The Lingering Mystery of ‘Undetermined Reasons’

When an accident’s cause is not easily found.

When the National Transportation Safety Board announces the probable cause of an accident, it often makes use of certain stock phrases. One of the most common is: “loss of control for undetermined reasons.”

Unlike “controlled flight into terrain,” which can often be traced to an incorrectly selected radio frequency or a sectional chart left behind, loss of control for undetermined reasons remains mysterious. These are not the loss-of-control accidents that result from icing encounters or from a non-instrument-rated pilot flying into IMC; in those cases, reasons can at least be plausibly guessed. Undetermined reasons really are undetermined.

In October 2007, a VariEze crashed shortly after taking off from a Virginia airport. Witnesses saw the airplane roll down the runway and lift off, then turn crosswind and downwind; the pilot announced each step on the unicom frequency. Then came a panicked cry for help. Flying erratically, the airplane turned at low altitude, caught a wing and cartwheeled across a divided highway, disintegrating and scattering a shower of debris that collided with the moving car of one witness. The 580-hour private pilot, 44 years old, died instantly in the crash.

Designed in the mid-1970s, the VariEze is a plans-built, two-seat, swept-wing airplane of canard configuration and glass-fiber and epoxy construction. While it is a fast and efficient cruiser, it is most notable for its benign low-speed handling characteristics and immunity to stalling. It and its larger successor, the Long-EZ, were, for a long time, among the most popular amateur-built designs; thousands were built.

The pilot had bought the airplane from its builder 16 months before the accident and flown 70 hours in it since. His companion told investigators the purpose of the accident flight was to test the airbrake, which he had tried only once before—during the first hour he owned the airplane. On that occasion, she reported, it “really scared him.”

This comment had the effect of making the airbrake a central focus of the accident investigation.

The airbrake is a more or less square panel located amidships under the fuselage and hinged to swing downward. It is actuated by a lever behind the throttle on the left armrest, and is used during approaches and landings to make the glide steeper and ground roll shorter. Because it does not disturb flow over the wings or canard, it does not affect stability or handling qualities. Maximum speed for deployment is 90 kias; at a higher speed, the brake retracts automatically. The VariEze will climb even with the airbrake deployed.

Following up on the friend’s remark about the airbrake, the NTSB investigator in charge queried the builder of the accident airplane, who had flown it almost 1,500 hours, and Burt Rutan, who designed it. Both—as well as Mike Melvill, who had been responsible for customer support at the by-then-defunct Rutan Aircraft Factory—advised him they did not see how the airbrake could possibly have been responsible for a loss of control.

However, Melvill recounted how, the first time he inadvertently exceeded 90 kias with the airbrake out, the actuating handle “smacked” his wrist as the brake retracted itself. He wondered whether the same thing could have happened to the accident pilot; it “might have scared/surprised him.”

I think the airbrake was a red herring from the start. If the pilot had previously found something frightening about the airbrake—perhaps the mild rumbling buffet it caused or a sudden unexpected retraction—it’s not likely he would have deployed it at the very start of his flight, at low altitude and in the traffic pattern.

Editor’s note: This article is based in part upon the NTSB final report for a given accident. The intention is to bring the probable and contributing causes of these accidents to our readers’ attention, so they can learn from them and apply them to their own flying.

In his statement to the NTSB, Burt Rutan suggested that, because the airbrake could hardly have been responsible for a loss of control, the investigator might want to look at other possible causes, such as distraction or a “control disconnect.”

In the aftermath of any accident, investigators routinely examine the components of the flight control system. In this case, continuity was reported to have been established “from the cockpit area out to the cable breaks and from the breaks to their respective control surfaces.” Most airplanes use cables for all flight controls, but the only cables in the VariEze’s control system are those actuating the airbrake and rudders; ailerons and elevators are controlled by pushrods or torque tubes, of which no mention is made in either the NTSB’s final report or the report from the materials laboratory to which the cables and airbrake mechanism had been sent for analysis.

One of the ingenuities of the VariEze design is the extreme simplicity of its flight controls. Particularly, the pitch control system, in contrast to the long and often-circuitous linkage between the cockpit and the empennage in a conventional airplane, consists of a single short link from the sidestick to the elevators, which are on the canard surface within arm’s reach of the pilot. Mention of this linkage or the aileron system—which uses torque tubes, not cables—appears in neither the final report nor the supporting docket, and it is unclear whether they were recovered and, if so, whether the investigator knew what to make of them.

The maintenance of homebuilt airplanes is subject to special rules. The builder receives a repairman’s license for the airplane. If the airplane is sold, that license remains with the builder, and the new owner must either get the builder to perform the required annual inspections or take the airplane to a regular A&P. (Most maintenance on certified engines must be performed by an A&P.)

When the pilot bought the VariEze in June 2006, the builder/seller helped him dismantle, inspect and reassemble the airplane. The builder then made a 30-minute acceptance flight and found everything to be in order. A year later, when the next annual condition inspection came due, the pilot asked the builder to help again, but because of time constraints, he could not. Although the docket mentions that an inspection took place in June 2007, four months before the accident, it does not say who performed it.

Pick up a copy of Peter Garrison’s book here.

The radioed call for help reported by two witnesses suggests, to me at least, the instinctive exclamation of a pilot suddenly confronted with some awful surprise he has no way to cope with. He certainly did not really expect someone to come up and help him. It wasn’t an engine problem; the engine was developing power, as investigators could infer from the way the wooden propeller blades had shattered. As I said, I think it unlikely he would have tried the airbrake almost immediately after takeoff. I think the fact that the airplane descended and the pilot failed to arrest the descent hints at a loss of pitch control.

It is conceivable, for instance, that a nut on some link in the control system was removed at some point and not properly safetied on reassembly and the associated bolt eventually worked its way out. The stick would have suddenly gone slack in the pilot’s hand—the sort of thing that could bring a panicked exclamation out of anyone. It’s not probable, but the very nature of the accident suggests that whatever caused it was something improbable.

There is no material evidence to support this hypothesis nor, for that matter, is there material evidence for the airbrake’s being a factor. If any clue existed, it was probably destroyed in the catastrophic disintegration of the airplane or disappeared in the subsequent cleanup. Like many loss-of-control accidents, this one remains a mystery, its reasons…undetermined.

This story appeared in the Jan/Feb 2020 issue of Flying Magazine


New to Flying?


Already have an account?