Re: [Paparazzi-devel] Wrong range for MPU60X0?

[Prof. Dr.-Ing. Heinrich Warmers]

Dear Chistophe,
tank you for the answer.
On our HB-Mini we have also the adx355  +-3g sensor and flu mutirotor and normal aircraft without any problems.
On the mikrokopter autopilot a +-2g sensor is working.
The question is, to cut the accelerations  values if there are lower of -4g and higher than 4g or another border.
Then most of the vibration energy is eliminated before filtering.
The DCM algorithm has inside a border to keep out  high values.
I will make same experiments in this field and report the results.

The vibrations must be damped also mechanically.

-Heinrich

Christophe De Wagter schrieb:

Dear Heinrich,

In most planes we have, saturation is a bigger issue than resolution. Vibrations are the biggest culprit as g-forces from turn rarely exceed 2g while a rigid carbon frame with an unbalanced prop can easily generate 10g, and on helicopters we measured over 100g frame vibrations easily. 

But since filtering is enabled, this saturation might not even be visible in the down-linked measurements so peak-tracking/saturation tests should be done with the internal sensor filtering off. 

Which peak values do you measure at full power in turbulence? We always measure raw unfiltered values of much more than 4g and through filtering it then becomes 2-3g too. 

-Christophe 

On Thu, Apr 25, 2013 at 6:06 PM, Prof. Dr.-Ing. Heinrich Warmers <address@hidden> wrote:

Hi,
Today we implemented and tested the temperature compensation for the MPU60X0.
It works fine.
By testing, we fond that the raw  value where  2000 for g  i.e. the range is +-16g.
If we want to correct the angel zero it would be much better   if the range  is +-4g is used since then the value is 8000 for g.
The resolution grows from 11 bit to 14 bit.
I think the default value should be set to +-4g. range.
Usually we have no higher values than 2-3g for our kind of aircraft.

Heinrich




_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel