Aftermath: Control Problems

(November 2011) Control Malfunctions — failures of the control systems to work properly — are among the most challenging and frightening difficulties that an airplane can present. One feels able to deal with almost any eventuality, as long as the flight controls work the way they should. But an aileron jammed in a hard-over position, an engine that has come disconnected from its throttle or an elevator from its push rod, a stuck fuel valve — these are the stuff that nightmares can be made of.

Two fatal accidents illustrate the range of forms that loss of control can take. In one, the pilot of a Citation 550 found the airplane fighting him in a way he could not understand; in another, the pilot of a Pilatus PC-12 allowed one wing to fill with fuel while the other emptied, and then could not manage the huge fuel imbalance as he maneuvered to land.

The Citation accident took place at Milwaukee on a June afternoon in 2007. There was a broken ceiling at 3,000 feet and an overcast at 3,500. The purpose of the flight was to deliver an organ for transplantation from Milwaukee to Ypsilanti, Michigan. The 14,000-hour captain, who was flying, was the chief pilot and FAA-appointed check airman for the Citation’s operator. In its account of the captain’s career, the NTSB relates a history of failed check rides and even a brief license suspension due to a felony conviction for illegally transporting methaqualone — Quaaludes — into the United States from Canada. Associates of the captain described him as a “capable” pilot who was, however, deficient in knowledge of airplane systems and somewhat casual about checklists.

The 65-year-old first officer, a new hire, was a part-time commercial pilot. He held a Citation type rating but was, according to a pilot who had flown with him, “a nice guy who had no idea how the airplane operated.” He was prone to careless mistakes, on one occasion, for example, turning off an airplane’s avionics when he intended to switch the engine ignition from “on” to “norm.”

The trouble began shortly after takeoff. Ten seconds after the captain called for the yaw damper, he said, “Why am I fighting the controls here?” The airplane wanted to turn left. The captain tried to analyze the situation, eventually concluding that something must be wrong with the rudder trim. In the meantime, the first officer made well-meaning attempts to adjust an unspecified trim — most likely rudder or aileron.

As the airplane gained speed and the out-of-trim forces grew, the captain’s utterances became increasingly urgent. “Tell ’em we’ve got to come back and land ... she’s rolling on me. Help me, help me.” Evidently suspecting an autopilot malfunction — although as far as he knew the autopilot had not been turned on — he asked the first officer to pull the autopilot breakers. “Where is it?” the hapless second pilot asked. At the same time, he was helping the captain fight the powerful roll forces.

The captain declared an emergency, asking the Milwaukee departure controller to “guide us in” to any runway. Repeating “I don’t know what’s wrong,” he asked the first officer to “hold it” while he tried to pull the circuit breakers himself. The Citation crashed into Lake Michigan nine seconds later.

Examination of the recovered wreckage did not provide clear insight into the cause of the trouble. The only physical anomaly was evidence of a short circuit in trim switch wiring within a control yoke. Cessna, as a party to the investigation, argued that, because of the way the trim circuit was designed, the chance that a single short could produce a continuous runaway to the limits of the Citation’s inordinately sensitive and powerful roll trim was infinitesimal; but that defensive claim remained in the docket and did not find its way into the report.

Noting, however, that the trouble began just after the captain called for the yaw damper, the NTSB offered an alternative to the trim runaway hypothesis. Possibly the first officer had inadvertently turned on the autopilot when he had meant to activate the yaw damper — the two were controlled by identical buttons set side by side on the central pedestal — and the mysterious resistance the pilot initially felt was actually the autopilot servo. Under this scenario, it would be the first officer’s subsequent mishandling of the manual trim controls — for which the only evidence was his anecdotal history of switch dyslexia — that did the rest.

Whatever the physical factors, however, the NTSB entirely omitted them from its finding of probable cause. The cause was “the pilots’ mismanagement of an abnormal flight control situation through improper actions, including failing to control airspeed and to prioritize control of the airplane, and lack of crew coordination.” (Educators should investigate the probable cause of the NTSB’s prose style.) Additional factors were listed, but they were administrative rather than mechanical: the customary stew of operator deficiencies and lax FAA oversight.

