Martin Hoeger

Turning Mid Turbine Frame Behavior for Different HP Turbine Outflow Conditions

The design of turbine frames with turning vanes, known as turning mid-turbine frames (TMTF), becomes of great importance for high by-pass ratio engines with counter-rotating turbines. To achieve a more efficient low-pressure turbine the overall diffusion and radial offset should be increased. One goal of the EU project DREAM is to analyse the flow through a TMTF and a downstream arranged counter rotating LP rotor. The investigation of these complex interrelationships has been performed in the unique two-spool continuously operating transonic test turbine facility at Graz University of Technology. The test setup consists of an unshrouded HP stage, the TMTF and a shrouded LP rotor. The shafts of both turbines are mechanically independent, so the test rig allows a realistic two shaft turbine operation. The TMTF flow field is highly complex. It is a turbulent and unsteady flow dominated by strong secondary flows and vortex-interactions. The upstream transonic high pressure turbine stage produces a complex inflow with high levels of turbulence, stationary and rotating wakes and vortical structures. Therefore the application of advanced measurement techniques is necessary. To describe the HP-TMTF interaction time-resolved pressure measurements have applied within the project. The TMTF was instrumented with 10 fast response pressure transducers; static pressure tap recordings on the strut and on the TMTF end-walls have been also applied. Five hole probe, total pressure and total temperature rakes have been additionally acquired in the planes just in front of the struts and downstream to evaluate the performance of the TMTF. The results of these conventional techniques are presented in this work and they represent the necessary starting point for the evaluation and the description of the flow field. The idea is to start the study analysing the mean quantities and the overall performance of the two stages for different conditions and to leave the analysis of the time-resolved results for further investigation. Detailed investigations will start from the data presented in this paper; indeed, the use of unsteady measurement techniques is time consuming and cannot be performed for such a large amount of flow conditions, radial planes and HP vane - TMTF relative positions. Three operating conditions for different clocking positions have been considered. The variation of the operating conditions has been achieved by varying the HP shaft velocity and pressure ratio, with a consequence change of pressure ratio in the LP rotor. For this analysis the LP shaft velocity was kept constant. The TMTF performance variations will be analysed in terms of total pressure loss coefficient and exit flow angle; the mean interaction between the structures coming from the HP stage and the struts will represent the interpretation key to explain these variations. This work is part of the EU project DREAM (ValiDation of Radical Engine Architecture SysteMs, contract No. ACP7-GA-2008-211861).Copyright

A Comparison of Three Low Pressure Turbine Designs

As part of the current research, three LPT geometries — which were designed with a common pitch, axial chord, inlet angle, and exit Mach number and to create the same nominal level of turning — are compared. Each of the LPT cascades was investigated under a range of Reynolds numbers, exit Mach numbers, and under the influence of a moving bar wake generator.Profile static pressure distributions, wake traverses at 5% and 40% axial chord downstream of the trailing edge and suction side boundary layer traverses were used to compare the performance of the three designs. The total pressure losses are strongly dependant on both the maximum velocity location as well as the diffusion on the suction surface. The importance of the behavior of the pressure surface boundary layer turned out to be negligible in comparison.Cases with equivalent operating Reynolds number and suction side diffusion level are compared in terms of the total pressure losses that are generated. It is shown that a relationship between loss and suction side maximum velocity location exists. An optimum suction side maximum velocity location depends on the Reynolds number, diffusion factor, and wake passing frequency.Copyright

Acoustic Comparison of Different Turbine Exit Guide Vane Designs Part 2: Experimental Analysis

In this paper the sound power levels of three different designs of turbine exit guide vanes (TEGV) of turbine exit casings are compared with a standard design of an TEGV. The comparison is made with respect to the LPT for the acoustically relevant operating point approach. Additionally a rough loss estimation is also given in this paper. It is shown that the acoustically optimised TEC reduces the sound power level of the main interacion modes by about 14 dB while the aeroynamically optimised TEC even increases the sound power level by 2 dB. All three TEGV designs show higher aerodynamic losses for this off design point (approach). The measurements have been conducted in the subsonic test turbine facility at the Institute for Thermal Turbomachinery and Machine Dynamics, Graz University of Technology.

