NASA Administrator Mike Griffin, answering questions on the eve of shuttle Discovery's launch Saturday, said today he firmly believes his decision to approve launch over objections from NASA's top safety manager and chief engineer was correct and based strictly on the technical merits of the argument.

Schedule pressure due to the Bush administration's 2010 deadline for finishing the international space station and retiring the shuttle played a role in the decision to proceed. But Griffin said the risk associated with so-called ice-frost ramp foam on the shuttle's external tank is relatively small and in any case, does not directly threaten crew safety.

If foam debris from the ice-frost ramps caused catastrophic impact damage during the climb to space - which Griffin said was a remote possibility - the astronauts could attempt repairs or move into the international space station to await rescue by another shuttle crew.

Even so, NASA's safety manager and chief engineer voted no-go for launch during a flight readiness review earlier this month after the engineering community formally classified the IFR foam as "probable/catastrophic." That means, safety chief Bryan O'Connor told CBS News, that over 100 flights - the original design life of a space shuttle - there would be a 50-50 chance of a catastrophic failure.

Discovery is the first shuttle ever cleared for flight with a system that is officially deemed an unacceptable risk. But Griffin said he was confident the decision to proceed was correct and based on sound engineering.

"Frankly, the decision to fly coming out of the FRR, which everyone seems to find so controversial, on the technical merits made itself," Griffin said today. "Flying the shuttle is not without risk for many reasons way beyond foam and in fact, I worry that we've spent so much time worrying about foam that we won't worry about other things which could get us. We've tried to address them all.

"Foam is a concern. But I very strongly feel that we are not risking crew for foam in this case, or I wouldn't feel comfortable launching. I believe I understand at a deep level the technical components that went into the decision and frankly, I think it makes itself. So I didn't regard it as (a difficult decision) at all.

"Now it was time consuming," Griffin said. "I personally, even as administrator, have spent weeks of my time on this issue to make sure I have heard everything everyone had to say and understood the analyses in a very deep way. So it was time consuming, but in the end, I think the data speaks for itself just fine."

Flight controllers at the Johnson Space Center in Houston today updated Discovery's launch time by a few seconds. The 10-minute launch window will open at 3:43:41 p.m. Saturday, but liftoff will be targeted for five minutes later, at 3:48:41 p.m., roughly the moment Earth's rotation carries pad 39B into the plane of the space station's orbit.

Astronaut Andy Thomas, a member of Discovery's crew for the first post-Columbia flight last summer, told CBS News today he agreed with the decision to proceed with flight. And like other astronauts, he said foam poses a relative risk, especially when compared to a shuttle's main engines, boosters and other high-energy systems.

"I think we are at the point where we do need to fly," he said. "I think Mike Griffin expressed that well, where you have to weigh the risks of where you are with the engineering of the foam against the bigger context of programmatic risk. And I think it's the right decision to go fly. We've got an obligation to the U.S. taxpayer, to our international (space station) partners and we need to meet those obligations."

Asked if the foam debate diverted attention from other risks that are routinely accepted for shuttle flights, Thomas said "You are absolutely right."

"You have to remember, this vehicle develops millions of pounds of thrust. When it leaves the launch pad, it's consuming fuel at a rate of 12 tons per second. There is turbo machinery in the back end of that spinning at unthinkable rates delivering unthinkable volumes of propellant. And burning it. This is very risky business. I think when you look at that, you come to recognize a fundamental truth about the shuttle. It really is, despite its problems, one of the most extraordinary human engineering accomplishments that has ever existed. It's quite amazing."

But the space shuttle, he added, is "very fragile and unforgiving of mistakes. And that's the problem it's displaying now."  

NASA's analysis indicates that in a worst-case scenario, foam weighing up to two-tenths of a pound could break away from an ice-frost ramp and cause catastrophic impact damage. But the analytical model does not match the shuttle's flight history. Of the tanks that have been photographed after separation from the shuttle, no pieces of IFR foam larger than about one-tenth of a pound have ever broken away and no heat shield damage has ever resulted. That does not mean it can't, just that it's never been observed.

