It's certainly different. I didn't mean to imply that that algorithm was
identical to the "intelligent" method. Several of my programmers, including a
couple I've built myself, albeit for later (28-pin) parts, use an algorithm
similar to what you describe, though it's still from before the "intelligent"
algorithm became popular. That was about concurrent with the availability of
CMOS EPROMs, IIRC.
more below...
Dick
----- Original Message -----
From: "Tony Duell" <ard_at_p850ug1.demon.co.uk>
To: <classiccmp_at_classiccmp.org>
Sent: Monday, May 06, 2002 5:03 PM
Subject: Re: 2708 Programming Algorithm?
> >
> > I don't think this approach to programming 2708's is at all valid. I've
been
>
> Well, it's pretty similar to the one in the NatSemi datasheet I've just
> looked at (I posted the corrections separately). I doubt you know more
> about programming 2708s that they do....
>
Certainly not. I bought a programmer that would to the 2708's much later than
I ever needed it. Of course it would handle other parts.
I used to let Data I/O handle the 1702's and 2708's for me.
>
> > rummaging in the basement lately and may be able to find one of the
> > "moldy-oldies" (literally, as they've been wet) that has a programming
> > algorithm description in it among my old databooks not yet tossed.
> >
> > The procedure you describe is a post-2716 algorithm (intelligent) which
was
>
> Not at all :
>
> The 'intelligent' programming algorithm is something like :
>
> Apply Vpp
> Set up address and data
> Apply a 1ms program pulse
> Verify the data
> Apply another 1ms program pulse
> Verify the data again
> Keep on applying 1ms pulses until the location verifies (OK, give up if
> it takes more than 50 pulses or something).
>
> THen apply 4 times as many pulses again to ensure that location stays
> programmed.
>
> Then go on to the next location and repeat the process.
>
> The 2708 algorithm is
> Apply 12V to CE/ to go into program mode
> Set up address and data
> Apply a 1ms program pulse
> Go to the next location (in other words change address and data)
> Apply a 1ms program pulse there
> Keep on until yuu've programmed all 1024 locations
> Now do the whole thing again, a total of 100 times (or to be exact so
> each locations sees a total program pulse width of 100ms).
>
> Note the differences. For the Intellegent algorithm you program one
> locations completely and go on to the nect. For the 2708 you apply a
> little pulse to each location, tben go on to the next location. For the
> intellegent algorithm you verify the data after each pulse and only give
> the chip as many program pulses as are needed. For the 2708 you give each
> location a total of 100ms of program pulse and only verify the location
> at the end of programming.
>
>
> -tony
>
>
Received on Mon May 06 2002 - 19:22:36 BST