BNO055 spike of Linear Acceleration and Angular Velocity

[Justina]

Dear all,

I'm trying to read the linear acceleration and angular velocity register in NDOF mode by using I2C bus. I think if I just put this sensor on the desk and do not move it, the results should alway be 0. But I can get some spikes. I do not know why. Can you help me to explain it?

I also searched the same problem in this community. I got this BNO055 Linear Acceleration But in my code, there is no "time delay" between reading LSB and MSB.

Here is my results. Thank you very much.

[Vincent]

When device / sensor put on desk still, then the output of linear acc and angular velocity should be zero.

BHO055 default ODR for NDOF mode is 100Hz.

If you time scale is second at x-axis of your plotter.  it looks like your table is not stable at all.  it has vibration time to time on the table especially on y and z axis. 

Did you try with different BNO055 samples and always get same result?  

Can you also provide data log of accleration sensor output means linear accleration, angular velocity and accleration sensor are turn on and logged at same time?   

[Justina]

Hello,

thanks for your reply. Yes, my time scale is in second. I do not think, my table is not stable, because I have already tested with other sensors on the other tables. I always get spikes of my results. I think this value is noise and from "FF".

Here is the results from the other sensor. And a part of my log data. Time is in second.

Thank you.

[o_o]

Hi Justina,

You data is ... suspicious ... to say the least. I am very confident that there is an error is how to interpret the data coming from the bus.

Therefore I recommend that you do the following:
1- Confirm with a logic analyzer that the data on the i2c bus is the same as the data in your MCU memory
2- Confirm that you do not have any CAN communication error
3- Confirm that you properly decode the MSB and LSB into a signed 16 bits integer.

After extracting your data, I converted it to decimal and removed duplicate value. As you can see there is a big jump of 256 LSB in the X and Y channel, and in the Z channel there is also a jump of 32 LSB. This data is not normal.

X Y Z
-256 -256 -256
-255 -255 -255
-254 -254 -254
-4 -5 -253
-3 -4 -36
-2 -3 -35
-1 -2 -34
0 -1 -33
1 0 -32
2 1 -31
3 2 -30
4 3 -29
254 4 -28
255 5 -27
251 -26
253 -5
254 -4
255 -3
-2
-1
0
1
2
3
4
253
254
255

[Justina]

Hello,

Thanks for your reply. Can you help me have a look at our schematic for the BNO055? Maybe there is some reasons for the spikes of my results.

Attached is our schematic for the BNO055. SCL and SDA are connected with other fan chip SCL and SDA ports.

Thank you very much.