NASA Traces Shuttle’s Leak to a Pump : Space: The launching of Columbia and its Astro observatory may be delayed again. It will depend on the timing of the Discovery mission.

The launching of the shuttle Columbia was thrown into confusion Thursday as NASA engineers traced the source of the hydrogen leak that postponed Wednesday’s liftoff to a recirculating pump in the shuttle’s aft engine compartment.

The space agency will replace the pump, delaying Columbia’s launching with the Astro observatory until at least early in the week of Sept. 17. But the craft may not get off the ground until much later than that.

Because NASA normally requires a three-week interval between launchings, this will bring the agency too close to the scheduled Oct. 5 launching of the shuttle Discovery with the high-priority Ulysses probe.

There is a possibility, however, that the Discovery launching will also be delayed by three or four days to repair a recently discovered leak in the ship’s Freon cooling loop. This would enable NASA to launch Columbia on the 17th and still have the required three weeks in which to prepare Discovery.

“That’s the one shot we have to safely get the Astro mission airborne” before Discovery, NASA Administrator Richard H. Truly said Thursday. But, if lengthy repairs to Discovery are not required and it appears that it can make the Oct. 5 date, added associate administrator William B. Lenoir, the Columbia mission will have to wait until after the launching of Ulysses.

Starting Oct. 5, the space agency has only an 18-day window to launch Discovery because the Ulysses probe must perform a slingshot maneuver around Jupiter to go into a polar orbit around the sun. If it isn’t launched during that period, NASA must wait 13 months until the planets are again in the proper alignment.

Space shuttle director Robert L. Crippen had said earlier that technicians here would perform a fueling test on Discovery before liftoff to ensure that it too does not have a hydrogen leak, a process that would add another five days to its preparation time before launching.

But in a Thursday afternoon press conference, Lenoir said that he believed the leak in the recirculating Columbia’s pump is a one-of-a-kind problem rather than a generic one and that the fueling test on Discovery would not be necessary.

The recirculating pump, about the size of a portable television set, pumps liquid hydrogen fuel through the shuttle’s three rocket engines to cool them before launching. Otherwise, they would receive a damaging thermal shock when the supercold fuel begins flowing through them at ignition. The pump is turned off after ignition and the fuel flows by gravity.

Physically replacing the pump and checking it will take only about two days once the process is begun Saturday, said Bob Sieck, launching director. The rest of the time will be used for normal countdown operations. Lenoir said that Columbia could possibly be launched before the 17th, but that it is “unlikely.”

Lenoir said that the leak was definitely not in the 17-inch “disconnect” valve that connects the massive external fuel tank to the shuttle. That valve was the source of a hydrogen leak that forced NASA to scrub both an attempted launching of Columbia on May 29 and a July launching of the shuttle Atlantis and led Truly to ground the entire shuttle fleet.

When engineers at Rockwell International in Downey disassembled that valve, they found that it was contaminated by microscopic glass beads that prevented a tight seal from being formed. Lenoir said that technicians had just disassembled another “disconnect” valve from the same batch of eight and found that it was also contaminated by the beads, suggesting that there was a problem with the entire batch.

Lenoir said there is evidence that at least one of the valves was in the same building where its manufacturer, Parker-Hannifin of Irvine, used the glass beads in a project unrelated to the space program. The agency will not use any of the other “disconnect” valves from that batch, he said, and “we think we no longer have a problem (with it).”

The new problems with Discovery and Columbia have capped what Truly called “a troublesome and vexing summer.” In addition to the shuttle grounding, the National Aeronautics and Space Administration has suffered through the discovery of a design flaw in the mirror of its showpiece Hubble Space Telescope, intense criticism about its plans for Space Station Freedom and the temporary loss of contact with the Magellan probe that is now orbiting Venus.

The discovery of another hydrogen leak has brought more predictable criticism of the troubled agency. Typical were the remarks of Robert Hotz, a member of the commission that investigated the 1986 Challenger explosion that killed seven crew members: “We discovered three years ago that NASA does not understand the problems of repetitive flying. The shuttle is just too complicated a bird.”

But Truly countered that the agency was doing its best and that flying the shuttle “is not an ordinary business.”