Experts Suspect Wiring Or Connectors At Fault; Panel To Make Recommendations Tomorrow

NASA Loses Bid For Four Shuttle Launches In 2007, But Gerstenmaier Says Space Station Will Be Completed

Putting astronaut crew safety first, NASA has mandated a weeks-long troubleshooting delay for a scheduled space shuttle flight.

A maddening intermittent malfunction in an external fuel tank sensor system finally forced NASA to abandon plans for its fourth space shuttle launch this year, pushing the liftoff of Space Shuttle Atlantis into next month, NASA leaders said yesterday.

That ended days of battling the frustrating intermittent problem, which can’t be replicated on demand so that it might be diagnosed and cured.

All had seemed well in the smooth and flawless countdown leading to a scheduled Thursday afternoon blastoff of Atlantis from Launch Pad 39A at Kennedy Space Center (KSC).

But then, within hours of T minus zero, two low-fuel engine cutoff (ECO) warning sensors in the liquid hydrogen portion of the external fuel tank malfunctioned, and the launch was delayed to Friday. The fuel gauges are critical, because if the shuttle runs out of fuel with the engine pumps still running, an explosion might result.

Two liquid hydrogen ECO sensors gave false readings, and a third sensor failed after the tank was drained of fuel.

As NASA engineers wrestled with the problem, they discussed whether it might be safe to fly without all four ECO sensors working properly, and whether it would be permissible to launch if, inexplicably, the malfunctioning sensors suddenly began to give correct readings of wet or dry conditions in the fuel tank.

As marathon discussions continued, and experts wracked their brains to see whether they might have overlooked something, they thought that perhaps a launch could safely be initiated.

The Thursday launch was rescheduled to Friday, then to Saturday, then to yesterday, as debate-filled hours-long meetings were held each day.

Finally, with a determination that it would be safe to fly with the sensors and their data lines in place, crews began early yesterday to fill the liquid hydrogen and oxygen portions of the external fuel tank. But before long, incorrect sensor data were detected again, and at that point NASA leaders threw in the towel on the long battle and decided to begin a major sleuthing and repair mission on Atlantis, hoping somehow to locate the problem, or at the very least to narrow down the possibilities of its cause.

Repairs probably will be done on the launch pad, without the time-consuming effort of moving Atlantis back to the Vehicle Assembly Building at KSC for the trouble-shooting and repairs.

Space Station Completion Seen

The delay of Atlantis lifting off means NASA will have executed just three space shuttle missions to expand the International Space Station (ISS) this year, when an average of four missions yearly are needed to complete construction of the space station before the scheduled 2010 retirement of the shuttle fleet.

Only the shuttles have the size and power required to carry huge structural components into space needed to build the orbiting artificial moon.

But the delay in the launch of Atlantis won’t knock NASA off schedule in its manifest of planned shuttle launches, and the space station will be fully built before the shuttles retire, according to Bill Gerstenmaier, associate NASA administrator for space operations, speaking to reporters at a weekend briefing to detail the no-fly decision.

There is ample slack in the schedule to accommodate unexpected problems, he said. And if a huge delay occurs, he said, "if something really bad happens to us, we’ve got the ability to remove" some of the last shuttle flights from the current manifest of planned missions.

Gerstenmaier also pointed out that delaying the Atlantis launch from December into January "won’t impact the next mission," called STS-123, when Space Shuttle Endeavour on Feb. 14 is to lift off to carry the Kibo Japanese laboratory into orbit for addition to the ISS.

"There’s enough margin in the system" of scheduled launches to accommodate the delay in the Atlantis STS-122 Mission, Gerstenmaier said.

"We’ve got the capability easily to fly four flights a year," he said. Even if troubleshooting the sensor malfunction takes some time, NASA "can move the next flight [in February] a little bit," he said.

He termed the delay "a slight disappointment, but overall not a big deal."

As well, Gerstenmaier pointed out that NASA almost achieved four shuttle missions this year despite a huge problem earlier in 2007.

That also involved Atlantis, and also occurred when the shuttle had made it all the way to the launch pad. As the shuttle sat there, a rogue thunderstorm swept in and machine- gunned hail onto the spaceship, shooting thousands of dings into the foam insulation on the external fuel tank.

That required repairs that delayed Atlantis launching for months. Gerstenmaier said the immense repair job shows that NASA technicians "can execute," he said. "The recovery from hail damage … was phenomenal."

The star-crossed Atlantis also had another run-in with bad weather last year, when a bolt of lightning struck nearby and forced technicians to examine myriad electrical circuits to see whether the bolt fried them. Later, a hurricane pounded toward KSC, only to turn away at the last moment.

And now, Atlantis again faces trouble in getting into space.

LeRoy E. Cain, manager of launch integration and mission management team chairman, outlined the situation as NASA confronts the will-o-the-wisp sensor data glitch.

