Hazelwood Lyle wrote:
> It was myself who sent in the reference to AN910 a few days ago,
> which is unclear at best when describing the need for multiple
> voltage verification.

There is a lot of confusion around this issue, and I'm not totally clear on
the real answer either.  When I started designing programmers the note about
production versus development programmers and verifying at the Vdd limits
was in just about every programming spec.  Newer parts don't seem to have
that clause.  But the really strange thing is that newer revs of old
programming specs have the clause removed.  That implies that Microchip
thought about it and decided it wasn't necessary, or maybe they just forgot
to remove the disclaimer originally when they went to the F parts.  But then
why didn't they remove it from the 16F84A, which is actually a newer part?
Does it really use different flash technology?

I largely didn't notice that the multiple Vdd readback clause was getting
removed from programming specs, so I kept setting up parts to be verified at
the Vdd limits.  This is probably not necessary for the majority of parts.
However I do feel a bit uncomfortable about removing the two verifies.
Doesn't it make sense that if something is really close to failing that a
error might not show up when at one limit condition but not the other?

I'm going to continue running two verify passes on anything I'm responsible
for shipping.  The cost of a single field failure is probably worth the
extra few seconds on tens of Kunits.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist