A relatively simple and inexpensive method of fabricating a temperature-compensation element for high- temperature strain gauges has been devised. This element, connected in the adjacent arm of a Wheatstone bridge, provides temperature compensation for an active strain gauge attached to the substrate. A method for accurately measuring structural static strains in harsh environments is an important requirement for future flight research of hypersonic vehicles and ground test articles. Sturdy, flight-worthy strain sensors must be developed for attachment to super-alloys, new composite materials, and thermal-protection systems. With little deviation from standard Rokide flame-spray installation procedures, preliminary tests indicate viable data can be produced to operating temperatures of at least 1,700 °F (927 °C).

Figure 1. The Active Strain Gauge and the Temperature- Compensation Element are labeled “RActive” and “RComp,” respectively. The straps that hold down the compensation element have been removed, and the gauge has been lifted for this photograph. Contact with the substrate must be maintained to ensure thermal conduction in the presence of transient heating.

In the present method, the temperature-compensation element is encapsulated and insulated in alumina by the Rokide flame-spray process and used as an inactive element in a half- bridge configuration. An inactive element, or gauge, is often also referred to as a “dummy gauge” because it does not sense surface strains; in other words, there is no mechanical strain transfer from the substrate to the gauge filament. The temperature-compensation element is mounted in close proximity to the attached, or active, strain gauge. Adequate surface contact of the compensation element to the test article must be achieved in order to maintain good thermal conductivity. However, unlike the active strain gauge, the temperature- compensation element is not rigidly attached to the substrate which is to be measured; instead, the temperature-compensation element (see Figure 1) is attached flexibly to the substrate using nickel/aluminum-alloy straps.

Configured as a half-bridge, the temperature-compensation element is connected in an arm of a Wheatstone bridge adjacent to an arm containing the active strain gauge. The temperature-compensation element does not sense mechanical surface strains, but it is subjected to the same temperatures as is the active strain gauge. Inasmuch as equal changes in adjacent arms of a Wheatstone bridge cancel, the equal temperature-induced components of the changes in the resistance of the active strain gauge and the temperature-compensation element cancel, leaving a Wheatstone-bridge output indicative of only the surface strain in the substrate.

The Flight Loads Laboratory at NASA Dryden Flight Research Center has evaluated and characterized many high-temperature strain-gauge assemblies over the years, maintaining rigorous focus on reducing the thermal output, or apparent strain, of these gauges. High-temperature strain-gauge alloys generate outputs indicative of large magnitude, nonlinear, apparent strains that depend on maximum operating temperature, time at temperature, and rates of cooling. The apparent-strain output of a high-temperature strain gauge consists of three main components: (1) the mismatch in coefficients of thermal expansion between the substrate and the gauge alloy, (2) the thermal coefficient of electrical resistivity of the gauge alloy, and (3) the change in gauge factor as a function of temperature. Characterization of strain gauges at elevated temperatures is critical inasmuch as correction curves must be generated and applied to raw data to determine true mechanical strains from indicated strains.

Prototype temperature-compensation elements, according to the present method, were wired with active high-temperature strain gauges as half-bridges. Both the temperature-compensation element and the active strain gauge were made of 0.002-in. (0.05- mm) Fe/Cr/Al-alloy wire. The active strain gauge was attached and insulated to the substrate using standard NASA Dryden plasma spray (precoat) and Rokide flame-spray procedures, while a modified version of the procedure was used in fabricating the temperature- compensation elements.

Figure 2. These Apparent-Strain Curves obtained by a half-bridge strain gauge utilizing the presented temperature- compensation element exhibit little zero shift, a low rate of drift at 1,500 °F (≈820 °C), less nonlinearity (in comparison with uncompensated strain gauge), a high degree of cycle-to-cycle repeatability, and no cycle-to-cycle slope changes.

Preliminary apparent-strain tests of the present method of temperature-compensation at temperatures up to 1,700 °F (927 °C) were performed. The compensated half-bridge outputs were more nearly linear and repeatable, and of less magnitude, than those of the strain gauges in the uncompensated quarter-bridge configuration. Early results indicate that effective cancellation of the effects of temperature-induced changes in the electrical resistance of the active strain gauge and the temperature-compensation element was achieved. Numerous undesired attributes of high-temperature strain gauges used in the quarter-bridge configurations were reduced when thermally compensated by present method (see Figure 2). These attributes include zero shifts (sensor non-return to zero) as a function of cooling rates, rates of drift during static holds, and uncertainties in the phase transformations of gauge alloys.

Two problems observed in bare-wire temperature-compensation elements have also been eliminated using the present method. These problems include slope change of the overall apparent-strain curve from one cycle to the next cycle, and excessive drift at high temperatures. A “cycle” refers to both the heat-up and cool-down portion of a test. These changes in slope from cycle-to-cycle and excessive drift rates do not occur in the gauges fabricated and used according to this method because the active gauge and the temperature-compensation element are under the same condition; they are both encapsulated in alumina, therefore, subjected to the same oxidation environment. In contrast, a bare-wire temperature-compensation element oxidizes differently than the active gauge since it is not encapsulated in alumina. In addition, heat conduction will often be quicker to a bare-wire element (lower mass) when compared to the encapsulated active gauge. This temperature lag in the active gauge becomes more pronounced as transient heating rates increase causing the electrical resistance cancellation of the half-bridge to be less effective.

This work was done by Anthony Piazza of Dryden Flight Research Center.

This invention is owned by NASA, and a patent application has been filed. Inquiries concerning nonexclusive or exclusive license for its commercial development should be addressed to the Patent Counsel, Dryden Flight Research Center; (805) 258-3720. Refer to DRC-96-74.