Experimental investigation of vane clocking effects on stall performance and unsteady vane boundary layer development in a multistage compressor

Natalie Rochelle Smith, Purdue University

Abstract

The objective of this research was to investigate the effects of vane clocking, the circumferential indexing of adjacent vane rows with similar vane counts, on stall margin and unsteady vane boundary layer development. Experiments were conducted at the Purdue 3-Stage Axial Compressor Facility. The blading in the facility is representative of the rear stages of a Rolls-Royce high pressure compressor. Vane clocking affects blade row interactions by modifying how the upstream, Stator 1 wake interacts with the downstream, Stator 2. While vane clocking effects on compressor efficiency have been documented in this facility, detailed investigations of the Stator 2 boundary layer were required to understand the underlying flow physics. The two clocking configurations, CL3 and CL6, investigated in detail at four loading conditions were half a passage out-of-phase and included clocking configurations with the best and worst Stage 2 efficiency. Due to the Stator 1 wake skew in the Stator 2 exit plane, impingement configurations (leading edge and mid-passage) were opposite between the hub and tip regions. At design point, clocking effects were most pronounced at 70% span; the wake of CL3 (leading edge impingement) was 2% vane passage thinner that the wake of CL6 (mid-passage impingement). At higher loadings, clocking effects were greatest at the hub. Between the high loading conditions, impingement conditions changed, and thus, so did the configuration with the thinnest wake. Changes in clocking behavior with loading conditions suggested the potential for variations in stall margin, another important aspect of compressor performance. Full-span, rotating stall originated in Stage 1. The variations in stalling mass flow rate between the six clocking configurations were found to be within the measurement uncertainty. To gain further understanding of how the Stator 1 wake impingement on the Stator 2 leading edge influences the boundary layer response to rotor wake passing, surface-mounted hot-film anemometry was implemented as a new measurement technique for the laboratory and used to evaluate the vane boundary layer response at 50% span on Stator 2. Quasi-wall shear stress (QWSS) was determined by relating variations in the vane boundary layer to the thermal boundary layer created by the heated sensor. The Stator 2 boundary layer transitions over a separation bubble. At the design point, the CL6 configuration (mid-passage impingement) had a larger separation bubble and higher QWSS than CL3 (leading edge impingement). While this could indicate that the CL3 configuration results in a thinner boundary layer, supporting the steady data, there was insufficient evidence to fully explain the affects of Stator 1 and Rotor 2 wake interactions on the Stator 2 boundary layer.

Degree

M.S.A.A.

Advisors

Key, Purdue University.

Subject Area

Aerospace engineering|Mechanical engineering

COinS