[sdiy] Thermistor cunductance curve question

[Ian Fritz]

At 07:23 PM 7/23/05, Paul Perry wrote:
>If the only data you have is 10K at 21 deg, then you don't have enough 
>data to do a curve.
>The curve probably IS exponential, but the parameters depend on the material.
>
>You could put it on ice & get a reading, put it in steam & get a reading, 
>then you have enough data
>to throw an exponential curve through the 3 points.
>
>paul perry Melbourne Australia
>----- Original Message -----


Hey Paul -- you only need two points to fit an exponential.  (Take the log 
of both sides and you have a straight line.)


>
>I have a thermistor and I tried to find its data sheet instead I found 
>that its part #'s were really just absis and ordinate points for its 
>conductance curve (10k @ 21degrees celsius or written 21T10K ). How do I 
>figure out the curve? Do I use some exponential formula or something?

Here's a site with all the math.  Yes, the behaviour is nearly exponential, 
but the coefficient in the exponent (beta) depends on material (and also a 
bit on temperature).

http://www.betatherm.com/betavalueb.htm

Oh Magnus, did we spoil your day?  So sorry!

   Ian