Mistrim suggests itself, but this was an extremely small and light airplane, and even if it were severely mistrimmed it could not have overpowered its pilot. A more likely symptom of mistrim would have been some overcontrol and porpoising, not an apparently unmanageable pitch-up.
“Failure to maintain pitch control” could be interpreted to mean “inept piloting.” But this pilot was experienced, had previously built and flown a high-performance homebuilt, and to all appearances was taking a reasoned, cautious and incremental approach to testing. So it is unlikely that he lost control of the airplane because he didn’t know any better.
A more probable explanation is suggested by the changes that had been made after initial taxi testing — the canard surface had been enlarged and the CG moved aft — in order to facilitate rotation.
It appears from photographs that the main landing gear was located slightly ahead of the trailing edge of the wing. Main landing gear axles are normally placed along a line sloping back at an angle of about 15 degrees from the center of gravity. The vertical axle-to-CG distance appears to be about two feet, and so the CG would be expected to be at least six or seven inches forward of the gear. This would roughly coincide with the quarter-chord point on the wing. In other words, the relationship of wing, landing gear and center of gravity seems to be the one that would be expected for a conventionally configured airplane.
But this airplane was not conventionally configured. Since it so strongly resembled the three-surface Piaggio Avanti, it is instructive to note that the landing gear of the Avanti is located approximately at the leading edge of the wing, implying that the Avanti’s CG is well ahead of the wing. This is typically the case with canard-equipped airplanes, unless, as on the B-1 bomber and some highly maneuverable fighters, the canard area is so small as to be almost negligible.
Air does not make the same distinctions as we humans do between wings, canards and tails. They are all lift-producing surfaces, all flying along together, and must be taken as a whole. The lifting canard surface behaves like a part of the wing and moves the airplane’s center of lift forward. The amount of forward shift depends on the relative areas of the two surfaces, on how far apart they are, and on their lift coefficients. For stability, however, the canard must be more heavily loaded than the wing, and so its contribution to the total lift is disproportionately large.
The conventional placement of N72TZ’s landing gear with respect to the wing, and the mention of ballasting the CG to 30 percent of the mean aerodynamic chord — presumably the mean aerodynamic chord of the main wing — suggest that allowance may not have been made for the forward shift of the aft CG limit due to the canard, especially after the latter’s area had been increased. Thirty percent of wing chord is pretty far aft, even for a conventional airplane; flight testing usually begins with the CG forward of 25 percent to ensure good static stability. When a canard surface is present, however, the usable CG range moves forward significantly, and 30 percent becomes very far aft indeed.
As the CG moves farther and farther aft relative to the aerodynamic center of the lifting surfaces, longitudinal stability diminishes, disappears and then becomes negative or “divergent,” meaning that rather than seek a trimmed attitude the airplane actively flees from one. The sudden extreme pitch-up is consistent with longitudinal divergence, and longitudinal divergence is consistent with a three-surface airplane whose CG has been positioned as though the canard were not there. The airplane would seem to behave normally as long as its wheels were on the ground, but once airborne it would be fatally tail-heavy. Because of the rapid loss of airspeed in a steep climb, the elevators would lose effectiveness — assuming that they were effective in the first place — and the pilot’s efforts to force the nose down might be in vain.
Like many other NTSB findings, the conclusion that this accident was due to “the pilot’s failure to maintain pitch control” does not so much answer the question as restate it. Read it as a concession that they — and we — can’t say precisely what happened. But the appearance of longitudinal instability is clear enough that the issue is at least worth discussing, if only to alert future amateur designers to the complexities lurking in configurations other than the conventional one.
This article is based on the NTSB’s report 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.