View Single Post
  #5   Report Post  
Posted to sci.electronics.repair
mike[_22_] mike[_22_] is offline
external usenet poster
 
Posts: 1,243
Default EPROMs nearing end of life?

On 4/3/2017 10:21 AM, John Robertson wrote:
On 2017/04/03 9:59 AM, tom wrote:
wrote in message
...
The EPROMs in one of my machines are at least 32 years old. Reading
online about EPROMs it seems that maybe my old devices are nearing the
end of the life of the memory contents. So I'm thinking I should copy
or have copied the data in these old devices. Am I right?
Thanks,
Eric


Definitely make a copy. Then program a second copy of the EPROMS and swap
them in to confirm the copy. If the copies are good, erase and
reprogram the
originals and keep as a spare.

Also replace any memory backup batteries while you are there.

You will probably be dead before you need the copies.



EPROMs are quite stable in my experience, however backing up and using
the second set of EPROMs to run the product seems reasonable if a bit of
overkill. Make a copy, archive it off-site and keep your burner in a
safe place. I would simply use the originals until they gave me cause
for doubt.

Also - do not erase the originals, simply reprogram (refresh) them after
telling the programmer that the chips are not blank.


I'd like to hear the theory behind that.
Are you saying that a failed bit is never zero?
Is programming a zero to zero the equivalent of twice the
program pulse width? Is that less stressful than an erase?

EPROMs have a
limited number of erase/burn cycles so why knock one cycle off? I do
find that a number of originally good EPROMs (that had valid data) fail
to reprogram after being erased.


If it's that close to failure, wouldn't you want to learn that now?

I burn many classic EPROMs (2716s and up) and PROMs a month, so have
some experience in this matter.

John :-#)#

I understand the desire to do whatever you can to keep old equipment
working,
but this is getting mighty close to, "if it ain't broke, don't fix it."