Ongoing space shuttle wing leading edge impact tests show tiny cracks or even damaged surface coating, coupled with internal delamination, can lead to catastrophic failure.

What is not yet known is how small a piece of debris can be and still cause the kind of damage that, if unrepaired, could bring down a shuttle; and whether chunks of foam insulation from the shuttle's upgraded external fuel tank will be above or below that critical size, whatever it turns out to be.

"You're asking the $64,000 question," said Don Curry, a senior engineer at the Johnson Space Center in Houston. "We're trying to establish that right now."

The issue is important because NASA's new external fuel tank, now at the Kennedy Space Center undergoing launch processing, was built to ensure no pieces of foam heavier than 0.023 pounds - 0.37 ounces - can break away from the upper portions of the tank during launch.

Debris from the upper part of the tank poses the greatest threat to the ship's nose cap and reinforced carbon carbon leading edge panels, which experience the highest heat loads during re-entry. The 0.023-pound limit is for foam covering the liquid oxygen tank at the very top of the structure.

Foam from the intertank area, where bipod struts attach the nose of the shuttle and where the foam originated that doomed Columbia, can be slightly more massive: 0.03 pounds, or 0.48 ounces. Pieces of foam from the bottom of the tank pose less of a threat and thus can be larger still: up to 0.75 pounds.

Here is a partial table of allowable debris as approved by shuttle program managers:

REGION...................................FOAM (pounds/cubic inches)

Oxygen tank....................................0.023/15.9
O2 tank to Intertank flange.................0.026/20.7
Intertank.........................................0.03/20.7
Hydrogen tank and intertank flange........0.03/20.7 to 0.075/51.8
LH2 tank.........................................0.075/51.8

Engineers using a nitrogen gas-powered cannon and a wing leading edge mockup at the Southwest Research Institute in San Antonio, Texas, have been able to create "entry critical" damage from pieces of foam as small as 0.044 pounds - 0.7 ounces - depending on impact velocity and angle.

But they have not yet been unable to determine the damage threat posed by 0.03-pound pieces of foam because the small, low-density foam test "bullets" are deformed too much when fired from the cannon.

NASA hopes to launch the shuttle Discovery on the first post-Columbia mission in mid May. Curry said the engineering community is working to refine test methods and computer models to verify that pieces of foam debris weighing 0.03 pounds or less pose no threat to the shuttle.

But the testing already has revealed that entry critical damage can be caused by much smaller impacts than previously thought.

Columbia's left wing was damaged 81 seconds after liftoff Jan. 16, 2003, when a 1.67-pound chunk of foam broke away from an aerodynamically shaped ramp insulating one of two bipod fittings used to attach the nose of the orbiter to the tank.

The briefcase-size piece of foam slammed into the left wing at a relative velocity of more than 500 mph. Enhanced video from the one camera that viewed the impact point indicated the foam struck the leading edge at or very near the lower side of reinforced carbon carbon panel No. 8, one of 22 such panels making up the leading edge of the left wing.

But the grainy video, unable to resolve anything smaller than two square feet, provided no direct evidence of actual damage. In the wake of the mishap, the Columbia Accident Investigation Board, working with NASA, decided to conduct a complex series of tests to find out whether impacts by low-density foam could, in fact, cause the kind of damage needed to bring down the shuttle.

Initial tests at SRI showed foam impacts could cause damage, but the results were not clear cut. Then, on July 7, 2003, a properly sized foam bullet was fired at the lower side of RCC panel 8 using an impact point, angle and velocity thought to more closely mimic conditions possible during Columbia's launch. The result was a gaping hole 16 inches across.

Few engineers believe Columbia's wing had a hole that size and, based on more recent test data, it likely was on the small end of the 6-to-10 inches cited by CAIB investigators.

The question today is, what is the threshold for entry critical damage, that is, what is the smallest size piece of foam that could cause the sort of damage that would have to be repaired before a crew could safely return to Earth?

To find out, engineers are running exhaustive tests and analyses to define, for each of the 22 RCC panels making up a leading edge, what sort of impacts cause entry critical damage. Anything at or above that level must be repaired.

Because the RCC panels are custom built to match the sweeping curvature of the wing, each panel is divided into six zones and each zone is assigned its own number to more accurately model real-world conditions.

To find out what level of damage leads to oxidation, or burning, Curry said two approaches are in use. In one, 2.8-inch disks of RCC material are deliberately damaged and then subjected to re-entry heating with an arcjet plasma. The arcjet can produce temperatures of 2,960 degrees, closely matching the maximum experienced during re-entry.

Initial tests showed even small cracks could produce severe damage if the cracks penetrated a surface coating down into the underlying material, or substrate. Intrigued, engineers ran tests using a piston mechanism to push on a disk to impart loads, or forces, that could cause interior layers of RCC material to delaminate, or separate.

The results showed that as long as the surface coating remained intact, the test disks survived the arcjet without any major damage, even if delamination was present. But if the surface coating was damaged or removed, and if delamination was present, the reinforced carbon carbon burned up.

And in all cases, Curry said, "if we break the coating, we will have a delamination."

As a result, coating damage alone may be classified as entry critical. A shuttle could, in theory, be brought down by an impact that did not cause a hole in an RCC panel, just a visible crack in the coating, along with internal delamination. But engineers have not yet identified the debris threshold for causing such damage.

It is a complex problem because the force imparted to a leading edge panel depends not only on the size of the debris, but also the impact velocity, the angle of the impact relative to the surface of the RCC panel, the precise location of the strike and the panel's location on the leading edge.

To improve accuracy, engineers now are running so-called "wedge" tests in which the RCC is mounted at an angle to the flow of the arcjet to more realistically mimic real world conditions.

In disk tests, where the arcjet hits head on at a 90-degree angle, damage grows in a circular fashion as one might expect. In wedge tests, which more accurately reflect reality, the damage is more elliptical in shape with the worst heating occurring toward the downstream area of the ellipse.

Engineers currently are working to model damage patterns in each of the six zones in each RCC panel of a leading edge. Computer models are being refined to predict damage based on multiple variables and so far, the models show good correlation with the disk tests. Wedge models are more complex and are not yet fully developed.

In any case, Curry said Discovery's astronauts, using a boom-mounted camera and laser sensor package to inspect the shuttle's wing leading edge panels in orbit, will be able to spot the sort of surface coating damage that might pose a threat.