brumster / EDTracker2

EDTracker2 Arduino sketches for Invensense MPU-xx50 based devices
107 stars 48 forks source link

Problem using MPU-9250 #20

Closed thementony32 closed 4 years ago

thementony32 commented 4 years ago

Hi all, Ive been using an old 6050 based EDTracker for years and decided to build a couple DIY versions for use with the Condor soaring simulator. This is the board https://www.phippselectronics.com/product/mpu9250-9-axis-motion-sensor-module-accelerometer-gyroscope-compass-motion/ that I've mated up to a 5V Arduino Pro Micro. The Pro Micro connects and flashes perfectly using the UI (version 4.0.4.0) and the magnetometer calibrates with perfect spheres. However after the gyro auto bias calibration the head tracking seems to have cross talk between the pitch and yaw axis. I've tried three different 9250 breakout boards and get similar issues, or occasionally no gyro values appearing in the three green boxes. I've quadruple checked the wiring and voltages, and for shorts, all seems good, including the reset switch which seems to work appropriately in the UI.

boards1 Mag calibration

Any ideas, surely not all three boards are bad ?.

thementony32 commented 4 years ago

Well, here's an example of how defective boards behave. I purchased new MPU9250 boards from a different supplier and they work perfectly.