Bill Gerstenmaier (search), the space agency official leading the investigation into the foam loss, said the shuttle's fuel tanks will need modifications, which eliminates any chance of launching in September.
The next available launch window would be November, then that would be it until next year because of strict lighting requirements needed to photograph any flyaway foam or shuttle damage.
A 1-pound slab of foam insulation broke off Discovery's external fuel tank two minutes after liftoff on July 26 and, unlike in the case of Columbia (search), missed hitting the shuttle.
Discovery's redesigned tank also lost smaller but still sizable and worrisome pieces of foam in four other areas, including the same spot where a large chunk came loose during Columbia's doomed liftoff in 2003.
"There's no immediate answer or problem that jumps out at us," Gerstenmaier said.
Two days after Discovery's safe landing in California, Gerstenmaier told reporters that of the 4,192 pounds of foam that was on the spacecraft's fuel tank, only about 1.2 pounds came off at undesirable — even potentially dangerous — times.
That was enough, though, to prompt NASA to ground future shuttle flights just one day after Discovery's launch.
Gerstenmaier, the program manager for the international space station, stressed that neither he nor other NASA officials are making excuses for the latest foam loss.
"Frankly," he said, "even the next time we fly the tank, I would expect to see a little bit of foam loss somewhere in the tank. I think it's an extremely difficult engineering problem to solve."
But the bottom line is "we're going to have to really understand why this foam came off."
The space agency's No. 1 worry is the big chunk of foam that came off a hand-sprayed section of the tank where cable trays and pressurization lines need to be insulated. The piece measured up to 36 inches long and 11 inches wide.
Also of concern is the next biggest piece of lost foam, an 8-inch-by-7-inch chunk that came off the same area from which Columbia's much bigger, deadly slice broke loose. Engineers suspect this foam loss may have been caused by a wire to a heater that was installed to prevent dangerous ice buildup during the loading of the super-cold fuel. The heaters were added to make up for the removal of the wedge-shaped blocks of foam in this area.
In February, just one week before Discovery's fuel tank was shipped from the Louisiana manufacturing plant to Cape Canaveral (search), workers repaired a small crack in the foam, in the same place where the biggest piece ended up breaking away.
Considerable work had been performed near that part of the tank, which may well have resulted in the crack, or indentation. The repair was standard and seemed to have been done properly, Gerstenmaier said.
"That alone probably wouldn't be enough to cause the foam loss that we saw," he said. "There's probably another underlying problem in there."
Nondestructive examination of that section of the foam, before the flight, showed what appeared to be thin lines or pockets of low-density areas, but the images are fuzzy, sort of like ultrasound pictures, he said.
Gerstenmaier said one short-term solution may be to remove the foam from this so-called PAL ramp area where the big piece came off, and reapply it in a better way.
Long-term options include removing the foam altogether from this location, adding fiber to the foam to make it adhere better, or attaching some sort of netting in select spots. But with a 2010 deadline looming for the retirement of the shuttles, it's uncertain how much time and money NASA is willing to spend.