A Flow Temperature Independent Hot-wire Calibration method and its Implementation in Turbomachinery Air Flow

Open Access
Kafaee Razavi, Abbas
Graduate Program:
Aerospace Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
May 23, 2014
Committee Members:
  • Cengiz Camci, Thesis Advisor
  • Turbomachinery
  • Turbulence
  • Hot-Film
  • Hot-Wire
  • Fluid Dynamics
The present thesis mainly deals with an implementation of a temperature dependent hot wire measurement approach using hot wires and hot films. Conventional hot wire measurements in isothermal flow systems could generate highly repeatable and accurate instantaneous velocity measurements. The temporal response of these sensors could be in a range from a few Hertz to a few hundred Kilo Hertz depending upon the diameter of the hot wire, thickness of the hot film, fluid type, flow Reynolds number and the type of the hot wire bridge employed. However, when the local temperature in a flow system noticeably varies during the experiment, instantaneous velocity measurements starts producing non-negligible measurement errors. This is especially true if the hot wire/hot film sensor is calibrated at a single "pre-selected" flow temperature. The present thesis uses a non-linear curve fitting approach for a temperature dependent calibration of the hot wire/hot film sensor. The method discussed in this thesis is geared towards obtaining accurate instantaneous velocity measurements in turbomachinery systems in which the mainstream temperature variations from the start to the end of a given test is strong. The current methodology was developed and tested in the Axial Flow Turbine Research Facility AFTRF at Penn State in which a typical temperature rise is about 15oC for every hour of operation. The specific non-linear curve fitting approach selected corrects the hot wire based instantaneous velocity measurements with great effectiveness as long as the mean flow temperature at the inlet of the turbine facility AFTRF is known. In general, another source like a thermocouple or thermistor based temperature probe is utilized in parallel to the hot wire sensor based velocity measurements. Cimbala & Park (1990) developed a new method for calibrating hot-wire/film anemometer that can be utilized in the incompressible flow, where there are significant ambient temperature variations. The current study implemented the specific non-linear curve fitting iv approach in a turbomachinery research facility where the operating temperature at the inlet varies at a non-negligible rate. After calibrating a hot-film sensor using the calibration method, a hot-film anemometer was utilized to measure turbulent flow characteristics at the inlet of the AFTRF which is located at the Turbomachinery Aero-Heat Transfer laboratory at Pennsylvania State University. The AFTRF turbine was run for twice; the first time was in a cold day (301oK) and the second run of the facility was done in a hot day (309oK). Then results from the both days, proved that by calibrating the output signal utilizing the non-linear curve fitting method, they will be almost the same and the uncertainty of the flow velocity will be minimized to about 1.25% of the mean flow velocity. Furthermore, the calculated velocity from the Cimbala & Park calibration method was used to obtain the turbulent flow characteristics, such as RMS velocity, mean velocity, turbulence intensity, turbulence length scale and etc. The present thesis explains the first implementation of the non-linear curve fitting based instantaneous velocity measurement method in a turbomachinery research facility where temporal and spatial temperature variations are significant.