EEPROM writecycle/life

P

Peter Hydes

Guest
I am using an X28C010 EEPROM device and I have a query that is not answered
by the data sheet. I have emailed the manufacturer with my query but haven't
had a reply.

The data sheet states that each byte of the device is guaranteed for a
minimum of 10,000 write cycles. The question I have is 'if the data being
written to the device is the same as the data that is already stored in the
EEPROM then does this reduce the life of the device for that byte?'.

At the moment the device is written to unconditionally before powering down
the system. A collegue is suggesting to me that we could extend the life of
the device by reading each currently stored value and only writing if the
new value is different. However the logic within the chip could be such that
writing the same value as is already stored does not have any life
decreasing effect.

--
Peter Hydes
 
In article <9a23133f.0311110701.3fd7e68f@posting.google.com>,
peterhydes@yahoo.co.uk (Peter Hydes) wrote:

I am using an X28C010 EEPROM device and I have a query that is not answered
by the data sheet. I have emailed the manufacturer with my query but haven't
had a reply.

The data sheet states that each byte of the device is guaranteed for a
minimum of 10,000 write cycles. The question I have is 'if the data being
written to the device is the same as the data that is already stored in the
EEPROM then does this reduce the life of the device for that byte?'.

At the moment the device is written to unconditionally before powering down
the system. A collegue is suggesting to me that we could extend the life of
the device by reading each currently stored value and only writing if the
new value is different. However the logic within the chip could be such that
writing the same value as is already stored does not have any life
decreasing effect.

--
Peter Hydes
Click to expand...
While nowhere near definitive, I'd be going with the idea that "A write
is a write, regardless of whether the data in the location changes".

My logic?
When you write, the chip has to do exactly the same steps, in exactly
the same order, regardless of whether the value you're trying to write
is the same as what's already at that location, or whether it's a
different value. Either way, exactly the same thing happens (since the
chip can't safely make any assumptions about what the contents currently
might be) on every write.

--
Don Bruder - dakidd@sonic.net <--- Preferred Email - SpamAssassinated.
Hate SPAM? See <http://www.spamassassin.org> for some seriously great info.
I will choose a path that's clear: I will choose Free Will! - N. Peart
Fly trap info pages: <http://www.sonic.net/~dakidd/Horses/FlyTrap/index.html>
 
Nonvolatile technology requires that the cells be
returned to a known condition before writing. For
this reason, the logic driver circuitry does not
have provisions to anticipate whether the updated
data will be different from the previous cells
value. This is not a failure in design, but simply
a pragmatic approach to the cost/feature tradeoff
in the EEPROM's architectural design. In my use of
this technology, I've implimented algorithms which
look for differences between the stored data and
the updated data. This imposes a small amount of
time to the process and extends the EEPROM's useful
life to a projected 350 years! If I didn't do so,
the EEPROM would have only had a useful life of 5
years...at best.

Peter Hydes wrote:
I am using an X28C010 EEPROM device and I have a query that is not answered
by the data sheet. I have emailed the manufacturer with my query but haven't
had a reply.

The data sheet states that each byte of the device is guaranteed for a
minimum of 10,000 write cycles. The question I have is 'if the data being
written to the device is the same as the data that is already stored in the
EEPROM then does this reduce the life of the device for that byte?'.

At the moment the device is written to unconditionally before powering down
the system. A collegue is suggesting to me that we could extend the life of
the device by reading each currently stored value and only writing if the
new value is different. However the logic within the chip could be such that
writing the same value as is already stored does not have any life
decreasing effect.

--
Peter Hydes
Click to expand...
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top