I'm trying to use a pic16c74 to measure very small changes in resistance
(0.01 % which is given by delta R/R * 100) over a 1 meg minimum (eg.
could go from 1 meg to 1.01 meg) using the a/d with high accuracy.
At the moment I'm hooking up the resistance changing sensor to a 1 meg
resistor (to act as a voltage divider) but the problem is that the output to
the a/d is very centered around 2.5 volts and I would like to use the
complete spectrum to obtain optimial accuracy (eg. 0-5 volts).  My
resistance never changes past 10% so what is the best way to do this?
Also does anyone know the max. gain I can get out of 741 op amp using
the above setup without the noise just overpowering the data?  Thanks for
any help.  Also, I have thought of using a 12-bit a/d.  Would this help?