DE4027511C1 - - Google Patents

Description

1. Field of technology

Silencers are predominantly made with "passive" sound absorbers lined. The interaction of the absorber lining with The sound field in the muffler duct is essentially through the acoustic surface impedance of the absorber lining. There is a so-called optimal surface impedance caused by the winding point the so-called absorber function between the first and the second silencer mode is clearly described. Realization efforts in the specialist literature have shown that there are fundamentally is not possible with purely passive absorber elements this winding point impedance over a larger frequency range to adhere to.

Another technique, the so-called "active" sound insulation, is the noise to be reduced by an electroacoustic System superimposed on a sound wave generated by this system, that it is an extinguishing in the direction of sound propagation Interference comes (method of anti-sound).  

2. Outline

In the following, the technical terms of the main claim are brief explained, then the state of the art is discussed, from which the Task of the invention is derived. Furthermore, the The idea of the invention is formulated and the steps for its implementation are exemplary treated. Then the technical progress of the invention is demonstrated and finally the claims of the invention are enumerated.

Where appropriate, mathematical formulas are used as Shortest and most precise representation of physical-technical relationships. Furthermore, exemplary schematic drawings serve to explain the Invention idea and its realization.

3. Explanation of terms

The explanations of terms are intended to help describe the invention and to justify their technical progress. For a more complete presentation the acoustics of silencers is referred to in literature [1].  

Silencer

Silencers are used to block the passage of sound through openings reduce. In most applications, these openings are the clear cross sections of channels. Silencer - and thus also the invention Silencers - are also used to reduce the passage of sound through joints (e.g. door joints) and through openings for one Material or for people passage (for example in soundproof capsules from Machines and in soundproof cabins). The openings are as follows collectively called "channels".

The characteristics of the hybrid sound absorber according to the invention must be adapted to the shape of the clear cross section of the silencer duct. The idea of the invention is developed here as an example for rectangular cross sections, as shown in Fig. 1, but is not limited to this shape. The invention can also be used for other cross-sectional shapes, for example in the case of round or annular channels.

Absorber lining

Silencers with a sound-absorbing lining basically have a sound-absorbing effect by fitting an absorber lining on the entire circumference or on part of the circumference of the clear duct cross-section, which (mostly only in parts, for example the air enclosed in porous materials) is excited to vibrate by the sound wave and thereby interacts with it, so that the sound wave as it propagates through the channel is dampened to intensify the interaction, the clear cross section of the channel is often divided by sound-absorbing backdrops. Such a backdrop silencer is shown in Fig. 2. The idea of the invention is also applicable to such forms of absorber lining. Each gap in the backdrop is then treated like a silencer according to Fig. 1.

Sound absorbers are called "passive" when the resonance of the Absorber lining by the sound wave only via mechanical (acoustic) Interactions. The strength and phase of the The resonance is then determined by the arrangement, the dimensions and the Material properties of the absorber lining determined. The selection of the Components of the passive absorber in the silencer according to the invention not restricted by the idea of the invention. These can be, for example: porous sound absorbers made of fibrous materials (mineral fibers, metal fibers, Plastic fibers, organic fibers, open-cell foams, porous sintered ones  Materials etc.), flexible plates, foils and membranes, perforated sheets, Fabrics, fleeces etc. These components of the passive absorber can become either in direct contact with one another or in a mutual one Distance from each other. For a more complete description of such Absorber components and their acoustic properties are based on the Reference [2] referenced.

Active subsystem

The active subsystem according to the invention consists of one or more Sound recorders, one or more signal conditioners and one or more Sounders (mostly: loudspeakers). The sound pickups are mostly microphones, can also be structure-borne noise sensors. The signal formers for that Output signals from the sound pickups are combinations depending on the task of amplifiers, frequency filters and other known electronic Components. The sound generator is determined by the electrical output quantity of the Waveformers according to the inventive concept set in vibration that its vibrating surface (for example a speaker membrane) one acoustic impedance according to the invention.

The electronic signal formers can in turn consist of passive and off active electronic (according to the common usage in electronics) Components exist; these can be known "analog" or "digital" Be functional units.

