Hysteresis and Instability in Some IPRT Sensors Within Temperature Ranges Extending from $$-196\,^{\circ }\hbox {C}$$ - 196 ∘ C to $$150\,^{\circ }\hbox {C}$$ 150 ∘ C

Abstract

Industrial platinum resistance thermometer (IPRT) sensors or probes suffer from some instability on cycling over significant ranges of temperature and, specifically, from hysteresis in which the resistance tends to follow different paths for increasing temperatures compared with decreasing temperatures. The effect is well known, and cases of quite large hysteresis have been reported in the literature. Therefore, in establishing calibration and measurement capabilities for IPRT calibrations it is important to include an assessment of the performance which can be expected of a ‘typical good’ IPRT and to include this in the overall uncertainty which the laboratory can expect to achieve in such calibrations, even though the effect itself is outside the laboratory’s control. This paper presents results which have been obtained in cycling IPRT probes from four sources within various temperature ranges of current interest at NPL, between $$-196\,^{\circ }\hbox {C}$$ and $$150\,^{\circ }\hbox {C}$$ , to see what levels of hysteresis may be expected. The cycles were carried out quite quickly in order to detect the hysteresis before it was mitigated by relaxation effects, but the time dependence was not itself studied. In most cases, hysteresis was $${<}0.0025\,^{\circ }\hbox {C}$$ between $$0\,^{\circ }\hbox {C}$$ and $$100\,^{\circ }\hbox {C}$$ , and $${<}0.0035\,^{\circ }\hbox {C}$$ when the range extended down to $$-80\,^{\circ }\hbox {C}$$ or up to $$150\,^{\circ }\hbox {C}$$ . Greater instability occurred when the sensors were cooled to $$-196\,^{\circ }\hbox {C}$$ .