Ulf Tapken

Optimisation of sensor arrays for radial mode analysis in flow ducts

For the assessment and improvement of noise reduction concepts and the validation of computational aero-acoustic (CAA) codes in turbomachinery applications, the detailed knowledge of the in-duct acoustic mode spectrum of tonal frequency components is of great interest. Radial mode analysis (RMA) is an experimental technique that delivers the complex amplitudes of higher order acoustic modes propagating through flow ducts. Thus, RMA enables the calculation of the acoustic power radiated in and against flow direction, though, requiring at high frequencies the acquisition of the sound field at a large number of positions in the flow duct. The quality of the analysis results is very sensitive to the arrangement of the measurement coordinates, the frequency, and the flow parameters. A simple and robust RMA realisation just consists of sensor rakes at a single duct cross section. This method, however, has the drawback of not being able to distinguish downstream and upstream propagating modes. Furthermore, the sound field as well as the flow field may be altered by the rakes. In the paper, a numerical study of RMA employing four substantially different sensor arrangements is conducted. The arrangements I and II consist of sensor rakes located at two or four axial measurement positions, respectively. Arrangement III is equipped with sensors mounted flush with the hub and the outer duct wall and in arrangement IV the sensors are mounted flush with the outer duct wall, only. The dependency of the RMA quality on the frequency, e.g. the number of cut-on modes, and on the number of axial and radial measurement positions was investigated by means of a condition analysis. Studies were carried out for the analysis of coherent modal sound fields in hard-walled cylindrical flow ducts of arbitrary hub-to-tip ratio with a constant mean axial flow profile. Additionally, the influence of a solid-body like swirl was considered. The paper reveals reasons for bad conditioning of the RMA-systems and gives guidelines for an optimum sensor separation in order to improve the overall system condition. Since the condition number is only a relative measure for the inaccuracies caused by the RMA system, simulations with synthetic sound pressure data were carried out. The RMA performance of the sensor arrangements I-IV is compared.

Turbine blade/vane interaction noise: acoustic mode analysis using in-duct sensor rakes

The sound field in the outlet duct of a high speed low-pressure turbine was studied to deepen the understanding of the sound generating mechanisms in a three-stage turbine. Special interest was given to the influence of the exit guide vanes (EGV) downstream of the turbine on the noise generation. Six radial rakes carrying ten Kulite-sensor probes each were mounted downstream of the EGVs in the cylindrical duct section of the turbine exit. The rakes were positioned at different azimuthal angles and staggered axially in pairs to avoid probe wake interference. The rakes were traversed azimuthally over 180 degrees in steps of 1.5 degrees to give a total of 240x30 measurement points. All sensor signals were acquired simultaneously with a sampling frequency of 22 kHz and stored digitally for later analysis in the frequency range 0-8.5 kHz. Measurements were made at operating conditions from 63% to 99% rotor design speed. Special attention was given to the blade passing frequencies (BPF) of the three turbine rotors. The chosen experimental setup permits decomposition of the sound field into azimuthal and radial modes. With this information, the sound power transmitted upstream as well as downstream can be calculated for frequencies up to 6 kHz. The results of the mode analysis provide a detailed view on the sound generation mechanisms and interaction processes between the various blade and vane rows. According to Tyler & Sofrin, the noise sources can be separated in rotor/statorand rotor/stator/EGV-interactions with associated azimuthal modes.

Turbomachinery Exhaust Noise Radiation Experiments - Part 2: In-duct and Far-Field Mode Analysis

