In June 2006 a Beech B36TC Bonanza, fresh out of annual inspection, lost power while cruising at 5,000 feet on the way from Kalamazoo to Ypsilanti, Michigan. Although conditions were day VMC, the 950-hour pilot had filed an IFR flight plan. He advised the controller of the situation and asked for a vector to the nearest airport.
The controller provided a vector to Carriage Lane, near Gregory, about 30 miles northwest of Ann Arbor. Gregory is a microscopic rural community, and Carriage Lane airport is proportionately tiny; besides, its 1,700-foot turf runway can barely be distinguished from its surroundings. A couple of minutes later, the controller offered another airport: Richmond, 69G, a couple of miles southwest of Carriage Lane. The runway there was turf as well, and equally well camouflaged; but at least it was 2,471 feet long and it was slightly closer.
Exactly where the Bonanza was when the engine failure occurred is not clear from the National Transportation Safety Board’s accident report, but it must have been roughly midway between Jackson and Ann Arbor, since the vector provided by the controller was 360 degrees. The local terrain height is about 920 feet and the plane had settled a bit as the problem first developed, and so the pilot had somewhat less than 4,000 feet of height from which to find his way to safety.
The B36TC version of the Bonanza has wingtip extensions that increase its span by about four feet. Although they are principally intended to boost high-altitude performance, they would improve the glide angle as well. With a windmilling propeller and reasonably good speed control, the airplane should have a glide ratio of 10:1 or better, which would allow it to glide at least six or seven miles before arriving at the ground. If the Bonanza was on V-221 due south of Carriage Lane when the power loss occurred, it was about seven miles from 69G.
The NTSB report does not make clear what exchanges, if any, took place between the controller and the pilot, other than to say that after the controller suggested 69G there was no response and no further communication. It’s likely that by that time the pilot had located 69G for himself using the “nearest airports” function on his GPS, because he was right on course for it when he struck trees 2,000 feet short of the threshold.
The pilot, who was alone, was killed. Most of the airplane was consumed by fire despite the efforts of neighbors to extinguish it with water from a nearby creek. The engine, however, remained largely intact, and it was sent to Teledyne Continental for bench testing. After several damaged accessories had been replaced, the engine ran normally. The propeller, although damaged, appeared to have been sound before the impact, and the liquid recovered from the distributor manifold was 100LL and was free of water contamination.
The NTSB gave the following probable cause for the accident:
“The non-mechanical loss of engine power for undetermined reasons during cruise flight and the unsuitable terrain the pilot selected for the forced landing. A factor was the tree that was impacted during the forced landing. An additional factor was the inadequate planned approach during the forced landing when the pilot over flew open fields on the way to the intended airfield.”
These probable causes often have an odd quality, as if they had been composed by robots unacquainted with the way things are here on Earth. The tree, in this case, seems out of place; the airplane came down in a wooded area, and so it is not surprising that at some point it might strike a tree. In any case, the tree was no more of a “factor” than lots of other things were; the major causal elements, obviously, were the engine failure and the pilot’s actions once it had occurred.
Since the reason for the engine failure is unknown — though the NTSB appears to have thought it had something to do with the fuel system — what concerns us — as candidates, by virtue of being pilots, for participation in some similar future scenario — is the pilot’s thinking. What options were open to him? Why did he choose to try to land at a tiny, almost invisible airport at the very limit of his gliding range?
The NTSB thought the pilot was wrong to try to make the airfield when he passed suitable unprepared fields along the way. But fields were not the only thing he passed up. When the engine failure occurred he was within a few miles of Chrysler Corporation’s automotive proving ground, which includes an immense oval with long multilane straightaways onto which it would be an easy matter to dead-stick an A380. Between him and it was Interstate 94, with its parallel runways of practically infinite length. On the way toward 69G he overflew long straight stretches, free of power lines, of lightly traveled State Road 52.
With the advantage of hindsight it is easy to see other problems with the decision to head for 69G. One was the overall pattern of the terrain. To the east and southeast all was farms and fields; the land in the direction of 69G, however, was hilly and largely forested, with many small lakes.
One unanswerable question is what the pilot could see. At Howell, about 12 miles north-northeast of 69G, the sky was clear, with wind out of the northwest at nine knots, gusting to 14; but it is still possible that there were clouds in the pilot’s vicinity and that they limited his view of the ground. The NTSB does not mention clouds, nor does it comment upon the pilot’s decision to turn north rather than to seek an emergency landing place in his immediate vicinity; instead, it seems to accept the initial decision to head for 69G, but then to fault the pilot for not opting for an open field along the way instead. In fact, there were some fields along the way, but by the time the pilot was low enough to begin to suspect that he might not make the airport, he was over hilly, wooded terrain with no very promising fields in reach. The best places for an off-airport landing were, in fact, back where the power loss first occurred. In other words, to the extent that one can pinpoint a single fatal error in the pilot’s decision-making, it was the original decision to make for the nearest airport.
A great deal goes through a pilot’s mind after a complete power loss. He does not necessarily think clearly. The apparent fearlessness of military pilots and test pilots is due in part to traits of character (without which they would not be in those lines of work) but in part to having at their fingertips a well-rehearsed sequence of pre-planned options. It is not necessary for them to reason out every decision in an emergency; the reasoning has already been done, and the bold-faced items contain it. Most civil pilots are much less well-prepared; there is likely to be an element of grasping at straws in their decision-making, and as a result any offer of help may take on more weight than it deserves. What a controller offers as information may be received by the pilot as advice.
When the controller told the pilot that the nearest airport was due north of him, he was merely providing one piece of information among many. Probably he stated the distance; the NTSB report doesn’t say. Now, if the nearest airfield had been 20 miles away it would have been out of the question, and the pilot would immediately have turned to his other options — all of which were actually better. But seven miles sounded feasible. The mental math gets tricky if you’re scared and in a hurry, but you’re likely to reason something like this: “I’m about a mile up, the glide ratio is certainly better than seven to one, so I can make the airport.” The caveats — you’re not really a mile above the surface, there’s a quartering headwind, the airstrip may be invisible, if you end up short you’ll be in trees — may be too complicated and uncertain to think about. It is such a relief to have made a decision that you don’t try to second-guess yourself. And somehow the idea of getting to an airport sounds so much better than any alternative that you just don’t question it. If you can land at an airport — well, naturally you want to. Who wouldn’t?
Apart from the pilot’s decision-making, there was also a technical problem that emerged during the post-mortem examination of the propeller. It appeared that the blades had been at their low pitch — that is, high rpm — setting. I wrote in Technicalities in January 2008 about the effects of propeller pitch and throttle settings on power-off gliding performance. For that article I did a number of tests, shutting off fuel and mags and varying prop and throttle settings during glides, and found that pulling the prop control all the way back and pushing the throttle all the way forward yielded by far the best gliding performance — 30 percent better than with flat pitch and throttle closed. In other words, the best control settings yielded several miles more gliding distance, per mile of altitude, than the worst.
It is saddening to reflect that, despite his questionable decision-making and his failure to recognize the nearby options for landing with little risk of damage to himself or even to the airplane, this pilot would probably be alive today — and would have a great story to tell — if only he had pulled the prop control back to coarse pitch.
This article is based on the National Transportation Safety Board’s report of the accident and is intended to bring the issues raised to the attention of our readers. 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.
