Failure aboard Hubble puts shuttle flight on hold
BY WILLIAM HARWOOD
STORY WRITTEN FOR CBS NEWS "SPACE PLACE" & USED WITH PERMISSION
Posted: September 29, 2008;
Updated with launch delay confirmation;
Updated following evening news conference
Living up to its "Perils of Pauline" heritage, a critical equipment failure aboard the Hubble Space Telescope on the eve of a long-awaited fifth and final shuttle servicing mission put astronomical observations on hold and forced NASA managers today to delay the mid-October flight of Atlantis.
Pending an engineering review, the long-awaited servicing mission is expected to slip from Oct. 14 to mid February - and possibly later - to give engineers and astronauts time to shoehorn replacement hardware into an already challenging five-spacewalk mission.
Weiler said NASA was lucky the electronics failure occurred now, on the eve of launch, and not after the final servicing mission was over.
"Think about if this failure had occurred two weeks after the servicing mission," he told reporters in an afternoon teleconference. "We'd have several single point failure, we could have lost the mission in six, 12, 18 months. So in some sense, if this had to happen it couldn't have happened at a better time. ... So I'm trying to look at the glass as half full today and I think it IS half full for us."
Shuttle Program Manager John Shannon said he expects a decision on how to proceed by the end of next week. Assuming the Hubble mission is, in fact, delayed to next year, NASA will press ahead with launch of the shuttle Endeavour on a space station assembly mission around Nov. 14, two days earlier than currently planned.
Endeavour is stacked atop pad 39B at the Kennedy Space Center to serve as a quick-response rescue vehicle for the Hubble astronauts in case of any post-launch problems that might prevent a safe re-entry. With a decision to delay the Hubble flight, Atlantis would be hauled off pad 39A and moved back to the Vehicle Assembly Building. Endeavour then would be moved to 39A for normal launch processing. NASA is in the process of converting pad 39B for use by the Ares rockets that will replace the shuttle. While the pad could be used to support Endeavour's launch on a rescue mission, it's payload changeout room is no longer capable of normal processing.
Whenever Atlantis takes off on mission STS-125, Endeavour will no longer be available for rescue duty. For a Hubble flight in February, the shuttle Discovery, currently scheduled for launch Feb. 12 on a mission to deliver a final set of solar arrays to the space station, would be pressed into service. That flight is known as STS-119.
But Shannon must take into account how the Hubble flight would fit into the overall manifest. The Russians plan two Soyuz launches to the space station next spring and temperature constraints due to the station's orbit will preclude shuttle visits for several weeks in that timeframe. One option would be to delay the Hubble flight until after Discovery's station assembly mission. In that case, Endeavour, scheduled for another launch next May, could once again serve as the rescue vehicle.
The problem aboard Hubble cropped up shortly after 8 p.m. Saturday when channel A of the telescope's control unit/science data formatter, or CU/SDF-A, began acting erratically. The telescope's flight computer, following pre-programmed instructions, then acted to "safe" the payload computer and science instruments. An attempt by ground controllers to reset the formatter was not successful. Troubleshooting continues, but engineers are not optimistic.
The telescope is not in any danger, but science operations are on hold until engineers can reconfigure the observatory to use channel B of the control unit/science data formatter. Engineers plan to make the switchover Thursday or Friday, after a detailed readiness review.
"All the testing and all the efforts so far to restore (the A-side electronics) indicates it has totally failed," said Hubble Program Manager Preston Burch. "Our only option at this point is to switch over to science data formatter B, which is the redundant channel. Unfortunately, switching to that side will require the switch over of the spacecraft data management system to the B side as well. ... So this is a major event for Hubble."
The backup, or "B side," of the data management system has not been powered up since the telescope was launched in 1990. Even if it works - and if five instrument subsystems successfully make the transition to their own B channels - NASA would still be faced with a loss of redundancy in a critical system and a subsequent failure would permanently disable the observatory.
"The transition to side B operations is complex," NASA said in a statement. "It requires that five other modules used in managing data also be switched to their B-side systems. The B-sides of these modules last were activated during ground tests in the late 1980's and/or early 1990, prior to launch. The Hubble operations team has begun work on the Side B transition and believes it will be ready to reconfigure Hubble later this week. The transition will happen after the team completes a readiness review."
Given CU/SDF-A worked normally for nearly two decades, one could argue the backup channel should work as required for years to come. But senior managers do not want to risk mounting a costly servicing mission and then leave the telescope without redundancy and no chance to carry out an additional servicing mission before the shuttle is retired in 2010.
A spare science instruments command and data handling system, which includes the needed control unit/science data formatter, is available at the Goddard Space Flight Center in Greenbelt, Md. But it is not flight qualified and it has not been powered on since 2001. Extensive testing and checkout will be required to upgrade it to flight status. The box weighs about 135 pounds and measures 21.5-by-32.5-by-9.5 inches. It will be mounted on the side of a payload carrier in Atlantis' cargo bay for the ride up to Hubble. Burch said the shuttle can easily accommodate the additional hardware and nothing will have to be removed to make room.
Installation of the box is expected to take about two hours to complete. The unit would be attached to the door of electronics bay No. 10 with 10 bolts and one electrical connector. Engineers have not yet decided where the work might go in the already tight spacewalk timeline.
But Burch said it's possible the additional task can be added to the crew's timeline without losing any other already planned task. The repair of Hubble's Advanced Camera for Surveys, for example, is currently spread over two spacewalks, EVAs 3 and 5. But if the work is completed in the first of those two excursions, the crew would be able to install the new data formatter hardware during the ACS block of time in the second EVA.
"Given the fact that we think this job can be done in under two hours, there is a possibility ... if he's able to complete the ACS repair on EVA day 3, that frees up a substantial amount of time on EVA day 5. So in theory, this may be a doable thing for us to have our cake and eat it too. But a lot of things will have to go right. And we certainly don't want to over extend the crew."