Oppie wrote:
"Robert Baer" wrote in message
net...
Thanks a lot; that does explain the matter.
Frankly i paid no attention to what pins could do what, except the
crystal clock pins and the programming pins (which they made almost
impossible to find due to their multiple naming schemes).
Why did i not pay any attention? Because it is a PITA to ramble thru
a ten thousand page document (not really quite that bad, but...) to
perhaps find what one is looking for (and easy to miss).
"Data sheet" ?? Horsefeathers..look at real datasheets on any simple
part and find: simple, understandable, non-conflicting pinning info,
DC specs and AC specs, maybe one app reference. No hundreds of pages
for other things that could be separate - like programming.
Since it works, i think i am going to ignore what pins could be
analog because i cannot use them.
Why? I know too little to do any real programming of that beastie -
the only way i am getting what i want is by learning ladder logic and
using LDmicro.exe .

Every micro manufacturer has a bit of a different scheme for writing
data sheets and establishing pin conventions. Microchip is a bit
different from the others but the information is pretty clear once you
get used to the format and design options. I've done a few designs with
their processors so I've got it down somewhat. First design, was
'WTF???!!!' but it got better with experience.

What you did in digital was the equivalent of ignoring the common mode
range of an opamp. Both are parameters that are very easy to miss if you
don't watch closely.

Oppie
I did not ignore the common mode range; note the biasing puts the
inputs exactly in the middle: 2.5V of the 5V supply.
And at 10 amps input, the drive to the 10K is 1V so at a gain setting
near 2, all is linear for a drive to the MCU as a "hi" for overcurrent
tripping.
As long as the opamp is running in the linear region (as noted in
example), it is well within the common mode range.