Martin N. Goodhand Robert J. Miller Whittle Laboratory, University of Cambridge, Cambridge CB3 0DY, UK Compressor Leading Edge Spikes: A New Performance Criterion Compressor blades often have a small “spike” in the surface pressure distribution at the leading edge. This may result from blade erosion, manufacture defects, or compromises made in the original design process. This paper investigates the effect of these spikes on profile loss, and presents a criterion to ensure they are not detrimental to compressor performance. In the first part of the paper, two geometries of leading edge are tested. One has a small spike, typical of those found on modern compressors; the other has no spike, characteristic of an idealized leading edge. Testing was undertaken on the stator of a single-stage low speed compressor. The time resolved boundary layer was measured using a hot-wire microtraversing system. It is shown that the presence of the small spike changes the time resolved transition process on the suction surface, but that this results in no net increase in loss. In the second part of the paper, spike height is systematically changed using a range of leading edge geometries. It is shown that below a critical spike height, the profile loss is constant. If the critical spike height is exceeded, the leading edge separates and profile loss rises by 30%. Finally, a criterion is developed, based on the total diffusion across the spike. Three different leading edge design philosophies are investigated. It is shown that if the spike diffusion factor is kept below 0.1 over the blade’s incidence range, performance is unaffected by leading edge geometry. DOI: 10.1115/1.4000567 1 Introduction Spikes in the surface pressure distribution often exist at the leading edge of a compressor blade; these are caused by large changes in surface curvature. Cumpsty 1 questions the potential effects that these spikes have on compressor performance. He notes that cascade tests by Andrews 2 exhibit nearly constant loss over a wide incidence range, despite a “very large” spike in the calculated pressure distribution at high incidence. Cumpsty states: “since there is good reason to expect the spikes to have been present in the tests, it is not clear why they do not lead to massive boundary layer separation and a severe degradation in performance.” This raises the question of whether the spikes affect blade performance. Evidence that spikes have a large effect on performance is found in the work published by Carter 3. He systematically changed the ratio of leading edge radius to max thickness of a compressor blade from 0.08 to 0.35. Conventional wisdom, founded in the aerofoil theory, predicts that the lowest value or “sharpest” leading edge would have the narrowest operating range because the spike would be very large away from design inci- dence. The opposite was found with the sharpest leading edge having the widest operating range. In the discussion of his paper, two other people reported similar trends. These large effects caused by small changes in leading edge geometry raise the ques- tion as to whether changes in spike height are important. It also raises the questions of why a “sharp” leading edge has the widest operating range and whether this is the result of the leading edge having either no spike, or only a small spike over a wide range of incidences. Spike height in this paper is quantified using a spike diffusion factor D spike , based on the inviscid surface velocity distribution, and is defined in Eq. 1. This is based on the principle of the local diffusion factor of Lieblien 4 and represents the magnitude of spike induced boundary layer diffusion from the peak to the trough of the spike. The inviscid spike is used because it is a unique function of blade geometry and is a direct measure of the detrimental effect of the leading edge on the boundary layer. The inviscid spike has the added advantage that, unlike the viscous spike, it does not collapse as leading edge separation occurs. A schematic showing the spike is presented in Fig. 1. D spike = u max - u min u max 1 Leading edges with no spike are often found on external air- foils, even when operating at moderate incidences. This is rarely achieved on real compressor blades, which tend to be blunter. This bluntness occurs because thickness constraints are often required. These are set to give adequate structural integrity and erosion/ impact tolerance. This limit is required because of their small size, typically 0.4 mm thickness. A second consequence of the small size of compressor leading edges is that they often deviate geometrically from the design intent, due either to manufacture variation or erosion. The spike height therefore varies between blades. It would be both prohibi- tively expensive and technically difficult to ensure that all com- pressor blades in service had no spikes and it would therefore be of use to understand what spike height could be practically toler- ated and thus develop a criterion to define this. Contributed by the International Gas Turbine Institute of ASME for publication in the JOURNAL OF TURBOMACHINERY. Manuscript received July 20, 2009; final manu- script received July 30, 2009; published online October 20, 2010. Editor: David Wisler. Fig. 1 Schematic of surface pressure distribution with en- largement of spike Journal of Turbomachinery APRIL 2011, Vol. 133 / 021006-1 Copyright © 2011 by ASME Downloaded From: http://turbomachinery.asmedigitalcollection.asme.org/ on 09/10/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use