The Effect of Tip Clearance on the Development of Loss Behind a Rotor and a Subsequent Nozzle

Abstract

In a previous work of a similar nature, the performance of a low speed axial turbine with a second stage nozzle was examined. Present findings suggest some interesting phenomena, including the effect of tip clearance on the flow within the rotor and show that poor resolution of a transducer and insufficient data points near the endwall were responsible for a number of incorrect conclusions in the original study. In terms of blade tip geometry, a standard flat tip shape was found to deliver only a marginally better performance when compared to two streamlined shapes previously investigated and a double squealer tip. Although contemporary opinion suggests that a streamlined tip should increase the leakage flow and hence cause greater mixing losses, the machine efficiency was not significantly reduced. This is an exciting result since it suggests that a streamlined tip shape can be used to alleviate the problem of blade tip burnout without significantly reducing machine efficiency. The flow behind the rotor (ie time averaged) was found to be in reasonable agreement with linear cascade data when time averaged even though the latter did not include any effects of relative motion. An increase in clearance was seen to reduce the Euler work and also to cause a deficit of mass flow across the remainder of the blade right down to the hub. As in the previous study, the second stage nozzle efficiency was seen to be independent of tip clearance or tip shape. However, the improvement in comparison with the first nozzle was not found to be as large, due to a previously undetected very thin ring of high energy leakage fluid. When this is taken into account, the efficiency of the second stage nozzle is comparable to the first. The second nozzle was seen to have a flow straightening effect on the undetected high energy leakage flow, causing a rapid mixing process within these downstream blade passages. The growth of secondary flow was reduced at both the hub and the tip and this is believed to result in a slight reduction in loss. The outlet flow was closer to design conditions than that of the first stage nozzle.