Tool/software:
I found two E2E threads accurately describing my issue, but neither one has a resolution.
In my case I have a custom board Implementing the lp876440C0.
We developed custom firmware and loaded it to the device. We have done this successfully on 6 prototype boards for this project.
While debugging other issues on the board I did one of dozens of power cycles, but after this power cycle the lp8764 reports 0x00 for all I2C register reads.
We have the lp87664 connected to an on board microcontroller I2C port we use for programming and monitoring. On every power cycle we verify that the lp8764 has the proper firmware before we enable its supplies.
This worked dozens of times until it didn't. We only programmed the NVM on this device once, and that was days ago and multiple power cycles ago. We do change the I2C address to 0x48 during NVM reconfiguration. This has always worked fine and helps us recognize un-programmed devices.
I can watch the I2C bus on a logic analyzer. The lp8764 is the only device on this i2c bus. The lp8764 still Acks the I2C writes and reads appropriately and drives SDA low at the appropriate time during reads, so it is not completely dead.
I can scan every address on the I2C bus. Addresses 0x12, 0x48, 0x49, 0x4A, and 0x4B are the only address that Ack a write. They all reply with 0x00 when I try to read register 1.
Did either of those other posts generate any answers to how it gets in this mode?
Any other things I can test while I have it instrumented, before we just replace the part?
