I have tried a workaround for VRS traps that worked twice so far after a near earlier disaster. I have had a total of 3 VRS incidents, the first was a near disaster which I will relate because it made me think of the workaround instinct to avoid this condition, and the second two were recovered successfully during a VRS condition.
In any case I came up with this response to the VRS that so far has peeled me out of it twice now when it has happened. When it enters this state it will very likely begin yawing one way or the other as it doesnt take much to cause a net torque on the airframe. I immediately release the right stick and push the left stick in the direction of the yawing spin if I am watching the Phantom2 flying without FPV. If I am flying FPV then I push the left stick in the opposite direction of the yaw while doing nothing else. Its instinctive to most of you I am sure but the key is to do NOTHING else it appears. No throttle as others have mentioned and no translational motion if you can avoid it initially. I hold that stick for what really amounts to a fraction of a second until the heading stabilizes even if still wobbling like mad. Then the craft is falling but now on one compass heading. Once that happens I release both sticks. During the two recovery incidents of VRS, the craft stabilized both times and I avoided the VRS CRASH condition. Plus there is an onset condition that I recognized too where the craft beings to go unstable and I have thwarted that a number of times before entering a full VRS condition as well. The onset condition is more of a 'feel' that you notice and it can only be noticed if you are not flying stabilized FPV but visual distance. However if the Phantom2 enters a full VRS condition so far what I am doing seems to be working which his why I am sharing it.
The very first time VRS happened was that near disaster I mentioned up top. It was after that event that I began to think about how the props are stalling and how the torque causes that yaw which is really the major problem in the VRS it seems. During that first scary fall, the Phantom2 began wobbling like mad and so violently in fact that one prop was cut down a full 1/2" by a prop guard as it turned out. As it fell out of the sky like a plastic out of control buzzsaw I knew this wasnt going to be good when I saw white plastic pieces flying off here and there... It also yawed madly BUT as it did that yaw, I gave it what it needed with the left stick to stop the YAW instinctively because of some past heli experience and then released the sticks. I was amazed to see that it had actually nearly fully stabilized but only just before it hit the ground, where it BOUNCED off an asphalt road and back into perfect, stable flight! As I picked my heart up off the roadway where I was flying, the quad simply sat there in the air pretty as could be, no worse for the wear apparently.
I landed and only then noticed that one prop was short at one end by 1/2" (the white plastic pieces flying off during the drop) . There had been no sign of imbalance after the prop cut and the bounce so the autopilot was able to compensate. I was impressed with that but not my bounce...
So from that incident I realized that if I could STOP that yaw, the NAZA had half a chance to stabilize itself because the yaw itself is compounding the stall as you enter what amounts to a flat spin condition. Give it more vertical airflow and the copter has a good chance of recovery I figured.
In two later incidents, one where a brisk headwind suddenly stopped as I was descending rapidly (using that wind to simulate a forward motion) the Phantom2 immediately entered a rapid onset VRS but I had practiced that instinctive response in my office to get muscle memory going and the yaw stopped followed by a full release of the sticks bringing me back to full stable flight. In the second VRS incident I was flying FPV and saw the landing gear suddenly violently appearing in the view followed by the motor pod on the opposite side as the world was spinning. There were literally NO references other than knowing which direction it was spinning by looking at the screen. I pushed the left stick in the opposite direction of the spin in FPV and the heading stabilized, followed by full stabilized flight returning approximatley 1-2 seconds later.
So if this is worth anything to anyone, this has been working. It may not stop all VRS crashes but its better to have a plan of some kind it seems.
I see lots of VRS discussions on the board so if this yaw thing has been discussed countless times then please dont flame me for mentioning it. I dont read every post out here on VRS but wanted to relay the experience I had.
Thanks,
Marc