Falcon 1 launch as a “partial success” because the primary objective of the mission was not to put the $700,000 satellite in orbit, but to gain flight data on the rocket and show that it could be made ready to launch in a couple hours or less.
“We obviously cannot say it was a complete success. That would be ludicrous. We also cannot say it was a complete failure. That would be just as inaccurate,” Musk said.
The U.S. Air Force and the Defense Advanced Research Projects Agency paid $8 million for the mission in order to evaluate the Falcon 1’s suitability as a highly responsive launcher. Musk said with additional work, a Falcon 1 could be prepared for launch in under an hour.
Musk said none of his customers have given any indication that they intend to flee in light of the Falcon 1’s unsuccessful debut, and many have gone out of their way to show their continued support.
Musk himself stressed that he is not discouraged and remains committed to the Falcon’s success. “Both me personally — and SpaceX — are in this for the long run. We’re not going to cut and run if we have a few issues. We’re not going to cut and run if we have a lot of issues. We’re going to see this through.”
During an April joint press conference here with Musk, Lloyd Feldman, the assistant director of the Pentagon’s Office of Force Transformation, said the office was not fazed by the Falcon 1’s botched debut and still intends to launch the experimental TacSat-1 on the rocket’s next mission, which is expected to occur this fall from Kwajalein.
“This is what taking risks tastes like,” Feldman said. “There is no such thing as a smooth learning curve. You only learn by failure.”
Musk said SpaceX will make a number of process changes in light of the launch failure, among them, a further automating of the Falcon 1’s vehicle health-monitoring system. He said SpaceX will increase “by an order of magnitude” the number of software aborts that can trigger a launch abort in response to abnormal sensor readings.
None of the fixes SpaceX intends to make are expected to materially impact the cost, Musk said, adding that the company intends to continue to offer Falcon 1 launches for $6.7 million.
Musk also announced during the symposium that SpaceX had signed up a new customer for the Falcon 9, the nine-engine Falcon 1 follow-on rocket the company still intends to introduce in 2007, albeit two to three months later because of the failure.
Musk said MacDonald Dettwiler & Associates of Canada had purchased a Falcon 9 launch slated for 2008, noting that the deal was completed after the Falcon 1’s unsuccessful debut.
MacDonald Dettwiller & Associates officials said the payload for the Falcon 9 launch is Cassiope, a Canadian mission to demonstrate an experimental store-and-forward data-delivery system for transmitting large data packages. The roughly 500 kilogram satellite also includes a science instrument and will be launched into low Earth orbit.
Musk also visited the U.S. Air Force Academy here April 5 where he offered to launch a future cadet-built satellite for free on a Falcon 9 and to pay the travel expenses for four cadets to attempt.
Musk, a 34-year-old technology entrepreneur, had his coming out before the traditional aerospace community in 2003 at the National Space Symposium as the featured speaker at a Colorado Space Business Roundtable lunch. He predicted that he would launch the Falcon 1 within 18 months.