Experimental investigation on spatial-temporal evolution of tip leakage cavitation in a mixed flow pump with tip clearance

• Published in: International journal of multiphase flow Vol. 164; p. 104445
• Main Authors: Han, Yadong, Tan, Lei
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.07.2023
• Subjects: Experiment; High-speed visualization; Mixed flow pump; Tip leakage cavitation; Tip leakage vortex; Tip leakage cavitation; Experiment; Mixed flow pump; Tip leakage vortex; High-speed visualization
• DOI: 10.1016/j.ijmultiphaseflow.2023.104445
• ISSN: 0301-9322

•A cavitation structure of double-hump PTLVC is firstly observed in mixed flow pump.•Double-hump PTLVC is classified: incepting, growing, merging, propagating stage.•Propagating velocity of the front hump of PTLVC reaches the maximum of 0.38 Utip.•Empirical function proposed for projected area, axial thickness, collapse position. Tip leakage cavitation remains an unsolved problem that threatens the safe operation of hydraulic machines and plagues researchers worldwide. The objective of this work is to investigate the classification and spatial-temporal evolution of tip leakage cavitation, and even to provide additional insights into the flow physics. Experiments are conducted in a mixed flow pump installed on a closed-loop test rig. High-speed visualizations are performed to capture the flow patterns of tip leakage cavitation at rated flow rate. It is demonstrated that tip leakage vortex cavitation can be categorized as primary tip leakage vortex cavitation (PTLVC) and secondary tip leakage vortex cavitation (STLVC). A new tip leakage cavitation structure, named as the double-hump PTLVC, is firstly observed in the mixed flow pump under severe cavitation conditions. The spatial-temporal evolution of the double-hump PTLVC is classified into four stages: incepting stage, growing stage, merging stage and propagating stage. The averaged propagating velocity of the front hump of PTLVC increases with decreasing net positive suction head (NPSH), and reaches the maximum of 0.38 Utip in the present experiment. Three empirical functions are proposed to describe the relationship between projected area, the maximum axial thickness, circumferential collapse position and NPSH, respectively. It is found that for every 0.1 m drop in NPSH, the projected area increases by about 2.1%, the maximum axial thickness increases by about 2.7%, and the circumferential length of the PTLVC increases by about 3.5%, respectively.