Acting NASA administrator William Graham said yesterday that the space shuttle's solid rocket boosters were considered so reliable that no provision was made to detect the sort of failure that appears to have led to the explosion of Challenger last Tuesday.

The solid-fuel boosters, Graham said on CBS' "Face the Nation," "were considered primary structures and not susceptible to failure."

The National Aeronautics and Space Administration, he explained, felt it was virtually impossible for such a failure to occur and therefore saw no need to equip the shuttle with a means of detecting it during the flight.

Had Challenger's commander or pilot been aware early enough of the impending disaster, Graham said, they could have taken steps to abort the flight by separating from the solid-fuel booster and the hydrogen tank and trying to glide back to the ground. A NASA spokesman said later, however, that the crews had never trained for such a maneuver. NASA documents state that such an escape is possible only after the boosters have completed firing. That would have been nearly a minute after Challenger blew up.

The fact that Challenger was in trouble at least 14 seconds before the explosion emerged publicly late Saturday night when NASA released a new film that showed the shuttle from a different angle than did the previously broadcast footage.

The new film, taken by an automatic camera that could see the right-hand booster, shows that at 58.32 seconds into the flight, flames from the booster had burned a hole through the booster's steel wall and were blasting like a blowtorch on the adjacent tank of highly explosive liquid hydrogen.

It was another 14 seconds, however, before the jet of hot exhaust gases caused the hydrogen tank to rupture and explode.

Discovery of the burn through, possibly narrowing dramatically the search for a cause of the disaster last Tuesday that killed Challenger's crew of seven, led some NASA sources to speculate that corrective action can be taken quickly enough to resume shuttle flights as early as June.

Asked about that timetable, Graham did not deny it but refused to speculate on how soon the remaining shuttles in NASA's fleet would be allowed to resume flying. Graham did say, however, that it would not be long before space shuttles begin flying again. "The overall configuration and design we believe to be fundamentally sound," Graham said, "and we don't think it'll take a very long time to get this problem corrected."

Graham defended the lack of sensors to detect a booster burn-through by comparing the boosters to the wings of an airplane. Just as airplanes cannot be designed to fly if a wing fell off, so the shuttle cannot be designed to fly without one of its boosters.

"These very heavy steel casings that constitute the structure of these solid rocket boosters," Graham said, "are some of the sturdiest parts of the entire shuttle system, were considered primary structures and not susceptible to failure. Of course we designed them that way. It wasn't just chance that they're not susceptible -- or we thought them not susceptible -- but they were designed with great care and great thought and everything that we could do to keep them from having any failure modes."

Graham, who made the rounds of all the Sunday morning talk shows, put NASA's position another way on NBC's "Meet the Press":

"There's no credible failure mode that we have identified in the engineering design of those [solid rocket boosters] and therefore there was no reason to instrument them against incredible failure modes."

"Credible failure mode" is engineering jargon for a failure that can be imagined as possible under realistic circumstances. A problem that no knowledgeable engineer believes possible is "incredible."

NASA sources said the boosters did have additional sensors during the first four test flights of the shuttle in 1981 and 1982. But, once it was felt the boosters performed well, they were removed from subsequent flights to save weight.

"If there's a problem with the solids or with the expendable tank at any time in the ascent," Graham said on "Face the Nation," "the pilot can detach the orbiter from the tank at the three mount points . . . and roll the orbiter in such a way that he is now able to . . . return to the launch site at Cape Kennedy."

"If they had had any warning," Graham said, "the normal safety plan would have been to detach at the mount points and return as an unpowered glider to the landing site at the Cape."

John Lawrence, a NASA public affairs officer at Houston's Mission Control Center, gave a different view. "As I understand the abort system," he told The Washington Post, "you are married to the solid rocket boosters through the point at which [they finish burning]. I've been through hundreds of simulations and I personally have never seen a scenario where you jettison the SRBs when they're still burning."

According to a NASA engineer in Houston, there is a rotary switch in the cockpit facing the pilot that says "Abort Engage." Throwing the switch triggers the explosive bolts that free the orbiter from the large external tank and the boosters.

The move is considered extremely risky, however, because it takes some time for the now-unpowered orbiter to maneuver itself away from the boosters. Freed of the load of lifting the orbiter, the boosters would move ahead as the orbiter slowed. Their exhaust flames, which reach hundreds of feet behind, could burn the orbiter.

There is no indication, however, that Challenger's crew knew anything was wrong. Flight Director Jay Greene in Houston has said he also did not know what was happening, and he speculated that he might have known had there been temperature sensors on the external fuel tank. One theory as to how the ruptured booster caused the explosion is that its heat warmed the hydrogen in the external tank so much that it expanded to a pressure that burst the tank walls.

"It would have been nice to have fuel tank temperatures during ascent," Greene has said, "but we've had no operational use for those temperatures and so we don't have them."

The investigation of the crash is being conducted by a NASA board, using data gathered from all the space centers, including the film footage made public Saturday night, and an examination of 13 tons of debris recovered from the ocean in a massive search operation.

Off Cape Canaveral yesterday, six Navy and Coast Guard ships assisted by seven helicopters and cargo planes searched a 6,300-square-mile area extending 80 miles offshore.