As tanking began, in the process of filling the external fuel tank, matters appeared to proceed well until "one of the sensors failed to the wet state," Cain said. That is dangerous, because if sensor data tell the shuttle crew and ground controllers that fuel remains in the tank when in fact it is running dry, a catastrophe could ensue because the crew wouldn’t be aware that they should shut down the engines to avert a possible explosion.

Because current criteria mandate that all four ECO sensors must be working properly before a mission, that violation forced a scrub of the Atlantis mission.

The tanks were drained and the countdown to launch was discontinued.

So engineers and experts are brain-storming where the problem might be. Rather than the fuel gauge sensors themselves, suspicion centers on the wiring, connectors, junction boxes and related items, the path through which sensor data must flow to be read by the crew and ground personnel.

The experts will report their findings to the program board tomorrow, Cain said.

One possibility is to begin filling the tanks again, see whether the glitch replicates itself, and also look at various indicators such as voltages flowing through the wiring at different points.

Just when Atlantis might fly depends on what the experts conclude, among other factors.

Problem Erupts Again

"We gave it a good try," but getting the shuttle off the ground proved impossible, Launch Director Doug Lyons said.

At first, as technicians began filling the external fuel tank, all seemed well. Sensors properly indicated that the empty external fuel tank was filling up, and a test of the sensors went well, too.

But then, "two or three minutes later, [ECO] sensor 3 failed," Lyons said, with voltages in lines suddenly rising, and indications registering that there was an open circuit somewhere.

And, since even one faulty sensor bars any launch, that was the end of attempts to get Atlantis off the ground.

Lyons said NASA team members, after the setback, were "disappointed but highly motivated" to figure out what is causing the problem, fix it and "get us back into a launch posture."

The NASA leaders said they will make every effort to determine just what is causing the sensor data malfunction.

But years ago, the same sort of problem arose, and NASA was unable to locate the root cause then.

Still, NASA leaders hope they will have better luck this time. "Our hope is that we can go do some trouble shooting," Cain said.

As to whether NASA might drop the current requirement that all four ECO sensors operate well before launch is permitted, "we’re going to look at everything," Cain said.

Whether the requirement for all four sensors to work without flaw will be retained is to be determined, Cain indicated.

If engineers could just see what is causing the failures, then the requirement might be changed, and "we might well go back to three of four" sensors working before liftoff.

"We’d like to find out what the actual problem is," Gerstenmaier said. But short of that, even finding out what does not cause the problem can be helpful, he added.

Best Of A Bad Situation

Meanwhile, NASA is left with unexpected weeks on its hands where there won’t be a space shuttle launching, or a shuttle crew working on the space station construction job.

So if life gives you lemons, make lemonade, Gerstenmaier indicated.

He said the idle time might be filled with a spacewalk by space station crew members, who might look at a problem with a solar alpha rotary joint (SARJ) that keeps a giant solar electrical generating panel correctly pointed toward the sun, for maximum electricity generating capability.

On an earlier shuttle mission, astronauts noticed filings or debris on the SARJ, and NASA wishes to determine the cause of the problem.

Space station astronauts may "see how they can optimize this [extra] time" by donning spacesuits and inspecting the SARJ, Gerstenmaier said.

How much time the space station astronauts will have on their hands, awaiting the arrival of Space Shuttle Atlantis, is unclear.

For example, less time might be required to fix the glitch if Atlantis can be diagnosed and repaired while it still sits on the launch pad, rather than having to transport it back to the Vehicle Assembly Building (VAB).

A rollback may not be required. "There is good access" to many parts of the sensor data path, while Atlantis sits on the pad, Lyons said. He sees little that can’t be done on the pad, just as well as in the VAB.

It is unclear, with experts holding differing opinions, as to whether technicians could replace a sensor on the pad, if that is required, Lyons said. That would require removing foam insulation, removing a panel, removing and replacing the sensor, and then replacing the panel and foam on it. But he noted that indications are the problem likely isn’t in the sensor.

"We have a lot of options," Cain said. Clearly, if a rollback is imperative, then "that does not support a Jan. 2 launch," he explained. He didn’t say how late in January the liftoff might occur if a rollback is required.

Meanwhile, the Atlantis crew has broken quarantine and returned to Johnson Space Center, where they will enter training sessions.

The Atlantis crew consists of Commander Steve Frick, Pilot Alan Poindexter, mission specialists Leland Melvin, Rex Walheim, Stanley Love and European Space Agency astronauts Hans Schlegel, from Germany, and Leopold Eyharts, from France.

They will take the European laboratory Columbus to the space station for attachment to the "Harmony" connecting Node 2 module that was installed on an earlier flight.

(Please see full story in Space & Missile Defense Report, Monday, Dec. 3, 2007.)