Severe vibration issues

[Max]

[bluuu]

yes, you are right Max, I thinked about Tilt RATE D

[hogster]

Just to bring this thread to a satisfactory conclusion, a LOT of these vibrations were caused by my Tilt Rate D being too high. I've lowered it down to 10000-12000 now (from 34800) and all the harsh 'scratching' noises and accompanying high vibration levels have gone

I'm not sure what caused the surge in vibration as shown towards the end of the attached log ... it's up for discussion in this thread:

viewtopic.php?f=35&t=3531

Cheers!

David

[hogster]

Hmmm it would seem that this better level of vibration changes back to the horrible +/- 10 level as soon as the wind picks up

I'm at a bit of a loss what to try next, all the props and motors are as balanced as I can get them, and yet as soon as I take off I get very high vibrations levels being picked up by the IMU.

I've got some rubber M3 mounts which I could try mounting the AQ on, but who knows if that will actually help. I'm starting to wonder whether this frame is inherently bad for vibration, perhaps partly due to the coaxial arrangement and partly due to the very stuff motor arms sending all the vibration through to the main chassis. But then again there are many videos of it flying fine with DJI flight controllers ... perhaps their IMU is isolated more from vibration (inside its case), given that it also gets stuck directly to the chassis ...

On the other hand maybe I've still got gain problems / a feedback loop which is causing the frame to resonate? Reducing the Tilt Rate D certainly made a huge difference, but I still can't remove a resonant 'zinging' in the frame. Interestingly I can spool up the props until the X8 is light on the skids and the vibration is always around +/- 2, but then as soon as I take off (in wind) it shoots up to +/- 10 ... Perhaps the undercarriage legs are resonating?

Any thoughts are very much appreciated.

Cheers, David

[hogster]

Here's another interesting problem - why would the ACCX values be higher (particularly in the negative direction) than the other accelerometers? Which direction is X? Fore/aft or left/right? Perhaps it's vibrating against the sides of the box the AQ is sitting in?

Many thanks for any and all input

David

[Max]

Hi David,

You could try the attached firmware and see if it helps at all. This has more filtering applied on the MPU6000 gyro/acc DIMU sensor (otherwise it's the same as the other experimental builds). With the current default filtering, the PID controller may tend to amplify small/high frequency transitions, which could appear as severe vibration issues (both in the logs and in how it behaves in the air).

Start with default PID Rate/Angle values. Set motor mix pitch/roll values to defaults (+/-100 for an X8). Test in manual mode and see if the altitude oscillations and/or motor surging is improved/eliminated. Check if controls are sufficiently crisp/precise.

This makes a big improvement on 2 of my systems when switching to DIMU, both using decent quality ESCs flashed with SimonK's firmware (HK Blue 30A). Both use 10 - 12" props, and hover is around 45% of potential motor output. The difference is less pronounced if the system is more loaded down (eg. 65% for hover), or if using slower motor/prop.

If you try it, please let me know how it feels.

-Max

[hogster]

Hi Max,

Many thanks for sharing this firmware with me. I will give it a try the next time I have a bit of spare time and I will certainly let you know how it feels.

When I upload the firmware will it set the PID values back to their default settings? I can't remember what the default values are otherwise.

Thanks again,

David

[s_sergiu]

Hi Max,

Can you please specify what exactly in the code you modified to get more filtering of Acc/Gyros ?

Thank you,
Sergiu

[kinderkram]

Max changed MPU6000_DRATE_SLOTS to 80 (100Hz) which was 40 (200Hz) before in mpu6000.h

[Max]