Pressure fluctuations from orifices in water filled ducts
Shravan Kottapalli defended his PhD thesis at the Department of Mechanical Engineering on January 29th.
For his PhD research, Shravan Kottapalli investigated turbulent wall pressure fluctuations due to flow through single-hole sharp-square-edged orifices in a water-filled pipe by means of measurements and incompressible Large Eddy Simulations (LES). He also looked at the influence of orifice thickness and chamfering of edges on the drag and pressure fluctuations.
The static pressure measurements and simulations show a Reynolds number dependency on the drag. However, Reynolds-averaged Navier-Stokes (RANS) and large-eddy simulations, which were undertaken as part of the work for PhD research, fail to estimate the drag coefficient of the upstream-chamfered orifice accurately.
Good prediction
LES simulations provide a good prediction of the near-field hydrodynamic wall pressure fluctuations, which dominate up to the first three pipe diameters downstream of the orifice.
Upstream and far downstream wall pressure fluctuations are dominated by acoustic pressure fluctuations. The source of the radiated sound field was estimated from the surface integral of the fluctuations across the orifice from LES. The sound field calculated by implementing the sound source in a one-dimensional acoustic model from the LES predicts the power spectrum density (PSD) of the acoustic pressures within a factor 3.
In the inertial range of the turbulence, a global decay of the pressure PSD and of the sound source show a power of the Strouhal number between -3.66 and -2.8. For higher frequencies the pressure fluctuations were dominated by acoustic resonances.
A simple non-dimensional source model (estimated from LES) for orifices using a power law predicts the acoustic pressure fluctuations reasonably well for thick orifices. However, the results deviate significantly for thin orifices.
Narrow fluctuations
For the narrowest orifice, the measured PSD of pressure fluctuations displays sharp peaks due to self-sustained oscillations (or whistling). At low frequencies, one observes a peak in the PSD hydrodynamic wall-pressure fluctuations at one pipe diameter downstream of the orifice. This is probably due to an acoustically silent "edge-tone-like" sinusoidal self-sustained motion of the jet.
At low Strouhal numbers (based on the orifice diameter and the cross-sectional averaged velocity in the orifice), the thin orifice behaves similarly to the thicker strongly chamfered orifices.
At higher Strouhal numbers, the measured and predicted acoustic-pressure fluctuations due to these orifices are two orders of magnitude lower than those of thick orifices with sharp square edges.
Transmission losses
Finally, the research also investigates the transmission losses of compact compliance-based resonators in water circuits. Experiments are performed to measure the anechoic transmission losses (TLan) of flexible-plate resonators and a gas resonator designed for frequencies between 10 Hz and 100 Hz.
The measurements are compared to theoretical results based on a lumped-element model and a finite-element model. TLan is measured using a robust form of the multi-microphone method. For the flexible-plate resonators TLan measurements agree with theory, except close to resonance where the transmission signals are below the detection limit. Low resonance frequencies are easier to reach with a gas resonator.
For the gas-resonator, the measurements agree with the theoretical predictions when assuming a significant damping due to friction. Theory predicts both resonators have quite similar performances except close to the resonance frequency. The flexible-plate resonator has a higher quality factor and higher TLan around the resonance frequency.
The gas resonator is more complex and needs more maintenance but allows fine tuning of the resonance frequency by varying the gas volume.
Title of PhD thesis: . Supervisors: David Smeulders, Avraham Hirschberg, and G眉ne艧 Nakibo千lu.