Re: AD9548 EEPROM corruption, Silicon rev "C6".
We are using the AD9548 on a board with success, but have found that two boards appear to have a corrupted EEPROM after operating successfully for some time. We suspect that the problem occurred during power up or power down.
Register 0x0D02 has the value 0x04, indicating "An error occurred while saving data to or loading data from the EEPROM", and I assume this is a checksum error.
The "corruption" has been fixed on one board by reprogramming the AD9548's EEPROM.
I have a second board which is still in the corrupted state. I have a dump of the current register values, and they appear to be the device defaults. This is presumably what happens when an EEPROM checksum error is detected.
The device is configured for the SPI interface, although our standard software does not interact at all with the AD9548 device. Therefore we are reasonably confident that the problem is not rogue or interrupted SPI commands.
Pins M0/1/2 are pulled low with 10K resistors, M3 is pulled high with a 10K resistor, but M4/5/6/7 are floating. Is that acceptable, or can that upset the EEPROM Controller ?
Has this problem been encountered before, and is there something we can do to fix it ?
Is there a way to interrogate the actual EEPROM contents - to see what the corruption looks like - as opposed to reading the register values (which are sitting at their defaults) ?
I'd appreciate any help.