By lex, on May 17th, 2010
When I was a junior officer at NAS Lemoore, California, we lost an FA-18 shortly after departure due to a loss of thrust. It was a hot day – if you’ve never been to Lemoore in July, you don’t know hot – and he was heavily loaded with external fuel tanks and air-to-ground ordnance.
He had both blowers staged when the right motor gave up the ghost and the ground started to come up. His mistake was to allow the angle of attack to get away from him. Past the upper limit of the L/Dmax curve with one engine in full grunt at high density altitude things get ugly quickly. Yaw coupled with roll, and he had to shell out at a few hundred feet. He survived, but there was an onion patch at the departure end that never quite recovered.
For several months after the mishap, this scenario became de rigueur for simulator evaluations. Proper procedures in such instances called for hitting the emergency jettison button, which slicks off everything below the pylons. Losing eight or so thousand pounds of external stores simplifies things amazingly.
But to emphasize the risk, many of the simulator instructors demanded that the pilot retain the external stores and attempt to fly the aircraft out. This required the deeply counter-intuitive step of lowering the nose while single engine and in close proximity to the unyielding turf – it was the only way to get enough airspeed to start climbing again. It also required a full stab of rudder opposite the good engine at slow speed, something prop pilots are used to practicing, but a control input unfamiliar to those used to flying a multi-engine fighter with centerline thrust.
Even when you knew it was coming – in a simulator – it could be hard to successfully save the machine. The real test, from the simulator instructor’s point of view, was whether the pilot ejected in a safe envelope. Saving the machine when this mode of failure occurred in a live flight would require almost preternatural reactions. When the failure was imposed upon me, I almost sensed a certain disappointment from my instructor when I managed to milk her out of ground effect and into a climb. He went ahead and failed the other engine just for fun.
Most mishaps have some element of pilot error. Sometimes the pilots just get put in a bad place.
Which is why I thought about that when I read about this:
Other aircraft had avoided the storms that Air France 447 attempted to pass through when it crashed. Weather at the time of the accident as depicted by infrared images seven minutes before and after the last ACARS message sent by the aircraft shows “the general conditions and the position of Inter-tropical Convergence Zone over the Atlantic were normal for the month of June,” according to investigators. But messages sent automatically by the Airbus A330 accident aircraft show the aircraft was providing unreliable or conflicting air data to the pilots. Investigators have publicly announced that experienced teams working in simulators struggled to maintain control of the aircraft at cruise in turbulence while working with faulty air data.
It’s one thing to lose a single-engine, single seat fighter on a 0.001 per cent malfunction at low altitude on a hot day. Quite another thing to have a failure mode put 228 people into the sea from cruise flight.