By Mark Ward and Diane Seligsohn INADEQUATE testing of one onboard system caused the Ariane 5 rocket to veer off course and break up on its maiden flight on 4 June. After six weeks of investigation and deliberation, the panel of inquiry appointed by the European Space Agency to investigate the accident delivered its verdict in Paris this week. As expected, the board blamed the inertial reference system, which controls the rocket’s trajectory. This keeps the launcher upright during liftoff, then, as the flight progresses, it gradually angles the engine nozzles to adjust the rocket’s course to reach the right orbit. Ariane 5’s inertial reference system is the same as the one used on Ariane 4, and it had worked flawlessly many times on the older rocket. But Ariane 5 weighs nearly twice as much as its predecessor and the three main engines generate more than twice as much thrust. So the rocket climbs quicker and its flight path is different. Although the inertial reference system was tested before 4 June on a turntable that provided an acceleration of 7g, many times the acceleration acting on it during the opening seconds of the flight, the tests failed to take account of the difference in horizontal velocity of the two rockets. An Ariane 4 rocket travels vertically during the initial stages of a flight, but the flightpath of Ariane 5 strays from the vertical much sooner. “On the turntable the mean horizontal velocity over time is zero but during the actual flight it builds up quickly,” says Colin O’Halloran of Britain’s Defence Research Agency in Malvern, a member of the inquiry team. Tests conducted after the disaster have confirmed that the inertial reference system would fail when the real figures for changes in horizontal velocity are used. Although the design flaw in the inertial reference system’s software is easy to fix, the bigger problem is ensuring that a similar fault never happens again. “The review and qualification procedures need to be tightened,” says O’Halloran. These changes are not going to come cheap. ESA believes it could cost up to 4 per cent of the £5.3 billion Ariane 5 budget to improve the review procedures. “Only in the course of the coming weeks will we be able to finalise the actions that must be taken and be able to assess the financial consequences,” says Jean-Marie Luton,
上一篇 ：A heavy responsibility - Removing lead from petrol is cheap and easy, so why are poor countries resisting attempts to phase out a substance that threatens the health of millions?