NASA managers met Thursday and agreed to relax a self-imposed post-Columbia daylight launch constraint, clearing the way for the shuttle Discovery's liftoff Dec. 7, around 9:38 p.m. EST, on a complex space station assembly mission. It will be the first night shuttle launch since 2002.

But agency managers attending a program requirements change board meeting reserved the option of requiring daylight launches for flights with external fuel tanks featuring significant design changes.

One such proposed change is a modification of the tank's so-called ice-frost ramps, aerodynamically shaped foam insulation used to prevent pre-launch ice buildups on fittings that support external pressurization lines and a cable tray.

NASA and contractor engineers have been working on a redesign to reduce the amount of foam used to make the ice-frost ramps and thus minimize the possibility of debris shedding during ascent. But the ramps performed well during the past two missions and some engineers now favor leaving the design alone.

Another meeting to discuss the topic is planned for early November.

The ice-frost ramps are made up of foam poured into molds and then shaped by hand. Because the ramps are built up on top of already existing foam, engineers believed they were susceptible to temperature-induced cracks that could lead to in-flight debris shedding.

The ramps were officially classified as "probable-catastrophic" prior to the last two shuttle missions. That means if no changes were made, one could expect a 50-50 chance of a catastrophic failure over the 100-flight design life of a space shuttle.

Before mission STS-121 last July, NASA Administrator Mike Griffin said he did not agree with the probable-catastrophic designation and Discovery was cleared for flight as is. Likewise, Atlantis was cleared for launch last month on mission STS-115 with the understanding that a new ice-frost ramp design would be implemented as soon as possible.

As it turned out, no significant ice-frost ramp damage or foam shedding was seen during the last two flights. An ongoing engineering analysis, based in part on in-flight video of the tank, indicates what foam does separate in flight comes off after the tank is out of the dangerous lower regions of the atmosphere.

Going into Discovery's flight last July, the odds of a catastrophic IFR failure were believed to be between 1-in-75 and 1-in-100. The odds currently are believed to be in the neighborhood of 1-in-575 or better, according to a senior manager familiar with the ongoing analysis. As such, the ramps likely will not be classified as probable-catastrophic for Discovery's upcoming flight in December.

At the Kennedy Space Center, meanwhile, engineers are assessing options for fixing a radiator panel mounted on the inside of the shuttle Atlantis' right-side payload bay door. The panel apparently was damaged when a piece of space debris or a micrometeoroid slammed into the radiator, presumably during the shuttle's flight last month, blasting .108-inch-wide hole in the upper surface and destroying the aluminum honeycomb material below before exiting the other side.

The impact did not threaten the crew and the damage can be repaired. But it illustrates the danger posed by micrometeoroid/orbital debris (MMOD) and the reason why NASA considers such strikes a high risk. The odds of a catastrophic impact-related entry failure range between 1-in-210 to 1-in-350, depending on whether the astronauts inspect the ship in orbit prior to re-entry.

At orbital velocities, even tiny pieces of debris pose a serious threat. An aluminum sphere just .4 inches across moving at 10 kilometers per second, or 22,370 mph, carries the same impact energy as a bowling ball moving at 300 mph.

A preliminary engineering analysis shows the impact in question was one of the most significant instances of MMOD damage in shuttle history, second only to a cargo bay door impact during shuttle mission STS-72 in 1996.

The shuttle's 60-foot-long payload bay doors each feature four radiator panels that are exposed to space once the doors are opened in orbit. The forward two radiator panels measure about one inch thick, feature Freon coolant tubes positioned about 1.9 inches apart and can pivot to radiate from both sides. The aft panels are fixed and only radiate from one side. They measure a half inch thick and feature coolant tubes separated by about 5 inches. The interior of the panels is made up of an aluminum honeycomb material.

The impact on Atlantis's right-side, or starboard, radiator was found roughly midway between two coolant lines on panel No. 4. The object blasted a .108-inch-wide hole and presumably broke apart on impact. The resulting spray of debris created a cone-shaped damage cavity immediately below the face plate, destroying the honeycomb interior to the full half-inch depth of the panel. The lower face sheet was pushed out in two places. A .26-inch crack and a .03-inch-wide exit hole were found.

As part of NASA's post-Columbia safety protocols, the shuttle and space station fly in an orientation that protects the shuttle's critical wing leading edge panels from direct, worst-case impacts in the line of flight. In addition, the astronauts now carry out detailed inspections of the shuttle's heat shield after reaching orbit and again before re-entry to make sure no MMOD impacts occurred during the course of the mission.

Areas of the shuttle that are not critical for re-entry are not inspected and the damage to Atlantis's radiator panel No. 4 was not discovered until post-flight servicing at the Kennedy Space Center.

A senior NASA manager said the impact damage was unlikely to prompt a change in the shuttle's orbital orientation. He said program managers understand the radiator panels and cockpit windows face a higher risk of impact damage because of the shuttle's orientation. But he said the risks associated with wing leading edge impacts are more severe.

The shuttle has two Freon coolant loops and while the loss of one would force a crew to return to Earth at the next available U.S. landing site, computer software is in place to immediately isolate a leak even if the event occurred when the crew was asleep or otherwise occupied. Assuming a leak was quickly isolated, the astronauts could implement contingency procedures and press ahead with a near-normal mission despite the damage to one radiator panel.

As for an impact on the shuttle's cockpit windows, engineers do not yet know if the debris in question would have penetrated the thick multi-pane glass. The three panes making up the six forward cockpit windows have a combined thickness of 2.55 inches. The three panes making up each overhead window have a combined thickness of 1.58 inches.

Some 11,000 objects 3.9 inches in diameter or greater are currently tracked by Air Force radars and optical systems. About 100,000 objects are believed to be present that are between .39 inches across and 3.9 inches in diameter. Most of those are not tracked and pose a threat to spacecraft in low-Earth orbit and to communications and weather satellites in higher orbits.