Discovery's launch delayed until at least mid-December
BY WILLIAM HARWOOD
STORY WRITTEN FOR CBS NEWS "SPACE PLACE" & USED WITH PERMISSION
Posted: November 24, 2010
NASA managers reviewing the progress of repairs to the shuttle Discovery's external tank and the rationale for making another launch attempt decided Wednesday to pass up an early December launch window, delaying the flight to at least Dec. 17 and possibly all the way to February.
"Clearly we're not ready for the December 3rd-through-7th window that's coming up," Shannon said. "We're going to leave the option open for the next launch window that starts Dec. 17. But a lot of data has to come together for us to support that. I don't think the team is worried about any specific launch date at all right now. They're just worried about getting the proper data so we can determine what our risk exposure is to this problem. Once we determine that and determine what the appropriate path forward is, then we'll go off and pick the right launch date."
A launch on Dec. 17 would occur at 8:51:53 p.m. EST. Docking with the space station would be expected around 5 p.m. on Dec. 19, followed by two spacewalks, one on Dec. 21 and the other on Dec. 23. Appropriately enough, the crew would enjoy a half day of off-duty time on Christmas Day, followed by undocking Dec. 26 and landing back at the Kennedy Space Center the afternoon of Dec. 28.
But Shannon cautioned that the Dec. 17 "no-earlier-than" date is little more than a planning target at this point, the first day a launch would be possible following a Russian Soyuz launch and docking Dec. 15 and 17 respectively. As it now stands, the shuttle could launch through Dec. 20 without any major impacts. Launches after that also are possible, but in those cases the shuttle would be in orbit during the New Year transition, requiring the ship's flight computers to be reconfigured. That would have to be done during a period of light activity, requiring close coordination with space station planners.
Even if the tank is cleared for launch by Dec. 17, mission managers may have to contend with reduced ascent performance because of the density of the winter atmosphere. Internal studies indicated cargo would have to be off loaded, but Bill Gerstenmaier, director of space operations at NASA headquarters, said Wednesday that was not the case.
In any case, if Discovery remains grounded through December, the flight would slip to February because of conflicts with other unmanned cargo launches and temperature constraints related to the station's orbit. The next official launch window opens Feb. 27 and closes March 6. But it may be possible to launch Discovery between Feb. 3 and 10 if a Japanese cargo ship, scheduled to arrive in late January, can be moved to another port on the space station.
A key element of the "flight rationale" discussion is an understanding of what caused cracks in the external tank stringers spotted after a Nov. 5 launch scrub and whether they represent a generic problem that poses a broader threat. While there are several theories, engineers have not yet identified the root cause.
Cracks became more common after the 1998 debut of new super lightweight tanks built with a more brittle aluminum-lithium alloy. Shannon said a review of manufacturing records showed that some 5,000 stringers were built for 43 aluminum-lithium tanks. Out of that total, 31 cracks were found that required repairs, including four in Discovery's tank and three in a tank scheduled for use by the shuttle Atlantis next summer.
But Discovery's cracks were the first to be discovered at the launch pad, the presumed result of exposure to ultra-low temperatures during fueling.
The problem was spotted after a gaseous hydrogen leak in a vent line attachment fitting forced a launch scrub Dec. 5. During de-tanking operations, engineers spotted a large crack in the foam insulation of the intertank flange area near the base of the upper liquid oxygen tank. When the insulation was cut away for repairs, four cracks in two underlying stringers were found.
Using a certified repair technique, the damaged stringer sections were cut out and replacements were spliced in, along with "doublers" to provide additional strength. Fresh BX foam then was sprayed on and shaped to blend in with the surrounding insulation.
Working in parallel, engineers also replaced the hydrogen vent line quick-disconnect fitting after tracing the leak to a misalignment issue. A suspect circuit breaker believed to be responsible for a subtle electrical problem that led to an earlier launch delay also was replaced.
