>> Anybody have any idea how accurate a sigma-delta A/D you can build
>> using the comparator in the PIC?
>
> AFAIR Scott Datallo achieved about 12 bits (or was that 10?) using PIC
> pins alone, without using a comparator.

(it's two t's and one ell).

Yes, I could achieve about 12-bits using a two I/O pin method. With a
single I/O pin I could get about 8. I think it'd be possible to get even
more if you're really careful.

The comparator helps in a couple of ways. First, the analog signal you're
measuring gets to see a high input impedance. With the method I used
(which I learned from Walter Banks) the analog signal sees a relatively
low input impedance. Second, I suspect (but haven't verified) that being
able to control (or even eliminate) hysteresis will allow one to achieve
even higher accuracy.

Another problem with I/O pins is that their input threshold has low power
supply rejection. This is something easier to control with a comparator.
Also, I suspect that the PIC power consumption increases slightly when the
I/O pin is held a voltage close to the switching threshold.

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