> I am just about to tackle the A/D convertor. Can anybody give me an idea
of
> how many times you should check the A/D for a single reading? Is simply
> taking one reading good enough or is it preferred to take multiple
readings
> and average them. What is considered "enough" samples for a single
reading?

This all depends on the noise content of your signal, what the maximum
frequency you care about is, available processor cycles, etc.  All else
being sufficient, I usually sample the A/D several times faster than I need
a new reading, then apply a 1 or 2 pole low pass filter.  This attenuates
random noise.  I set the step response time of the filters based on the
highest signal frequency I care about.  This also has the advantage that a
software value is always kept "live".  When the software needs a new
reading, it reads the current output of the filter, which is always
available immediately.  The A/D routine and the analisys routine run
independently of each other.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olin@embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics