Rapid and Efficient Sound Source Location, Measurement and Analysis
The AcoustiCam sound-field holography system consists of software fromAkustikforschung Dresden together with MSX16 measuring hardware and microphone array from SINUS and was developed on the basis of the SMT.
A common problem, especially in the acoustic examination of complex objects, is the efficient and correct location of sound sources: traditional measurements with only one or a small number of microphones often lead to unsatisfactory results. The method utilized in the AcoustiCam offers a solution for this problem. The simultaneous measurement with a large number of microphones enables the rapid localization and separation of sound sources.
The operational principle of AcoustiCam is based on signal processing of the synchronously collected phase-exact sound pressure levels of the individual microphones. The algorithm mathematically replicates the function of a concave mirror which scans the sound field under examination utilizing the sound traveling time.
The microphone array can be focused on any point in front of it by varying the calculated amplitude and phase correction terms. In this way, an image of the sound pressure distribution on a single image plane is obtained without mechanically moving the array.
The separation of the sound sources in terms of position and frequency depends on the selected microphone geometry. Any sound situation can be reproduced as a colored two-dimensional image of the absolute sound power distribution. For the better visualization of the sound situation the localization result is superimposed on a photograph of the object under examination.
Typical applications for AcoustiCam are:
Two MSX16 units controlled by a notebook or a standard PC are used as hardware for the data acquisition. In cases where it is necessary to improve the position and frequency resolution, the hardware may be extended to 64 measurement channels. Compared to other concepts, our solution has the following advantages:
- Localization and separation of sound sources for reducing noise emission of vehicles, machines, white goods and electric tools
- Source analysis of wind tunnel models, complex vibrating structures, screeching or rattling structures as a basis for the acoustic design.
- The hardware for the data acquisition may also be used for other measurement tasks.
- The system allows data to be recorded continuously on HDD from 32 channels over 8 hours.
- The entire measuring hardware may be independently powered by buffer batteries.
- The system works with various array geometries depending on the requirements.
- The solution features a high accuracy for a relatively low price.
Technical data (for Ring 32)
- Orthogonal beamforming
- This new algorithm of AcoustiCam is based on the decomposition of the localization result into uncorrelated, i.e. independent sound sources. These are based on different source mechanisms which may result in strongly differing source sound levels. The orthogonal beam forming leads to separate images of the individual sound sources which enables the separate localization of not only the high-level main sound sources, but also of the lower-level masked sound sources without having to enclose parts or perform several measurements. This method increases the signal-to-noise ratio to > 25 dB, which is a clear improvement in comparison to the 10 ... 15 dB achievable by methods currently in common use.
- Source analysis
- The AcoustiCam measuring system is able to analyze single sound sources by determining the source-typical sound pressure spectra related to specific points. Even under acoustically unfavorable conditions (e.g. in a wind tunnel) the existing sound sources are localized reliably. No special acoustic rooms are necessary.
- Boundary layer array
- The use of the so called boundary layer microphone array suppresses the sound propagating from behind the microphone array. For this reason, no special acoustic rooms are necessary and various geometries may be used for those measurements. Special inexpensive microphones are integrated in the array.
- Ring array
- The considerably lighter ring array (in comparison with the boundary layer array) consists of 32 microphones on a metal ring which can be mounted on a tripod. Standard measuring microphones with ICP supply and BNC connectors are used, which are easy to mount and which may also be used for other measuring tasks. Easy positioning is possible by using a wheeled tripod with a swivel head.
- Aperture angle
- As in an optical system, increasing the angle between object direction and the camera alignment causes distortions of the object’s mapping. In the acoustic far field these distortions can be neglected up to a maximum aperture angle of ±30°. The larger the dimensions of the test object, the greater the distance between the microphone array and the object has to be.
The AcoustiCam measuring system is based on the near- field beamforming algorithm. This algorithm allows acoustic examinations of test objects at small distances because the suggested maximum aperture angle increases with decreasing distance to the test object. The minimum distance which must be maintained is approximately 25 cm. Smaller objects can be placed very close to the microphone array, whereas larger objects require a greater distance in order to obtain an optimal image. For the application of AcoustiCam there are virtually no constraints regarding the dimensions of the object under examination.
- Spatial resolution
- Various resolutions can be obtained depending on the microphone array applied. Apart from the geometric arrangement of the microphones, the frequency and the distance to the sound source also determine the spatial resolution of the system. The spatial resolution of the array increases with increasing frequency, i.e. with decreasing wavelength.
||1 m x 1 m
||1 ... 5 m
||300 Hz ... 8 kHz
||octaves, 1/3 octaves
||approx. 30 s
||approx. 1 s
||28 cm @ 1 kHz, 14 cm @ 2 kHz, 7 cm @ 4 kHz, 3 cm @ 8 kHz
||12 dB @ 1 kHz, 12 dB @ 2 kHz, 12 dB @ 4 kHz, 12 dB @ 8 kHz
|Microphone ring array
||with 32 x 1/4" microphones on tripod
||2 x MSX16 with ICP inputs; CardBus interface, D-SUB cable
||P4 2 GHz, 80 GB HD, 1 GB RAM, WindowsXP
||using centric point sound source or 1/4" calibrator
- Sound pressure distribution as colored map superimposed over photograph
- Triggered time data on all channels
- Transfer functions between all channels
- Listening-in to the measurement at one point of the image area