And your direction of bank was to the left I presume...

I've experienced something very similar in Challenger ultralight, climbing under power.

In hindsight I wish I had experimented with seeing what would happen if the bank angle were perturbed significantly in one direction or the other. If find myself flying a sailplane that exhibits these characteristics, I'll be sure and check that out. Would be good to know, if one is contemplating using that technique for a blind descent in a situation with some turbulence. My concern is that if the bank angle is somehow perturbed past wings-level into a right bank, you might end up in a really tight spiral to the right. But I don't know for sure.

Btw this technique wouldn't work in a 2-22. I'm sure you can guess why. The ailerons are too powerful, in relation to the rudder.

It's interesting to think through the dynamics at play here. Why is this situation stable? What are the dynamics that act to restore the bank angle, after a slight perturbation? Are they significantly different than the dynamics of a benign spiral? I suppose the basic dynamics of a stable benign spiral are-- increased bank angle also increases sideslip which interacts with dihedral to create more rolling-out torque. Decreased bank angle also decreases sideslip, which allows other competing rolling-in torques to increase the bank angle.

In the cross-controlled spiral you'd have something kind of similar going on-- you have a strong sideslip (skid?) from the rudder (yaw string streams toward inside of turn, which is toward the high wingtip), creating a roll torque that tends to increase the bank angle. Any decrease in bank angle will increase the turn rate which will tend to increase the drag of the outboard (low) wingtip and decrease the slip (skid?) angle, decreasing the roll torque toward the high wingtip, and allowing other competing factors to drive an increase in bank angle. Any increase in bank angle will decrease the turn rate which will tend to decrease the drag of the outboard (low) wingtip and increase the slip (skid?) angle, increasing the roll torque toward the high wingtip, and driving a decrease in bank angle.

In a normal spiral, the high descent rate creates a requirement for roll toward the low wingtip just to hold the bank angle constant, so roll damping tends to drive a decrease in bank angle. That's why opening the spoilers helps keep the bank angle from increasing.

But in this cross-controlled spiral, the direction of turn is toward the HIGH wingtip, not the low wingtip. The high descent rate creates a requirement for roll toward the HIGH wingtip just to keep the bank angle constant, so roll damping tends to drive an INCREASE in bank angle. (The opposite would be true if the maneuver were performed while climbing under power.) So opening spoilers would NOT be expected to create a stabilizing effect in roll-- this helps to explain how a similar dynamic could exist even in powered flight. Likewise the drag from the sideslip, and the resulting high sink rate, would NOT be expected to contribute a stabilizing effect in roll.

In a normal spiral, the outside (high) wingtip is moving faster than the inside (low) wingtip, which tends to generate a rolling-in torque, increasing the bank angle. In the cross-controlled spiral, the high wingtip is also the inboard wingtip, and it is moving slower than the low wingtip which is also the outboard wingtip, creating a roll torque that tends to decrease the bank angle.

It's a curious animal. I wonder what other gliders show exhibit this stable cross-controlled spiral behavior-- and how stable it really is?

S