DE102005048905B3 - Active channel noise attenuator having several acoustic sensors to detect the noise field parameters at the loud speaker - Google Patents

Abstract

The The invention relates to an active duct silencer, comprising at least one Speaker (2) with back soundproof housing (6), a first acoustic sensor (3) for detecting the sound field size on the speaker (2) and a linear signal amplifier (4), which is formed is, with the sensor (3) detected sound field size the speaker (2) to supply inverted, and characterized by the fact that further acoustic sensors (3) for recording the sound field size on the loudspeaker (2) are available, the of the speaker (2) and the flow and sound incident direction (7) are positioned differently in the channel (1), wherein the detected Sound field size of the linear signal amplifier (4) the speaker (2) can be fed inverted.

Description

The The invention relates to an active channel silencer according to the preamble of claim 1.

Spatially and structurally, active silencers concentrate eg on the type of reactive silencer EP 0898774 , the hybrid muffler DE 4027511 or the active Helmholtz resonator DE 4226885 have in channels advantages over conventional anti-sound systems (Nelson, PA, Elliott, SJ: Active Control of Sound and Vibration., Academic Press Limited, London: 1992), as they also have a low frequency compared to passive mufflers Soundproofing effect. However, this applies only to low flow velocities up to about 10 m / s, since at faster flows despite fleece covers (flow resistance) or the like, the turbulent pressure fluctuations at the sensors (mainly microphones) lead to so-called pseudo-sound coming from the speakers is emitted. If this self-generated sound level exceeds the noise level that is actually to be damped in the channel, no attenuation and in some cases even a level increase will occur. This problem significantly limits the applicability of (re) active mufflers, as newer fans with reduced noise emission can produce high volume flows and flow rates.

In the practical application of reactive mufflers has been found that the attachment of the microphone ( 3 ) in a suitable form on the mounting flange of the speaker ( 2 ) is possible, 1 , It showed that the orientation of the microphone ( 3 ) to the sound incidence direction ( 7 ) has a significant influence on the attenuation spectrum. 2 shows measurement results with a microphone position according to 1 (solid curve) and a microphone position on the side of the mounting flange of the speaker ( 2 ) (dashed curve), that is perpendicular to the sound incidence direction ( 7 ). The influence manifests itself in a shift of the attenuation maximum by about one octave and is therefore at least as clear as the use of other components (eg loudspeakers (eg 2 ) with different membrane mass) for shifting the resonance frequency.

Task of invention

task of the active according to the invention channel silencer is it, the problems mentioned at high flow velocities, standing waves in the channel and Partialresonanzen the speaker membrane considerably to minimize and at the same time the adaptation of the attenuation spectrum to a given noise spectrum significantly simplify.

Walkthrough

to solution The object is a device and a method according to the independent claims proposed. An active duct silencer, the at least one speaker with rear soundproof housing, one acoustic sensor for recording the sound field size on the loudspeaker and a linear signal amplifier which is formed, the detected with the sensor sound field size the speaker feed inverted and characterized by the fact that more acoustic sensors for Recording the sound field size on the loudspeaker are present, the re of the speaker and the flow and sound incidence direction positioned differently in the channel are, wherein the detected sound field size of the linear signal amplifier the Speaker inverted feedable is solved the problem.

The Sound field size on the speaker sits down from the source of noise in the canal as well as other noise sources, such as flow noise, generated sound and from the sound generated by the speaker together. The sensors must be arranged so that they the sound field size on the speaker, ie in usually on the diaphragm of the speaker. The distance to the loudspeaker, especially to the diaphragm of the loudspeaker, should therefore be a maximum of about 1 cm. Cheap is one possible low distance. It is also possible the acoustic sensors directly to the diaphragm of the loudspeaker. This would be from the acoustic Optimal position; However, the attachment designed directly at the diaphragm difficult, since the lasting oscillations the attachment to be able to solve.

The use of multiple acoustic sensors leads to a surprising improvement in the detection of the sound field size and thus to a significant improvement of the generated in the channel muffler Damping. It is an effect that goes beyond the usual improvement in measurements achievable through the use of multiple sensors. This is mainly due to the following reasons:
In spatially concentrated active mufflers according to the prior art, the sensor, as viewed in the channel longitudinal direction, detects the sound field at exactly one point in the channel wall. It has now been recognized that in many cases sound waves occur in these channels due to multiple reflection on cross-sectional or directional changes. This is associated with pronounced local sound pressure minima. If the sensor is precisely at such a minimum (in terms of frequency), a minimum sound pressure level is detected and a response of the active muffler is eliminated. A problem solution could indeed be to place the sensor differently, but this would require the exact knowledge of the standing wave field. In practice, this is given only in very few applications. The present invention overcomes this problem by using multiple sensors.

