3.2.1 Autotune sets impossibly low PIDs

I experienced a crash due to extremely low PIDs after doing Autotune with V3.2.1 on my ArduPhantom (frame info here: diydrones.com/profiles/blogs/ard … m-airframe).

The resulting PIDs were a small fraction of the minimum value specified in Mission Planner, and the result was the copter couldn’t control roll or pitch at all, and tumbled immediately to a crash. I should have switched back to the old PIDs to prevent the crash, but my instinct to stop the fall was to switch to stabilize and increase throttle instead, which was the wrong thing to do in this particular situation.

Autotune dataflash log attached.

Any insight into what’s wrong would be greatly appreciated.

can you do that with IMU , and RCOUT logging enabled ?

Hi Andre, here is an Autotune log with IMU and RCOUT turned on. The Autotune didn’t complete though, because I ran out of battery and had to land. It only has enough battery to finish the Autotune on a perfectly calm day when little re-positioning is required. Sure would be nice if Autotune had an option to pause and continue after a battery change.

Hey David,

Did you ever figure this out?
I had the exact same thing happen to me, insanely low PIDs that I fought to control instead of just switching back to the old PIDs… It was mostly easy to fix, but it did rip the pins off of my power module, that was unfortunate.

No answers or replies yet, it would be nice if someone could take a look at the log.

I’ve the exact same problem. Mine is a 430mm frame which flies fine with the default PIDs. However, the PIDs after Autotune are very low (lower than lower limit of 0.08 on P Gain). The result is uncontrollable.