MPU9250 generate different values after every RESET, why? about m5stack HOT 11 CLOSED

It is best to calibrate the magnetometer on startup, and the magnets in the bottom will interfere with it.

from m5stack.

My guess is you are converting the the high and low bytes as small endian. 6050 / 9250 is big endian.

from m5stack.

from m5stack.

Hi my friends.
can anybody explane me, why MPU9250 generate different values after every RESET?
i use a basic example for MPU9250 with no changes, and my M5Stack with built-in MPU9250.
friends. my mpu9250 gyroscope sensor is constantly restarting M5 stackable .

from m5stack.

Hi Sterpa ,
did you solve this problem ?
Still I have problem with my M5 stack

from m5stack.

Hello,

I've the same issue. Someone on m5stack forum pointed me this issue and this article to calibrate it: https://appelsiini.net/2018/calibrate-magnetometer/
I will see if it solve the problem.

from m5stack.

Any updates on this issue? Has anyone tried to port the https://github.com/kriswiner/MPU9250/blob/master/MPU9250_MS5637_AHRS_t3.ino to the M5Stack?

from m5stack.

Hi Folks, finally figured it out. I have been using the mpu9250 on esp32 for a while now and just started using the m5stack.

The sample mpu9250.cpp driver in the stack is not resetting the mpu9250 device during startup. The modification is to the mpu9250.cpp file as follows.

In the init routine change the value 0x00 to 0x80. This will reset the device to a set of know defaults at startup.

void MPU9250::initMPU9250() { // wake up device writeByte(MPU9250_ADDRESS, PWR_MGMT_1, 0x00); // Clear sleep mode bit (6), enable all sensors delay(100); // Wait for all registers to reset

to

void MPU9250::initMPU9250() { // wake up device writeByte(MPU9250_ADDRESS, PWR_MGMT_1, 0x80); // Clear sleep mode bit (6), enable all sensors delay(100); // Wait for all registers to reset

from m5stack.

Hmm may have spoke too soon. The calibration routine resets the device, so an Init followed by a calibration followed by an Init should work..but it is giving inconsistent values back.. hmm Referencing the issue on the developers repository kriswiner/MPU9250#232

from m5stack.

I am having same issues. Testing now with processing script to figure out.
Guess GPS or the magnet on base is causing interference.

from m5stack.

Hello, tested the following code: https://github.com/m5stack/M5Stack/blob/master/examples/Basics/IMU/IMU.ino
Seems like there isn't any problems, it probably was a software issue that got fixed. make sure you have the latest M5Stack library version before testing it, regarding roll pitch and yaw, it takes time to stabilise so eventually after few seconds it will go back to the position as before the reset, it might require calibration. closing for now.

from m5stack.