For active silencers after EP 0898774 and DE 4027511 , but also after the controlled acoustic waveguide EP 1141936 , adds another difficulty, which limits a higher effectiveness. The adjustable gain of the sensor signal depends on the stability of the closed loop in the entire frequency range, which is based on the fact that no unwanted phase shifts occur. At low frequencies, the speakers used behave like piston emitters and cause a linear transmission (magnitude, phase). At high frequencies, however, natural vibrations, so-called partial resonances, of the loudspeaker membranes occur, which lead to a sound field distribution which is uneven in magnitude and phase in front of the loudspeaker diaphragm. Even without consideration of the acoustic ambient conditions (eg channel wall opposite, canal internals), this results in a dependent on the specific sensor position influencing the phase response of the control loop. As with the channel-related standing waves in front of the membrane, the changed sensor positioning is virtually impossible here as well; but can be overcome efficiently by the present invention.

A efficient choice the individual sensors results when the signals of the acoustic Sensors over a switching device selectable and subsequently summarized by means of a filter, totalizer and amplifier unit and the speaker fed are. This can be proven and proven in the art and proven means a suitable selection of the sensors can be achieved.

The Sensors can both simultaneously and sequentially depending on operating conditions of to be damped noise source selected become. It is also conceivable that groups of sensors exist which are selected simultaneously or sequentially. In general, probably but the simultaneous selection of a variety of sensors for the best Result lead.

Around the damping To further improve it is possible, if necessary, with several active channel silencer different positioning of the acoustic sensors side by side flat in the canal wall, a flow channel or in silencer scenes to arrange.

To the Protecting the speaker can provide an acoustically transparent cover be placed in front of the speaker.

When Acoustic sensors offer microphones, as they are very inexpensive too are preserved and powerful are.

The Invention will be described below with reference to the figures

there the figures show:

1 Schematic representation of a reactive muffler after EP 0898774 with positioning of the microphone ( 3 ) on the mounting flange of the speaker ( 2 ) at the location of the sound incidence ( 7 ) in the channel.

2 Measured insertion loss of a reactive muffler after EP 0898774 with positioning of the microphone ( 3 ) on the mounting flange of the speaker ( 2 ) at the location of the sound incidence ( 7 ) (solid curve) and a microphone position on the side of the mounting flange of the speaker ( 2 ) (dashed curve), that is perpendicular to the sound incidence direction ( 7 ).

3 Schematic representation of the active channel muffler according to the invention with a plurality of microphones ( 3 ), which relate to the loudspeaker ( 2 ) and the sound incidence direction ( 7 ) are in different positions.

4 Schematic representation of the active channel muffler according to the invention with a plurality of microphones ( 3 ) at different positions, wherein the microphone signals via a switching device ( 5 ) optionally a filter, totalizer and amplifier unit ( 4 ) whose output signal at the input of the loudspeaker ( 2 ) is present.

5 Schematic representation of the active channel muffler according to the invention with a plurality of microphones ( 3 ), regarding loudspeakers ( 2 ) and sound incidence direction ( 7 ) in the channel ( 1 ), different positions with a protective cover ( 8th )

The core of the active channel silencer according to the invention is the use of several microphones ( 3 ) and their targeted different positioning with respect to the loudspeaker ( 2 ) and the sound incidence direction ( 7 ). It should be mentioned that in comparison of the material costs the microphones ( 3 ) represent by far the cheapest component. The microphones ( 3 ) detect in the channel ( 1 ) propagating sound field in front of or at the diaphragm of the loudspeaker ( 2 ), the rear side together with the housing ( 6 ) encloses a closed, possibly damped with absorber material volume of air. The output signals of the microphones ( 3 ) arrive at a switching device ( 5 ), the selected microphones ( 3 ) to a correspondingly multi-channel filter, totalizer and amplifier unit ( 4 ) switches through.

This means that both the signals of several microphones ( 3 ) are switched through synchronously and successively in response to operating conditions of the noise source to be damped. The output signal of the filter, totalizer and amplifier unit ( 4 ) finally lies at the entrance of the loudspeaker ( 2 ) at.

