Abstract
This paper seeks to highlight the unexploited potential of industrial platinum resistance thermometers by investigating calibration measurements over the range of temperature between ice-point and zinc-point. Indications are that adherence to the IPTS-68 definition might depend to a greater extent on assembling thermometers in such a manner as to realise a prescribed unique value of the coefficient δ than on requiring the temperature coefficient a to be in excess of a particular specified value. Moreover, implications arise that apparent deviations from an IPTS-68-type characteristic, and the consequent observed variability in δ, are attributable to heat conduction along thermometer stems.
No attempt has yet been made to elucidate the reason for the above empirically inferred constancy of the ratio of the coefficients of terms of first and second degree in the quadratic equation relatitig resistance to the modified temperature t'. It appears that the 'fixed' correction polynomial used to convert t' into t 68 , which was defined to achieve compatibility with the thermodynamic temperature scale, represents the behaviour of a certain type of thermometer rather than of the sensing element itself; although of academic interest only, the 'bare' element would require a different, individually assigned, numerical factor in the polynomial in order to restore agreement. It might also be feasible to adopt the same procedure as a means of overcoming any discrepancy arising for a thermometer whose materials and manner of construction have resulted in a 'non-ideal' value of δ.
Get full access to this article
View all access options for this article.