A diagram of the active subsystem with the sound pickup ( 1 ), the signal former ( 2 ) and the sound generator ( 3 ) is shown in Fig. 3.

Impact on the passive absorber

In a preferred embodiment of the inventive concept, the loudspeaker of the active subsystem is arranged on the side of the passive absorber layer facing away from the channel and represents an acoustic terminating impedance Z _b for this. This arrangement, hereinafter referred to as "series connection", serves as an example below for the development of the inventive concept. It is shown schematically in Fig. 4. However, it is not the only possible arrangement according to the invention.

In another exemplary embodiment, the channel-side surface of the sound-absorbing lining consists of an arrangement of "passive" and "active" partial areas. The passive subareas are formed by the surface of passive sound absorbers, the active subareas are formed either by the surface of the loudspeakers or by the mouths of loudspeaker chambers, which in turn can be completely or partially filled with passive absorber components. This arrangement is also referred to as "parallel connection". Fig. 5 contains a schematic representation.

Acoustic impedance

The acoustic impedance Z _{w of} a surface is the ratio of the sound pressure on this surface to the surface normal sound velocity. Occasionally, the reciprocal of the impedance, namely the (acoustic) admittance G _w = 1 / Z _w , is also used below.

The product U = k₀h · Z₀G _w from the wall admittance G _{w of} the absorber lining with the frequency variable k₀h from k₀ = 2πf / c₀ with the frequency f and the speed of sound c₀ as well as from half the width h des (assumed here in the form of a column for simplicity of description) clear duct cross-section and finally the wave resistance Z₀ of the air (c₀ and Z₀ are material constants of the air or the sound-carrying medium in the duct) is the so-called "absorber function" U. It essentially determines the effectiveness of a specific absorber lining in a silencer (for more see below and in [1]).

To simplify the spelling, it is assumed below that all impedances Z and all admittances G with the characteristic impedance Z₀ des sound-conducting medium are standardized. This is done by replacing Z → Z / Z₀ and G → Z₀ · G.

Silencer modes

Modes are basic forms of vibration of a sound field. They have following properties:

•  they "fit" into the boundaries of the sound field, that is, they meet the boundary conditions each,
•  one can overlay any (even complicated) Synthesize sound field within this boundary,
•  the active power transported through such a sound field is simply that Sum of the sound powers of the modes (what if the Sound field into components other than fashions would not be so easy).

Let x be the axial coordinate of a channel in the direction of sound propagation and y is a transverse coordinate (in the gap channel perpendicular to the absorber surface), then a suitable representation of the sound pressure field is:

p (x, y) = e ^{- Γ x} · q (y) (1)

Here Γ = Γ ′ + j · Γ ′ ′ is the complex (longitudinal) propagation constant and q (y) a suitable transverse distribution function, which defines the boundary conditions at the Surface of the absorber lining must meet.

The real part Γ ′ of Γ is the damping in Neper per unit length. The attenuation D _h due to the sound level drop in decibels (dB) per half the gap h is customary (and used in the following). This is:

D _h = 8.68 · Γ′h [dB] (2)

The transmission loss D _{d of} a silencer of length L is then:

The longitudinal propagation constant Γ is obtained from the frequency variable k₀h and the auxiliary variable z from the formula:

where the auxiliary variable z is a solution of the determination equation

z · tan z = jU (5)

is with the one introduced above (and known from the absorber data) Absorber function U.

The calculation method outlined here applies to the selected example of one gap-shaped rectangular channel with symmetrical modes (in-phase Vibration of the sound field on both sides of the symmetrical surface of the gap channel). The Case of antisymmetric modes (antiphase oscillation on both sides of the Symmetry area) leads to the equation for z (see [1]):

z / tan z = -jU (6)

For other cross-sectional shapes, reference is made to [1]. With a round Channel cross-section with an axially symmetrical field is, for example Determination equation:  

with the Bessel functions J₀ and J₁ zero or first order. The Values in the winding point between the first and the second mode are:

U _wp = 2.98 + j · 1.98 and z _wp = 2.2285 + j · 2.358.

