An Interview with
Dr. Harvey Tananbaum on the Chandra Launch Delays

February 11, 1999, in Cambridge, MA.

After staying on schedule for an August 28, 1998 launch date for five years, the launch of the Chandra X-ray Observatory has slipped several times. We talked to Dr. Harvey Tananbaum, Director of the Chandra X-ray Observatory Center about these delays.

Q: When did you first realize that the Chandra Observatory would not be launched on the originally scheduled date?

TANANBAUM: In the Fall of 1997, TRW* encountered problems with the software they had developed to test the spacecraft. In spite of added manpower, incentives, and a reorganization of the management, these problems caused a slip in schedule to a December 3, 1998 launch date.

Q: We talked to TRW and other Chandra officials in April of 1998, and they were confident they would make the December ‘98 launch date. What happened?

TANANBAUM: Several things. In June of ‘98, during the course of a routine test, the door to one of the science instruments, the CCD Imaging Spectrometer, failed to open. It stuck and a fail-safe disk within the wax actuator burst, so the actuator could not open the door. Exactly why the door stuck was never determined. Most people believe that the O-ring seal on the door stuck, and the problem was compounded because we tried to open the door cold, which would not have happened in orbit. The testing could not find the reason for sticking. They ruled out every possibility. You would have concluded from the tests that the door must have opened!

Q: What has been done to ensure that this problem doesn't recur in orbit?

TANANBAUM: The mechanism that opens the door has been rebuilt and modified, and the door will be opened gradually to prevent the safety disk from bursting even if the O-ring sticks a little. In addition, the door will be opened at a warmer temperature.

Q: Were other problems cropping up at this time?

TANANBAUM: Yes. More software delays, related to safe mode testing. For example, you want to keep the telescope from pointing to the Sun in event of a failure, or keep the solar panels pointed toward the Sun. Software is used to simulate a failure, for example a gyro failure, and then the onboard software is supposed to solve the problems that may arise from this failure. All the software needed for this activity turned out to be more complicated than expected. The net effect of the software and the door problems was that the launch slipped to January 23, 1999.

Q: Then in October the launch date slipped again.

TANANBAUM: Yes, this was just before the spacecraft was due to be shipped to Kennedy Space Center to prepare for launch. The safe mode testing still wasn't complete and there was a problem with some electrical switching boxes that switch power on and off between various components of the spacecraft. In addition, some timing synchronization problems cropped up in the interface unit which handles communications between the onboard computers and other parts of the spacecraft. NASA's Associate Administrator for Space Science, Ed Weiler, decided that these problem should be fixed at TRW rather than at Kennedy. This pushed the launch back until April15. Everything was fixed ahead of schedule. TRW got ready to ship in early January and then other problems surfaced.

Q: What were the nature of these problems?

TANANBAUM: They were discovered while testing another satellite built by TRW. This satellite has some of the same components as Chandra. Small printed circuits boards were found to be faulty. It turns out that Chandra has 129 of these boards. Fortunately only four required replacement. New boards were fabricated at the subcontractor, B. F. Goodrich. Two of the boards have been installed in their unit and are undergoing testing at Goodrich. The other two boards will be installed in their unit soon. After testing, all the boards will be re-installed in Chandra at Kennedy Space Center in early March. In the meantime the observatory was put in a protective bag and shipped to Kennedy Space Center on February 4.

Q: Do these delays pose any danger to the instruments or other components?

TANANBAUM: The major worry is contamination, which is why Chandra was put in a protective bag, and why all the work on it is done in a clean room.

Q: What is the anticipated launch date?

TANANBAUM: We are now working toward a July 9 launch.

[ Press Index ] [ Interviews ]