A Thematic Network For Quality and Trust in the Industrial Application of CFD
ADVANCES IN CFD ANALYSIS FOR TRANSIENT TURBOCHARGER FLOWS
Fred Mendonça, Stephen Ferguson and Dean Palfreyman CD adapco Group London, UK Contact firstname.lastname@example.org
1. INTRODUCTION Transient CFD analysis on turbomachines should be strongly encouraged. It opens up a new chapter of possibilities in the understanding of the workings, efficiency and design optimisation of rotating components. Turbocharging increases the power output from reciprocating engines by utilizing the waste energy in the exhaust gases. The exhaust gases drive a turbine, connected via a shaft to a compressor, which pressurizes the air at the engine inlet thus allowing more fuel to be burned for the same air/fuel ratio. Computational analysis on turbocharger turbines and compressors can be performed separately. Engine exhaust flow enters the turbine through a scroll volute connected to the exhaust manifold. The volute discharges high temperature gas, either directly onto the turbine blades or through variable pitch vanes. Interaction between the fixed geometries and wheel creates dynamic flow modes, which are predominantly at the first (single rotation) order or higher (multiple excitations per rotation). On the compressor side, the decelerating or accelerating vehicle migrates the turbocharger performance towards ‘surge’ or ‘choke’. Under-bonnet space management usually means that the oncoming flow at the compressor face is non-uniform. Slightly eccentric installations or blade damage often results in modal excitation. Both factors lead to a noise problem. In this article, we explore the benefits of performing transient CFD calculations on both the turbine and compressor sides. • Improved understanding of the effects of pressure pulses from the engine on the turbine assists in the energy extraction from the exhaust flow. • Time varying aerodynamic and thermal loading on static and rotating components provides a useful insight into structural excitation. • Simulation of the acoustical radiation and flow-excited modes inside the engine bay provides a means to help reduce a predominant source of noise. Enormous value is derived from time-accurate CFD computations, both in their own right and also in the way they are linked with other key design features such as fluid-structure interaction and aeroacoustics. Combined advances in processor
speed and turbulence modelling (especially LES-based methods) make it feasible to compute multiple wheel rotations: recent experiences of the present authors using STAR-CD are presented below, and include an example of more than 50 limit-cycled wheel rotations using Detached Eddy Simulation (DES). In these examples we have employed many Best Practice procedures, including second-order spatial discretisation, y+ insensitive near-wall treatments and advanced turbulence modelling. 2. PULSED TURBINES There is a lack of understanding of the turbine aerodynamics under pulsating conditions. There are two main driving mechanisms, which have different timescales. First, there is the mass exhausted from each cylinder in turn, which convects through the exhaust manifold and volute into the turbine. Secondly, pulsations for the exhaust-valve opening event, propagating at the local speed of sound, are superimposed over the exhaust mass-flow. The period of this pulse is usually an order of magnitude or more faster than the engine period. Therefore the flow and thermal loading on the turbine blades is highly irregular, and also a function of the exhaust and volute geometries. Some designs include guide vanes and volute splitter plates upstream of the blades to even out the oncoming flow. In a recent study together with Imperial College , insight has been gained into the pulsating dynamics of the turbine.
Figure 1: Turbine wheel and volute
QNET-CFD Network Newsletter Volume...