MPU9250 results are not stable

Moderator: Benj

Post Reply
stefan.erni
Valued Contributor
Valued Contributor
Posts: 654
Joined: Fri Aug 19, 2016 2:09 pm
Location: switzerland
Has thanked: 182 times
Been thanked: 179 times
Contact:

MPU9250 results are not stable

Post by stefan.erni »

Hi Ben

I have a problem with the component MPU9250.

When I read data from the MPU it's reading 10 times correct values and then some wrong data.
I tried the Accelerometer alone or Giroscope or Compass. All the tree have the same problem.
Even if I only read the temperature, it jumps to wrong values.
I tested this on two different boards from different manufacturers.
I suspect results are mixed up.

regards

Stefan

stefan.erni
Valued Contributor
Valued Contributor
Posts: 654
Joined: Fri Aug 19, 2016 2:09 pm
Location: switzerland
Has thanked: 182 times
Been thanked: 179 times
Contact:

Re: MPU9250 results are not stable

Post by stefan.erni »

Hi Ben

I wrote a smal program to save all data from the MPU to an excel file
(Temp;Accelerometer x,y,z; Compas x,y,z; Giro x,y,z)

I get 100 value with the maximum X-axis (from the accelerometer)
I turned the hardware and get 100 value with the maximum Y-axis
I turned the hardware and get 100 value with the maximum Z-axis

There is a problem with this MPU and jumping data.

regards

Stefan
Attachments
Flip-and_Click_Pic32efh100-display_D_MPU9250_A_1.fcfx
(61.51 KiB) Downloaded 168 times
test10-100x100y100z.xlsx
(45 KiB) Downloaded 150 times

User avatar
Benj
Matrix Staff
Posts: 15312
Joined: Mon Oct 16, 2006 10:48 am
Location: Matrix TS Ltd
Has thanked: 4803 times
Been thanked: 4314 times
Contact:

Re: MPU9250 results are not stable

Post by Benj »

Hello,

Thanks for sharing your findings, the accelerometer does seem to be rather noisy. Have you tried turning down the sensitivity from +/-2G to say +/- 16G and seeing if you get more sensoble values.

The compass initially looks bad but then sems to get much better and the gyro looks good. I wonder if the compass was initially picking up nearby electrical noise?

stefan.erni
Valued Contributor
Valued Contributor
Posts: 654
Joined: Fri Aug 19, 2016 2:09 pm
Location: switzerland
Has thanked: 182 times
Been thanked: 179 times
Contact:

Re: MPU9250 results are not stable

Post by stefan.erni »

Hi Ben

One of the bad signal is the temparatur. it's jumping.

stefan.erni
Valued Contributor
Valued Contributor
Posts: 654
Joined: Fri Aug 19, 2016 2:09 pm
Location: switzerland
Has thanked: 182 times
Been thanked: 179 times
Contact:

Re: MPU9250 results are not stable

Post by stefan.erni »

Hi Ben

I used the Interface Board from you. That is a great help. With the csv writer it's very comfortable to save data.
https://www.matrixtsl.com/mmforums/view ... 26&t=22354

The result from the demoboard MPU9250 looks much better with your board. But it is still not perfect.

I will now look at my software again to see if I can make the values more stable. I will also try to connect your board to my hardware and turn off the I2C bus (from my hardware) and only use the I2C bus from your board for the MPU9250.

regards

Stefan
test1.csv
(139.29 KiB) Downloaded 154 times
Grove_MPU9250_I2C.fcfx
(39.99 KiB) Downloaded 171 times

stefan.erni
Valued Contributor
Valued Contributor
Posts: 654
Joined: Fri Aug 19, 2016 2:09 pm
Location: switzerland
Has thanked: 182 times
Been thanked: 179 times
Contact:

Re: MPU9250 results are not stable

Post by stefan.erni »

Hi Ben

I connected your Interfaceboard to my hardware and read the temperatur from the MPU9250 . It is stable and correct all the time.

With my board (PIC32MZ and MPU9250) is reading the temperatur not stable.
With a demoboard PIC32MZ and a demoboard MPU9250 is reading the temperatur not stable.

But with your Interface and the demoboard MPU9250 reading the temperatur is stable and correct all the time.

On my board is a I2C RTCC and reading from there is always working fine.

regards

Stefan
20201016_115244 (3).jpg
20201016_115244 (3).jpg (137.54 KiB) Viewed 3954 times

Post Reply