Throttle Oscillations at ANGLE_MAX/MOT_BAT_CURR_MAX

I’m finishing up a copter build and all I have left to iron out are these oscillations near max allowed angle and/or motor current max. I thought I solved this in previous builds with pilot_xy_p or vel_xy_p, but they don’t appear to be doing anything anymore (and the descriptions and options values are removed). I also can’t find throttle_accel variables.

Can anyone point me in the right direction? Here’s the log. Can see the oscillations clearly in volt/curr graphs.

I took a look at the logs to see if I can spot anything, but I’m not very experienced.

To me it looks like once you reach about 70amps of current draw the oscillations start and continue until current draw drops right off to almost nothing. I’d guess at an issue with your power distribution or measurement board, or even possibly your battery or a connector. Even if you had a 90amp power brick, 70amps is still a lot of stress.

Also I’d be inclined to check for motor/ESC desync, although RCout didn’t look especially crazy to me, and this is getting into uncharted territory for me.

I was about to say try setting these params:
MOT_BAT_VOLT_MAX
MOT_BAT_VOLT_MIN
MOT_BAT_CURR_MAX
But I see you have already. It looks like the CURR_MAX is being reached and everything begins oscillating as arducopter tries to limit output bu maintain attitude.
Maybe reduce the max lean angle?
http://ardupilot.org/copter/docs/parameters.html#angle-max-angle-max

Maybe set MOT_BAT_CURR_MAX to 60 and/or ANGLE_MAX to 3500 centidegrees (or at least 4500, you’ve got 6000 in there)

xfacta, The culprit was my yaw d term. I performed this same test using an autotuned profile and it didn’t oscillate. It hovered like crap though, the autotune just isn’t working on this build. So I manual tuned and applied some d yaw and that oscillation started. Removed it and it still yaws beautifully and no more oscillation near max throttle. Maybe that’s why autotune never applies yaw d.