Passive Controls of Tip Clearance Flow in a Transonic Compressor for Stall Margin Improvement: Common Flow Physics

Abstract

Abstract Two passive tip clearance control devices for stall margin improvement in a transonic compressor are compared in detail. The primary objective of the current study is to examine the overall performance in terms of stall margin increase and any associated performance penalties near design operation with axial casing grooves (ACGs) and circumferential casing grooves (CCGs). The underlying flow physics of these passive flow control devices are examined and compared to identify any relative advantages in the development of optimized casing treatment designs. To understand the underlying flow mechanisms of ACGs and CCGs, available measured data and both steady and unsteady flow simulations from Large Eddy Simulations are used. The measured data illustrate changes in the overall time-averaged performance and provide very useful snap shots of the flow fields at specific times. The simulated steady (CCGs) and unsteady (ACGs) flow fields from LES show changes in the entire flow structure, which reveal the underlying flow physics. The current LES captures measured changes in the flow field due to the applied casing treatments well. Detailed comparisons of the calculated flow fields with and without CCGs or ACGs show that flow blockage near the casing is reduced by both treatments for near-stall operation. This reduction of blockage generation extends the compressor stall margin. Therefore, the root fluid mechanism of stall margin increase by these casing treatments must be the underlying method of reduction of blockage generation. Tip leakage flow occurs due to higher pressure on the pressure side of the blade than on the suction side at the blade tip. The tip leakage flow moves counter to and collides with the incoming main flow, creating a tip leakage vortex (TLV). The formation of the TLV due to this collision is the main source of flow blockage generation. The unsteady flow fields from LES near the casing with ACGs show that flow blockage in the tip region is periodically reduced when the blade passes under the ACGs. Detailed examination of the flows near the casing indicates that effective tip leakage flow below the casing is reduced by about 26% as some of the tip leakage flow moves into the ACGs. Most of the flow into the current ACGs is reinjected into the main passage upstream of the leading edge. The reduced effective tip leakage flow colliding with the incoming main flow produces a much smaller TLV with ACGs. This is the main mechanism for the blockage reduction with ACGs. With CCGs, part of the tip leakage flow moves into the rear portion of the grooves and is reinjected into the main flow near the front side of the grooves. The radially reinjected flow from CCGs blocks upstream movement of tip leakage flow below the casing, resulting in roughly 16% less effective tip clearance flow below the casing. Near stall operation, a much smaller TLV is formed with ACGs as compared to CCGs due to a larger reduction of effective tip leakage flow. With the smaller blockage, the overall incoming axial velocity is smaller for the given flow rate and larger pressure rise is obtained with increased inlet flow angle. The increase in stall margin by ACGs and CCGs is due to the reduction of blockage generation. The reduction of the blockage generation is due to the decrease of the effective tip leakage flow under the casing by ACGs and CCGs. It has been believed that casing treatments always incur small additional losses near the design condition due to the mixing of the flow exiting the grooves and the main flow, as well as to an increased wetted area due to the grooves. Near design operation, tip leakage flow typically initiates at about 25–30% axial chord downstream of the leading edge. ACGs or CCGs located near the leading edge barely affect tip leakage flow development. Since the current ACGs are located near the leading edge, the tip leakage flow development is not affected by the ACGs. An additional efficiency penalty is incurred by mixing losses near the ACGs. Measurements and LES indicate 1.1% and 1.4% efficiency penalties due to ACGs at design condition. CCGs near the trailing edge pull the TLV toward the blade suction side at the design condition, eliminating double leakage flow in the current transonic compressor and resulting in higher efficiency. The measured efficiency penalty due to the CCGs is within the measurement uncertainty and LES shows the same result. An integrated arrangement of ACGs and one or two rows of CCGs toward the trailing edge could eliminate any efficiency penalty near the design condition with a significant increase of stall margin.