Uncertainty propagation in the calibration equations for NTC thermistors

Abstract

The uncertainty propagation problem is quite important for temperature measurements, since we rely so much on the sensors and calibration equations. Although uncertainty propagation for platinum resistance or radiation thermometers is well known, there have been few publications concerning negative temperature coefficient (NTC) thermistors. Insight into the propagation characteristics of uncertainty that develop when equations are determined using the Lagrange interpolation or least-squares fitting method is presented here with respect to several of the most common equations used in NTC thermistor calibration. Within this work, analytical expressions of the propagated uncertainties for both fitting methods are derived for the uncertainties in the measured temperature and resistance at each calibration point. High-precision calibration of an NTC thermistor in a precision water bath was performed by means of the comparison method. Results show that, for both fitting methods, the propagated uncertainty is flat in the interpolation region but rises rapidly beyond the calibration range. Also, for temperatures interpolated between calibration points, the propagated uncertainty is generally no greater than that associated with the calibration points. For least-squares fitting, the propagated uncertainty is significantly reduced by increasing the number of calibration points and can be well kept below the uncertainty of the calibration points.

Design of a Negative Temperature Coefficient Temperature Measurement System Based on a Resistance Ratio Model

“…The Hoge equation was employed to perform polynomial fitting on 1/T and lnR to achieve higher precision. Considering the calibration equations' fitting accuracy and the system's computational complexity, n = 3, i.e., the Hoge-2 equation, was chosen for fitting in the calibration equations [29][30][31]. Its fundamental form is described by Equation ( 5): In order to verify the accuracy and stability of the current output, the output value of the circuit loop was measured with the standard resistor box instead of the NTC thermistor.…”

Computer-Integrated Device for Acidity Measurement Monitoring in Greenhouse Conditions with Compensation of Destabilizing Factors

“…The articles Liu et al [31] and Vovna et al [32] substantiate the main scientific and practical approaches to assessing metrological characteristics of non-destructive testing using the concept of uncertainty and the error theory. Laktionov et al [10,33] and Dar et al [34] developed the main provisions for planning experimental tests of computerized means for monitoring the greenhouse microclimate.…”

Calibration approach to quantify nonlinearity of MEMS pore pressure sensors using optimal interpolation

“…Using LI, it is possible to solve nonlinearity calibration when order of polynomial is higher than three. It is discussed that higher order regression represents more precise fit for calibration data since it will get very close to original function, meaning higher convergency [25,31,33,64,65]. However, higher orders also have tendency to oscillate which may give erratic value between the points.…”