NASA originally planned to follow the June flight of Endeavour by launching the original shuttle Columbia in July on a 16-day microgravity research mission featuring the first Israeli astronaut. Atlantis was to follow suit Aug. 22 with Endeavour delivering P1 in mid October.

But during routine inspections of the shuttle Atlantis on June 17, engineers discovered a small crack in a liner inside the 12-inch-wide liquid hydrogen feed line leading to main engine No. 1. Two more cracks in the same flow liner were found the next day.

These images show a crack found on a liner in shuttle Discovery's main propulsion system. The top photo is magnified 30 times and the bottom photo is magnified 100 times. Photo: NASA

Similar cracks then were found aboard Discovery and then Columbia, at which point shuttle program manager Ronald Dittemore suspended work to ready Columbia for the July flight. Cracks then were found in flow liners aboard Endeavour and the shuttle fleet was grounded.

The concern was that if a crack worsened in flight and a piece of debris broke off, it could get sucked into a main engine turbopump at high velocity, triggering a catastrophic failure.

"My concern from a safety point of view has been diminished because of (tests and analyses) over the past several weeks," Dittemore said during a news briefing on July 12. "That's not to say I'm ready to go fly. I still need to understand more about these cracks and I still need to understand more about the potential for these cracks to grow."

Until then, he said, "we will not fly."

Eleven cracks were found in all, three each aboard Atlantis, Discovery and Columbia and two aboard Endeavour. Seven of the cracks were found in flow liners leading to a shuttle's No. 1 main engine, the one located directly under the ship's vertical stabilizer. The other four cracks were found in liners leading to the No. 2 engine position.

Six cracks were circumferential a five were axial. Five were discovered visually, two by ultrasound and four using eddy currents, an electrical test that can find areas of weakness in an alloy.

Here's the breakdown (all measurements in tenths of an inch):

ORBITER...DATE FOUND..LOCATION...SIZE..DIRECTION........INSPECTION

Atlantis..06/17/02....SSME1/LH2 .2"....Circumferential..Visual
..........06/18/02....SSME1/LH2 .2"....Circumferential..Eddy current
......................SSME1/LH2 .2"....Circumferential..Eddy current

Discovery.06/24/02....SSME1/LH2 .15"...Axial............Visual
......................SSME1/LH2 .20"...Axial............Eddy current
......................SSME1/LH2 .10"...Axial............Eddy current

Columbia..07/03/02....SSME2/LH2 .2"....Circumferential..Visual
..........07/08/02....SSME2/LH2 .2"....Circumferential..Ultrasound
......................SSME2/LH2 .2"....Circumferential..Ultrasound

Endeavour.07/09/02....SSME1/LH2 .25"...Axial............Visual
..........07/10/02....SSME2/LH2 .3"....Axial............Visual

In addition, engineers found a similar crack in a main engine test article used to test fire shuttle engine clusters many years ago at the Stennis Space Center in Mississippi.

"Even though we've found cracks, it's not age related," Dittemore said. "Whether I've flown the vehicle 16 or 17 times or 30 times, the data appear to show the cracks are present and relatively about the same size."

He said the cracks could be the result of stress when the flow liners were initially installed or the result of some as-yet-unknown environmental factor that is present when the engines are running and supercold liquid hydrogen is flowing through the lines.

Scott Minnick, with United Space Alliance, places a fiber-optic camera inside the flow line on Endeavour. Photo: NASA

In the end, Dittemore approved a plan to repair the cracks using a welding technique and to polish out tiny defects that could evolve into cracks over time. The weld repair was approved during a meeting Aug. 1. At another meeting two days later, Dittemore formally approved a revised shuttle launch schedule that called for launching Atlantis Sept. 28 and Endeavour around Nov. 2.

Columbia's microgravity research flight ultimately slipped to mid January while Atlantis was delayed to Oct. 2 because of problems with NASA's crawler transporters used to haul shuttles from the Vehicle Assembly Building to the launch pad.

"From the operational standpoint, ascent will look just as it always has," Engelauf said of Atlantis' repaired flow liners. "We haven't changed the instrumentation of the vehicle in any way, shape or form. (But) I think from an engineering standpoint, there will be an increased amount of attention on engine performance and post-flight inspections."

For their part Ashby and Melroy had nothing but praise for the way NASA handled the fuel liner crack issue.

"I personally was very impressed with the way NASA approached it, with very little pressure to get back on a launch schedule, but giving the crew, the team that worked it, plenty of time to properly work through the fault analysis, to decide what had happened and where to go from there," Ashby said. "From what I saw, it was a very professionally completed process and I feel very comfortable that we're better than we were a few months ago as far as those engines."

Said Melroy, who flew to the Cape at one point to meet the welders: "They were unbelievably professional. ... Those people really know what they're doing."

CONTINUE TO NEXT PAGE -->