Nervous micro-oscillation

Hello all,

I was wondering if someone could point me in the right direction how to solve this micro-oscillation I see in the flightlog data. I am experimenting with a newly build copter with fairly small (4014) motors and although the copter is fairly ‘locked in’, stable and flies pretty well I can hear a slight ‘nervousness’ in the motors when flying.

In the flightlogs I can see a small micro-oscillation in the ‘actual’ vs ‘desired’ roll & pitch.

And I see a similar nervous effect in the ‘rates’ envelope.

Does anyone have any suggestions how to reduce/eliminate this nervousness?