NASA will delay the launch of its next space shuttle mission until July in order to replace potentially faulty fuel sensors inside the orbiter’s massive external tank, the agency’s shuttle program chief said Tuesday.

Wayne Hale, NASA’s shuttle program manager, told reporters that the shuttle Discovery and its STS-121 return to flight mission will now launch no earlier than July 1, weeks later than its earlier May 10-22 flight window.

The extra time will allow for an invasive, three-week swap of four engine cut-off (ECO) sensors inside the liquid hydrogen portion of the orbiter’s 15-story fuel tank, he said.

“This is not a decision about schedule,” Hale said of the delay during a press conference at NASA’s Johnson Space Center in Houston, Texas. “This is a question of safety.”

Discovery’s STS-121 mission, commanded by veteran astronaut Steven Lindsey, is the second shuttle flight set to fly since the 2003 Columbia disaster that killed seven astronauts. It is also the second test flight before NASA returns to its International Space Station (ISS) construction launch schedule.  

Hale said he remained optimistic that NASA could launch three shuttle flights – STS-115 aboard Atlantis in late August and STS-116 in the fall – by the end of 2006.

ECO sensors are designed to monitor shuttle fuel tank levels and shut down an orbiter’s three main engines before its fuel tank runs dry.

But recent studies found that wiring defects in the manufacturing of some sensors could lead them to falsely report a dry tank, which could force an early engine shut down before a shuttle reaches its proper orbit, Hale said, adding that one such sensor on Discovery’s external tank has also been returning some errant readings.

“If a number of these sensors fail to the ‘dry’ state they would shut the engines down early, prematurely, which is not a good thing in spaceflight,” Hale said, adding that the chances of that occurring is admittedly remote. “This is what we call a criticality one, life or death, kind of situation in that you want those sensors to work properly either way…we need to have a good set.”

Errant ECO sensor readings scrubbed the attempted July 13 launch of NASA’s first post-Columbia mission – STS-114 also aboard Discovery – though the orbiter launched 13 days later without incident. Problems with the sensors also cropped up on a separate tank during an April 2005 fueling test.  

Hale said that unrelated problems have led shuttle engines to shut down early on two previous orbiter launches; STS-51F in 1985 and 1999’s STS-93 mission.

Tank engineers will replace all four liquid hydrogen fuel ECO sensors on Discovery’s External Tank-119 (ET-119) which now sits inside NASA’s 52-story Vehicle Assembly Building at the Kennedy Space Center in Cape Canaveral, Florida.

The tank will be hoisted into the vertical position, where engineers will remove its foam covered surface and pry open a manhole at its bottom to gain access to the liquid hydrogen tank, Hale said. After swapping out the old ECO sensors – which were built in 1996 – with new ones, engineers will then close out the tank and set it back in a horizontal cradle to reapply the foam insulation, he added.

“There are certain risks that you might damage the tank,” Hale said of the sensor swap operation.

Discovery’s launch window, which runs from July 1 to July 19, also allows shuttle workers about six additional weeks to close out several issues that have cropped up in the last few weeks.

Engineers will now be able to repair Discovery’s damaged robotic arm – which was dinged inside the shuttle’s hangar-like Orbiter Processing Facility on March 4 while workers tried to clean up broken glass from a light bulb that fell into the orbiter’s payload bay. The arm is instrumented to evaluate its performance in the upcoming flight, and would have had to be replaced with an un-instrumented arm with out the extra work time, Hale said.

Hale also said shuttle workers will be able to replace outer windows on Discovery’s cabin and remove additional foam from ice frost ramps on the orbiter’s fuel tank.

NASA has been working to reduce the amount of foam insulation aboard shuttle fuel tanks since 2003, when a chunk of foam fell from Columbia’s tank and breached the heat shield along its left wing leading edge, leading to the orbiter’s Feb. 1, 2003 destruction. The problem cropped up again during Discovery’s launch when cameras caught unacceptably large pieces of foam falling from the tank, primarily from a protective ramp that has since removed from current and future tanks.

All of the work to be completed on Discovery and its external tank should be completed well before the July launch window, especially since the agency was pushing hard to make the May launch window, Hale said.

“We were racing very hard to get to the mid-May launch period,” Hale said, adding that the extra six weeks could allow some shuttle workers to take weekend rests in what has been a daily effort to reach the May launch window. “I would expect it will allow us to slacken the pace is other areas.”