Comparison of the Experimental Results Between a 2D EGV Cascade Test and a Rig Test Under Engine Representative Conditions

Today the exit guide vanes (EGV) of the turbine exit casing (TEC) of a bypass-engine have to fulfill three major functions. Firstly, they have to support the rear bearing of an aero engine and to provide space to lead through different supply lines, secondly the engine mount is supported and therefore they have to be rigid and large in thickness. Thirdly, the EGVs have to reduce swirl of the last stage LPT (carrying the aerodynamic load) in order to transform it into thrust for highest propulsive efficiency. Further, if available they have to provide the correct flow field for the following component e.g. a mixer as used in low by-pass ratio engines. Such a state-of-the-art EGV is subject to a 2D cascade test performed at the Institute for Fluid Mechanics at Technical University Braunschweig and a rig test performed at the Institute for Thermal Turbomachinery at Graz University of Technology. This work presents the differences in the results between these two tests due to three dimensional effects such as incoming wakes, turbulence and radial variations in swirl, total pressure and yaw angle over the passage height.Copyright

Darmstadt Rotor No. 2, II: Design of Leaning Rotor Blades

For Darmstadt University of Technologys axial singlestage transonic compressor rig, a new three-dimensional aft-swept rotor was designed and manufactured at MTU Aero Engines in Munich, Germany. The application of carbon fiber–reinforced plastic made it possible to overcome structural constraints and therefore to further increase the amount of lean and sweep of the blade. The aim of the design was to improve the mechanical stability at operation that is close to stall.

Impact of vane clocking on the TEC loss in Rig

The 30% span profile section of the H-TEC highly loaded Turbine Exit Casing was transformed into a plain cascade configuration with identical pressure distribution and loading level. At increasing levels of complexity the incompressible flow in cascade is studied first at different inlet turbulence levels at the Technische Universitat Braunschweig. Unsteady inlet conditions were then generated using rotating bars at the cascade wind tunnel of the Armed Forces University Munich at compressible flow speeds and varying Reynolds numbers. Finally the TEC configuration is investigated in the TU Graz STTF 1.5 stage turbine rig using a conventional 5-hole-probe, a miniature Pitot-probe and a hot wire anemometer. A special test- and evaluation concept allows for highly accurate data. The results show a strong vane-TEC clocking, which for the TEC total pressure inlet profile may be approximated by a sine-function. Downstream the TEC, between the wakes, still a sinus shaped total pressure variation is found. Two loss coefficients are evaluated (i) by a classical control volume approach between TEC inlet and exit plane and (ii) by a so-called viscous wake approach. This method compares the viscous flow in the wake region with a hypothetical potential flow deduced from the flow in the same exit plane between the TEC wakes. The viscous wake method compared at 30% span quite nicely to the cascade data. However, the control volume approach yields more than twice the cascade loss, which indicates further loss sources to exist, e.g. unsteady losses to rise from a vane wake - rotor - TEC interaction or turbulence to impact the probe reading.

Acoustic Comparison of Different Turbine Exit Guide Vane Designs - Part 1: Design Philosophy and Numerical Predictions

In a number of publications of the past years, the authors have presented detailed descriptions of the STTF acoustic turbine test facility at the Graz University of Technology and highlighted its importance for tool validation and technology verification. In this context, multiple configurations have been designed and tested at the cold flow LPT rig and their results with respect to aerodynamics and acoustics have been published. The present paper completes this approach by summarizing the results of a selected number of acoustic design modifications to the Turbine Exit Guide Vane (TEGV) in a comparative way and complementing it with an aerodynamically motivated design optimization. In total, a number of four TEGV designs will be presented and discussed with respect to the respective design objectives, effects on noise generation as well as scattering effects within the LPT (and TEGV) control volume, and noise reduction potential. Within this paper, the corresponding numerical predictions by the MTU in-house LEE code Lin3d will be presented. In addition, in a companion paper also to be presented at the 21st AIAA/ CEAS Aeroacoustics Conference, the experimental results of the four different TEGV designs will be assessed in detail. In addition to the modal acoustic measurements, also several aerodynamic results will be discussed.