At 11:03 AM 5/21/2008, Vasile Surducan wrote: >On 5/20/08, Forrest W Christian wrote: > > I need a low cost (in qty) way to sense room temperature. > >1N4148 I spent quite some time a few years ago building a reasonable but inexpensive temperature sensor using the delta I / delta V technique. Initial results using a bunch of 1n4148 diodes were just awful. Changed to a diode-connected transistor (2n4401 with B tied to C) and got GREAT results. I had a choice: no calibration needed but using really accurate resistors (0.1%) for the delta I part of the circuit or single point calibration using dirt cheap 1% MF resistors. I tried both techniques and wound up using a single point software calibration, based upon the current room temperature expressed in degrees Kelvin. Overall accuracy matched the best thermometer I had available to me over the range of -40C through about +40C. This was a couple of years before the modern SM buss sensor conditioning circuits became available. The stuff I did was derived from a magazine article (EDN or Electronic Design) by Woodward. I thought it worked rather well! dwayne -- Dwayne Reid Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax www.trinity-electronics.com Custom Electronics Design and Manufacturing -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist