Sound emission abatement measures for aircraft can only be developed with a knowledge of where the noise is being generated how it contributes to the overall sound signature of the aircraft. Therefore, it is necessary to locate the sources and rank their contribtions to the total emitted sound. Using microphone arrays in fly-over tests, the sound sources of the airframe and the engines of an aircraft in flight can be analysed
Source localisation and separation inside an aircraft engine
Localising sound sources in an aircraft engine is challenging since there is a multitude of sources present in the engine (combustor, turbine, ...) and each source emits a complex sound field. The complexity of the total generated sound structure is even more enhanced by multiple reflections at the inner duct walls. In order to resolve these difficulties, we develop localisation methods that use information from a multitude of microphones inside the engine. Specifically, the signals of a compact microphone array are combined with the signals of individual reference sensors placed near the dominant noise sources compressor, combustor and turbine (see figure). The microphone array can be used to analyse the highly complicated spatial structure of the local sound field by decomposing it into characteristic waves or modes. By combining this information with the signals from the reference sensors, the sound field can be separated into the components generated by the different engine stages. Because of the difficulties involved in installing reference sensors inside aircraft engines, we also develop methods that do not rely on reference sensors. The loss of information is compensated by proper modeling of the sound sources in the engine. There are currently two lines of development, one using a general approach based on point sources and one more elaborated approach using line sources. The latter allows us to model the sound generated by blade edges. The source localisation and separation methods can be applied to both tonal and broadband sound sources, and they can be particularly useful in detecting damaged engine components.
Static noise tests of aircraft engines
The noise emission of an aircraft engine is composed of contributions of the individual components of the engine. The dominating sources that radiate in the direction of flight are the fan and the first compressor stages. In the rear arc, the main sources are the fan noise that propagates through the bypass duct, sources from the engine core, mainly the combustor and the last turbine stages, and jet noise.
In order to validate the acoustic effects of modifications to an engine, we need diagnostic tools that separate the contributions of the different sources to the overall sound field of the engine. The standard procedure are free-field acoustic measurements on large open air static test stands. The acoustic certification of commercial engines is being performed in open air test facilities under the rules laid down in the document Annex 16, Volume 1 of the International Civil Aviation Organization (ICAO).
We are developing and applying measurement techniques where we use large linear microphone arrays with up to 250 microphones to analyse the sound field of an engine. The inverse method SODIX (Source directivity modeling in the cross-spectral matrix) was developed here in order to determine the absolute power and directivity of the individual sound sources of an engine. Because of the complexity and high costs of free-field noise tests, we are working to extend the measurement and analysis techniques from free-field tests to measurements inside engine test halls. The difficulties in projecting the sound field measured in an engine test hall to free-field conditions lie in the compensation of room acoustic effects and in reducing the turbulence levels of the flow into the engine to levels where fan noise remains at levels typical for the conditions in flight.
Fly-over acoustic tests
Low noise aircraft and aircraft engines can only be developed on the basis of a detailed knowledge of the sound sources. Using individual microphones, like in the fly-over measurements that are required in the acoustic certification process of an aircraft, the spectral composition and the directivity of the noise signature of an aircraft can be analysed. However, there is no direct information on where the sound sources are located on the aircraft. Only when the sound sources are known, the underlying acoustic mechanism can be analysed and constructive or operational measures can be employed to reduce noise.
Using sound source localisation methods based on micropone arrays, a technique also known as the acoustic camera, the sources of vehicles at rest as well as in motion or in flight can be analysed. Noise sources of aircraft can only be anlysed in flight because they either occur only in flight, like most aerodynamic sources, or they are modified by the relative motion in flight, like the jet noise or interaction noise between aircraft components.
The department of engine acoustics has been using microphone arrays to analyse sound sources on moving objects since the year 1978, initially for high-speed trains and later on aircraft in flight. The method has been continuously improved and applied in various European and German research projects. Curently, we use large microphone arrays with up to 250 microphones that are distributed over a large area, typically with a diameter of 35 m. The flight path and velocity of the aircraft have to be known in order to track the aircraft with the focus of the array and also compensate the Doppler frequency shift on the microphone signals. The array data is analysed using our programme ProSigMA. It contains an implementation of the classical beamforming algorithm in the time domain for moving sources coupled with a hybrid deconvolution method that calculates the source amplitudes by approximating the point-spread function of the array in the frequency domain.
Further information
Every two years, the "Berlin Beamforming Conference" (http://bebec.eu) is jointly hosted by DLR, the TU-Berlin and the GfaI. At this conference, internatinal experts on beamforming and source localisation methods present their latest developments. On the internet site of the BeBeC, you can find a literature data base on the subject:
http://bebec.eu/Downloads/Beamforming_Repository/beamforming_literature.html