Model scale tests to investigate the sound propagation through the engine nozzle system and the jet shear layers were carried out. The present paper focuses on fan noise radiation from the bypass nozzle into the far-field. For the synthesis of representative tonal in-duct sound fields with distinctive modal contents an actuator ring providing 30 loudspeakers was operated in the bypass duct. Radial mode analysis was accomplished in order to assess the quality of the mode synthesis and to deliver data for the validation of numerical methods. 60 microphones were employed in a rotatable bypass duct section to detect the in-duct sound field at 2400 positions. Tests were conducted without and with flow in both clean nozzle configuration and nozzle configuration with installed pylon, respectively. The azimuthal and polar structure of the radiated far-field was examined with help of a large circular microphone array, which was equipped with 80 microphones. Polar radiation angles between 25° and 115° could be detected by traversing the array in the anechoic chamber along the jet axis. Azimuthal mode analysis was performed in the far-field ring to explore in more detail the impact of the mean flow on the sound propagation and to assess installation effects. The influence of signal-to-noise ratio, microphone mal-positioning and microphone failure on the quality of the far-field mode decomposition are discussed. Further covered in this paper are experiments with loudspeakers placed parallel to the jet axis outside the nozzle. The experiments were carried out to investigate the acoustic shielding of sound waves due to the coaxial core and bypass jets, which is of interest with regard to reflection effects at the lower wing surface in engine under-wing installations.

Acoustic Mode Decomposition of Compressor Noise under Consideration of Radial Flow Profiles

§To explore the sound generating mechanisms of turbomachinery like turbines or compressors in rig tests, the sound field radiated in the inlet or outlet duct is acquired by means of a sensor array and fitted to a theoretical model of the modal sound field. For this purpose, an analytical model with a uniform radial flow profile is usually deployed. This paper presents the derivation of an alternative sound field model under consideration of a realistic radial flow profile. The new model, which has to be solved numerically, was fitted against experimental data acquired in the inlet of a three-stage low pressure compressor, yielding the amplitudes of all radiated modes. The results are compared with the outcome of the classical radial mode decomposition approach basing on the analytical model.

Turbomachinery Exhaust Noise Radiation Experiments - Part 1: Polar Directivity Measurements

The test setup and first experimental results of model scale tests to investigate the sound radiation through the aero engine nozzles and the jet shear layers are presented. Fan rearward noise was simulated by a mode generator and loudspeakers. Tones and broadband noise were generated. Radial mode analysis in the bypass duct allowed characterising the generated sound field. An aerodynamic source and a siren were used to generate sound in the hot core where loudspeakers could not be applied. A polar microphone array in the downstream anechoic chamber and an azimuthal array that could be traversed along the jet axis delivered data to determine both the polar far-field directivity and the azimuthal modal breakdown. Various configurations were tested.

A New Modular Fan Rig Noise Test and Radial Mode Detection Capability

A new major large-scale fan rig test facility, UFFA (Universal Fan Facility for Acoustic), has been designed with the objective to allow test bed changes for engine representative OGVs and bypass duct annulus and liners, for reduced build times, and higher fidelity investigation of aft fan noise technologies. An important enhancement consisted in the implementation of three Radial Mode Detection (RMD) devices in the bypass duct and further downstream in the nozzle equivalent plane. High effort was spend on the realisation of a wall-flush mounted sensor array, which has the advantage not to disturb the flow and the acoustic field. However, the separation of different radial mode contributions is realised only implicitly by the analysis of the axial wave number spectrum, which is particularly challenging if sensors are installed only at the outer duct wall. More robust from the numerical point of view is the established technique to directly measure the radial structure of the sound field with sensor rakes. It is one of the main objectives of this paper to verify whether both techniques deliver the same experimental results also at the high targeted frequencies up to kR=75. As the examination of recently obtained data revealed, the sensor rake measurements were influenced by aerodynamic perturbations originating from the fan rotor wakes. The radial mode analysis could be significantly improved by incorporation of appropriate aerodynamic eigenfunctions. Further investigated was the sensitivity of mode detection with sensor rakes against manufacturing and installation tolerances. I. Introduction A major new large scale fan rig test facility, UFFA (Universal Fan Facility for Acoustic), has been designed, manufactured and commissioned by AneCom AeroTest GmbH, Wildau, Germany. The new modular fan rig test facility builds on previous Rolls-Royce large scale fan rig design experience 1 but extends the experimental capability to allow test bed changes for engine representative OGVs and bypass duct annulus and liners, for reduced build times, and higher fidelity investigation of aft fan noise technologies. In order to meet the challenging noise targets in a timely manner for future aircraft, the industry requires engine and nacelle representative high Technology Readiness Level validation test vehicles to deliver the technology ready for implementation in to the full scale product. An important enhancement consisted in the implementation of three Radial Mode Detection (RMD) devices in the bypass duct, i.e. radial sensor rakes directly at OGV exit and radial sensor rakes respectively a wall-flush mounted sensor array further downstream in the nozzle equivalent plane of the UFFA test facility. The sensor arrays were developed by DLR within the frame of the EU FP 6 project VITAL 2 , based on predicted sound field characteristics provided by Rolls-Royce. At the maximum targeted frequency of kR=75 more than 2500 modes of azimuthal mode orders up to m=85 and radial mode orders up to n=10 should be resolved. Major objectives of future RMD measurements are (1) the accurate assessment of fan rig design changes, (2) the provision of high quality data

