and here's another log after the tune. It didn't perform the pitch tuning (forward & back). What could be the possible problem? appreciate your help.
Embarrassingly I did not save the log files and for some reason, I am not seeing all my logs on the APM anymore... is there a reason why the APM would not store all my logs? I think I've done a total of 10 flights and all I see on there now are the last 4 which are post autotune. I do however have my parameters from pre and post autotune and the ones that I am currently using. Thanks for getting back to me on this issue.
By the way, it does not "over pitch/roll" while in Autotune, but as soon as I am done with AutoTune and give it extreme stick commands, it went ape and managed rolls and pitches near 90 degrees.
Could you try reducing ANGLE_RATE_MAX to 9000 and tell me if you see the same problem?
The problem seems to have disappeared with my new PID settings and some softer foam under the APM. How would you like me to test it? Run AutoTune again?
What settings did you change?
just the Rate PIDs for pitch and roll and a soft foam under the APM. I avoided Autotune as it pushed the PIDs up high to the point where it would over correct and flip and become unstable. Are the PIDs reflected in the Parameter files? (sorry, quite a novice question I know).
After the Autotune resulted in a crash with the Autotuned PIDs I just decided to go to very low PID settings preferring a lazy copter to a hyperactive one with suicidal acrobatic tendencies. I believe you can see the difference in the Rate Roll, Rate Pitch between the PreAutoTune.param and PostAutoTune.param files or?
Yeh, that is a little strange. Autotune normally reduces Rate P. I would need to see autotune logs and pictures of your frame to help you further.
If you are happy with the defaults then you may prefer to not do auto tune though.
Ehh... I'm sorry, it's not the Rate Roll and Rate Pitch that went wacko, it was the Stabilize Pitch and Stabilize Roll value it seems. Before Autotune, Stab P and Stab R were 4.5, post were 5.25. I since lowered them to 2.0 and the copter was nice and sluggish to lateral inputs. Yesterday I boosted them to 2.8 to make sure that Roll In and Roll were closer to each other and so far that seems to have made it nice to fly.
I have also changed the THR_ACCEL_P from 1.5 to 1 and the THR_ACCEL_I from 0.75 to 0.5. THR_RATE_P has gone from 6 to 5 as well. Sorry... It was difficult to find the "problem" so I started changing a variety of settings and don't know which one was/is responsible for the erratic behavior that I was experiencing. I just wanted to get the platform to fly without incident.
Here is the latest log file:
Hi Leonard, embarrassingly, I misread the Parameters. It did reduce the Rate P by 0.01 but it pushed the Stabilize P values up to where Roll was higher than Roll In. After changing the foam under the APM to a softer foam and reducing the THR_ACCEL P and I rate by 33% I have not been able to replicate the sudden extreme pitch and roll and altitude change that happened while flying in Alt Hold and Loiter.
I would consider setting ANGLE_RATE_MAX to 9000 and see how you like it then increase the stab values again.
If this all feels good then I would consider doing autotune again.
What you are doing sounds pretty sensible from a manual tuning perspective. The ANGLE_RATE_MAX parameter is an additional parameter that lets you use larger Stab_P values without overshoot. So you get a sharp feel without flipping over.
I am not pushing you to do any of this. After all, if you are happy with the way it is flying, why change it. Let me know how you go!
Leonard, does changing the ANGLE_RATE_MAX to 9000 mean 90 degrees max angle?