After a Taurus XL launch vehicle failed to loft the Orbiting Carbon Observatory (OCO) in February 2009, NASA used another Taurus XL to launch the Glory climate-monitoring spacecraft despite a recommendation from its own engineering safety office to ground the Orbital Sciences Corp. rocket until key components could be requalified.

The agency accepted a risk of a similar mishap on the March 2011 launch attempt that was calculated as high as 50%, a gamble that resulted in the loss of the $424 million mission when the vehicle's payload shroud once again failed to open and pulled the satellite into the ocean off Antarctica.

Since then, NASA has decided against using a Taurus XL to launch the replacement OCO-2 mission. Other Orbital vehicles, including the air-launched Pegasus and a new Antares rocket, use a version of the same fairing separation system that is most likely responsible for the combined $700 million loss of two key climate-study satellites. Orbital's original name for Antares was Taurus II.

So far, NASA has not accepted the Antares shroud-separation configuration for operational flights. Dulles, Va.-based Orbital says it has made a number of changes to its frangible joint fairing separation system in the wake of the Glory launch failure, including modifications to the frangible rail used on Antares. The company is developing that rocket under NASA's Commercial Orbital Transportation Services (COTS) program to carry cargo to the International Space Station (ISS).

The findings of a Glory Mishap Investigation Board review reached NASA headquarters in September 2011, but a public version will not be released until early next year. William Gerstenmaier, NASA's associate administrator for Human Exploration and Operations (HEO), says the failure analysis could not identify a root cause of the Glory accident, though investigative teams continue to explore possible root causes.

In the meantime, the fairing-separation system that currently equips Pegasus and Taurus—and which is planned to fly on Antares next year—has yet to be fully qualified, an issue that came to light in May 2010 when the NASA Engineering and Safety Center (NESC) recommended grounding the Glory mission until the frangible joint system could be sufficiently certified.

“The team recommends that the Glory mission not fly until the frangible joint system is adequately qualified per existing project requirements for its intended-use environments,” NESC stated in a May 27, 2010, report on its investigation into Taurus XL and other Orbital launchers that use similar frangible joint systems.

The NESC investigation was initiated by NASA's Science Mission Directorate (SMD) in March 2010 after a mishap investigation board failed to find the root cause of the February 2009 OCO loss. The NESC team's findings are detailed in a 35-page report that references more than a dozen supporting documents chronicling the system's engineering development and test history dating to the early 1990s.

According to the document, when NASA first considered Taurus XL to launch the OCO and Glory missions in the 2005-07 timeframe, a certification plan was established by the agency's Launch Services Program (LSP) and a Taurus XL qualification process was set in motion to certify the vehicle for NASA missions. This process was aided by the fact that Taurus had already launched seven military and commercial payloads to low Earth orbit at the time.

As a result, much of the qualification process involved a review of engineering drawings and other paperwork associated with the rocket, including results of a 2002 test requested by the Missile Defense Agency (MDA) that qualified a frangible joint system similar to the one on Taurus XL and Pegasus.

NASA uncovered a number of issues with the 2002 MDA test as it related to Pegasus. This led to the first recommendations for requalification of the Pegasus frangible joint system being suggested in February 2006, when NASA's Safety and Mission Assurance office (SMA) recommended “pursuing the analysis of the failures and a subsequent requalification of this frangible ring assembly.”

In March 2006, the launch service program's chief engineer said “the main goal is to reconvene to set requirements for requalification to fully retire this risk.” One year later, in April 2007, SMA stated: “Qualification should be reperformed to account for the missing low-temperature test. This test would also serve to fix the test setup errors that may have caused the incomplete fractures during the ambient and high-temperature tests.”

On Dec. 1, 2008, NASA SMA recommended Orbital requalify the frangible ring—part of the frangible joint system—to show “appropriate thermal margins and clear up any test setup errors that may have contributed to the qualification discrepancies that this risk was written for.” Less than two months before NASA flew the OCO mission in February 2009, the agency's technical authority echoed this position.

It is worth noting that the frangible joint system in question has flown successfully on Orbital Sciences rockets, including Pegasus, Taurus XL and Minotaur. Despite the fairing-separation system's apparent flight-worthiness, however, it has never been qualified by NASA for Pegasus or Taurus XL missions.

Ultimately NESC determined the likelihood of another Taurus XL launch mishap as a result of the unqualified frangible joint system to be as much as 50%, though the report says mitigation efforts could have reduced or prevented the risk.

“Probability of occurrence is moderate,” NESC states, characterizing the frangible joint system's status as “flightworthy versus qualified (meaning less of a known quantity).”

Before launching the Glory mission, NESC advised NASA's LSP to qualify the Taurus XL frangible joint system through additional tests, modeling and analysis, and a review of earlier qualification protocols to address concerns with the 2002 qualification test.

NESC also called for NASA and Orbital to develop a system-level model of the fairing deployment system to identify and evaluate additional risk factors and potential failure scenarios. Agency managers rejected the recommendations on cost and schedule grounds.

“There was a very overt discussion of what was the risk that we believed was in there versus the risk and the cost that would be tied up with a very long launch delay,” says NASA Chief Engineer Mike Ryschkewitsch.

Ryschkewitsch and Jim Norman, director for launch services in the HEO mission directorate, stress that there is no way to duplicate a shroud separation of the size in question on the ground, so flightworthiness decisions must be based on analysis, engineering judgment and experience.

Norman says NASA engineers have good insight into Orbital's engineering work on the Antares shroud separation flightworthiness. “Significant progress has been made,” he says. “We're on a good path. We are not yet done.”

Under the COTS contract, Orbital Sciences will have the final say on the first Antares launches. NASA won't start having a go-no-go decision role until operational cargo missions begin under the company's $1.9 billion, eight-flight Commercial Resupply Services contract, Norman says.