Endeavour to go Nov. 14; Hubble slips deeper into '09
BY WILLIAM HARWOOD
STORY WRITTEN FOR CBS NEWS "SPACE PLACE" & USED WITH PERMISSION
Posted: October 30, 2008
NASA managers today cleared the shuttle Endeavour for launch Nov. 14 on a space station assembly and servicing mission. But work to test a component needed by the Hubble Space Telescope will not be finished in time for launch aboard the shuttle Atlantis in February. That flight, Hubble Servicing Mission 4, originally was scheduled for launch Oct. 14 but it was delayed when a critical science data relay unit aboard the observatory failed in late September. NASA managers decided to delay the Hubble flight to mid February to give engineers time to test replacement electronic gear, but detailed checkout and problems with the equipment require additional troubleshooting, officials said today, delaying the long-awaited flight to May at the earliest.
"Today, after a thorough review at the Goddard Space Flight Center of the work to go preparing the spare science instrument command and data handling system, the science mission directorate has informed the space operations mission directorate and the space shuttle program that the spare unit will not be available to support a February launch date with acceptable schedule margin and technical risk," Jon Morse, director of astrophysics at NASA headquarters, told reporters.
Engineers now plan to remove Atlantis from its external tank and solid-fuel boosters and give that "stack" to the shuttle Discovery for launch Feb. 12 on a flight to deliver a final set of solar arrays to the international space station. If all goes well, the Hubble mission, aboard Atlantis, could take off around May 12, two weeks before launch of a Russian Soyuz spacecraft May 25 that would carry three crew members to the space station, boosting the lab's crew size to six.
But that schedule assumes engineers at Goddard can resolve anomalies in the replacement Hubble command and data handling subsystem and complete an exhaustive series of tests to verify its flight readiness. Preston Burch, Hubble program manager at Goddard, said he's confident the hardware will be ready to fly in May.
"I'm very confident, personally, that we can troubleshoot the anomaly that we're currently working on and get it ready to fly," Burch said. "We have all the design documentation, all the schematics and everything, we have a very smart group of engineers and technicians that are working on this. This is a glitch on a black box that's well known, well understood, flight proven, so I don't see this as a huge, insurmountable deal. It's not like we're doing something experimental in quantum physics or anything like that. So I suspect we're going to have success with this."
The Hubble trouble began Sept. 27 when the telescope's control unit and science data formatter, or CU/SDF-A, suffered a "hard" failure, preventing ground controllers from receiving data from the science instruments. The A and B channels of the redundant science instrument data handing system are located on the same electronics tray and NASA managers decided to replace the entire unit with a flight spare to restore lost redundancy.
But the flight spare must be tested and recertified, forcing NASA to delay Atlantis' launch from mid October to mid February. In the meantime, after initial problems, engineers successfully activated the B-side electronics aboard Hubble and resumed science operations.
But they have not been so fortunate with the spare hardware on the ground. During initial checks, engineers were able to activate the spare unit's B-side but they had problems with its counterpart. Those problems prompted speculation and reports of potentially serious trouble.
"There's been a lot of speculation in the media and I guess they've been going to sources that don't have any direct knowledge of the condition of the hardware," Burch said. "That's really unfortunate."
He said the spare science instrument command and data handling system "was delivered to NASA in the early 90s and at that time, the acceptance test program had not been completed."
"There was some open paperwork and there were some anomalies that had not been resolved at that time," Burch said. "In the early 90s, we were focused on preparations for the first servicing mission and so did not want to spend the resources and the time on the SIC&DH spare at that time. So we set it aside with the understanding that we would be able to address any of the open issues associated with it and get it in a flight worthy condition in the future should it be needed. And now that time is here. So it's been quite a few years since we've looked at this particular unit."
Along with the spare SIC&DH, Goddard also has an engineering unit used for testing. Here's how Burch explained the issue:
"We have an engineering model SIC&DH that's a single string unit, it doesn't have two sides, an A and a B, but it's built just like the flight unit," Burch said. "But it hasn't gone through the same type of program that the flight unit has gone through and so it's not considered to be a flight unit. But we have had a need over the past 15 years to occasionally use an SIC&DH on the ground for testing instruments and for testing the NICMOS (infrared camera) cooling system that you'll recall we developed in preparation for servicing mission 3B, which we launched in March 2002.
"As a result of that, we availed ourselves of the spare unit and we took some equipment off of it. We removed the B side CU/SDF, the control unit/science data formatter, because we needed one to use on the engineering model SIC&DH. We took off one of the computer processors and put that on and one of the (data interface units). So there was some disassembly of that unit, but it's otherwise in a pristine condition. So when we found ourselves with a need to fly it, we immediately went back and rounded up the (needed hardware). And now we are operating it on the ground.
"We tried to fire up the A side CU/SDF Friday and it didn't handle the commands properly that were being sent to it to turn it on," Burch said. "So after a little bit of trying that, we thought well, we might have a problem with our test configuration. So we thought, why not try the B side and see how that works? So we switched over to the B side and that came up and worked just fine. We tested the B side for a few days and then decided to go back to the A side to try to troubleshoot that.
"Initially the unit did not respond properly but with repeated attempts at commanding the A side, the A side slowly came around and started handling the commands properly. At one point, it was performing as well as the B side, it was handling multiple format-type commands, it was handling data from multiple science instruments ... it was behaving beautifully. Then it stopped behaving like it should and we restarted it. And so it's been kind of a back and forth deal with it.
"What we're trying to do now is to understand what the sensitivity of that unit is to things like temperature. There are a number of suspicions about what could be the cause of the problem. What we suspect is, there's a workmanship or a parts problem on that unit, which is causing this glitch and we're going to need to try to troubleshoot that."
Burch said he was confident engineers will track down the problem, fix it and press ahead with testing.
"Our plan overall takes something on the order of about six-and-a-half months from now," he said. "There's about a month or so devoted to inspecting and resolving any of the performance issues associated with it, about three months for environmental tests and then about two to two-and-a-half months to do final testing and shipping down to Kennedy Space Center."
The primary goals of Hubble Servicing Mission 4 are to install new batteries and stabilizing gyros, two new science instruments, a replacement fine guidance sensor and new insulation. The astronauts also plan to repair two of Hubble's instruments, the Advanced Camera for Surveys and the Space Telescope Imagine Spectrograph.
Three of Hubble current set of six gyroscopes have failed, its six nickel-hydrogen batteries are nearly 20 years old and only hold half the charge they were designed for and its data subsystem no longer has redundancy. Even so, Burch said the observatory should be able to ride out the latest servicing mission delay with no major problems.
"Right now, we are doing science on two gyros and we have one gyro remaining that's a spare that is off," he said. "We should be able to easily go another year with the current gyros we have."
Hubble's batteries currently are operating at half their rated capacity. While that is sufficient for normal operations, engineers have no idea how long they will last.
"Nobody has ever operated nickel-hydrogen batters in orbit as long as Hubble has," Burch said. "The way we use the batteries, we've been able to get smarter and come up with a better regimen for operating them on a daily basis. But that still doesn't get around the fact that they're well past their design lifetime. The prudent thing to do is replace them. But I think we'll be fine with batteries for the next several months."
As for the no-longer-redundant science data handling subsystem aboard Hubble, "I am very confident the B side will continue to operate normally," Burch said.