The function tan z is periodic in the real part z 'of the complex argument z = z ′ + j · z ′ ′. Their stock of values is repeated along z ′ at intervals of π. The leads to the fact that the equations (5) and (6) are (infinitely) many Have solutions z. Each of these solutions then represents a silencer mode It can be seen from Eq. (4) that the different modal solutions for z too lead to different modal damping. The fashion dampening grows in generally with the ordinal number m of the modes, that is, with the running number of the Period stripe along z 'from which the solution comes.

Which silencer mode prevails, which mode damping is too use depends not only on the design and dimensioning of the Muffler, but mainly from the transverse distribution of the Muffler impacting sound field, accordingly from the sound source (blower, Compressor, flame etc.) and from the defects (corners, branches, Fittings etc.) in the duct in front of the silencer. One through use The justified rule is therefore the silencer for the lowest damped To interpret fashion. This was recommended in the early works [3] and [4]. This rule is confirmed because the lowest subdued fashion is below the first two modes and because higher modes are difficult to excite, so rarely dominate the sound field. Should they still exist, then they would have a higher damping than the design damping.

Winding point between modes

For the solutions z = z ′ + j · z ′ ′ of the determination equation, curves with constant real part z ′ and constant imaginary part z ′ ′ are carried over the complex level of the absorber function U = U ′ + j · U (which is assumed to be known) '' On, then the curve networks of Fig. 6 and Fig. 7 emerge . Each of these figures covers approximately one of the period strips over z ', that is, a mode. Fig. 6 the first fashion of Eq. (5) and Fig. 7 the next higher mode. Other forms of representation can be found in [1], [3], [4].

At the bottom of the range in these figures, in which the Curve networks are (almost) rectangular, you can see points around which the "Wind around" curves. These are the "winding points".

What is important in connection with the invention is that shown in [4] The fact that a fashion has its highest damping in its upper winding point Has.

Also important in connection with the invention is the fact that neighboring modes have these winding points in common. There the two solutions z of neighboring modes coincide. So if the absorber lining of a silencer has an absorber function with the value U = U _wp in the winding point of the first two modes, then there is no mode of oscillation in this silencer with a lower damping, since the least damped mode is among the first two modes.

Approaching the turn point

[4] describes how the values of the absorber function U _wp at the winding point and the associated solutions z _{wp are} determined and numerical values are given there. For the winding point between the two lowest modes, the following applies for a symmetrical sound field in the rectangular gap channel: U _wp = 2.05998 + j · 1.65061 and z _wp = 2.1062 + j · 1.12536.

Fig. 8 shows the frequency curve of the damping D _{h of} the lowest damped mode in a rectangular gap channel when the absorber function U of the absorber lining is at the winding point U _wp at all frequencies. This curve is obtained from Eqs. (4) and (2), if one in Eq. (4) z = z _wp . The abscissa of this (and all of the following) representation of the damping D _h is the product f · h in [Hz · m], because silencers can be summarized that meet certain "similarity laws". Note the logarithmic plot of the damping.

In order to clarify the statement in Fig. 8, consider a silencer with a gap H = 2 h = 0.2 m, i.e. h = 0.1 m. This is a frequently used gap width in silencers for technical systems. Then a silencer according to Fig. 8 with a length L = 1 m at low frequencies would have an attenuation of D _d = 190 dB, a truly extraordinary value!

This extremely high damping is technically hardly fully usable, because then other sound transmission mechanisms, such as structure-borne sound propagation in the silencer housing, would limit the effective noise level reduction by the silencer. Therefore, one does not need to set the value U _{wp of} the absorber function exactly in a silencer according to this invention; it is sufficient to approximate this value in order to achieve significant improvements in sound attenuation compared to conventional silencers.

Desired frequency range

With purely passive absorber linings, it is not particularly difficult to approximate the value U _{wp of} the absorber function in individual frequencies. The data for a simple absorber layer made of mineral fibers are given in [1, Figure 19.13] (d / h = 1; Ξd / Z₀ = 0.4), the absorber function of which even passes through the winding point at two frequencies. Fig. 9 shows the associated damping curve (albeit with today's absorber characteristic values of the mineral fibers instead of the characteristic values used in [1]; therefore only an approximation to the optimal damping curve) . Typical for such purely passive realizations of the optimal wall impedance of the absorber lining is that the high Attenuation value is only achieved in narrow resonance peaks and is bought at other frequencies with deep attenuation drops.

