Unsteady pressure measurement around aerodynamic bodies: Development of a calibration apparatus and wind tunnel testing

Abstract

Separated flows are complex but interesting to study because they are variable and unsteady. They are present for every bluff bodies and stalled streamlined bodies (at high angle of attack). Experimental aerodynamics is able to study these types of flow, using pressure sensors. Due to sensor size, pressure tubes are used to connect the pressure scanner to the tap (where the pressure is effectively measured). When an unsteady flow is studied, the signal measured by the sensor is perturbed by the tube. The Transfer Function of the tube has to be computed, to correct for the pressure measure using an inverse Fourier Transform and to obtain the pressure effectively present at the tap. The correction is made on the fluctuation amplitudes (around the mean) and the phase of the signal. The synchronization is important when vortex shedding is studied. This Transfer Function is computed by comparing the pressure measured at the begin and at the end of the tube. For that purpose, pressure with a frequency content has been applied on the tube entry (periodic for KTH calibrator and aperiodic for ULg calibrator). The ratio between these pressures gave the desired correction, showing resonance peaks for some frequencies. When a simple tube is used, theoretical models from fluid equations give very similar results to experimental ones. A parallel with electricity has also been made, replacing the pressure tube by an RLC circuit or a transmission line. The longer and the narrower the tube, the higher the signal distortion.

3D printed models are nowadays commonly used in experimental aerodynamics, allowing not only to build complex shaped models easily, but also pressure taps directly on the model and pressure channels into the structure. These more complex measurement systems have also to be experimentally calibrated. Indeed, diameter restriction on tap or shrinks in tube channels highly distorts the signal. We used this calibration to correct the pressure on a stalled wind turbine wing (at high incidence). The stall is linked to viscous effects, the flow becoming separated and turbulent. The fluctuations and phase of pressure taps signal have been studied to understand the Reynolds effect on a stalled wind turbine wing. Experiments were compared with CFD and theoretical models to validate the results.

Another application of unsteady pressure that we have studied the vortex shedding process, occurring around bluff bodies (in particular for rectangular cylinders). The synchronization and amplitude fluctuations of these vortices have been corrected using the dynamic calibration device. Fluid-structure interaction (vortex induced vibration) has then been studied: when vortices were ejected at the resonance frequency of the cylinder, the structure entered auto-excitation and vibrated a lot. When the cylinders were closely spaced in the flow (assembled into a grid), they interfered with each other and the vortex shedding process was changed compared to a single cylinder. To understand deeply this grid, theoretical and numerical models have been used (FEM and CFD) in parallel with experimental sensors: accelerometers (for vibration), pressure sensors connected on taps by tubes, Cobra Probe (velocity in the wake of cylinders), Hot Wire (free stream velocity). The study of this process in function of the incidence and the cylinder spacing allowed us to predict airspeed that induces instability. This is crucial in order to find parameters that minimize vibrations occurring on a real grid, with undesirable noise. In conclusion, this work can be used to take into account unsteady effects when pressure is measured around streamlined and bluff bodies.