High-precision temperature measurement provides basic data for product development and industrial automation applications to improve product quality and ensure production safety. Due to its excellent linearity, measurement repeatability and stability^[1], a resistance temperature detector (RTD) is widely used therein. However, for remote measurements, the electrical resistances of long connecting lead wires between the RTD and the control room instrument produce an appreciable error in measurement. This unwanted error varies not only with the length of the lead wires but also with ambient temperature variations where the lead wires layout ^[2]. Therefore, methods to reduce or eliminate measurement errors caused by lead wire resistance have been studied in many literatures.

Currently, the aforementioned problem is addressed by adding lead wires. For example, in Reference^{[3] [4]}three-wire and four-wire techniques were used separately. However, the cost of one additional lead wire in a three-wire RTD and of two additional lead wires in a four-wire RTD will be extremely high and with extra difficulty of wiring, especially in industries where a large number of process points are to be monitored from a control room located at a remote place ^[5], such as chemical, thermal power, electric plant and other industries. In the bridge-based measuring system, lead wire resistance also exerts an adverse effect on measurements. Authors in Reference^[6][7] presented a method to eliminate lead wire resistance for quarter- and half-bridge interface circuit respectively, which employed several operational amplifiers (OPAMPs) and made use of their high input impedance. This method was able to eliminate the unfavorable effect on the measurement caused by lead wires, but it cannot obtain the specific lead-wire-resistance values, in addition, its interface circuits and power circuits were complicated due to the OPAMPs. Similar methods using OPAMPs were also published in Reference ^[8][9][10]. All of these bridge-based interface circuits used a three- or four-wire method, except for Reference^[6].

Therefore, researches have been carried out on the compensation method of two-wire sensors. In Reference ^[11], a compensation resistor is selected having the same value and of the same material of the lead resistances combined with three operational amplifiers and a constant current source to compensate the lead wires error. Although the circuit of this method is simple, the compensation resistance must be adjusted manually if the lead wire length is changed due to reconstruction. In addition, the compensation effect due to temperature drift of the lead wire resistance remains to be validated. The authors of [5] proposed a novel lead-wire-resistance compensation technique using a two-wire RTD. This technique employed two diodes, one of which was in series with the RTD and the other was in parallel with the series circuit composed of the RTD and the first diode. A current source, four analogy switches and four sample-holding circuit were used in its interface circuit, and an operational amplifier is used to output the voltage with respect to the RTD resistance. In principle, this technique completely eliminated the measurement error caused by the lead wire resistance and its temperature drift, but its measurement accuracy depended on the consistency of the forward voltage drop of the two diodes. Therefore, in order to achieve a high precision measurement, it is indispensable to measure and pair the diodes in advance. Besides, this technique cannot obtain the true value of the lead wire resistance. This method was also seen in other references, and some improvements had been made in the interface circuit. In the Reference ^[12], three single-pole double-throw (SPDT) analog switches are used to realize the lead-wire-resistance compensation, which was independent of the voltage reference. The authors of ^[13][14]used a voltage-to-current converter to provide the current and reduced the number of SPDT analog switches to two. A Chinese patent invented a method that employed the transient characteristics of resistor–capacitor (RC) circuit to measure the lead wire resistance and compensate measurement error of the thermal resistance. In this method, the measurement precision is not affected by the characteristics of the additional capacitor, and the lead-wire-resistance value could be obtained in real time, which avoided the negative influence of the temperature drift of the lead wire resistance. According to the patent, the update rate is less than 5 Hz, and it is not suitable for the applications where temperature changes need to be tested dynamically. In References ^[15][16], the interface circuits for four-wire resistive sensor were designed by the similar method of multiple charge and discharge to a capacitor, and the update rate reported was 25 Hz. The authors in References ^[17][18] combined the method for two-wire sensors reported in Reference ^[5] and also proposed the method using a capacitor, and the their minimum complete measurement cycle required 5.3 ms. These methods have the same problem of limited update rate.

This entry is adapted from the peer-reviewed paper 10.3390/s20092742