MY 15kg quad not perfectly tunned even after enabling notch filter and tunning again

[neel_chhatrala]

this is yaw

this is pitch

this is roll
i am unable to tune my 15kg quadcopter with 30 inch propeller and i enabled the notch filter and reference frequency is settle up using test flight. i am unware why the quad heavily unstable at speed of 6 and higher m/s.

here are my pid response for the last tunned copter also attaching the .bin log file for further analysis
i also tried wind speed estimation but it got unstable and suddenly the vibrations go upto 60 also attaching the log file for the same

https://drive.google.com/drive/folders/12SA_Y16vbJwWJEbQQQXMGHHxfjIBMrXr?usp=drive_link

[Janno (Jan Ole Noack)]

Read and follow the How to methodically tune (almost) any multicopter using ArduCopter 4.4.x guide. Download the .zip folder with the param files from the guide, edit each file to suit your copter and upload them one by one as you read along. The guide tells you what to do and which file is needed at any given time. Do not skip any steps! All the steps are essential for all of the following steps (i.e. it is pointless to do notch filter setup and autotune before you did the IMU temperature compensation). Anything that might seem like a shortcut will result in a worse tuned drone in the end.

[neel_chhatrala]

i definetly followed this step but how to overcome this is there any way?

[Janno (Jan Ole Noack)]

The way to overcome this is by doing it properly from the start.

@neel_chhatrala

i definetly followed this step

let me look at your logs… No, I don’t think you did.

According to your tuning_response_flight.bin log file you didn’t do IMU temp compensation.
Or you did do it and you disabled it afterwards… Either way, it’s missing.

Also MAG-Fit is missing

And you have a lot of Z vibrations:

Did you just enable the notch filter or did you tune it for your quad?
Because this doesn’t look tuned:

If you have ESCs with telemetry I recommend to use ESC telemetry as INS_HNTCH_MODE instead of throttle.
And what is that spike at 800 Hz? But I digress.

I don’t know what I do differently, but when I put your tuning_response_flight.bin log through the PID web tool the plots don’t look as bad as yours.
Roll:

Pitch:

Yaw:

Regarding your barocompensation.bin file:

Your BCOEF values seem too high:
EK3_DRAG_BCOEF_X = 516.85 at 15kg results in a frontal area of 0.029m²
EK3_DRAG_BCOEF_Y = 544 at 15kg results in a side area of 0.027m²
For reference a 30" long 25mm carbon tube has 0.019m² and the lowest cross-section of a typical 12S (2x 6S) 22Ah battery is 0.011m². These two alone are higher than your values and that doesn’t include the rest of the frame, motors, landing gear, sticking-up GPS sensor(s) or any other payload.

Also EK3_DRAG_MCOEF is 0, so I guess you skipped MCOEF?

I was absolutely serious when I said “Do not skip any steps!” (hint: that’s why it is bold )

[Janno (Jan Ole Noack)]

Also isn’t this basically the same question as here?

Underdamp roll and aggresive pitch in 30inch propeller quadcopter

We build a 30" propeller quadcopter. We first manually tune the drone so that it will ready for Autotune. I did Autotune for several times even after I enable notch filter and again did Autotune but the results are same. Our frame is stiff enough so that vibrations never occur. Here I attached my log file of the flight . Please share valuable suggestion to us. https://drive.google.com/drive/folders/1uEXy0oglCZsjggTbztSrYSUAB8gReqeT

Please don’t double post because you didn’t like the answer. What do you expect to get by asking the same question again?

Do you want a drone that flies or a drone that flies great?
If you want a drone that flies, congratulations you are done.
If you want a drone that flies great, you will have to put in the work to get it right.

[neel_chhatrala]

we tried this but at final stage of the wind speed estimation my quad goes heavily unstable so after this i didn’t took any test flight for it

[neel_chhatrala]

i will explain what i have done firstly i ran autotune to the quad and then i enabled the notch filter further this i tunned the roll,pitch and yaw axis with 0.075 aggression.then i again autotunned the craft at a aggression of 0.1 also i first tunned all three axis and then i tunned roll and pitch again after yaw as discussed in the methodologically tunned article

[Janno (Jan Ole Noack)]

I will not answer you again before you did IMU temp cal and magfit. Refer to my previous answers and the tuning guide to know what is wrong and how to fix it.

Doing autotune out of order is pointless at best and stupidly negligent at worst.

[neel_chhatrala]

Okay then we are truely sorry for that but please give us your valuable time and solution for the same

[amilcarlucas]

@Neel Good news, I improved the Blog post text: It now tells you 5 times to download the .zip file.

I hope that makes it explicit enough that you need to download the .zip file, edit the provided files and upload them. By doing that you will efficiently use your time, and ours.

[Pimpelpimpel]

Hi, as far as I know, the best results for cuts down vibration issue is using physical dumpers. I did some tests and while I used soft rubber dumpers and added some weight to pixhawk I got 10 to 20 times less vibration than without it. Of course, I’m still writing about copters with prop size bigger than 24 inch.

As far as I tested notch filters, harmonic filters doesn’t work properly for my copters,

[amilcarlucas]

Yes, the best vibration reduction method is to use physical dumpers

And … incorrectly configured harmonic copters do not work well on any copter.

Post a .bin file so that we can take a look at it.

[neel_chhatrala]

Yeah, for that I tried to attach the physical damper in my quad I will be attaching the log file for the same