At 10:07 PM 6/5/01 -0500, Nick wrote: >Ok, I've been browsing trying to find the right solution. I'm needing to >sense temperatures in a very resonable range, like (-20F,100F). No big >deal. I would LIKE to simply use a thermistor and an A/D but I'm having >trouble finding any information on the operation of thermistors in general. >I've also seen the Dallas Semi 1-wire devices and those seem nice, but >overkill if I can simply use a thermistor. The chip I'd LIKE to use is the >16F870, and seems adequate for both jobs. Anyone have any other >suggestions/experiences they could lend to this? You first need to define your mission requirements: 1. What accuracy do you need? 2. What range of temperatures must be measured? 3. How durable must the sensor be? 4. What noise environment must be experienced. A thermistor in a voltage divided with an equal midrange resistor is reasonably linear and accuracy over about a 50 C range. (Reasonable = 1 C). Also reasonably linear and accurate is a diode-wired metal can transistor driven by a 10-50 uA current source superglued to the object being measured. You use the -2.1mV/C to measure temperature. Both of these are limited to -20 C to 150 C max, with the error degrading above 100 C. Wider temperature ranges are best measured with thermocouple wire. Type T is easiest to work with, since it uses one copper wire and has a high temperature coefficient. It's good to about 400 C, although you'd need teflon insulation to accomplish it. Type K is good over 1000 C, but that's uninsulated, or with glass insulation. With plastic, the same limits as Type T. Thermocouples need 200x amplifiers to get to reasonable measurement voltages. For a narrow range of temperatures near 25 C, an unlinearized thermistor is very accurate (0.2 C) and very sensitive. Many solid state sensors are also available for this scenario. Before you get excited about 1 C accuracy, be aware that doing better than this with ANY method is very difficult, except at a single measurement point temperature. Ask yourself: How am I going to calibrate my thermometer when I'm done, and how accurate is the tool I'm calibrating with? Thermocouples and commercial solid state sensors have the convenience of out-of-the-box calibration to within 1 C or sometimes better. Thermocouple wire is good to 1 C over a fairly wide range (0-100 C) of temperature. For most purposes a linearized thermistor (i.e., thermistor voltage divider or bridge) is sufficient for all reasonable purposes. ================================================================ Robert A. LaBudde, PhD, PAS, Dpl. ACAFS e-mail: ral@lcfltd.com Least Cost Formulations, Ltd. URL: http://lcfltd.com/ 824 Timberlake Drive Tel: 757-467-0954 Virginia Beach, VA 23464-3239 Fax: 757-467-2947 "Vere scire est per causas scire" ================================================================ -- http://www.piclist.com hint: To leave the PICList mailto:piclist-unsubscribe-request@mitvma.mit.edu