Experimental and numerical quantification of cavitation aggressiveness
Abstract
Cavitation is a major issue in hydraulic machinery, it causes performance drop, vibrations increase and damage. This PhD is dedicated to cavitation erosion and propose tools to quantify the erosive potential of a cavitating flow.
Experiments were carried out on a symmetrical hydrofoil based on a NACA0015 with a flat area between 27% and 70% of the chord length for an easy instrumentation with pressure sensors. The study is focused on partial cavitation that detaches from the leading edge of the hydrofoil and periodically sheds vapor clouds. All experiments are conducted in the cavitation tunnel of the LEGI laboratory, partly renovated during this PhD. Additionally, numerical calculations are performed using 2D cavitating unsteady code IZ developed at LEGI.
Cavity dynamics, cavity length and shedding frequency are deduced from analyses combining numerica simulations and high-speed video. Influence of various hydrodynamic parameters (flow velocity, hydrofoil angle of attack, and cavitation number) on the cavity behavior is studied. Strong 3D effects observed experimentally make cavity dynamics behaviour predictions difficult with a two-dimensional code. Nevertheless the maximum cavity length and the shedding frequency are well predicted numerically.
To measure the pressure peaks resulting from collapses of vapor structure a matrix of eight sensors is flush mounted in the hydrofoil. The matrix has an active area of 2x2 mm2 and its position can be adjusted between 30% and 67% of the chord length. To reveal the fast dynamics of collapsing events, those eight sensors are polled simultaneously at a 10 MHz sampling rate.
On the experimental side, the flow aggressiveness is deduced from the peak frequency distribution as a function of the peak amplitude. Numerically, an instantaneous and a mean aggressiveness at the hydrofoil surface are derived from an aggressiveness parameter based on a model previously developed at LEGI. The origins of the most aggressive area are identified from local studies. For low incidence experimental and numerical results are in good agreement. Yet, numerical study is overestimating the aggressiveness close to the hydrofoil leading edge . Additionally the most aggressive area are located at the cavity closure and the flow velocity heavily influence the aggressiveness level.