This also applies to the efforts to implement passive absorbers in the literature reference [6], which was dedicated solely to this task. Also the Successes in the implementation efforts in [7] are not significantly different from that.

As already explained in [4], the reason for the failure of realizing U _wp over a wide frequency range is that no passive component is known which connects the reactance of a spring with the frequency response of an inertial mass. The problems with the purely passive, broadband implementation of the optimal impedance of the absorber lining are therefore of a fundamental nature.

It is the crucial advantage of the invention over having silencers purely passive absorber linings that the optimal impedance over a wide frequency range can be kept.

In principle, it would certainly be desirable to set the optimal impedance over the entire frequency range, i.e. to implement the attenuation curve in Fig. 8. However, this would increase the implementation effort considerably. It is a further advantage of the invention that the frequency range and the level of the improved sound attenuation can be set, as a result of which compromises can be made between expenditure, technical constraints (such as space requirements, pressure loss, etc.) with the sound attenuation to be aimed for.

Most technical applications of silencers is common that a high attenuation at low frequencies (approx. 30 Hz to approx. 250 Hz) is required. This is due to the fundamental course of the frequency spectra the noise of flow generators with an increase towards low frequencies. The problems of achieving high, broadband are almost as fundamental Sound attenuation values at low frequencies with purely passive Absorber linings. This is due to the principle of passive acoustic Components that all dimensions are measured in units of sound wavelength which grows towards low frequencies. As a result conventional silencers for low frequencies a large volume, the Absorber thicknesses d have to become large, which then leads to an increase in Flow pressure loss leads. An example below shows that the resulting operating costs of a silencer Acquisition value can far exceed.

For this reason, special attention is paid to the damping low frequencies.

4. State of the art

A description of the state of the art must address two techniques: the technology of "passive" sound insulation with purely passive components and the Technology of "active" sound insulation with only electroacoustic, "active" Components.

The problems with the implementation of high sound attenuation with purely passive absorber linings - especially at low frequencies - were mentioned above in part in the explanations of the term. The state of the art in a good design of a passive broadband silencer is described [1, Fig. 19.21], which is repeated here as Fig. 10 with the designations used here. A similar designation also contains [8, Fig. 12/10] and [9, Fig. 6.138], whereby in the latter reference the upper limit for technical realizations is not set to D _h = 3 dB, but to D _h = 1.5 dB is. The revision of the VDI directive on silencers [10] also repeats this information as the state of the art for good broadband silencers.

The invention presented here does not actually have to deal with passive silencers, since they are comparable neither in terms of the solution concept nor the damping achieved. A modern development is briefly presented to illustrate the current status in this area. In [11] an absorber element made of so-called "membrane absorbers" for silencers is described, which has been painstakingly tuned by the combination of multiple resonance systems in such a way that over a certain frequency bandwidth (usually approx. 1.5 octaves) the highest possible damping with as much as possible shallow depth. Fig. 11 shows damping curves of different types of this absorber element, which are tuned to different frequency ranges. The above-mentioned technically feasible attenuations of D _h = 3 dB are hardly exceeded. The advantage of the smaller overall depth remains. But the price of such elements is three to five times the price of conventional shock absorbers with otherwise comparable performance.

DE 34 25 450 A1 [20] also deals with the same goal, namely to reduce the overall depth of the link elements. This solves a particular problem with link silencers, which arises from the fact that the entire width of the channel in which the links are installed (see Fig. 2 of the present description) is constantly changing because this channel width is in the available space must adapt, but that on the other hand, the width of a certain backdrop arrangement can only change abruptly by the width D + H with the addition or omission of another backdrop. The cited document indicates that with a backdrop arrangement of backdrops of the same thickness with backdrop gaps of the same width in the intermediate region of such a jump in the channel width, either sound insulation is lost if the backdrop column is increased proportionally with increasing channel width, or the pressure loss of the Muffler increased inadmissibly if you add another backdrop with increasing channel width and the backdrop column is reduced so that the added backdrop finds space. According to this document, the problem is to be solved by using scenes with different thicknesses. A suitable combination of these makes it easier to realize a predetermined overall width of the link arrangement without the ratio D / H, which is decisive for the damping and the pressure loss, having to be changed. Suitable combinations are given in detailed lists in this document.

