At 12:07 PM 5/22/02 +0100, you wrote:

>Depends how accurately you want to measure the temperature, and over what
>temperature range. For most uses (like measuring the ambient temperature) it
>is probably not worth individually calibrating them.

If you measure the voltage *difference* of a diode-connected transistor
at two currents 10:1 different, you get a predictable output voltage from
most GP transistors of 200uV/K. This is about 10% of the output sensitivity
of the junction itself, but it is predictable to within about 1'C typically
at room temperature without calibration using garden-variety 2N3904
transistors. This method is used to measure CPU chip temperatures. IIRC,
the main error term is the base spreading resistance parameter.

>As an aside I suspect that the cheap LCD temperature probes you can get that
>measure inside and outside temperatures use this method.

No. They use a very nonlinear thermistor and some smarts- no voltage
measurement
at all, nor a reference. The clinical thermometers are essentially the same,
but narrower range.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
9/11 United we Stand

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads