Update for 6:20 p.m. EST: Officials decided late today to press ahead with the repair option that calls for a clamshell-like sleeve to be welded around the line break. Mission managers will meet on Friday morning to select a new target launch date. A Sunday attempt is now considered highly unlikely. Below is the full story as posted at 4:20 p.m.:

The hydrogen vent line on the side of the launch platform develops a leak during fueling of Atlantis today. The large vapor cloud from the leak is easily visible. Photo: NASA TV/Spaceflight Now

The shuttle Atlantis' launch on a space station assembly mission is off until at least Sunday - and possibly longer - because of a dramatic hydrogen leak at the launch pad that raised the heart rates of reporters, if not shuttle engineers.

The leak occurred shortly after 9 a.m., about an hour and five minutes or so after engineers began fueling Atlantis for a 5:13 p.m. launch attempt. A few moments after switching from slow fill to fast fill, an alert engineer monitoring video cameras focused on the pad saw a cloud of vapor erupt from a hydrogen vent line.

The 16-inch aluminum line is one of two that carry excess hydrogen gas from the shuttle's external tank to a "flare stack" near the pad where the gas is released into the atmosphere in a controlled burn. The line only carries gas, not supercold liquid hydrogen.

"It was during fast fill on the hydrogen side that a facility vent line leaked," Leinbach said. "This was on the side of the mobile launcher, it's a pipe external to the mobile launcher. All leaking gaseous hydrogen was vented to atmosphere, there was no problem with the vehicle itself, there were no leaks inside the mobile launcher."

Video from remotely operated cameras on the east side of the pad showed a cloud of vapor spewing around the circumference of the double-jacketed vent line, presumably due to a cracked weld. The weld in question dates back more than 20 years and is a veteran of more than 80 shuttle fueling operations.

Close-up view of hydrogen vent line rupture. Photo: NASA TV/Spaceflight Now

"About a minute after he spotted the leak, we terminated flow on hydrogen," Leinbach said. "We talked about the situation for another 12 or 13 minutes while we continued to load oxygen. But we determined very quickly the leak was way out of spec and we were going to have no launch attempt today."

Hydrogen gas can spontaneously ignite in the atmosphere in concentrations greater than about 4 percent due to interactions with other components of the air. While the vapor leaking from the vent line appeared to be fairly dense, Leinbach said there were no sensors nearby to determine the actual concentration.

In any case, no fires broke out and Leinbach downplayed the danger posed by the leak.

"We didn't ever consider this a threat to the vehicle," he said. "Of course, during external tank load there's no one at the pad so there was never any safety issue with people. We just really don't feel it was an issue."

He confirmed that engineers are studying three options for repairing the line. If enough room is available, technicians will grind down the crack, re-weld it and use X-rays or penetrating dyes to confirm the integrity of the seal. That's the best-case scenario, one that would result in a launch attempt as early as Sunday.

A second, more complex, option calls for welding a clamshell-like sleeve around the crack. But that would add a day or two to the repair timeline. The worst-case scenario would require engineers to detach a 60-foot-long section of the vent line and lower it to the pad surface for repairs. The line then would have to be reattached and retested. When another launch attempt could be scheduled under that option is not yet clear.

Technicians move into position to get their first inspection of the vent line leak. Photo: NASA

But no decision on which option to pursue is expected before Friday, after engineers have a chance to carry out a thorough inspection. Access to the crack, high above the pad on the side of the shuttle's mobile launch platform, is another issue.

Under NASA's post Sept. 11 security plan, the agency will not announce a target launch time for Sunday or any other day, saying only that Atlantis will take off, if the pad can be repaired in time, between 2 p.m. and 6 p.m. The exact launch time will be revealed 24 hours before takeoff.

As reported here earlier, however, satellite tracking software shows a Sunday launch would be targeted for around 5:02 p.m. EDT. The launch time moves about 24 minutes earlier each day in accord with the space station's orbit.

Assuming Atlantis gets off the ground Sunday, here's a rough timeline of major mission events (in EDT and mission elapsed time):

DATE/EDT...DD...HH...MM...EVENT
   
04/07/02
05:02 PM...00...00...00...STS-110 launch
   
04/09/02
11:38 AM...01...18...36...Station docking
12:47 PM...01...19...45...Hatch opening
   
04/10/02
11:22 AM...02...18...20...Spacewalk No. 1 begins
           to attach truss to station
   
04/12/02
10:52 AM...04...17...50...Spacewalk No. 2 begins
   
04/13/02
10:52 AM...05...17...50...Spacewalk No. 3 begins
   
04/15/02
10:52 AM...07...17...50...Spacewalk No. 4 begins
   
04/16/02
12:22 PM...08...19...20...Farewell ceremony
12:42 PM...08...19...40...Hatches closed
03:22 PM...08...22...20...Station undocking
   
04/18/02
06:52 AM...10...13...50...Begin deorbit timeline
10:54 AM...10...17...52...Deorbit rocket firing
11:59 AM...10...18...57...Landing