I tried rotating the motors very slightly to counteract the torque, but it induced a wobble in the AQ. In my gut, I knew that the AQ should be almost effortless to hover, so I knew there was something causing the drift.
Putting on my detective hat, I swapped the board out with my known good AQ. To my surprise, it also had the same yaw drift. So that basically ruled out the AQ board entirely
So the variables left are:
1) motors
2) escs
3) something else
I am using these new pancake TMotors and SimonK flashed escs, which have been rumored to have some compatibility issues. Simon's test is to secure the craft and go from 0 to full throttle as quickly as possible to see if there are any sync issues (stalled or slowing motors). Putting a car tire over each side of the landing gear, I performed the test, and the motors/escs behaved normally. I was relieved to know that these new motors were not a waste of money, but now I had to look elsewhere.
I looked at the monitor on my Tx and after scrolling around the menus, noticed that the rudder was just barely left of center (had to really look), even though the trim was in the center position. I checked the Tx for mixes and trim offsets but all were 0. I then fired up the groundcontrol software and checked what the radio was really sending to the AQ. I then could see that the rudder was reading about 48% instead of 50% (center). All that was left to do was to add subtrim until it read 50%.
Test hovered and drift is gone
So it was a Tx trim issue, but somehow at neutral trim my Tx is off. Hope this helps someone with a similar issue.