But before Discovery can be cleared for launch, engineers must show that any additional cracks that might form in the tank stringers after the shuttle is re-fueled won't cause foam to pop off while the shuttle is climbing out of the dense lower atmosphere. During the first two minutes or so of flight, the density of the atmosphere is still high enough to quickly slow lightweight foam debris, allowing the rapidly accelerating shuttle to fly into it at a high relative velocity.
At a more fundamental level, the crack issue raises questions about the overall structural integrity of the tank and the ability of weakened stringers to handle the forces experienced during launch. The stringers are not considered load-bearing components, but engineers need to verify the tank's safety margins. Whether cracks in the aluminum-lithium ribs represent a generic problem remains to be seen.
"As we started looking at it, we have had a history on these stringers during assembly occasionally of getting cracks in them that are caused, we believe, by the assembly process," Shannon said. "If you look at the design -- and we've reviewed over the last week and a half very closely the design of these stringers, how they're assembled, how they're attached, the loads they would take in the initial cryo loading and also during launch -- they have plenty of design margin. So a properly assembled stringer in the expected flight environment will not crack."
But problems during assembly, including fastener misalignment or mishandling, could have caused a problem that escaped detection and later showed up under the thermal stresses of fueling.
"So the question was ... did we have an escape in our inspection methods that allowed a flaw to get into the assembly of the intertank that was exposed when we did our cryo loading? The other question was, did you have some kind of assembly misalignment that could have added additional stress to that piece that when you added the stress of cryo loading, you would overload that piece and you'd get this crack?
"The hope last week, and why we were kind of sequentially slipping launch dates, was that we would show that there was a flaw in this material, that you could go find some material defect that got through our process," he said. "And we also hoped the initial cryo loading stress on that part would be the maximum stress that that part saw during the entire flight phase. As we're going through the investigation, neither one of those points is as clear as we need it to be to commit to go fly."
However the discussion plays out, NASA's options are limited. The agency only plans three more missions: Discovery's, a flight by Endeavour at the end of February and a final flight by Atlantis next summer. There are no spare tanks and little time to make major changes without significant downstream impacts.
Even so, Shannon said his team will not be driven by schedule pressure and that Discovery will only be launched if the community agrees it is safe to do so.
"It's a complex problem," he said. "If you think about it, you're putting it through a very difficult environment where you have thermal stresses, you have bending stresses, you have tension and compression when you're lifting off. The team has done a great job so far modeling all of that and helping us understand where our vulnerabilities are in that stringer design. The main focus for the program right now is to understand the exposure.
"We need to understand our risk, what is the risk that something else could have gotten through our process and gotten out to the launch pad. There are a lot of things on the table. We could potentially instrument the vehicle and do another tanking test so that we would understand the stresses in that area even better and further improve our models to know what is our exposure to risk?
"We may end up deciding that we need to get some X-rays of the back side of the tank. That's not available to us on the launch pad so if we had to do that we'd roll the vehicle back to the VAB (Vehicle Assembly Building) and put the platforms around it and get that data.
"So we're going to develop the stress models and do all our testing and look at all of our assembly methods and pull together that whole story to understand what is our risk exposure here," he said. "If the we can't get enough data to show that our risk exposure is acceptable, then we're going to have to look at other options."
Those options might include additional work to strengthen the stringers or additional testing and inspections. Shannon would not rule out eventually switching to another tank, but he pointed out that the three in NASA's inventory were built the same way and with the same materials.
Another issue facing NASA planners is the health of the U.S. oxygen generation system aboard the International Space Station. The system currently is shut down and the crew is relying on a Russian generator and stored air pending the return to Earth of a critical component that failed earlier. While a spare component has been installed in the U.S. system, engineers do not want to restart the oxygen generator until they know what caused the original problem. In the meantime, NASA plans to deliver additional oxygen with Discovery.
Because of a variety of complex technical issues, the amount of oxygen Discovery can deliver to the station is directly related to when the shuttle takes off. And as far as the December opportunities, sooner is better than later.
But developing a flight rationale that is acceptable to the community could push the mission into February depending on how the analysis plays out.