juan manuel garofalo wrote...

>Does it make sense to build an ADC using a PWM output, low pass filtered=
 and
>a comparator from a 16f627?

>My idea is to conect the voltage i want to measure to one input of the
>comparator and the pic's PWM output to the other input. Then i sweep the
>duty cycle from 0 to the point  at wich the comparator changes. My guess=
 is
>that it's not a very fast ADC, but that's not a concern. Is this a =
workable
>solution or are there any gotchas i should be aware of  ?

This should be workable, though you're right about it not being very
fast. =20

As for the "gotchas", keep in mind that the comparator's input
common-mode range goes up only to Vdd-1.5V; also, you'll need a filter
that reduces the worst-case PWM output ripple by at least 60 db to get
full 10-bit resolution.  You probably would be best with a filter
response which has no time-domain overshoot, i.e. a Bessel response.

As a matter of strategy, how about arranging the software to implement a
"tracking" converter?  You'd have an interrupt on each PWM cycle; in
this interrupt, you'd query the comparator to determine whether the
current PWM output is too high or too low, then increment or decrement
the PWM accordingly.  This arrangement would not be good at following
fast input changes; but it would be superior for slowly-changing inputs
since it continually keeps itself within a count or two of the current
input value.

Just an idea...

Dave

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body