And therein lies the problem. That uncertainty, plus a new understanding of a possible failure mechanism, drove engineers to classify the IFR foam as probable/catastrophic. But NASA's models are overly conservative and if they err, Griffin said, it's in the right direction.

"Their performance in flight has been better than the analysis would lead you to believe," he said of the ice-frost ramps. "That's good, I mean if you're going to not understand it fully it's better if the performance is better than the analysis. But nonetheless, you look at them and you say I'd like to get rid of them, I'd like a better design. When the analysis doesn't really replicate what's happening in the flight history, that tells you there's at least something about it you don't fully understand.

NASA is working on a redesign of the ice-frost ramps that engineers hope to implement by the end of this year or early next. But Griffin said he decided to accept the risk of launching now with the old design because of the potential impact of a long delay. NASA wants to launch the next two flights in daylight to ensure good photo documentation of the tank. But launching in daylight, ensuring external tank separation in daylight and launching into the plane of the space station's orbit greatly restricts the available launch windows.

"If we stand down for longer than three or four months, then we can't fly by this fall, which means we can't get daylight photography throughout ascent and at MECO (main engine cutoff)," Griffin said. "We want to be able to see the external tank throughout the flight, throughout ascent, and after we jettison it. We need to fly in daylight to get some of the kind of test data that we have been talking about over and over again. We need data. We need to be able eventually to resume night operations, but we can't legitimately take the risk to resume night operations until we understand how the tank behaves and see it in daylight.

"So if we stand down more than a few months, then we have to stand down until next spring," Griffin said. "So that would be the better part of a year.

"Now I've said repeatedly, it is a matter of national policy above me, we are flying the space shuttles for one reason and one reason only, and that is to finish the space station project, which we believe to be valuable. It has been subject of a presidential decision and it's been ratified by Congress.

"I've been asked repeatedly, can we do this? And I've said repeatedly, that in the 25-year history of the space shuttle program we've averaged, with accidents, with down time for cracked wiring and flow liners and all the other issues we've had over the years, we've averaged four-and-a-half flights per year. So if we can resume flying this summer and then just execute our average rate, just business as usual for the next four years, we can finish the space station project quite comfortably.

"I do not want to make decisions now, attempting to be ultra cautious now, when I believe it is not warranted for crew safety as I do," Griffin said. "I do not want to make decisions now which will back risk up into the latter years of the program by having us have to fly six flights (for example) a year in order to complete the station. And again, I keep saying I want to do a Hubble repair if that's technically possible. I believe that's worth it.

"So we are trying to balance the risks of one flight and one crew against the necessity of recognizing that we're flying these shuttles at all in order to finish the space station program. And I do not accept that it is good management on my part, I do not accept that the right thing to do is be extremely cautious now at the expense of something later on in the future."

Asked if he was "playing the odds" with Discovery's launch, Griffin said: "You're not going to like this and I'm sure I'm not going to like how it sounds in print, but we ARE playing the odds."

"As taxpayers, you pay us to play the odds," Griffin said. "It's called risk management. I can't accept that as a criticism. If someone were to say we were doing it incorrectly, I would be all ears. I'd want to fix that."

As for schedule pressure, Griffin said "there are no activities that humans undertake that don't have a schedule associated with them. It matters whether you finish a job this year or in the next decade."

"I would struggle to think of an activity that the government undertakes, and this is a government activity, that doesn't have a desired performance level, a desired allowable expenditure and a desired time frame to complete it," he said. "Managers get paid to balance performance, cost and schedule against risk.

"We have a schedule for flying out and retiring the shuttle and assemble the station. Just taking our average flight rate, it's a schedule we can meet. But we need to get on with things. I do not think, I absolutely do not think and do not accept we are being unduly influenced by schedule pressure. But we pay attention to schedule because time is money. And that matters."