[sdiy] EPROM has gone defected ?

[Czech Martin]

Well, an EPROM (and also FLASH) works by capturing electrons in an isolated
area, usually some MOS gate, the isolator is in this case SiO2 or Nitride.

Data retention is a big problem if you develop such a memory.
There is always some probability that electrons will hopp
over the barrier, and this will get worse if the temperature
is elevated. Also defects in the isolating barrier can lead
to "bleeding" of the charge.

Silicon processing has much advanced in the last 20 years,
so the memorie quality in the average got better and better.

For critical applications we do retention test (elevated temperature
to speed up the aging). We do this because still there is a number of
FLASH chips that will not keep the promised life time of 100000
hours. Perhaps 1 out of 1000 will come up.
After all extensive electrical testing has been done.

In the past it seems that the quality of such memories was
not that good. Depends also on the supplier and the testing
philosophy. Testing costs money.

So the quality depends on a lot of things, today such failure
is certainly seldom, but in the past I am not so sure.

m.c.

> -----Original Message-----
> From: owner-synth-diy at dropmix.xs4all.nl
> [mailto:owner-synth-diy at dropmix.xs4all.nl]On Behalf Of 
> WeAreAs1 at aol.com
> Sent: Freitag, 17. Oktober 2003 21:31
> To: synth-diy at dropmix.xs4all.nl
> Subject: Re: [sdiy] EPROM has gone defected ?
> 
> 
> 
> In a message dated 10/17/03 8:37:41 AM, gilwein at walla.co.il writes:
> 
> << Can EPROM programming go defected if you power a synth up 
> without the back 
> battery in the circuit ? >>
> 
> No.  EPROM's can indeed go bad, but not because of a missing 
> or dead backup 
> battery.  BTW, EPROM failure is VERY rare.  Look elsewhere 
> for the problem 
> first.
> 
> Michael Bacich
>