A New Approach for Calibration of Hot-Wires for use in Uncertain Environments

Hot-wire sensors are known to exhibit sensitivity to multiple fluctuating quantities, particularly at transonic conditions, variation of calibration from sensor-to-sensor, and calibration shifts due to environmental effects, such as changing cable resistance. This paper addresses a new calibration technique using Response Surface Methods (RSM) to achieve rapid calibration of sensors intended for use in uncertain environments. A TSI 1201 hotfilm sensor was set up in the exit plane of an ASME nozzle supplied with compressed air. A LabVIEW program was developed which includes a driver for a TSI IFA-300 anemometer, permitting rapid selection of different sensor overheats and operating velocity. Typical calibration procedures involve acquisition of anemometer outputs at a minimum of five levels of both velocity and overheat. Using least squares regression, a polynomial prediction model is developed. Using the prediction model, a “response surface” of output voltage, velocity, and overheat (sensor temperature) is then created. Exploiting the characteristic form of the response surface, it is possible to identify changes in operating environments that may have caused output voltage shifts, such as change in cable resistance.