Experimental investigation of cavitation-induced erosion around a surface-mounted bluff body

More Info
expand_more

Abstract

The objective of this study is to investigate the collapsing behavior of cavitation, which leads to the erosion of material. An experimental examination was conducted in a channel with a semi-circular cylinder obstacle, which serves as a “vortex cavity” generator. Cavitation was achieved by employing a range of pressure differences over the test section and a high-speed camera was used to observe the cavitation behavior. The flow field behind the semi-circular cylinder was investigated as a characteristic example of bluff bodies that exhibit a distinct, separated vortex flow in their wake. The cases with the bluff body were also compared to the ones without the bluff body. Erosion tests were performed using paint (stencil ink). The intensity of cavitation is characterized by the cavitation number (σ); the lower the cavitation number, the higher the cavitation intensity. The erosion (removal of paint) after 40 min of operation revealed distinct and repeatable results. For a high cavitation number, a large number of von Karman-vortex-like cavities are shed downstream of the obstacle. This results in a higher number of collapse events and, ultimately, more erosion. On the other hand, at lower cavitation numbers, the erosion took place at the cavity's closure line. It was seen that with the increase in cavitation intensity, the erosion area increases. Moreover, the bluff body obstacle promotes and localizes cavitation-induced erosion on the sample plate compared to the cases without the bluff body. This ultimately means that in the cases with the bluff body, less power is required in the system to cause erosion. The erosion patterns caused by the bluff body cavitation are more repeatable compared to the cases without the bluff body due to the localized cavitation load. The erosion pattern from the paint test is also compared with a material loss test (30 h of operation). A very good qualitative agreement is found between the two tests, with the paint test requiring approximately two orders of magnitude less running time of the facility. We demonstrate that paint tests, combined with this geometry, provide an efficient and economical way to investigate erosion patterns compared to expensive material loss tests.