A DMSP weather satellite atop a Titan 2 rocket at the launch pad. Photo: Lockheed Martin

A U.S. military weather satellite that was seconds away from blastoff one year ago this week has remained Earth-bound due to a variety of technical problems, the latest of which technicians recently uncovered as they prepared to fuel the spacecraft for a February 1 launch.

The stage appeared set to finally launch the Defense Meteorological Satellite Program F16 spacecraft atop a refurbished Titan 2 missile from Vandenberg Air Force Base in California. However, the satellite's propulsion system failed to pass a routine vacuum decay leak test on January 16, satellite program officials said in a response to written questions.

The leak check was being performed in advance of loading hydrazine into the satellite for the upcoming launch attempt. Given the unsuccessful test, officials ordered the fueling be delayed until the source of the leak could be found.

"Further propellant loading preparation was halted in lieu of troubleshooting and resolution of this anomaly. Successful spacecraft hydrazine loading is a necessary step to continued booster/satellite launch processing," the Air Force statement said.

"Since the discovery of the anomaly, a team of government (DMSP Program Office), spacecraft contractor (Lockheed Martin Space Systems Co.), propulsion system subcontractor (General Dynamics), booster contractor (Lockheed Martin Astronautics), the Aerospace Corporation, and Vandenberg AFB personnel have worked to develop and execute testing to identify the cause of the anomaly and recommend corrective actions.

"We identified the source of the vacuum leak to be the valve on the spacecraft's hydrazine thruster No. 4. We are currently conducting testing to verify the integrity of this valve and system readiness for hydrazine loading."

Testing underway this week could lead to hydrazine loading as early as Saturday, January 26. The Air Force says, under that scenario, launch could be rescheduled for February 13.

"We are postured to immediately resume booster/satellite processing at the launch minus 16 day (L-16) point in the schedule after we successfully load the spacecraft hydrazine."

Officials say the thruster glitch is not related to the year's delay in launch.

The saga to launch this satellite began on January 20, 2001, when the countdown was halted at T-minus 3 minutes due to a ground support equipment problem.

A second countdown the following day reached T-minus 28 seconds when computers detected one of the rocket's fuel valves had not opened as planned. In the end, the valve worked just fine but a sluggish indicator switch didn't register the opening fast enough for the computers' liking.

The most serious problem occurred after the scrub when the DMSP's Inertial Measurement Unit -- the craft's navigation brain -- began acting erratically.

The gyroscopes within the IMU suddenly dropped out of flight mode; then the AC and DC power supplies switched from primary to backup systems on their own.

"If those anomalies occurred during the ascent phase of the launch, the likelihood is we would have lost the satellite and the mission," Col. Randy Odle, the DMSP program office director, acknowledged in an interview last year.

"Looking back perhaps fate was with us more than we thought. It is always nicer to fix problems on the ground than in space," Odle said.

"Whether we would have launched on the first attempt, perhaps the anomaly wouldn't have even manifested itself. But once we saw it, clearly we could not launch with that catastrophic potential."

After removing the satellite from the rocket and returning it to a processing building, engineers discovered a cable on the craft had broken due to poor workmanship during assembly.

By late June the spacecraft was fixed. But scheduling conflicts with other national security missions delayed the launch into 2002.

The U.S. military uses DMSP satellites in orbits around Earth's poles to collect the data that meteorologists need to generate forecasts for strategic and tactical planning.

Air Force officials say the cost of this launch -- including the satellite and rocket -- is now approximately $450 million. That's up $20 million from the price tag given before all of the delays.