Microchip has a really outstanding video on cleaning up noise on A/D
inputs.  There are a lot of steps, tho.
http://www.microchip.com/webinars.microchip.com/WebinarDetails.aspx?dDocNam=
e=3Den528480

--McD

On Thu, 2015-01-29 at 01:50 -0500, Josh Koffman wrote:
> Hi all,
>=20
> A few years ago I made a few copies of a project of mine. One of the
> things that it did was to read a slide pot via ADC and send the value
> via serial. When I built them, they were solid, meaning that if I
> didn't move the slider, the value stayed consistent.
>=20
> A few months ago I saw a few of them still in action, and the values
> were...fluttering a bit. So even if the slider was not being moved,
> the value transmitted fluctuated a bit around where it should be (say
> by about 1-2% in each direction). I didn't have much time to
> troubleshoot them, so I didn't take very good notes. Consequently I
> don't remember precisely what happened. I know I replaced the slide
> pot with an used one, but that didn't fix the problem. I replaced the
> slide pot with a fixed resistor, but I don't recall what happened
> there. A vague recollection is that it fluctuated a bit too, but I
> can't be certain. I tried adding capacitance to the power bus, but
> that didn't do anything. I tried a cleaner power supply, but no dice
> there.
>=20
> In the end I solved this with firmware, I basically filtered the
> results. However that lost me a bit of precision on the slider, and it
> meant it had to move more before it was registered.
>=20
> I originally had this set up as a simple voltage divider with a series
> resistor to protect the input pin if the slider was at one extreme or
> the other and the pin was accidentally set to output.
>=20
> I'm about to use the ADC again, this time to read a rotary pot. I
> could connect it the same as last time, but before I do, I'm wondering
> if there's a better way to do this that will result in longer term
> stability. I can't help shake the feeling that the way I connected
> things was partially to blame.
>=20
> Any thoughts?
>=20
> Thanks!
>=20
> Josh
> --=20
> A common mistake that people make when trying to design something
> completely foolproof is to underestimate the ingenuity of complete
> fools.
>         -Douglas Adams


--=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
.