> Yes, but that won't save you from getting -0.3V at the PIC pin..
> may this cause problems? For MCLR it is definitely out of specs.

> It isn't for I/O pins, but then again how do we know that component
> tolerances don't make the PIC conduct instead of the Schottky diode
> (both are rated -0.3V, but the former may be -0.3 and the latter
> -0.31 because of tolerances.. who knows?).

While using a Schottky may indeed violate the data sheet specs you can be
reasonably confident that it is OK based on the physics involved.
(Violating datasheet specs in any way places you outside guaranteed
operation, but doing so intelligently may work if you actually make the
correct assumptions about the reasons for the specs. Make the wrong
assumptions and anything can happen :-) ).

The on chip protection diodes are silicon. While at typical currents these
have about a 0.6v voltage drop, they will still conduct much lower currents
at much lower voltages. The law is an exponential one. The 0.3v spec is
(IMO) an attempt to set a level so far away from the full conduction
threshold that the current passed will be truly insignificant. They could
have set the spec at 0 volts but decided that some minuscule current was in
fact OK. The Schottky diode will draw orders of magnitude more current at
this voltage and can be ALMOST guaranteed to prevent the silicon diode
getting anywhere near significant conduction. YMMV BUT I have never heard of
anyone having a problem when using Schottky diodes in this manner.



        RM

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu