Axial compressor cascade

Application Challenge 6-10 © copyright ERCOFTAC 2004

Description

Introduction

Experimental investigations on the boundary layer and loss behaviour on a high turning compressor cascade up to 0.90 inlet Mach number were performed in the High Speed Cascade Wind Tunnel of DFVLR Braunschweig. The main objective of these investigations was to obtain accurate data for the validation of theoretical methods. The V2 cascade has a design point inlet Mach number of 0.85 and flow turning of 50°.

CFD results available include calculations using the STAR-CD commercial CFD code. The steady state compressible Navier-Stokes equations using a variant of the k-ε model for closure were solved using the SIMPLE algorithm. Discretisation was undertaken using the second-order MARS scheme with the default compression level of 0.5. The computational domain covers one blade pitch with the blade located centrally in the domain and an axial extent of -0.8 to 1.8 axial chords from the leading edge.

Relevance to Industrial Sector

The ECA2-V2 cascade was tested over a wide range of operating conditions, from high incidence/stall to choke flow and from low Mach number up to high subsonic flow. The transonic flow range and the high level of turning make it a challenging case, which is relevant to modern turbomachinery flows. Increased turning and higher blade loading are required as designers attempt to reduce the weight and size of gas turbine components whilst maintaining high pressure ratios. This application challenge is an ideal test case for assessing the ability of CFD codes to predict the detailed flow features and overall performance of a compressor blade row.

Design or Assessment Parameters

The design and assessment parameters can be grouped into firstly, overall performance indicators and secondly, parameters used to evaluate more detailed flow behaviour. The operating point of the cascade is defined by the inlet Mach number, inlet flow angle and Axial Velocity Density Ratio (Ω). The latter quantity defines the variation of streamtube thickness through the cascade in the axial direction. The ability to vary the streamtube thickness enables the endwall effects present in cascade testing to be controlled.

The overall performance parameters used for evaluation were exit Mach number, exit flow angle (hence flow turning) and pressure loss. The overall pressure loss, ζ₁, was defined as follows:

To evaluate blade surface pressure distributions a non-dimensional pressure coefficient was defined as:

For more detailed evaluation of the pressure loss, profiles of pressure loss coefficient were plotted across the wake. The pressure loss coefficient was defined as:

where p₀ was measured across the pitch at 30% of true chord axially downstream of the blade trailing edge.

Boundary layer measurements were carried out at different positions on the blade suction side at the midspan position.

Flow Domain Geometry

The V2 cascade is a high subsonic compressor cascade with a conventional double circular arc profile. The tunnel had a test section width of 300mm and an adjustable height of 250 to 500 mm depending on the inlet angle (see Figure 1.1). Changes in Axial Velocity Density Ratio (Ω) were obtained by varying the amount of air removed through porous side walls. The cascade consisted of 14 blades with a true chord length (l) of 80mm, resulting in an aspect ratio of 3.75 and a pitch-chord ratio of 0.45. The blade geometry is given in Table 2.1.

Figure 1.1 Experimental Rig of the ECA2-V2 Compressor Cascade.

Flow Physics and Fluid Dynamics Data

The V2 cascade has a high level of turning and is designed for a transonic inlet Mach number. The tests carried out were for an inlet Reynolds number of 5E5 based on blade chord. A range of inlet Mach numbers from 0.3 to 0.85 and a range of operating points from high incidence/stall to choke flow were investigated.

The working fluid was air at a temperature of 313K and with a freestream turbulence value of 5% at inlet. The flow over the aerofoil is characterised by an initial laminar region followed by transition to a fully turbulent boundary layer, which then thickens and eventually separates from the blade surface towards the trailing edge. Due to the high level of turning there is a strong adverse pressure gradient. At high subsonic inlet Mach numbers there are locally supersonic regions near the blade surface.

Contributors: Michael Dickens; Alex Read - Computational Dynamics Ltd