AW: Re: Tempco Resistors

[Haible_Juergen#Tel2743]

> Bernie Hutchins mentioned on one issue of the Electronotes Newsletter that
> it was entirely possible to use the proper sign (+ or -) Thermistor in 
place
> of a Tempco resistor as long as it had the proper temperature specs (3000 
 -
> 4000 ppm/C).  If the Thermistor was off the opposite sign, then that could
> be dealt with by the opamps involved.  (Inverting or not).

The only question is if it is *linear* enough. If this is the case, every
factor above 3300 should work - just combine it with an ordinary
resistor. If you should find a 1k type with 6600ppm, just put an
ordinary 1k resistor in series, and you have your 2k / 3300ppm
stuff. You get the idea. The series resistor would even *linearize*
the thermistor a bit - so maybe if you find a 200R / 33000ppm thing,
you wouldn't have to care much about linearity, just combine it
with a 1k8 resistor. Well, that's just theoretical, of course - I don't
have any clue if such things do exist.

Another thing which I'd like to see is a discrete ( ... array ...) version
of the 3340's multiplier that compensates for temperature. I would't
be surprised if someone showed me a simple ota-based circuit
that scales the CV over temperature some time ... anybody?

JH.