ORIGINAL: HarryC
Mick, in the other great thread about gyros. I think it was entitled the ABCs of gyros, I said that although the problem is started in yaw, and stopping the yaw will stop the rolling action, the roll reaction can be much higher than the yaw action and thus it can be more profitable to start with an aileron gyro and stop the very noticeable rolling reaction, when the fishtailing is of low amplitude. With very highly swept wings, the gain from yaw to roll can be high so the slightest fishtailing which may be below noticeable level, can produce a noticeable wing rock. In those situations it can be hard to eliminate the fishtailing sufficiently to stop the wing rocking. If only fishtailing is obvious use a rudder gyro, if wing rocking is obvious use an aileron gyro first, and then only consider a rudder gyro if a noticeable fishtailing remains after the wing rocking has stopped.
The full size use yaw dampers which is a sophisticated version of our rudder gyro to kill dutch roll. But when you sit in a full size you feel any slight fishtailing and it is not comfy, whereas slight fishtailing on a model that is too small to be noticed from the ground is not a problem to us, hence for us the cosmetic value on the flightpath of the aileron gyro has more appeal.
H
Interesting thread, had to think a few seconds before I got it. Harry C does understand the dynamics. At first glance the yaw 'seat of pants' effect is counter intuitive. But physics is physics! Using an INS in an autopilot the software would have to take care of this since yaw and pitch start to mix during bank. If 90 deg bank was possible without departing level flight pitch would be yaw and yaw would be pitch in world coordinates?
The dutch roll effect and yaw roll coupling would determine how severe the "yaw" effect is I suppose.
Wonder if the likes of ACT gyros could be set to respond to only the rate of rate-change? That way a prolonged turn would not be affected, only oscillations?