> I thought the software fix was mostly ok, but it troubled me that
> it was a problem that cropped up later, after use.

You'd expect (hope !!) s/w to behave exactly as it did when first
written. If the ADC was set up correctly then it should work as
well under the same electrical conditions

My first prime suspect would be the pot. You could even try a
smaller value, say 1k, and see if the noise changes. In good
condition, a wide range of pot values will give the correct ADC
result. A very small value, eg 100 ohms, is obviously going to
draw more current on Vcc, a much larger than spec one should
still give the correct value, although leakage will start to be a
significant issue (result will be lower than correct) and it could be
more susceptible to external influences, such as noise from
switching components like relays, FETs + load etc.

Has there been any change in, for example, positions of wires
that carry a signal coincidental with the time of measurement ?

If in a socket, is the PIC seated firmly ?

Joe


-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2015.0.5645 / Virus Database: 4273/9021 - Release Date: 01/29/15

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.