Joe McCauley wrote:
<snip>
>Does anyone know how the e2prom fails if wou continously write to the
>same bytes?
>Do the bytes being written to not work any more or is the entire e2prom
>wrecked?

Joe:

Look at the Microchip EEPROM Endurance software, it can help some. Most (but
not all) failures only fail on bits or bytes.

What you write to the EEPROM can have a lot of effect. Each write (even
writing a 0xFF to a cell already set with 0xFF) causes a full erase (setting
the cell to a 1) of the byte followed by a write cycle. Stress per cell
(bit) is related to the number of erase cycles plus write (1->0) cycles
done.

Continuously writing all "0"s to the EEPROM is the worst thing you can do.
Continuously writing all "0xFF"s to the EEPROM is the best. If you are
saving say 4 bits of data it is best to set the unused bits to "1" than "0".
It also would be better to alternate the which nibble you are saving into.

Some of the older EEPROM devices allowed the programmer to separately
control the erase and write cycle. This allowed you to set the byte to all
1s and then selectively clear each bit. This allowed you to count to 8 with
only 1 full stress cycle on the EEPROM byte.

I have an application where I am saving BCD data. To minimize wear I write
inverted data since there is about 60% 0 bits in a digit, plus the data I am
writing tends to have more 0 digits (4 0-bits) than non-zero.

However having said all this I don't have any history with this product to
know how effective this method really works. The product should start
wearing out in about 25 years.

Chip Weller