KENNEDY SPACE CENTER, Fla.--The shuttle Discovery's planned launch early Wednesday on a space station resupply mission was called off during fueling Tuesday when a valve in a liquid hydrogen feedline apparently failed to close properly. Launch was tentatively reset for Friday, assuming the problem can be resolved in time.
Engineers plan to test the suspect hydrogen fill-and-drain valve Wednesday to determine if the valve or a sensor system that measures its position is to blame for the readings that forced NASA to order Discovery's second delay in a row. Bad weather blocked a launch attempt early Tuesday.
If it turns out the position sensor was to blame--and if NASA managers can get comfortable launching Discovery without full instrumentation in a critical system--then a launch attempt Friday at 12:22 a.m. EDT might be feasible.
But if engineers are forced to open the shuttle's engine compartment and replace any suspect components, launch could be delayed to around October 17.
Discovery's current launch window closes after Sunday because of upcoming Japanese and Russian space station launches and because of a scheduling conflict with the Air Force Eastern Range, which provides tracking and telemetry support for all rockets launched from Florida.
Telemetry indicates the position indicator is to blame for Discovery's latest problem and "right now, I'm pretty confident...we'll be really good to go on Friday if we come out with a good technical story that says we can fly without instrumentation," said Mike Moses, chairman of NASA's pre-launch Mission Management Team.
"So we both have to first prove it is instrumentation and then prove we're OK to fly without instrumentation," he said.
Engineers plan to begin testing the valve Wednesday afternoon or early evening after Discovery's external tank is fully drained and free of any residual hydrogen gas.
"Heading into the midnight timeframe tomorrow, we'll probably know whether we are going to keep pressing forward (for a Friday launch) or we've learned we're not going to make it," Moses said.
Because the space station launch window moves about 23 minutes earlier per day, Discovery has two launch opportunities Friday, the first at 12:22 a.m. and the second at 11:59 p.m.
The Mission Management Team plans to meet at noon Thursday to assess the progress of troubleshooting and to make a formal decision on whether to proceed with launch.
The problem developed around 5:52 p.m. Tuesday when the 8-inch-wide liquid hydrogen inboard fill-and-drain valve in Discovery's aft engine compartment was commanded to close during fueling, a routine step to slow the rate at which hydrogen flows into the shuttle's external tank.
One position indicator showed the valve was no longer wide open, but the closed indicator never provided a reading and engineers were unsure of the valve's actual position.
The valve must be fully closed for launch. In the event of a delay, the valve also must be able to open back up to drain the giant external tank. In addition, the valve is opened in orbit to blow residual propellant out of the engine plumbing and into space.
NASA flight rules forbid cycling the valve for troubleshooting once fueling has started out of concern about galling, debris creation and the possibility a failing valve could get stuck or break in the closed position. Following the rules, NASA managers called off the countdown and ordered engineers to drain the external tank.
Based on two past failures, engineers found galling can create debris in the line that can cause a valve to jam.
Moses said engineers "really do think this is just telemetry. ... We have the pressure traces, we know it looks like the valve is going its full range of motion. But we have a much bigger database at ambient temperatures when we do valve checkouts on the ground pre-launch than we do here at cryo conditions. So we want to get back to that pre-launch state where we know it's an inert system, go run some valve cycle tests, compare those two sets of data and show we really do have just a position indicator problem.
"That's postulating," he said. "The teams are going to spend the next two days gathering the technical analysis that says if that's truly the situation, we're comfortable launching in that scenario. And then go gather the data to make sure it really matches with what we think.
"If we find that's not the case...or the team just doesn't get comfortable that we really can tell truly that this valve is actually open or actually closed without this position indicator, then we would probably not be in position to launch 48 hours from now," he said.
As if engineers didn't already have their hands full, telemetry during Tuesday's fueling indicated elevated levels of hydrogen gas in the tail service mast on the left side of the shuttle where the liquid hydrogen feedline attaches to the orbiter.
Launch Director Pete Nickolenko said engineers planned to take advantage of the launch delay to troubleshoot that issue as well.