Mars Observer: piecing together the puzzle.

Astronomers had hailed the mission as the first U.S. venture to Mars in 17 years. And for nearly 11 months, as the Mars Observer spacecraft hurtled through space to keep its planetary appointment, it seemed destined to explore the Red Planet in unprecedented detail. But last Aug. 21, three days before Mars Observer was to begin orbiting the planet, the craft mysteriously fell silent (SN: 9/4/93, p. 149).

Now, a panel of independent investigators reports the likely cause of the craft's disappearance. Leaky valves probably allowed Mars Observer's oxidizer and fuel to mix and spontaneously ignite at exactly the wrong place: inside the system's fuel lines instead of at its rocket thrusters. Such premature ignition would have ruptured the fuel lines, causing fuel to spew out in all directions and spinning the craft out of control.

Several experts say that if the panel's conclusions are correct, NASA might have averted the catastrophe had it not decided, just months before launch, to delay pressurizing the craft until it reached Mars.

Panel leader Timothy Coffey, director of research at the Naval Research Laboratory in Washington, D.C., announced the investigations results at a press conference last week. In their report, Coffey and his coauthors emphasize that their suggested scenario, though "probable," is not conclusive. since ground controllers had turned off the craft's transmitter just before the Observer was lost. As a result, says Coffey. the panel lacked a "smoking gun" that would indicate exactly what went wrong.

After evaluating some 60 explanations for the craft's loss, the panel eliminated all but a few possibilities and focused on flaws in the propulsion system. Tests revealed that a group of valves, known as check valves, were leakier than expected. Common in sprinkler systems, check valves ideally allow material to flow in only one direction.

Aboard the Mars Observer, two sets of check valves allowed high-pressure helium gas to rush through, pushing out from their respective tanks the desired amount of either the oxidizer nitrogen tetroxide (NTO) or the fuel monomethyl hydrazine (MMH). At the same time, the valves were designed to prevent the two liquids from flowing backward and thus inadvertently mixing and igniting in the fuel lines.

Experiments revealed that during the craft's 11-month journey, as much as 2 grams of NTO could have seeped through the check valves, condensing on the fuel lines' cold titanium tubing and setting the stage 1or disaster.

On Aug. 21, ground controllers took final steps in preparation for Mars Observer to fire its thrusters and enter an orbit around the Red Planet. Temporarily turning off the craft's transmitter as a safety measure, they commanded the craft to blow open several pyro valves. With these rigid barriers out of the way, helium gas could flow though the check valves and pressurize the fuel and oxidizer tanks.

The absence of the pyro valves also meant that any NTO that had leaked through the check valves could mix with MMH, the panel conjectures. The enormous heat generated by the NTO-MMH mixture would have made titanium as soft as butter, rupturing the system and rendering the craft dead in space, Coffey says.

The panel charges that NASA's Jet Propulsion Laboratory (JPL) in Pasadena, Calif., which supervised the Mars Observer project, and its main contractor, Martin Marietta Astro Space, relied too heavily on designs for satellites orbiting Earth. Check valves function well, with little leakage, for short journeys in the warmer environment near Earth. But using them for a long interplanetary journey can prove problematic, says panelist Peter G. Wilhelm of the Naval Research Laboratory.

Indeed, NASA originally planned to rely on the check valves only for the first few days of the mission, says Mars Observer project manager Glenn Cunningham of JPL. The agency intended to pressurize the propulsion system five days into the flight rather than wait until the craft reached Mars 11 months later, when its thrusters had to be fired.

But those plans suddenly changed about six months before launch, Cunningham said. Scientists recalled that one of the Voyager spacecraft in 1976 had suffered a leaky regulator soon after its fuel system was pressurized, so they decided to hold off pressurizing Mars Observer until necessary.

Had JPL officials gone with their original plan, the NTO leak would have been far smaller (five days' worth of seepage rather than 11 months' worth), and the craft might have survived, Wilhelm says. Wilhelm adds that in the long run, debating the cause of the Observer's failure has less significance than ensuring that NASA addresses the myriad other problems found by the panel. If NASA doesn't, he cautions, "then the next time [they launch a similar craft], it will be something else that gets them."