The aforementioned publication has with the present invention only the general goal in common, for backdrop silencers increase damping and decrease pressure loss. The type of absorber structure of the backdrops used - completely open according to the task at hand; consequently there is no reference to the one to be realized Optimal impedance of the backdrop surfaces that form the basis of a forms muffler according to the invention. The font mentioned contains also no mention of an electroacoustic system to achieve such an optimal impedance. So everyone is missing there Reference to the basic concept for the silencer according to the invention and also on a crucial basic component of the same, namely on the active subsystem.

The comparison with the so-called "active" silencers is more important. This development started with the patent specifications [12]. After a long time in between started a vigorous research activity on the Methods of "active" noise abatement. In [13] is a short one Given overview. A compilation of literature from 1988, [14], contains 1708 relevant citations from publications in this field. This lively one Research activity was triggered by natural restrictions passive acoustic components that only work with the sound field according to the specified acoustic laws for their impedances and transfer functions in Can interact. Another impetus for the lively efforts the technical development of the "active" noise abatement electronic components, which allows practically any mathematical Formulated impedance or transfer function with a reasonable Realize costs electronically.

The principle of active noise reduction, often referred to as "noise abatement by means of anti-noise", is shown in Fig. 12. From the left, the sound wave from a noise source hits a microphone ( 1 ) in a channel. Its output signal is sent to a loudspeaker ( 3 ) via a signal processor ( 2 ) (with amplification, filtering, delay time shift, etc.), which essentially emit sound that is intended to interfere with the sound of the noise source in a way that cancels out. In fact, this interfering extinction is effectively only possible in one direction (forward direction); in the backward direction the sound tends to be amplified in most active systems. In a further development, according to Fig. 13, a microphone ( 4 ) is also arranged behind the loudspeakers, the output signal of which serves to adaptively improve the parameters of the signal processor ( 2 ) via a controller ( 5 ) in the direction of minimizing the sound pressure at the location of the Microphones ( 4 ).

Most studies of the research mentioned deal with the development of control algorithms for signal processing and for adaptive rework. Another part deals with the development suitable speakers and their arrangement.

As impressive as the principle of active noise abatement initially appears, some fundamental and technical difficulties are clearly evident. First of all, it is easy to see with what accuracy the sound field compensation must work across the entire channel cross-section. In order to achieve a transmission loss D _d of 40 dB (a requirement of the usual order of magnitude in the medium frequency range), the residual error of the sound field compensation must only be 10% on a residual area of 1% of the channel cross section or a fixed error of approx. 3% of the compensation on a partial area of 10%. This explains why broadband sound attenuation of 40 dB with active systems was only realized under laboratory conditions (exact channel cross-sections, exactly flat sound waves, no flow), since the loudspeakers for the sound at the edges produce sound fields that are difficult to control. Another fundamental problem arises from the fact that air flows prevail in the silencer ducts. However, the active system cannot initially distinguish between turbulent pressure fluctuations in the flow, which do not emit sound and are therefore harmless, and the sound pressure fluctuations to be reduced. This is the reason why the originally held hopes of obtaining a silencer for low frequencies (below approx. 100 Hz) could not be fulfilled, as the turbulent pressure fluctuations become stronger with falling frequency. A technological problem is that most flows in technical silencers transport hot gases and / or chemically aggressive vapors and / or dirt, since both the control microphones and the loudspeakers are directly exposed to the sound field and thus to the flow. Another loudspeaker technical problem is that the speed of sound of the loudspeaker diaphragms must be approximately the same size as the speed of sound of the noise, but that the sound pressure is almost zero due to the desired compensation. The loudspeaker thus works on a sound field with a low field impedance. As a result, the mechanical impedance of the loudspeakers must also be very low; the membranes must be light and soft and must make large strokes (at least at low frequencies). Such loudspeakers are in turn not suitable for being exposed to turbulent flow. A technical problem arises from the fact that the control microphone ( 1 ) on the side of the noise source not only picks up the sound field to be suppressed, but also the anti-sound emitted to the rear by the loudspeakers. In fact, a large proportion of research has been used to prepare the interference signal from the superimposed signal using suitable control algorithms. Another fundamental problem in the technical application of active silencers is that both the control microphones ( 1 ) and control microphones ( 4 ) can only scan the sound field in a few points, and the anti-sound field can only be accurately measured in a few points by the loudspeaker arrangement can be specified. This can work as long as the noise field has a constant transverse distribution. However, this transverse distribution changes continuously with technical sound sources and with the large transverse dimensions of the channels. All of this may be the reason why active silencer systems have barely got beyond the laboratory stage; at least they have not yet gained any technical significance despite the intensive research activity. In the case of a technical application, it would also make it more difficult that if the active system failed, there would be virtually no sound attenuation at all.

