my company has ordered a few samples of your MPU9250-breakout-board and we experienced quite alot of difficulties getting an i2c-connection established properly.
Our hardware-setup has a standard uart, a csr1010-btle-chip with battery-holder and a connector for an external power supply. The MPU9250-breakout-board is attached through a pin-bracket to the base-board.
We made numerous attempts testing the boards with/without soldering the pull up pads, with no pulls/weak pull up & strong pull up on software-side and different voltages from 3.3 to 5 V.
Apperently the only setup we got to work, was without soldering the pull up pads, a strong pull up on software side and 5 V power. Trying the setup with a lower voltage, resulted in a “no pull down”-effect for the clock-signal of the i2c-scl. The same result happened for the described soldered pull pads, no software-side pull up and voltages vom 3.3 to 5 V.
In a separate test we tried a gy521-breakout-board for the mpu6050, which worked smoothly even with battery at 3 V and weak-pull-up on software side. The problem seems to be the different dimensioning of the resistors for i2c-sda and i2c-scl. While the gy521-board works with 4.7k-resistors your board seems to use 10k-resistors making it difficult to operate at any voltage below 5 V.
Could you provide us with a how-to-tutorial on what options we have to consider in order to get i2c-connection established? Would be realy great if we could get it running even at 3.3 V.
Looking forward to your response!