LOS ANGELES - The investigation into the crash of NASA's $264 million Genesis spacecraft in Utah last month is focusing on the discovery of a design error that resulted in improper orientation of switches that were supposed to detect its entry into Earth's atmosphere and deploy its parachutes.
"The design drawings are not correct," Michael G. Ryschkewitsch, chairman of the Mishap Investigation Board, said Friday. He emphasized that the findings were not complete.
Genesis captured atoms and ions of the solar wind during nearly three years in space and was supposed to carefully bring them to Earth on Sept. 8. Instead it smashed into a Utah range at nearly 200 mph.
The design drawing, described as backward, was produced by Lockheed Martin Astronautics of Denver, which built Genesis for NASA, Ryschkewitsch said.
How the faulty design escaped detection during many reviews was under investigation, he said.
The discovery of the likely cause of Genesis' crash raised questions about review processes at Lockheed Martin Astronautics, which were increased at the time it was built because of two failed Mars missions.
NASA's Mars Polar Lander, also built by the company, was lost in December 1999, probably when a sensor cut off its descent motor too soon. And a few months earlier, the Mars Climate Orbiter was lost when no one noticed that Lockheed Martin Astronautics was giving NASA navigation data in English units rather than metrics.
The Genesis design flaw also raised concern about NASA's similar Stardust craft, which was built by Lockheed Martin Astronautics, launched in 1999 and is designed to parachute samples of the comet Wild 2 to Earth in 2006. It has the same type of switches but the installation is different and a preliminary assessment found that they are correctly oriented, Ryschkewitsch said.
Stardust will nonetheless undergo a thorough review at a cost in the "single-digit millions," said Orlando Figueroa, NASA's deputy associate administrator for programs in the office of Science Mission Directorate.
Genesis' sample return capsule was supposed to deploy a small drogue parachute after entering the atmosphere and then deploy a parafoil. A specially equipped helicopter was to have then snagged the parafoil in flight over Utah and gently lowered the capsule to the ground.
Instead, the capsule failed to deploy its parachutes and plunged to the ground, shattering the collector arrays with samples of the solar wind. The pieces now fill more than 3,000 containers that have been sent to Johnson Space Center, where scientists are optimistic that the samples will be useful
Genesis was built with four so-called gravity switches, a pair in each of two separate electronics boxes. Each switch is a tiny plunger on a spring in a case about the diameter of a pencil and three-quarters of an inch long.
"What is supposed to happen is that when the switches are under an acceleration in the right direction it's supposed to pull the plunger down against the spring force and make contact with an electrical contact, closing a circuit," Ryschkewitsch said.
"That is supposed to happen initially when the spacecraft first hits the top of the atmosphere. And then when it slows enough, the deceleration drops off, the spring is supposed to push the little plunger back away from the electrical contact. And at that point a timer is supposed to start and that timer sequence then successively was supposed to have released first the drogue and then the parafoil."
Only one pair of switches in one box was needed to work, so the system should have been more reliable because of redundancy.
"In this case, since there was a design error and all four switches were in an incorrect orientation, the system could not have worked," he said.
Finalizing the design and assembly of those electronics occurred in 1999 and early 2000, he said. Assembly and testing of the whole spacecraft was a nearly yearlong process before its Aug. 8, 2001 launch.
Ryschkewitsch said the mishap board was examining the review process that was used on Genesis.
"Since Genesis was being assembled around the time of the Mars failures there were a number of additional reviews and we are trying to understand in detail what was looked at and exactly what happened there, and we're not yet prepared to comment on that," he said.
He said he was unable to answer the question of whether there were any similarities in the people or review processes involved in the Mars Polar Lander and Genesis missions.
"The Genesis MIB has overtly not looked at any previous reports because we didn't want to color our thinking," he said. "We have been looking at the relationship between Genesis and Stardust because they were so close in time and because there were many systems that were inherited directly between the two. But we have not looked at any of the others."
Ryschkewitsch said he did not know if anyone raised concerns about the gravity switches before the accident.
"That is one of the things we are looking into, looking back at all of the documentation. So far we have not found any overt description for any of the late reviews that were added that this had been raised specifically as a concern," he said.
NASA's Jet Propulsion Laboratory, which managed Genesis, was also examining the spacecraft's battery, which experienced an overheating problem early in the mission.
"It is going to get a lot of attention in its own right but we are not prepared to make any conclusions on the battery," Ryschkewitsch said.
