We have just completed another series of flight tests on N4EZ to test its departure (loss-of-control) resistance. What prompted this is reports from two VariEze pilots in Texas that they experienced a partial or full snap roll at about 80 knots. These occurred below pattern altitude and fortunately the pilots managed to recover in time to avoid an accident. The maneuver was described as follows: Full aileron and partial rudder steady sideslip, then full rudder was applied which caused the airplane to yaw excessively and abruptly roll, experiencing negative g. Recovery with neutral control was prompt, but several hundred feet of altitude was lost.
The most surprising thing about these incidents to us was that control was lost at such a high speed - 30 knots above stall. Re-inspection of our stall/spin test data and the NASA tests indicated no susceptibility to departure. We then initiated a new test program to investigate this. Dick performed full rudder sideslip with N4EZ at all speeds and experienced no departures. Concentrating on the 80-knot speed range he then aggravated the yaw with abrupt rudder inputs while in a rolled attitude. On one of these he experienced a departure - a roll off in the direction of the slip. He then tried to repeat the maneuver and could not get a departure in over 20 attempts. We then adjusted the aileron and rudder rigging, moved the cg aft, and repeated the tests.
Dick found that by learning a specific technique he could cause a departure nearly every time, if speed were above 75 knots and an excessive side-slip angle were generated. The departure generally consisted of an uncontrolled roll away from the rudder input direction. Recovery with neutral controls was prompt. However, on several of the maneuvers the yaw angle was extreme at departure causing a massive stall of the winglets and blanking of the upstream wing. The airplane then yawed past 45 degrees, abruptly rolled, and entered a 1 to 2 turn inverted incipient spin.
The airplane always promptly recovered with neutral controls. If aileron or rudder were applied for recovery it could cause a further "snap" departure and delay recovery. Altitude loss on the worst of these maneuvers was as much as 1500 ft. Why did we not find this departure when we did the original tests and the tests with cuffs in 1978 (CP #19)? The most probable reason is that most of those tests were done at high angle of attack (full aft stick) which was thought to be the worst case. However, we have found that at lower angle-of-attack i.e., 80 knots, the rudders can generate more sideslip than at high angle-of-attack, and thus can be powerful enough to stall the winglets in an accelerated yaw maneuver.
We were then faced with a decision on what to do:
(1) caution pilots that the airplane can be departed when using excessive yaw inputs or
(2) fix the airplane to improve its departure resistance.
Since we feel strongly that good departure resistance is an important asset and design goal for the VariEze, we set out to attack # (2).
We have always known that the EZ has more rudder power than needed for normal maneuvers - a full aileron steady sideslip at low speeds requires only 60% of the available rudder to hold heading. The available travel is 3-1/2 inches, measured at the top of the rudders trailing edge. We then limited the rudders travel on N4EZ in various increments, 3", 2.6", 2.3" and 1.8". At each increment we flew tests to determine departure susceptibility and the necessary rudder authority for crosswind landings. As expected, the departure susceptibility reduced as rudder travel was limited.
After extensive testing and evaluation by three pilots we have N4EZs rudder now rigged for a two-inch full travel. With this rigging, the following characteristics exist: crosswind landings up to a component of 24 knots are possible without tire scrub. The aircraft is not susceptible to departure during any normal maneuver. Thus, we are now recommending a mandatory rigging change to limit the rudders to two-inches of travel. It must be noted that this may not guarantee total departure resistance. This may vary from one airplane to another, due to expected tolerances in things like winglet leading edge finish and shape, fuselage shape etc. Also, even at 1.8 inch rudder travel, Dick was able to induce a departure by learning an unusual and aggressive combination of control inputs: full left aileron, full left rudder at 30 degree bank, then at 1000 bank abrupt full right rudder.
When this was done a departure was possible (not probable) even though the rudder was limited to 1.8 inches. The important thing to note is that, even though this design is not as susceptible to loss-of-control as a conventional aircraft it should not result in over-confidence on the part of the operator. Assume your aircraft is susceptible to departure until proven resistant during your stall tests with lots of altitude and a parachute. Refer to the plans-changes section of this newsletter for a caution note to be added to your owners manual and for instructions on rigging rudders to two inches.