Uncoordinated roll inputs in a Schleicher Ka-6 sailplane
August 8, 2007 edition
Here is an interesting test I carried out on a Schleicher Ka-6 sailplane: I took my feet off the rudder pedals, and made a substantial left roll input with the control stick. I repeated this test several times using slightly different rates of stick movement and slightly different airspeeds. The nose always adverse-yawed to the right as the glider rolled into a left bank. The yaw string always deflected far to the right. The rudder floated over to the full right position as it streamlined itself with the slipping airflow (i.e. the rudder was acting like a yaw string and showing the direction of the airflow.) This suggests that the rudder was free to "unload" itself and probably had little effect on the glider's yaw dynamics: only the fixed vertical fin and the rear fuselage would have contributed to the glider's yaw stability. If I relaxed my roll input the glider would tend to roll back toward wings-level: I had to maintain a substantial left roll input on the control stick to keep the glider in a left bank at a constant bank angle immediately after the initial rolling motion. Interestingly, in some cases the glider would "stick" in a non-turning slip with no leftwards curvature of the flight path at all, despite the fact that the glider was banked to the left. This non-turning slip was not a temporary state of affairs-- the glider was happy to continue flying along the original direction of the flight path, with the nose yawed to the right and the yaw string deflected to the right. To keep the glider from rolling out of the left bank, I had to keep the control stick deflected to the left. When the glider was "stuck" in the non-turning slip in this way, to make the glider actually turn to the left, I generally had to relax the left stick input a little bit. As the ailerons moved a bit closer to the centered position, the yaw string would move a bit closer to the centered position, and the flight path would start to curve to the left, even though there was also a slight decrease in the left bank angle.
In similar tests where I kept my feet on the rudder to hold the rudder in the centered position, a left stick input generally always led to a leftwards curvature in the flight path, though initially there would be a great deal of adverse-yaw to the right and the yaw string would swing far to the right and the leftwards curvature in the flight path would initially be very gradual.
Discussion: we can see a lot of interesting things in this experiment, especially in the case where the rudder was allowed to float freely. For example:
1) The deflection of the ailerons must have created a large adverse yaw torque even when the glider's bank angle was constant. This contradicts the idea that adverse yaw is created mainly by the way that the relative wind, and thus the direction of the local lift and drag vectors, "twists" when the aircraft is rolling to a different bank angle.
2) The aircraft's dihedral geometry is clearly creating a strong right roll torque whenever the aircraft's nose is pointing to the right of the actual direction of the relative wind, thus the need for a strong roll input to hold the bank angle constant whenever the aircraft was slipping.
3) We are stuck in an interesting "loop" of inter-related causes and effects: the deflected ailerons are making an adverse yaw torque, and the adverse yaw torque is keeping the nose displaced to the right of the direction of the relative wind. The resulting sideways (slipping) airflow interacts with the dihedral to create a right roll torque, so we must keep the ailerons deflected to the left if we want to hold the bank angle constant, which maintains the adverse yaw torque, and we are "stuck" in the slip with no way to actually turn. As noted above, in some cases relaxing the left roll input actually allowed the flight path to finally began curving the left!
4) In aircraft with weak yaw stability we can see a kind of "hysteresis", especially when we leave the pilot's roll inputs "in the loop". In other words for a given bank angle, and assuming that the pilot is not using the rudder, and assuming that the pilot is making whatever roll inputs are needed to keep that bank angle constant, we can have the following situation: if the pilot rolled the glider into the bank with an abrupt roll input, the glider will "stick" in a non-turning slip, while if the pilot rolled the glider into the bank with a gentle roll input, the sideslip will be more moderate and will then decrease some more after the target bank angle is reached, allowing the flight path to curve to the left with only a small amount of sideslip. For a given constant bank angle, and assuming that the pilot is not using the rudder and is making whatever roll inputs are needed to keep the bank angle constant, the amount of sideslip and the rate of curvature of the flight path can be dependent on how quickly the pilot rolled the aircraft into the bank, i.e. on how much sideslip was created as the pilot rolled the aircraft into the bank. Similarly, as the pilot is making whatever roll inputs are needed to hold the aircraft at a constant bank angle, if the pilot uses the rudder to center the yaw string and then removes his feet from the rudder, he may see the aircraft come to equilibrium at a very different sideslip angle than if he kicks the top rudder pedal to increase the slip and then removes his feet from the rudder pedals.
5) When the pilot is not using the rudder, if a left roll input can cause the aircraft to "stick" in a non-turning sideslip, it follows that in another aircraft with slightly more cross-sectional fuselage area, producing slightly more sideforce in a slip, a left roll input might actually cause the flight path to curve toward the right.
6) It seems quite likely that during an intentional non-turning slip in the Ka-6 or a similar aircraft, e.g. during a crosswind landing or to steepen the glide path on final approach, in some cases little or no rudder pressure in the usual direction, or even some rudder pressure in the direction opposite from the usual direction, might be required to keep the glider in the non-turning slip. In other words the drag from the displaced ailerons might create enough yaw torque to keep the glider in the slip without any help from the rudder. As the rudder seeks to streamline itself with the airflow, it will always tend to deflect away from the centerline in the usual direction (i.e. to the right, if the pilot is lowering the left wing), but little or no force on the right rudder pedal, or possibly even a touch of force on the left rudder pedal, might be required to keep the flight path from curving in either direction when the left wing is kept low. In other words, as the pilot lowers the left wing, and maintains a steady left roll input to prevent the bank angle from decreasing, the adverse yaw torque (to the right) from the deflected ailerons might make the aircraft adopt a slip angle that is severe enough to prevent the flight path from curving to the left even without the pilot exerting any pressure on the right rudder pedal. (This is exactly what I saw during the experiment described above.) In some cases the adverse yaw torque (to the right) from the deflected ailerons might even make the aircraft adopt a slip angle that is severe enough to make the flight path curve to right, unless the pilot exerts some pressure on the left rudder pedal. This is opposite to the rudder input we usually expect to give in a non-turning sideslip.
7) Most of the above dynamics would not be observable in an aircraft with more yaw stability, i.e. with a larger fixed vertical fin. Most of the above dynamics would not be observable in an aircraft with a smaller wingspan, so that deflecting the ailerons created less yaw torque. Many of the above dynamics are easily overlooked during normal flight-- I've flown sailplanes for many years without being aware that one could "stick" in a non-turning slip in this manner, even without the pilot applying any pressure on the top rudder pedal.