These darned 12CExxx parts bite me every chance they get....

We've started to find in our beta units that occasionally they perform
a reinitialization of the I2C EEPROM, as if this was the first time =
powered
up since programming.  (We use a "magic cookie" type byte saved
to EEPROM address 0x00 to determine if this is the first power-up or
not.  If the code thinks this is the first powerup and it is not, there
is obviously a corruption of EEPROM address 0x00.).

In looking for root causes, I fould the piclist.com page:

http://www.piclist.com/techref/mem/eeproms.htm

in which Ian Chapman talked about AN709.  After reading it, and
looking at the 12CExxx spec, it seems that the 24x00 that is in
the 12CExxx parts does include pull-up resistors, so the first part
is moot.

However, the second part, about forcing internal reset via
software has me intruiged.  At first, it seems unlikely that the
internal EEPROM die would power up any differently than the
processor die itself.  However, there's not much else in the
possibility of EEPROM corruption of address 0x00. (Multiple
peer reviews have uncovered no problems.)

I plan on following the AN709 software reset of the serial
EEPROM, but I ask the PICLIST if others believe this is
necessary or not.

Thanks for reading, and I hope this helps others out as well.

John

(P.S. I've worked through sooooo many problems with the
I2C internal EEPROM access that I'm about to write a FAQ
on it... any interest?)

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads