The crumpled Bonanza lay in a field of corn stubble. The eastern Colorado terrain was flat for miles around. It was hard to see how even a complete loss of power should have led to anything worse than a little scraped metal, but evidently the airplane was out of control when it hit the ground. It had cartwheeled before coming to rest, its cabin crushed. Its two occupants died: the owner of the airplane, a nonpilot who had bought it three days before, and an instructor who was taking him to a business meeting that day and intended, in the future, to teach him to fly his newly acquired machine.
The log of the instructor pilot, 35, showed 467 total hours. Pilots who had flown with him, including the seller of the accident airplane, had only praise for his flying. But his experience level in Bonanzas is unclear: The National Transportation Safety Board’s report on the accident states (at different points on the same page) that he had two and a half hours or 131 hours in the make and model of the accident airplane. In another location, however, it refers to his “limited experience” in Bonanzas.
The accident occurred about 8 miles north of Weld County Airport (GXY) at Greeley, Colorado. The wind was gusty, 10 to 20 knots, out of the north. There were pilot reports of moderate to extreme turbulence in the area. Approaching from the southeast, the Bonanza had descended to 400 feet as it overflew Runway 35 at GXY, but then continued northward, climbing back up to 1,100 feet agl and maintaining around 90 knots groundspeed — that is, taking the altitude and the north wind into account, an indicated airspeed of somewhere around 95 to 110 knots. The pilot did not give a position report or state his intentions to unicom, but a person monitoring the frequency heard an exclamation of an expletive. The exact timing of the utterance is unclear. Perhaps the pilot exclaimed over some onboard problem that made him abort the landing and fly northward for several minutes without any apparent reason, or perhaps the exclamation coincided with the loss of control.
Several witnesses on the ground reported hearing the engine “sputtering” before the Bonanza began a left turn to reverse course and then dived. The NTSB, having examined the engine and what was left of the airframe and finding nothing to account for the power loss, concluded that, although there was ample fuel in the airplane, the tank feeding the engine had run dry. The fuel-tank selector was set to the right main tank. The tank, which had not been breached in the crash, was empty.
This analysis gained plausibility from the design of the fuel quantity indicating system. The Bonanza, built in 1956, had two main tanks, two wing aux tanks and two tip tanks, but only a single fuel quantity indicator with its own separate selector switches. It was the pilot’s responsibility to ensure that he knew which tank’s fuel level the indicator was reporting. The setting of the indicator’s selector switches could not be determined from the wreckage, but investigators hypothesized that the indicator might have been set for the wrong tank.
Since the left turn coincided with a steady loss of altitude, and both coincided with witness reports of what sounded like engine trouble, it seems apparent that the power loss occurred while the airplane was still tracking northward, into the wind.
Given the strong north wind and the fact that the terrain ahead consisted of a series of farm fields all equally suitable for a forced landing, the pilot’s decision to turn back southward seems puzzling. Probably he instinctively turned toward the airport with the hope he might be able to restart the engine or at least milk some power out of it. He probably believed he was feeding from a tank that had fuel in it, and so his first hypothesis would have been that there was a mechanical reason for the power loss. He had about a minute to figure out what it was.
A pilot’s thinking in such a situation — loss of power at low altitude, far from an airport — is seldom orderly or systematic. Unlike the pilot of the 1920s, who considered forced landings in farm fields normal, the modern pilot is unprepared. Unless the Air Force has trained him to rattle off the “boldface items” from memory, he does not go down a list of possible causes, weighing the likelihood of each, as we do when we reflect on such a situation in the security of our armchairs. He is more likely to find himself unable to think analytically at all.
The NTSB suggested that the airplane stalled because it was turning downwind in gusty conditions. This is plausible, although the physics of transitory fluctuations of airspeed are rather complex because it is angle of attack, not indicated airspeed, that causes a wing to stall. The mere fact of turning from upwind to downwind would not, in itself, entail any risk. But the powerful visual impression of sliding sideways and vertiginously gaining speed during such a turn at low (and rapidly decreasing) altitude can lead a pilot to cross controls and unconsciously raise the nose. The low altitude downwind turn requires attention to the instruments; a distracted pilot who is searching for the cause of an engine failure might mishandle it.
I am inclined to think that the pilot’s decision — or perhaps I should call it an impulse, since it probably lacked the deliberateness of a decision — to turn back toward the airport, while understandable, was the root cause of the accident. He could have continued straight ahead while trying to identify the engine problem. If he figured it out, he would be able to return to Greeley at his leisure. If he didn’t, he would at least encounter the ground with the least possible forward speed. Turning downwind, even if he had not stalled the airplane, put him at risk of arriving with twice the kinetic energy he would have had facing north. By the time he had turned 180 degrees, he was down to 350 feet agl; he would not have time to turn back into the wind.
There was, however, another decision that might have contributed to the accident. It was made at a time in the 1940s or ’50s, when some engineers who are probably no longer with us weighed the advisability of separating tank selection from quantity indication. There was an obvious and relatively easy solution to the problem: a cam on the fuel selector handle, and some microswitches and relays, to keep the indicator in sync with the selector. Someone probably argued — not without reason — that such a system, besides costing a few dollars, would introduce various new modes of failure, including the possibility that the pilot could be left with no fuel quantity indication at all. Which is better, a fallible system or a fallible pilot? In the end, they shrugged and said, “If the guy is smart enough to be a pilot, he can keep track of his fuel.”
The gospel on turn-backs after a power loss just after takeoff is to continue straight ahead, no matter how uninviting the terrain looks. Though it has been demonstrated, at least theoretically, that successful returns are possible given enough altitude and a sufficiently skilled and disciplined pilot, it has also been demonstrated that in practice the outcome is far more often an unrecoverable, and unsurvivable, low-altitude stall.
The accident at Greeley had many of the elements of a turn-back, even though it took place miles from an airport, and at the end, not the start, of a flight. The same rule should have applied: After a power loss at low altitude, land — or crash — as nearly straight ahead as you can.
