The Columbia broke up at 7:59:32 a.m. CST on February 1, 2003 and after 8:16 a.m. it was considered overdue for touchdown at Kennedy Space Center.  It was well before 10:00 a.m. when the first media reports began suggesting that a piece of debris, most likely insulation from the External fuel Tank (ET), was responsible for damaging the Columbia's thermal tiles leading to the breakup of the orbiter.  The first detailed news reports indicated that either the leading edge of the orbiters wing may have been seriously damaged or possibly some number of tiles on the underside of the wing were damaged or loosened.  Television news began showing the video of Columbia's launch on January 16^th where 82 seconds into the launch some piece of debris can be seen traveling underneath Columbia's left wing.

 Foam Impact During Ascent

View: Observational Analysis E1A (Examining The Official RCC Panel Debris Impact Test)

View: Observational Analysis E1B (The Probability of Damage to the Space Shuttle From Foam Debris Impact)

What hit the Columbia:
Speculation as to the nature of the debris said that it was possibly ET insulation foam, insulation foam soaked with water and then frozen, (making it much heavier than dry foam), or the debris itself was a piece of ice.  It was also thought possible that the debris could have been a piece of one the SRB's.  Between January 16 and February 1 NASA's own studies on the nature of the debris ranged from between a single piece of foam 20" x 10" x 6" and three pieces of foam ranging up to 20" x 16" x 6" in size.  The density of the foam is 2.4 lbs./Ft³ so the weight of the largest piece of foam would be 2.1 lbs.  The estimated speed of impact was estimated to be between 500 and 750 Ft./Sec.

This is the amount of kinetic energy being carried by a 2.1 Lbs. piece of foam traveling at 700 Ft./Sec.  It is approximately equal to dropping a 20" computer monitor 267 feet.  However, the debris did not dump all its energy into the wing of the Columbia.  The debris hit at a very shallow angle and lost only a tiny bit of its momentum.  Perhaps only 1% or less of this energy was inflicted upon the Columbia.  This is only intended to show that the interest in E.T. foam hitting the shuttles tiles is of real concern.

Where the Columbia was struck:
The debris appears to pass under the left wing without stopping or coming into contact with Columbia.  However, NASA reports that impacts occurred underneath the left wing outboard of the landing gear door.  The debris impacted the wing at an angle of no more than 21° but in most places it was a lot less.  The following NASA reports conducted between January 21 and 24 indicate that Boeing engineers did a very thorough study of what was shown on the tape,  COL_DEBRIS_BOEING_030121, 030123, 030124.

The official damage assessment:
The space shuttle orbiter uses several different materials to make up the Thermal Protection System (TPS) but the only two involved in the impact area are the black ceramic tiles which cover the aluminum skin of the orbiter and the Reinforced Carbon Carbon (RCC) molded sections that makes up the leading edge of the wing.  All of the following charts and diagrams are from COL_DEBRIS_BOEING_030121 and _030123.  This page is simply a review of those documents but is critical being that it paved the way for Columbia's final reentry.  Fig. E1 is the initial predicted impact area as determined through post launch analysis of ascent video from different camera angles.

Fig. E1

Fig. E2 provides a distribution of potential velocities and impact angles depending on exactly where a debris strike occurs.  These values are determined based on the trajectory of the debris.

Fig. E2

The engineers first analysis using the "Crater" computer program gave what were assumed to be overly conservative values.  According to the results one of the impacts would have penetrated 4.7" into the surface of the shuttle leaving a hole 27.2" long and 5.8" inches wide.  Fig.E3 is a spread sheet of the Crater results for the largest estimated piece of foam.  Fig. E2 can be used to find the locations shown in Fig. E3 using the X and Y locations.

Fig. E3

The engineers then looked at data from past shuttle missions and studies done by impacting actual test specimens with foam and other material, Orbiter Tile Impact Testing.  The following summary, Fig. E4, from, COL_DEBRIS_BOEING_030123, shows a more realistic prediction for tile damage on STS-107.  (NOTE: This data is still quite conservative)

Fig. E4

The following chart, Fig. E5, also from, COL_DEBRIS_BOEING_030123, is predicted damage to the RCC material on the leading edge of the wing based on impacts from solid ice.  Since the foam insulation is much softer it was assumed that the damage would be far less for the same angle and velocity of impact.  The following statement, "RCC is clearly capable of withstanding impacts of at least 15 degrees...", is from the same document as the chart.

Fig. E5

Conclusion:
This analysis showed only surface damage done to the tiles.  Chips of varying depths and sizes that would destroy the integrity and therefore the lifespan of the tile but should not jeopardize the shuttle during a single reentry.  Such damaged tiles and RCC panels are routinely either repaired or replaced after each mission.  Engineering methodology dictates that you go with the most reasonable solution to a problem which was the past shuttle mission knowledge base and the impact studies, and not the grossly over conservative data produced by the "Crater" software.

The following is the conclusion of the analysis done by Boeing engineers and published in COL_DEBRIS_BOEING_030123 and was the major reason the Columbia reentered the atmosphere with little or no concern for its safety.  There does not appear to be anything grossly negligent about the post launch debris impact analysis performed by Boeing engineers.

Fig. E6 shows what the expected damage should be to the wing leading edge and lower tile surface while Fig. E7 summarizes what the results of that damage to the TPS would be during reentry of the shuttle.

Fig. E6

Fig. E7