Discovery sat poised to begin the 100th shuttle mission today. But the mission was scrubbed in the early afternoon. Photo: NASA TV

Launch of shuttle Discovery on a space station assembly mission has been delayed at least one day -- and possibly longer -- because of concern about a suspect bolt in the ship's external fuel tank.

At the same time, engineers are assessing a sluggish valve in Discovery's main propulsion system that is used to help minimize vibrations during ascent due to the flow of propellants. Technicians may be ordered into the shuttle's engine compartment for an inspection or repairs. If so, the launch would be delayed until Monday at the earliest.

But in the meantime, liftoff of the 100th shuttle mission is tentatively targeted for 9:16:10 p.m. Friday, assuming the valve issue can be resolved and assuming engineers can convince themselves the external tank bolt in question will work properly in orbit.

If not, NASA managers could be forced to keep Discovery on the ground for additional analysis or, in a worst-case scenario, roll the shuttle back to the Vehicle Assembly Building for detailed inspections and/or repairs.

"We think it's prudent to stand down for a day to give our engineers time to review the data, to reveiw the rationale for flight and to move slowly and with all due concern for this issue," said James Halsell, director of shuttle operations at the Kennedy Space Center.

Jim Halsell, NASA's director of shuttle operations at the Kennedy Space Center, explains the possible consequences of the protruding bolt at a news conference. Photo: NASA TV

"In other words, we do not was to get 'go' fever, we want to give our people the opportunity to research this potential problem and to understand it fully before we press on."

During a routine engineering review of film shot by a camera aboard the shuttle Atlantis during launch last month, engineers noticed that one of the external fuel tank's two aft attachment bolts failed to work properly when the tank was jettisoned.

The 14-inch-long 2.5-inch-wide bolts extend from the tank into the shuttle's belly where they are secured with explosive nuts torqued with some 275,000 pounds of force. At tank separation, the nuts are detonated and the bolts are blown down into a housing in the tank.

During Atlantis's flight last month, the explosive nut worked properly but the right-side bolt remained extended some two-and-a-quarter inches. The concern is that a hung bolt could remain inside the shuttle attach point long enough for the orbiter to impart forces on the tank, potentially causing it to hit the shuttle.

"We do not know, and we have no reason to believe, we have any problems on STS-92," Halsell said.

The troublesome bolt is seen in this frame from a 35mm camera mounted in the belly of shuttle Atlantis. See an enlarged image with labels. Photo: NASA

But three teams of engineers are assessing the issue to determine if similar problems have occurred in the past; to re-asses the separation system's basic design; and to determine what, if anythine, would happen should another bolt fail to fully retract.

It is not yet known what, if anything, can be done at the launch pad to inspect the bolt in question aboard Discovery or to make repairs if any are required. If the shuttle must be hauled back to the Vehicle Assembly Building, launch likely would be delayed 10 days to two weeks at a minimum.

The next major space station milestone is the Oct. 30 launch of the lab's first full-time crew aboard a Russian Soyuz rocket. At least two days are needed between Discovery's landing and the Soyuz launch to give NASA engineers supporting the shuttle flight time to fly to Moscow for the Soyuz mission.

NASA presumably could roll Discovery back to the VAB, make inspections or repairs as needed and still launch the 100th shuttle mission in time to support the Soyuz launch on or close to Oct. 30. But at this point, that is little more than idle speculation.

This status report will be updated as warranted. In the meantime, the flight plan based on a launch Friday has been posted below, along with a revised ascent timeline.