Yes. We orginally discovered this problem with our own I2C host which is a STM32 uP. In order to further understand the issue, we were able to reproduce this problem without our host (by keep it in reset state) and use an external I2C master. With external master, we also able to issue different I2C commands (since it's more flexible). We know our BME280 is address 0x76. That's address our own I2C master accesses. We tried 0x77 with external host just to make sure there isnt a chance that BME280 power up as the wrong address.
... View more
Hi, We use BME280 in one of our boards. We use I2C communication scheme. We discovered that there are small percentage of the boards have intermitted I2C communication issues. With those problematic boards, some power cycle the communication is just fine and some power cycle we can not talk to BME280 and receive NACK on I2C SD line. This is power up dependent. If the board power up good. It will stay good. If the board power up bad, it will stay bad until we cycle the power. This seems to happen randomly. In our test setup, we have isolated to only access BME280’s chip ID register. We took our uP out of the picture by keeping it in reset so the I2C bus is free, and connected an external I2C tester to issue chip ID read commands. Once BME280 responses with NACK with some power up, We even tried to read both address 0x76 and 0x77. We still got NACK. This looks like BME280 is lock up. Is there a certain condition BME280 is more accepted to lock up condition? Below is the schematic our BME280 hook up.
... View more