Originally Posted by: crun Don't really agree. I would say PT100/1000 thermistors would be the most common industrial sensors I have seen.
I am still re-calibrating Philips PTC silicon thermistors in a telemetry system, made in 1994. Of the 100 or so still in service, none have detectable drift. (as an aside, the generation of fixed point binary polynomials with truncation optimised fitting, was done in Mathcad)
High beta NTC thermistors had the best resolution, accuracy, heat-injection, thermal leakage, and the lowest noise by a couple of orders of magnitude when I worked with ultra precision sensing. Not so good when hit with a hammer, I will grant...
=============================================
Your question was about plotting pairs of data, do you agree it works ?
No matter the "grand Savants" measuring temperature, they will never do exactly.
That beautiful note:
"On ne pourra jamais déclarer les Wr du thermomètre théorique à résistance de platine;
d'une part inférant à l'incertitude des mesures, d'autre part concluant le canevas
résolvant. En d'autres mots: les Wr ne pourront que refléter la coïncidence résolvante.
Dans les faits, le thermomètre théorique à résistance de platine n'existe pas par
privation de la nature de se reproduire elle-même exactement et infiniment au grès de
la transformation de l'homme. La nature ne donne le platine que sous forme de pépite
ou par extraction. Le Créateur a donné la température [état et sensation physique]
mais pas le thermomètre."
What about Experimental Quantic Computers that must operate @ max 0.02°C from absolute ?
The three attached documents concern industrial temperature measurements.
Types J, T were developed at time of Mathcad DAEP [Jean & Robert]
NI [National Instruments] were delighted of the method as I could see their next
Mathcad based Technology.
The Platinum Pt 100 exists nowhere else than in this Smath document.
The method to the coefficients is secret to only myself. I doubt it could be
reproduced in Smath [didn't attempt].
The previous Thermistor exercise is a piece cake. Some "Cryogenic sensors"
are nearly discontinuous, we fit a model via 'linfitCheby' piece of cake for Smath.
Please, don't hesitate to attach your raw data pairs ... from lab, school ...
Inst_Type T Copy.sm (53kb) downloaded 48 time(s). Inst_Type Pt100 Copy.sm (59kb) downloaded 44 time(s). Inst_Type J Copy.sm (112kb) downloaded 42 time(s).