What means increasing Rate.POut during flight?

ArduCopter Copter 4.6
1

Hi,

I was (Re)setting up my copter using AMC program today and found an advice at 7.1.1 Check for Motor Output Oscillation that if there is audible vibrations/oscillations during flight, review the RATE.*out values.

I got another copter which has been experiencing audible vibration near the end of long flights (30mins plus). When it is coming back, I can hear pulsing humming sound (but vibration level wasn’t increased.)

There is higher(around 20~30) X-axle vibration when it is flying into wind but it is not too bad without wind at around 15.

Rate.Pout is increasing like this. If I understood correctly, it increases when error between Desired and actual pitch increases. Then, why would the error gradually get bigger? There wasn’t increase in Rate.Rout. Would it have something to do with X-axle vibration? I also found Desired Pitch graph is not as smooth as that of Roll.

image
image2047×1159 234 KB

image
image2040×1149 278 KB

PWM ranges of all 4 motors look slightly wider at the end than begining of flight as well. This is only an example of motor4.

image
image2044×1149 232 KB

Throttle input and scaled throttle output values decreased. Would it have something to do with sagging battery voltage?

image
image2041×1152 233 KB

Someone said symptom could have something to do with MOT_THST_EXPO parameter. It is set as 0.65 now but Hobbywing X9 ESC might already has a linear output which needs 0.01 instead of 0.65 in that case. I couldn’t find a right answer for it yet. Or could it be MOT_BAT_VOLT_MIN/MAX params?

The copter wasn’t tunned tightly and correctly at the moment. I know it needs to be started from scratch but yet to be done. For now, I just want general idea what possibly can cause the copter makes an oscillating humming noise at the end of each long flights. It is completely gone everytime I restart copter with new batteries for the next flight.

Any idea or advice is greatly welcomed.

In case you need the Log,

2

You have terrible oscillations in your motors. reduce pids and that will get better. your high vibrations are probably a result of the oscillations.

image
image1613×446 125 KB

Also upgrade to the latest 4.6.3 firmware…

image
image529×377 11.1 KB

1 Like
3

Thank you for taking time to look into logs.

You’re right. It needs proper setting from the scratch.

I was just wondering why oscillation wasn’t present at the beginning but gets worse as flight continued.

4

PIDs are scaled up based on your voltage drop, that’s might be why.

5

Thanks Mike.

That’s what I was suspecting too and that’s why also I wanted to set out correct MOT_BAT_VOLT_MIN/ MAX values. Is there any other params which can affect PID change on battery sagging?

I also was thinking MOT_THST_EXPO might have something to do with the symptom.

Still, I don’t think any of our guess would fix the issue basically before the copter is tunned up properly again. I have no doubt that it needs to be restarted and am not trying to solve the problem with few param changes either.

6

I’ve not seen evidence for that.
I predict
MOT_THST_EXPO,0.79
will be correct.

1 Like
7

Thanks xfacta for your input.

I referred to THIS thread first and also tried myself with data table on their website which resulted as below.

1779700191499
17797001914991292×1365 66.2 KB

Later, on the other hand, when I asked hobbywing, I got only one graph from them.

Screenshot_20260525_172017_KakaoTalk
Screenshot_20260525_172017_KakaoTalk1080×930 131 KB

And the staff I was talking to said yes when I asked him whether 0.9 would correct value for 34 inch copter. I couldn’t tell which is correct yet.

8

LOL. The graph from the Arducopter Tuning Guide. Thanks for the help Hobbywing!

1 Like
9

I tried MOT_THST_EXPO 0.01 today and it worked. After several long flights, copter didn’t show any symptom of noise and visual vibration at all compared to previous value, 0.65.

It seems Hobbywing X9 motor/ESC package does come with output linearization.

There will be lots of chance to verify whether it really is fixed in the next few days and I will update if there is any problem found with flights with current thrust expo value, 0.01.

I still do not clearly understand though

  1. Why the motor thrust scaling only can be noticed at the end of long flight when voltage was sagged. or Could it be kind of certain error accumulation during the flight?
  2. I could hear fluctuating noise and see vibration on arms/legs but vibration level shown on the screen was still same, low. It felt like FC was just part of oscillation and didn’t log those as vibration.