Does a top heavy/ bottom heavy copter has separate auto tune algorithm? how does Copter knows this difference

Attached is the log of the flight after autotune. even after the autotune, copter feels unstabe and sometimes locks in pitch axis oscillations. can someone check which part i am missing.

https://drive.google.com/file/d/1RMdlgC_VO14sssx5ZrIZDsNN9j9PSDds/view?usp=share_link

1 Like

Top heavy or bottom heavy won’t matter.

Connect to MissionPlanner and go to Setup / Mandatory… / Initial Parameters
Put in your prop size and battery cells, accept everything it offers including the suggested battery voltage and failsafe settings.
If you’ve already done this, accept any that you hadn’t previously.

Also set these:
INS_ACCEL_FILTER,10
INS_HNTCH_ENABLE,1 ← set this then refresh params to see the rest
INS_HNTCH_MODE,1
INS_HNTCH_REF,0.16
INS_HNTCH_FREQ,75
INS_HNTCH_BW,37
INS_HNTCH_ATT,40
INS_LOG_BAT_OPT,2

The PIDs look quite low, what motors, props and ESCs do you have?

1 Like

MN7005 230kv, 24inch prop and 40AMP Alpha ESC

Use the settings I suggest and if stability seems OK you could probably run Autotune with
AUTOTUNE_AGGR,0.075

EDIT
Lower ATC_ANG_YAW_P,5.427387 somewhat to about 3.0 for now

If it was me, I’d put back the calculated ACCEL values and try these PIDs
ATC_ANG_RLL_P,6.5
ATC_ANG_PIT_P,6.5
ATC_ANG_YAW_P,3.0
ATC_RAT_RLL_P,0.100
ATC_RAT_RLL_I,0.100
ATC_RAT_RLL_D,0.005
ATC_RAT_PIT_P,0.100
ATC_RAT_PIT_I,0.100
ATC_RAT_PIT_D,0.005

1 Like

I have already enabled HNTCH and after FFT log after this flight, so are you telling this value after FFT analysis?

In the log HNOTCH wasnt enabled and I was able to check the FFT to come up with those values.
What values did you get?

Frequency - 73
Bandwidth - 21
INS_HNTCH_REF, 0.2510964(MOT_THST_HOVER)*May I know where did you took this value from?

Use my values, they’ll be OK - yours were pretty close too.
INS_HNTCH_REF = hover_thrust * (min_freq / hover_freq)^2

1 Like

Ok, Thanks for the valuable feedback.

I did autotune first, and run FFT flight, enabled HNTCH. was that the correct sequence? if not is it ok to do one more auto tune after a auto tune!

It’s OK to run Autotune more than once, that is a good test to see if new autotune values change much.
It’s best to test fly and check for vibrations, do any manual tuning required for basic stability.
Then check FFT and enable HNOTCH
Then Autotune.

I expected the HNOTCH frequency to be about 30 to 35HZ , so you might have to still use that if necessary, even though it’s not obvious in the FFT.

But do set INS_HNTCH_BW,40 ← I checked the FFT graph again

1 Like

Shawn, thanks for the value, but i am confused how values for min_freq and Hover_freq came or from where to take

From the FFT gyro graph

1 Like

ok, defintely will try these PID values. my copter is a perfect symmetric in roll and pitch axis but after autotune see this difference in

ATC_ACCEL_P_MAX,38553.31
ATC_ACCEL_R_MAX,65136 almost changed to double times.

I noticed that so I suggested to put the calculated ACCELs back in place instead of those Autotune values.
Those PIDs are just “generic” and not specific to your aircraft but should be quite safe.
These days it’s more typical to have higher ANG Ps and lower RAT PIDs, whereas older firmware and components you would probably have low ANG values and highish RAT values, especially with larger motors and props.

1 Like