Tonal Noise Radiation from an UHBR Fan { Optimized In-Duct Radial Mode Analysis

mounted sensors was applied to carry out a complete acoustic radial mode decomposition at high frequencies. In this case, Radial Mode Analysis (RMA) is in particular challenging due to the { compared to ducts of larger hub-to-tip ratios { large radial mode content. In this paper it will be shown that with the help of a regularization technique RMA results of high accuracy can be achieved with a minimized number of measurement positions. Further, the acquired test data will be used to verify that a partial RMA is feasible, which delivers the amplitudes of the dominant radial modes using the input of only a few seconds sound pressure measurements at a small number of sensor positions. In that way, i.e. the partial RMA enables an online-monitoring of the rotor-stator interaction modes. An application of the partial RMA technique is of high interest to monitor the active control of individual radial modes and respectively the assessment of the tonal sound power e.g. during fan rig accels or decels.

Validation of turbine noise prediction tools with acoustic rig measurements

Well-defined measurement data are of key importance for the validation and calibration of numerical methods. Accordingly, an acoustic test rig resembling the last stage of a low pressure turbine (LPT) of a modern turbofan engine has been established at the Graz University of Technology in Graz, Austria within the frame of the European Research Project VITAL. It has been designed to allow for a clear separation of the contributors to turbine noise by a systematic selection of the individual blade numbers of the subsequent rows. This setup permits the classification of the measured or computed tones with respect to their origin and source mechanism. Within the VITAL project, several acoustic test campaigns have been carried out: a datum (or reference) configuration as well as variations of the axial gapping between the stator, rotor, and the EGV. These give insight into the relative weighting of the above mentioned source mechanisms and the influence of distance variations on noise generation.

Experiments on an Axial Fan Stage: Time-Resolved Analysis of Rotating Instability Modes

Rotating Instability (RI) occurs at off-design conditions in axial compressors, predominantly in rotor configurations with large tip clearances. Characteristic spectral signatures with side-by-side peaks below the blade passing frequency are typically referred to RI located in the clearance region next to the leading edge (LE). Each peak can be assigned to a dominant circumferential mode. RI is the source of the clearance noise and an indicator for critical operating conditions. Earlier studies at an annular cascade pointed out that RI modes of different circumferential orders occur stochastically distributed in time and independently from each other, which is contradictory to existing explanations of the RI. Purpose of the present study is to verify the generality with regard to axial rotor configurations.Experiments were conducted on a laboratory axial fan stage mainly using unsteady pressure measurements in a sensor ring near the rotor LE. A mode decomposition based on cross spectral matrices was used to analyze the spectral and modal RI patterns upstream of the rotor. Additionally, a time-resolved analysis based on a spatial Discrete-Fourier-Transform was applied to clarify the temporal characteristics of the RI modes and their potential interrelations. The results and a comparison with the previous findings on the annular cascade corroborate a new hypothesis about the basic RI mechanism. This hypothesis implies that instability waves of different wavelengths are generated stochastically in a shear layer resulting from a backflow in the tip clearance region.Copyright

Noise Source Location in Turbomachinery Using Coherence Based Modal Decomposition

Coherence based source analysis techniques can be used to identify the contribution of turbomachinery core noise sources to pressure measurements in the far-field. The usual approach is to locate a ...