I don't think that would work. I've cooked
EPROMs for months
without any change in their functioning.
So did a partner of mine.... using a homemade eraser made from
a UV sterilizer unit taken from a scrapped dialysis machine.
It was an Intel eprom, a 2732 or 2764, ceramic case, quartz
window. The embedded system we were developing was an auto-
mated knife sharpener (not to be a consumer product).
During debugging, we had the prototype out of the case. Things
were working fine and it was time for the first install into
the van (which would drive from restaurant to restaurant doing
the knife sharpening thing). Put unit in case, fire up, no
operation. Take unit out of case, fire it up, everything works.
We looked for warping of the PC board, kinking of cables, and all
sort of things that coould glitch the unit out. Nada. Finally,
thinking that we were only going to see how the case was screwing
with the PC board from inside the assembled unit, we cut an access
panel in the aluminum, and shined a flashlight inside. It happened
to hit the EPROM window and voila! the unit starts running.
After a long call with Intel engineers, it was determined that
too much UV could fry the EPROM such that it would only function
correctly when light was shining onto the substrate through the
window!
An EPROM that was afraid of the dark. To get the prototype out
the door, we built it a night-lite, and everything was fine.
-doug quebbeman