NASA officials troubleshooting a sensor malfunction in the space shuttle Discovery said yesterday that the orbiter could launch on July 26 at the earliest and suggested they may have to reload its external fuel tank with supercooled liquids in order to duplicate the glitch so they can fix it.

"We've been dealing with launch vehicles" for some time, said Wayne Hale, deputy shuttle program manager. "A few days' delay to figure out what is going on is always the right answer. This is a business where you have to have patience."

NASA's launch team halted Discovery's countdown less than three hours before liftoff last Wednesday when a sensor malfunctioned in the bottom of the tank that holds the shuttle's hydrogen fuel. The sensors trigger an emergency shutdown of the main engines if they detect a fuel shortage during launch.

The scrub disappointed a host of NASA officials and dignitaries who had come to Kennedy Space Center to see the resumption of the U.S. human spaceflight program after the agency stood down for 21/2 years of self-examination and safety improvements following the Columbia tragedy.

NASA officials have gathered 12 teams of experts to diagnose the sensor problem and to try to fix it in time to launch Discovery within a window that ends July 31. If they miss the cutoff, they may have to wait until the next window in September.

"Right now we're still looking for the problem," shuttle project manager Bill Parsons told reporters during a televised Kennedy Space Center news conference. "We're going to continue down this path, trying to make this opening window."

Parsons said the launch team was examining several options if the troubleshooters fail to find a solution, including the possibility that planners may ask for and obtain a rule change allowing them to fly with only three of the four hydrogen sensors working.

Hale said the shuttle's original "launch commit criteria" allowed for only three functioning sensors, but the rule had been changed after Challenger exploded on launch in 1986. Now, however, planners have had "preliminary discussions" about whether to return to the original requirement.

Hale said NASA could also decide to extend the launch window for up to four days -- to Aug. 4. NASA has planned for a daylight launch so cameras and other imaging devices can get a complete picture of the orbiter's condition from liftoff until the external tank is jettisoned over the Indian Ocean 81/2 minutes later. Lighting conditions downrange begin to deteriorate after July 31.

Before the launch team makes a final decision, however, Parsons said the troubleshooters will work their way through an as-yet unsuccessful regime of tests and analysis at "ambient temperature" -- with Discovery's external tank empty of the supercooled liquid oxygen and hydrogen that boosts the orbiter into space.

If nothing comes of this work, "we'll have our conversation about what to do next," said Edward Mango, deputy director of the Orbiter Project Office. This will probably mean reloading the external tank with 1.6 million gallons of supercooled liquid to duplicate the conditions that existed last Wednesday.

Hale said "tanking" can take place July 26 at the earliest and will either be a test or part of a countdown that -- if all goes well -- could lead to a launch later that day. NASA lost a similar gamble on the launchpad Wednesday after engineers failed to resolve sensor malfunctions that had occurred during an April tanking test.

"We've told the team to concentrate on ambient temperature for the next day or two," Hale said. "At that point, if we still have no answers, we will talk about what we would do, why we would do it and when we would do it."