Olin Lathrop wrote:

> If the AC p-p is below 5V, then you have to amplify it before passing it to
> the PIC.  In either case, you are still subject to sudden spikes in the DC
> voltage (your power supply, I presume).  I therefore consider this a very
> bad way to get a power line clock signal.


Not enirely true, if the PIC input is a ST (schmidt) type,
the signal has low/high thresholds of 0.2v/0.8v so a
1v peak/peak waveform will work fine. These are worst
case figures too. If it is a standard PIC input pin (TTL
buffer type) the low/high is 0.8/2.0 volts. There is
no need for a 0v/5v signal and it's important to understand
this when designing input circuits for the PIC.

I connect all sorts of weird analogue signals directly into
the PIC inputs and although it borders on bad manners
to connect an analogue signal into a non-schmidt pin I
have never had a problem. Provided you understand these
threshold levels and work on the correct side of them
the PIC digital inputs are very reliable. :o)

20 years ago logic chips were very fussy but with
modern logic chips all digital inputs have a decent
hysteresis and stability and will handle an analogue
input well, as long as they are not left "floating".
-Roman

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