The selection of the switching device ( 5 ) switched through microphone signals can be made at any time and done case-specifically different. In any case, however, the signals of multiple microphones ( 3 ) through. This is crucial for suppressing flow-related turbulent pressure fluctuations in high velocity channels. The surprisingly working approach is that of eg two microphones ( 3 ) detected and summed signals x ₁ and x _2, depending on the origin and type of signals lead to a different result of the summation. The rms value of the sum of the signals is: Ex 1 (t) + x 2 (T)] 2 } = E {[x 2 1 (t) + 2x 1 (T) x 1 (t) + x 2 2 (T)] 2 } = X ~ 2 1 + X ~ 2 2 + E {2x 1 (T) x 2 (T)}

With the same origin (correlation or coherence) and almost identical rms values of the signals x ₁ and x _{2, the} following applies:

This can be readily assumed for the sound wave in the channel. For the square root of uncorrelated or incoherent signals with the same effective values, the following applies:

This, in turn, applies to flow-related turbulent pressure fluctuations in the channel. The detection and merging of a plurality of microphone signals in the active channel silencer according to the invention thus means that incoherent signal components (due to the flow) in the combined signal are significantly suppressed compared with the coherent signal components (sound). Exactly this is an essential goal and can be achieved by using and merging other microphones ( 3 ) still increase. A further immediate advantage results in a reduced position sensitivity of the active channel silencer according to the invention with standing waves in the channel ( 1 ). In view of several microphones ( 3 ) in slightly different positions reduces the risk that a single microphone ( 3 ) just recorded a local sound pressure minimum and a corresponding control signal to amplifier ( 4 ) and speakers ( 2 ) dominates.

Moreover, this effect persists and can be used to advantage when multiple microphones ( 3 ) with respect to the sound incidence direction ( 7 ) are concentrated. The already mentioned influence on the attenuation spectrum can be achieved by simply switching the switching device ( 5 ) instead of by exchanging acoustically relevant components (eg membrane mass of the loudspeaker ( 3 ) or size the back air volume) or rotation of the housing.

Finally, by using and considering multiple microphones ( 3 In the case of the active channel silencer according to the invention, the type of averaging also reduces the influence of the natural oscillations of the loudspeaker membrane occurring at high frequencies. The improved electro-acoustic stability of the control loop is the result.

Claims (7)

Active duct silencer, comprising at least one loudspeaker ( 2 ) with rear soundproof housing ( 6 ), a first acoustic sensor ( 3 ) for recording the sound field size on the loudspeaker ( 2 ) and a linear signal amplifier ( 4 ), which is formed with the sensor ( 3 ) recorded sound field size inverted to the speaker, characterized in that further acoustic sensors ( 3 ) are present for detecting the sound field size on the loudspeaker, with respect to the loudspeaker ( 2 ) and the flow and sound incidence direction ( 7 ) in the channel ( 1 ) are positioned differently, wherein the detected sound field size of the linear signal amplifier ( 4 ), the speaker is inverted fed. Active silencer according to claim 1, characterized in that the signals of the acoustic sensors ( 3 ) via a switching device ( 4 ) and then by means of a filter, totalizer and amplifier unit ( 5 ) and the speaker ( 2 ) can be supplied. Active channel silencer according to one of the preceding claims, characterized in that the signals of a plurality of sensors ( 3 ) can be selected either simultaneously or sequentially depending on operating conditions of the noise source to be damped. Active channel silencer according to one of the preceding claims, characterized in that a plurality of active channel silencers with different positioning of the acoustic sensors ( 3 ) side by side flat in the channel wall ( 1 ), a flow channel or silencer scenes are arranged. Active channel silencer according to one of the preceding claims, characterized in that an acoustically transparent cover in front of the loudspeaker ( 2 ) is provided. Active channel silencer according to one of the preceding claims, characterized in that the acoustic sensors ( 3 ) Are microphones Sound attenuation method in a channel ( 1 ) in which several acoustic sensors ( 3 ) the sound field on a loudspeaker ( 2 ) and with an inverting linear signal amplifier ( 4 ) the speaker ( 2 ), the acoustic sensors ( 3 ) with respect to the loudspeaker ( 2 ) and the flow and sound incidence direction ( 7 ) in the channel ( 1 ) are positioned differently.