Developments in turbomachinery internal air systems

Abstract

Bayley and Conway's 1964 paper [1], motivated by shortcomings in industrial design methods and understanding, was one of the first investigations of flow and heat transfer in turbomachinery rotating disc cavities. At the time of the study, a theoretical or numerical treatment was considered intractable and so experiments were undertaken. These paved the way for a large body of further research in this area. Today, use of computational fluid dynamics (CFD) in industry for internal air flow prediction is now commonplace. In this review of Bayley and Conway's paper it is shown that the unshrouded disc cavity flow considered is still challenging for modern CFD methods, and so the experimental data remains of interest to researchers in the field.

Flow mechanisms in axial turbine rim sealing

“…The first evidence of inherent unsteady flow features in rim seals was, to the present authors' knowledge, CFD predictions reported by Autef [39] in 2002, as mentioned by Chew et al [40]. A simple axial seal configuration (without vanes and blades) was simulated with 2D RANS and 3D URANS models with a weak axisymmetric cross flow in the outer annulus.…”

Large-eddy simulation of unsteady turbine rim sealing flows

“…The first published experimental evidence of the large-scale unsteady flow features was reported by Cao et al [4] in 2003, who also showed some agreement with unsteady RANS (URANS) solutions. Prior to this, in 2002, Autef [5] reported URANS solutions showing unsteady flow structures of a rim seal configuration without vanes and blades, as described by Chew et al [6]. In 2004 Jakoby et al [7] reported URANS studies showing large-scale unsteady flow structures unrelated to the blade passing, and supported by measurements from Bohn et al's rig.…”

Numerical Studies of Turbine Rim Sealing Flows on a Chute Seal Configuration

“…These are (1) disk pumping, (2) periodic vane/blade pressure fields (3-D and time dependent), (3) 3-D geometry within rim seal region, (4) asymmetries in the rim seal geometry, (5) turbulent transport in the platform overlap region and (6) flow entrainment. In addition to the mechanisms stated above, Smout et al [2002] and Chew et al [2003] reported large-scale low-frequency flow structures, even in an axisymmetric geometry without vanes and blades. The URANS simulations on rim seal geometries of Boudet et al [2005] revealed that those structures are inherently 3-D and unsteady, and that reasonable predictions of those features might improve the estimation of sealing effectiveness.…”