BMX055 gyro x-axis vibration

[hedazhuang]

Hello team,

I used BMX055 sensor on board. At some moving position, there will be a huge vibration noise on the gyro x-axis on some sensors, but others are good. 

What could be the potential issue?

In the graph blue line is the x-axis readings, when I change another board doing exactly same movement, there is no such noise.

I also tried to change the filter settings, but the noise is still there.

I also check the bit4 at the register 0x3C rate_ok, which is always 1 in real time when the noise appears.

A BiST is enabled and it returns bist_fail = 0, bist_rdy = 1 at register 0x3c. 

Is there anything I can check more?

Thank you.

[Vincent]

What is the meaning of the x-axis in your plotter?  is the the Hz value indicate the vibration applyed to sensor? 

Can you give the settings of sensor during the test> 

[hedazhuang]

Thank you for your reply.

1. My apology for unclear description. X-axis in the plot is the time and actually is not vibration but spike readings. We did a periodic movement. The vibration I mentioned is the peak value on the gyro reading, which is shown in the blue waveform (gyro rate_x reading), green (gyro rate_z reading) and red (gyro rate_y reading). We have a bunch of PCBs, but two of them have this kind of high peak values on the x reading of gyro. I update the comparison of the normal reading with the huge spike reading in the following two graph. Both are running the same movement for our testing. Graph one (left) is the normal reading, graph two (right) is the spike reading. x-axis is time. The sampling rate is 2kHz.

                                             graph 1                                                                                                        graph 2

2. the settings of the gyro is as follows

                                                                                                reg addr                    data

• BMX055_GYRO_BGW_SOFTRESET                  0x14                       0xb6
• BMX055_GYRO_RANGE                                         0x0f                         0x00   GFS_2000DPS
• BMX055_GYRO_BW                                                 0x10                        0x04   G_200Hz23Hz

I also reset the offset by register 0x21.

[Vincent]

Thanks for the detail information. 

So you do soft reset to the gyro sensor,  then put it into 2000 DPS range and 200Hz ODR. 

Then put different sensor in same position and do periodic movement. 

From 2 of those sensor,  you find a peak output from gyro x-axis (around 8 dps) looks like a vibration plotter. 

Is my understanding correct? 

Can you give a more specific description about your periodic movement? 

If you put both sensor on table and keep them still,  did you still see this abnormal output from gyro x-axis? 

[hedazhuang]

Thank you for your reply.

If it is put on the table still, the reading is good with no abnormal readings. 

The test we did is to fix the IMU on the shelf at the same spot, while the shelf also connect to a motor which rotates in the specific pattern. It will generate some vibration to shelf and transferred to the IMU for sure. But the question is, why different IMU has the different performance, which I didn't expect.