>Actually the temperature needed for this process is between 80C and 150C..
>So that makes it a lot more achievable. 0-150 +/-2 is 1.3% error, if I
>offset the conditioning circuit to start at 80, that means there is a 70
>degree range so the error is about 2.8%.

This changes things quite a bit, but I still don't have a drop-in solution
for you.  The type K thermocouple is actually surprisingly linear.  You
might be able to get away with the near-linearity through the 70 deg. C
range without using any computations at all; just take the A/D and scale it
through the 12bit range.  To make things simpler, you can use an 8-bit A/D,
which would give you +/- 0.6 deg. C resolution.  Also, if you went to a
non-flash PIC, you can use the 16C7xx with integrated A/D.  I would check
out this avenue first, as it would be an extremely simple implementation.

Alternately, the suggestion for the Analog Devices AD594 seems like a good
fit for this application... the output is 10mV/C and the chip actually is
laser-calibrated and acts as a cold junction compensator.
It might be costly, however.

Another option, if you can find one that's suitable, is to use a TelCom
voltage output temp. sensor.  Their standard product line includes ranges up
to 125 deg. C, but they might have an extended temperature line that gets
you to 150.  These things only cost ~$1.00 in low quantity.

Or how about just a simple NTC/PTC thermistor?  They work up to 150+ deg. C
and it's simple to just feed a few mA through a voltage divider to measure
the resistance change for your temp reading.  Normally they are +/- 1.0 deg.
C.  More expensive and precise varieties are available, like +/-0.2 deg. C.

Lots of options...

--Andrew
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com

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