EPROM erase times and lifespan

[Pete Turnbull]

On 29/01/2009 16:11, Dave Dunfield wrote:
>> But, yeah, I guess I'll try setting the timer for longer now, to ensure they're good and blank. I
>> was always told that too much UV would damage the chips - I've never left one in for more than say,
>> 15 minutes before. Next time I need to erase some, I'll put 'em in for longer than three or five
>> minutes...
> 
> Which is why a timer is a good thing ... I can tell you from personal experience
> that forgetting a tray of EPROMs in an untimed eraser overnight leads to non-
> functional devices. The ones I left in for an hour seemed to be OK, but I wouldn't
> do that on a regular basis - after about the third time I left the tray in
> overnight, I ripped the timer out of a dryer we were disposing of and made it an
> integral part of my eraser.

Yup, my home-made eraser uses a timer and my bought (by someone else, I 
merely inherited it) also has a timer.  And despite having different 
tubes, they both take about the same time to erase typical EPROMs - 
around 20 minutes is what I usually give.

I remember in the dim and distant past being told that you can sometimes 
recover over-erased EPROMs by baking them in an oven.  Maybe that was 
just 1702s or some similarly ancient device, I've never needed to find out.

And to answer Ian's other question, yes the UV exposure is cumulative, 
and giving 5 minutes followed by three minutes is exactly the same as 
giving one exposure of 8 minutes.  What you're doing is exciting the 
atoms in a floating insulated gate so the electrons leak away, and the 
longer the exposure, the more leak away.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York