The crux of that finding is the phrase “failing to control airspeed.” Control forces increase rapidly with speed. An airplane that is seriously out of trim should be held to approach speed while autopilots and dampers are turned off and all trims are centered. There was, as far as investigators could determine, nothing wrong with the Citation that would have kept it from flying if its trims had been properly set and its autopilot off.

The case of the Pilatus was, in a sense, the opposite. The precipitating factor was the pilot’s failure to ensure that ice inhibitor be added to his fuel when he topped the airplane off before launching with 14 aboard — four more than there were seats for, but they were children — on a March 2009 flight from California to Montana. Jet fuel absorbs water, which, at the low temperatures of turbine altitudes, forms ice crystals that can cover fuel filters and clog valves and pumps. Ice inhibitor is a routine protection, and it was not clear why the pilot had not requested it.

In the Pilatus, fuel pumps are turned on or off, manually or automatically, as needed to feed more fuel from the fuller tank. (There is no provision for pumping fuel directly from one tank to the other.) The electronic fuel-quantity indicators display two bands of bars, like the signal-strength bars on a cell phone, except that each band contains 28 bars representing the 1,370-pound fuel capacity of each wing. The POH recommended controlling the pumps manually if a discrepancy of more than three bars appeared, and landing if the discrepancy could not be resolved.

The nonvolatile memory of the airplane’s central advisory and warning system retained a record of fuel levels and pump operation during the accident flight, and made clear to investigators that fuel had ceased to flow from the left tank about 80 minutes after takeoff. At the same time, fuel returned from the engine was going to both tanks. As a result, the left tank became gradually fuller as the right tank emptied.

Belatedly, the pilot decided to divert to Butte, which was only slightly closer than his destination of Bozeman. By the time he got there the left tank was completely full of inaccessible fuel and the right nearly empty. The aileron trim was at the right-wing-down stop. Because of weather and altitude limitations, he had to descend steeply, and he used full left rudder trim to maintain a right slip to kill altitude. He was still too high to land, and it was during the go-around that he lost control of the airplane and crashed in a cemetery beside the airport. All aboard died.

By that time, nearly an hour had passed since the fuel imbalance had reached the point at which the POH advised landing immediately. Why the pilot chose to continue, passing many usable airports, some of them quite close to his track, only he knew. The NTSB listed three causes of the accident in chronological order: the failure to get ice inhibitor with the fuel, the decision to press on rather than land en route and, finally, the loss of control.

Unlike an airplane that is mistrimmed, an airplane that is unbalanced, either laterally or longitudinally, needs to keep speed up in order to preserve control authority. It must also be maneuvered gingerly, to minimize inertial forces. Statically, the Pilatus can supposedly maintain level flight at 90 knots with a 1,300-pound fuel imbalance, but with full left rudder trim (which tends to push the left wing down) and in dynamic maneuvering, considerably more control authority is needed.

Ironically, a check pilot who had flown with the accident pilot described him as possessing a “very high level” of competency and “superb” professional judgment. Even Homer sometimes nods.

This article is based on NTSB reports of the accident and is intended to bring the issues raised to our readers’ attention. It is not intended to judge or to reach any definitive conclusions about the ability or capacity of any person, living or dead, or any aircraft or accessory.

Send reader mail to: or P.O. Box 8500, Winter Park, FL 32789.

Peter Garrison taught himself to use a slide rule and tin snips, built an airplane in his backyard, and flew it to Japan. He began contributing to FLYING in 1968, and he continues to share his columns, "Technicalities" and "Aftermath," with FLYING readers.

Your email address will not be published. Required fields are marked *

Subscribe to Our Newsletter

Get the latest FLYING stories delivered directly to your inbox

Subscribe to our newsletter