Investigation of Numerical and Physical Modelling Effects on the CFD Simulation of the Unsteady Flow in a HPT Stage

In order to investigate the unsteady flow behaviour in an HPT stage and the effects on the CFD solution of some numerical and physical modelling assumptions usually undertaken by the engineering community, an ITP in-house unsteady CFD code called Mu2 s2 T is first validated and then run under different configurations. The code is a fully unstructured code which solves the Reynolds averaged Navier-Stokes equations with a k-ω turbulence model. Hybrid meshes are used by having semi-structured meshes along the profile wall and fully unstructured triangular meshes on the inviscid region. The VKI Brite Euram transonic turbine stage experimental test case is used for the investigation. This turbine is representative of a state of the art HPT and presents high potential interaction due to the vane shock waves. After validating the code in this case, the influence of typical engineering assumptions is investigated. First the influence of the rotor stagger angle is analysed, resulting in a high sensitivity of the predicted pressure level at the front part of the blade and a better matching with the experimental data when an opening of 1° is applied. The influence in the solution of applying an integer airfoil count ratio compared with the solution with exact number off computed by means of phase lagged boundary conditions is also investigated. Additional Euler and Navier-Stokes computations are presented and the influence of the viscous effects is discussed. Finally a simulation including vane trailing edge cooling is performed so that conclusions about the influence of the cooling can be drawn.Copyright © 2003 by ASME