Experiments aimed at investigating the possible factors affecting the temperature performance of thick-film resistors are presented. Particular emphasis is given to the temperature coefficient of resistance (TCR) of thick film strain gauges printed on both alumina and stainless steel substrates. The results confirmed that the resistance versus temperature curve is nearly parabolic, but showed that Tmin, the temperature at which the TCR changes to zero, is largely affected by the choice of resistor and substrate materials and also the thickness of the thick-film resistors. A possible explanation is proposed for the observed relationship between resistor thickness and TCR. Other factors, such as the thickness of the substrates, the choice of conductor materials, and whether single- or double-sided printing of the substrate was employed in fabrication were found to make little difference to the temperature performance of the thick-film resistors.

Results are presented from a programme of research aimed at establishing the mechanisms behind the effects of fabrication parameter variation on the performance of thick film strain gauges on steel substrates. The research is aimed at describing the effect on the repeatability of the device characteristics due to different choices of materials, thicknesses of printed layers, firing regimes and geometry of the gauges. In particular the effects of load and temperature on the offset and gain characteristics of a variety of different sensor constructions have been explored. The sensors described here are of a type where the applied strain is parallel to the measured resistance path but orthogonal to the substrate (k33). It has been found that these devices exhibit different characteristics to conventional thick film strain gauges that can help explain the mechanisms affecting gain and offset changes caused by temperature fluctuations and mechanical deformation.

Earlier study shows that resistor thickness of the thick film strain gauge can affect its temperature characteristic, which is usually a roughly parabolic curve. Thicker resistors tend to exhibit a higher positive temperature coefficient of resistance (TCR) and a lower T min , the temperature at which the TCR changes to zero in the curve. This paper presents a possible explanation of this observation based on an analysis of strain profiles and resistivity behaviour difference in resistors with different thicknesses subjected to temperature variation.

This paper presents results of work aimed at characterising the zero offset stability in novel thick film strain gauges. The devices studied are z‐axis (k33) load sensors fabricated on insulated stainless steel substrates and include examples of novel commercially developed force sensors. Devices loaded with compressive strains using a purpose designed test jig were found to exhibit a significant zero offset shift, which is negative up to a certain level (typically 1,000 micro strains) and then increasingly positive when strained beyond this point. Repeated cycles of loading then produced a certain level of stability until the previous maximum value of applied strain was exceeded. Temperature coefficient of resistance (TCR) measurements showed the devices to exhibit characteristics that depend significantly on the device geometry. The TCR was found to increase positively with increasing device thickness and surface area. The effect of overglazing the devices was found to decrease the TCR.

Novel thick film strain gauges have been constructed using a z‐axis orientation on insulated stainless steel for a variety of force sensing applications. These devices exhibit high gauge factor and good thermal stability compared with conventional x‐axis devices and offer other mechanical advantages due to their mode of operation. The work reported here investigates the characteristics of different types of stainless steel substrate and different types of insulating material used in the construction of the sensors. Both ferritic and austenitic steels have been investigated, together with different resistive and insulative compositions. The temperature coefficient of resistance of the devices has been shown to be a complex function of device thickness, surface area and the difference between the thermal coefficients of expansion of the various materials employed.