Hysteresis of thin film IPRTs in the range 100 °C to 600 °C

Abstract

Abstract. As opposed to SPRTs, the IPRTs succumb to hysteresis when submitted to change of temperature. This uncertainty component, although acknowledged as omnipresent at many other types of sensors (pressure, electrical, magnetic, humidity, etc.) has often been disregarded in their calibration certificates' uncertainty budgets in the past, its determination being costly, time-consuming and not appreciated by customers and manufacturers. In general, hysteresis is a phenomenon that results in a difference in an item's behavior when approached from a different path. Thermal hysteresis results in a difference in resistance at a given temperature based on the thermal history to which the PRTs were exposed. The most prominent factor that contributes to the hysteresis error in an IPRT is a strain within the sensing element caused by the thermal expansion and contraction. The strains that cause hysteresis error are closely related to the strains that cause repeatability error. Therefore, it is typical that PRTs that exhibit small hysteresis also exhibit small repeatability error, and PRTs that exhibit large hysteresis have poor repeatability. Aim of this paper is to provide hysteresis characterization of a batch of IPRTs using the same type of thin-film sensor, encapsulated by same procedure and same company and to estimate to what extent the thermal hysteresis obtained by testing one single thermometer (or few thermometers) can serve as representative of other thermometers of the same type and manufacturer. This investigation should also indicate the range of hysteresis departure between IPRTs of the same type. Hysteresis was determined by cycling IPRTs temperature from 100 °C through intermediate points up to 600 °C and subsequently back to 100 °C. Within that range several typical sub-ranges are investigated: 100 °C to 400 °C, 100 °C to 500 °C, 100 °C to 600 °C, 300 °C to 500 °C and 300 °C to 600 °C . The hysteresis was determined at various temperatures by comparison calibration with SPRT. The results of investigation are presented in a graphical form for all IPRTs, ranges and calibration points.

Modelling of Temperature Coefficient of Resistance of a Thin Film RTD Towards Exhaust Gas Measurement Applications

“…This model was applied tothe raw experimental data and the results obtained are shown in Fig.4.The best fit line showed a correlation of R 2 = 0.9999999835 and returned (7). ρ(T)= -0.0000229505T 2 + 0.4876336188T + 135.1721950209 7Putting this in the correct format provides, ρ(T)= -0.000000581366T 2 + 0.003868504198T + 1.00000000000 (8) This shows that the required characteristic equation for Pt/Rh sensors was again returned and agrees well with the conclusions put forward by [13].In addition to the responses for Pt/Rh sensor obtained using (6) that exhibited resistivitytemperature profiles, the TCR valuealso coincides with experimental results. Therefore, the results obtained with the two-dimensional model are in conformity with experimental findings.…”

“…Many papers reported on the design and development of platinum thin film RTDs and among them few researchers have analyzed the thermal hysteresis effects [5][6][7][8], thermal strain effects [9], thin film thickness effects [2], stagnation temperature [10], degradation effects at high temperature [11] and the long term stability [6,12] of the platinum RTDs. However temperature coefficient of resistance (TCR) plays an important role in determining the sensitivity of the RTD elementas it characterizes the average temperature chance of a 1Ω RTD.…”

Characterization and Corrections for Clamp-On Fluid Temperature Measurements in Turbulent Flows

“…The polynomial correction function derived on the basis of this data reduces the remaining deviation to below ±0.05 K. For two following testing cycles in the calibrator the polynomial correction function from the first calibration cycle is applied. The sensors show deviations to the calibrator reference within ±0.05 K as illustrated in Figure 4 for one sensor (according to the observations of [12] and [13]). Prior to a last (4th) testing cycle in the calibrator, the 400 mm long sensor sheath was bent by 90° at its centre with a bending radius of 30 mm.…”

“…Uncertainties in between ±1 K (1σ) are expected for a class A PT100 as defined by the IEC 60751 at absolute temperatures around 400°C. Gam et al [12] and Zvizdić et al [13] investigate the thermal hysteresis effects in industrial platinum resistance temperature detectors. Both groups observed a hysteresis around 0.1 K for absolute temperatures of 400°C.…”

Thermal-Feature System Identification for a Machine Tool Spindle

“…On the other hand, wireless real-time temperature monitoring facilitates the inspection and protection of machines [ 13 , 14 , 15 ]. Experiments [ 16 , 17 ] have shown that the resistance temperature detector (RTD) is ideal for temperature sensing because of its linearity, accuracy, long-term stability, and wide-range of temperature measurement.…”