Math error equals loss of Mars orbiter.

Two summers ago, NASA knew the thrill of victory when its tiny robotic spacecraft landed on Mars within kilometers of its target. Last week, after failing to properly use the metric system, the space agency learned the agony of de-feet.

NASA reported Sept. 30 that it had lost the $125 million Mars Climate Orbiter because the force exerted by the orbiter's thrusters remained in the system of units based on pounds and feet rather than being converted to metric.

The problem, believed to have originated before the craft's launch last December, wasn't caught until days after Climate Orbiter vanished on Sept. 23 (SN: 10/2/99, p. 214). It had dipped 100 kilometers lower than planned into the Martian atmosphere.

"Truly, it is just dumbfounding, flabbergasting--all those superlative adjectives--that this could possibly happen," says space-policy analyst Marcia S. Smith of the Congressional Research Service in Washington, D.C.

A preliminary review has now found that the problem doesn't plague the Mars Polar Lander, scheduled to arrive on the Red Planet on Dec. 3, says Carl Pilcher, NASA's director for solar system exploration. Two NASA committees and an independent panel are investigating why the Climate Orbiter blunder went unnoticed.

The problem arose because two teams working on the Mars mission weren't using the same units of measure. Scientists at NASA's Jet Propulsion Laboratory in Pasadena, Calif., had assumed that thrust data they received from Lockheed Martin Astronautics in Denver, which built the craft, were expressed in metric units, as newtons. Although propulsion engineers typically express thrust as pounds of force, it's standard practice to transform these to newtons when integrating the information into the design of a spacecraft, says Noel W. Hinners, vice president for flight systems at Lockheed Martin Aeronautics.

Somehow, no one did that. "We should have converted," he says.

One pound of force is roughly 4.45 newtons. Moving from one set of units to another boosts the chance for miscommunication, and "there are very few software packages that would avoid such an error," says Peter G. Neumann of SRI International in Menlo Park, Calif.

In 1985, he notes, controllers calculated distance in feet rather than nautical miles and inadvertantly pointed a mirror on the space shuttle Discovery away from Earth instead of toward a laser on Hawaii's Mauna Kea.

Pilcher notes, "The particular nature of the [Orbiter] error is less important than the fact that it was not recognized and corrected." Neumann and his colleagues are developing software that can check for consistency and reliability of data in spacecraft systems.

"Twenty years ago, we went through this whole hassle of, Should the U.S. go metric?" says Hinners. "I wish we had."