Studies on Reverse Flow in Centrifugal Impeller

1987 Fiscal Year Final Research Report Summary

• Project/Area Number: 61460099
• Research Category: Grant-in-Aid for General Scientific Research (B)
• Allocation Type: Single-year Grants
• Research Field: Fluid engineering
• Research Institution: Faculty of Engineering, Yokohama National University
• Principal Investigator: TOYOKURA Tomitaro Faculty of Engineering, Yokohama National University, Professor, 工学部, 教授 (00017857)
• Co-Investigator(Kenkyū-buntansha): KANEMOTO Toshiaki Facluty of Engineering, Yokohama National University, Associate Professor, 工学部, 助教授 (90092642) ; KASHIWABARA Toshinori Kochi Technical College, Associate Professor, 助教授 (40044218)
• Project Period (FY): 1986 – 1987
• Keywords: Reverse Flow / Centrifugal Impeller / Cavitation / Centrifugal Pump / 非定常流

Research Abstract

In order to avoid the choke resulted from the cavitation and to improve the performance of centrifugal pump, it is more often to take a large blade inlet angle and a big inlet diameter. In this case, on the contrary, there has a tendency to cause the cavitation based on the reverse flow in the suction side of impeller blades. Consequently, it needs to clarify the mechanism of this reverse flow which has hardly been investigated, although it is very complicated. So, this paper aims at making the mechanism occurence of the reverse flow clear and acquiring the knowledge in connection with the cavitation.
Since an axisymmetric property of the blade inlet flow disappears when the reverse flow happens, the position occurence and the scale of the reverse flow can not be obtained by the steady measurement. Accordingly, at first an unsteady measurement with use of pitot tube was established. Then, the pump performances and the relation between the reverse flow and the pressure distributions in t he impeller passage were investigated by using the open impeller. The achievements were gained as follows. As the blade inlet angle is large, when the flow rate is 95% of the maximum efficiency point (equivalent to the flow rate 45% of the shockless flow condition), the inlet flow has already changed considerably in the pitch direction, which seems that the reverse flow happens easigy, and this reverse flow can be observed when the flow rate becomes 90%. The scale of the reverse flow is so large that the whole region of one pitch between the blades near by the blade tip is occupied by the reverse flow but its region decreases gradually in the blade height direction and is limited near the leading edge at the place of 1/4 blade height. Furthermore, even the flow rate decrases much more, the reverse flow region on the blade inlet section hardly changes and enlarges to the upstream side. When the reverse flow occurs, the pressure increases greatly on the wide region of the negative pressure side near the blade inlet and the pressure increase can be observed locally even near the pressure side. As the reason of those phenomena, it can be considered that the former is affected by the secondary flow running to the suction cover side along the blade surface while the later is caused by the movement of stagnation point of the leading edge toward the pressure side. This fact has been confirmed also by the visible result derived from the oil film method. Besides, the pressure discributions on the blade surface show a same tendency with the one derived from the singularity method, except the leading edge neigbourhood, but the experimental value on the negative surface is smaller than the calculating one. Moreover, the influence of blade tip clearance width on he reverse flow scale and on the pressure distributions between the blades seems to limited.