Twenty years ago, Feb. 1, 2003: Engineers at Kirkland Air Force Base, New Mexico, capture this image of Columbia returning to Earth using a small telescope. Note how the left wing (towards the bottom) appears different from the right wing. The airflow around the leading edge appears deformed, and a plume extends behind wing.

*****

It’s February 1, 2003 and no one knows.  Columbia is completing its 28th flight, a 16-day science mission, STS-107.  It has performed nearly flawlessly.  At 2:30 a.m. EST, the entry flight team led by Flight Director LeRoy Cain comes on duty at Mission Control, Houston.  At 4 a.m., the hatch to the Spacehab research module is closed.  All is set for a landing at the Kennedy Space Center at 9:16 a.m.   The weather is looking good, after early morning ground fog dissipates.  Nothing that anyone knows can hold back Columbia.

The crew suits up and takes their seats.  Up on the flight deck, Mission Commander Rick Husband takes the lefthand seat, Pilot Willie McCool on his right.  Behind them, Mission Specialist Laurel Clark takes the right seat, and to her right, centered behind the pilot’s stations, serving as flight engineer, sits “K.C.” — Kalpana Chawla  Seated down in the middeck, Dave Brown, Mike Anderson and Ilan Ramon have nothing to looking at but a row of lockers, although a monitor shows the view of flight deck from a camcorder Clark is operating.

It’s 8 a.m. and no one knows.  Flight Director Cain goes around the room polling controllers, go/no go for the deorbit burn.  Capcom Charlie Hobaugh calls up, “I guess you’ve been wondering, but you are go for the deorbit burn.”  At 8:15:39 a.m., on its 255th orbit, flying tail-first over the Indian Ocean, Columbia fired the twin Orbital Maneuvering System engines for 2 min. 38 sec. 

No one knows that a 8-10 inch hole had been punched in the leading edge of the left wing on launch.  The perfect burn is sending Columbia to it’s fiery destruction. 

Once you know this, you don’t need to know what happens after Columbia reaches Entry Interface, the point where it first encountered the first thin touches of atmosphere.  That point occurs at 8:44:09 a.m. at an altitude of 76 mi., northeast of Honolulu.  You don’t need to know that super-hot air molecules entering the hole in Reenforced Carbon Carbon (RCC) panel #8 quickly begin burning insulation and attachment fittings, spreading left and right in the cavity behind the RCC panels.  Columbia continues on course, beginning a turn, the first of a planned three, to the right.  

The hot gasses impinge a spar, made of aluminum honeycomb, running the length of the wing behind RCC cavity.  At 8:52:16 a.m., the plume burns a hole in the spar which quickly enlarges, spreading hot gas into the interior of the wing.  Columbia is still 300 mi. from the California coast.

You don’t need to know that increased drag caused by the disruption of the flow of air around the wing pulls the nose to the left.  Dutifully, Columbia’s computers command the elevons to push the nose to the right.  The descent continues normally, the crew unaware of any problem.

At 8:53:28 a.m., the Shuttle crosses the California coast at an altitude of 42 mi.  Adhesives holding tiles and insulations blankets to the exterior wing surfaces begin to break down due to the heat loads.  Columbia begins shedding tiles and insulation.

The wheel box for the left main landing gear is just 4.5 ft. from the point where the spar is breeched.  The outboard wall of the wheel well begins melting. 

Shortly after 8:54 a.m., the onboard computers detect a change in the left wing, as if it’s lift has increased, pushing the nose to the right.  The trusses and spars in the wing have soften to the point where air pressure pushes up the bottom of the wing, actually increasing its lift.  The computers adjust the elevons again.  The descent continues to appear normal to the astronauts and Mission Control.

At 8:54: 25 a.m., Columbia crosses from California to Arizona.   The plume burns through the wall of the wheel well and strikes the main landing gear strut.  Columbia continues perfectly on course, initiating a “roll reversal,” a turn to the left.

Columbia crosses from Arizona to New Mexico.  The computers initiate increased elevon action to counter the roll to the right.  Pressure in the outboard main tire begins increasing.

You don’t need to know that 8:58:03, Columbia experiences a sharp change in aerodynamics, perhaps from a radical change in wing shape.  Yet the elevons still are able to counteract it.  The crew receives an alarm indicating a loss of tire pressure readings.  Husband makes his first call during the descent, “And, ah, Hou . . .” — and is cut off.  Communications are ratty, to be expected at times.

Capcom Hobaugh radios, “And Columbia, Houston.  We see your tire pressure messages, and we did not copy your last call.”

Husband calls, “Roger, uh, buh . . .” And is cut off.

Columbia is approaching Dallas.  The elevons have reached their limits.  Right-side jets, first two then four, fire in an effort to maintain a proper attitude.  Columbia’s nose swings further  to the left as the Shuttle begins rolling right.  At 8:59:46 a.m., a large piece of debris, possibly the wing, breaks free.  Then the vertical tail and possibly left engine pod break free.

At 9:00:02 a.m., two seconds of clear data are received showing that the fuselage is still intact, power and life support systems operating, but the vehicle is rapidly turning.  Likely in a flat spin.

You don’t need to know that when Columbia then breaks apart, the crew module falls intact, probably for 38 sec. before breaking up due the extreme forces and heating.

You don’t need to know — but you do.  You’ve known for twenty years. 

****

Go back 20 years before that time.  It’s February 1983 and the efforts to prepare Challenger for its maiden flight continue to hit delays due to maddening hydrogen leaks in its Main Engines.   By the end of the month, cracks are discovered in all three engines.  The year’s efforts, a schedule of five launches, a demonstration that the program is operational, is in danger.  Challenger’s launch date keeps slipping toward late March.

Finally the cause of the trouble is determined.  It stems from a modification to improve the engines.  Challenger’s engines were designed to provide 9 percent more thrust than the ones used on Columbia.  This will induce added vibration that could cause hydrogen lines to chafe against other parts in the engine.  So a protective sleeve was added to protect the lines.  Unforeseen, the sleeve made the lines more rigid and hairline cracks formed during the Flight Readiness Firings. 

In March, all three engines are removed and repaired.  Launch is set for early April.  This time Challenger looks good to make it.

There’s a lesson here:  Anytime you make changes, in an engine or, say, External Tank insulation, beware.