A method and a device for active sound attenuation are described in the published patent application DE 27 12 534 A1 [21]. It also defines the principle known as active damping, which consists in "that the energy content of (a) wave can be reduced by combining the primary wave with a specially generated secondary wave in such a way that the dilutions of the secondary wave with the Compression of the primary wave collapse and vice versa ". The aforementioned publication emphasizes that the secondary wave must be generated exactly in relation to the primary wave to be canceled by it. It describes the difficulties in the precise generation of the secondary wave according to the principle of active damping and deals exclusively with methods and devices for obtaining the control signal for the sound source of the secondary wave. The basic structure used in the aforementioned publication is shown in Figs. 12 and 13 of the present description. One of the difficulties mentioned in the cited document is the reaction of the sound from the secondary source (( 3 ) in the figures) to the microphone for the primary wave (( 1 ) in the figures), and the known teaching is accordingly primarily concerned with the problem of avoiding this reaction by subtracting an appropriately formed alternating voltage from the output signal of this microphone.

It is one of the main advantages of the present invention that this problem does not occur with the muffler according to the invention, since the sound wave of the loudspeaker does not have to be subtracted from the control microphone of the active subsystem according to the invention, so that the complicated preliminary methods and described in the cited document Devices for avoiding the reaction of the secondary source on the control microphone are not required and the known teaching cannot suggest the invention. In the cited document, a second microphone (( 4 ) in Fig. 13 of the present description) is also used to improve the signal processing from the first control microphone (( 1 ) in Fig. 13). This so-called backward regulation is generally the state of the art in control technology and is not in itself the subject of the silencer according to the invention. What is important for all control engineering tasks is how the signal of the reverse control is processed and introduced into the control loop. The implementation of the reverse control is completely different from one another in the above-mentioned laid-open publication and on the other hand in the muffler according to the invention in accordance with the fundamentally different task positions. The use of a backward regulation also in the aforementioned publication does not therefore anticipate the inventive concept.

As this publication confirms, the principle of active noise reduction and are also based on the problems of this active sound attenuation that the extinguishing secondary wave "must be generated exactly". This is exactly what is necessary with the silencer according to the invention avoided by use of a passive subsystem that already has a certain Soundproofing generated and which by an active according to the invention Subsystem is influenced so that this sound absorption is improved becomes. There is no reference to this disclosure on a passive sound absorber. This already shows that it cannot suggest the present inventive concept.

As outlined below, most of these problems are related to submitted invention avoided from the approach.

Finally, the state of the art should refer to the literature [15] are received, in which a silencer is described, the there is called "hybrid". In fact, it’s a consecutive Arrangement in the direction of sound propagation of a purely passive silencer and an active silencer. Both silencers are in their Frequency curves are only tuned so that the passage of sound both silencers together should result in a favorable frequency curve. A Interaction of a passive subsystem of an absorber liner with one active subsystem of this lining in the same silencer as in the submitted invention does not take place. While the effect of the two  Muffler sections is additive in the mentioned literature, the Effect of the two subsystems in the silencer according to the invention multiplicative. The literature reference has only that with the presented invention Endeavor to improve sound absorption and the word "hybrid" common.

5. Task and inventive concept

The object of the invention is thus the absorber lining of a passive subsystem and an electroacoustic to combine the active subsystem so that both interact compared to a significant improvement in sound absorption Provide dampers of known design.

To solve this problem, a hybrid silencer consists of a passive absorber lining (passive subsystem) and an active electroacoustic system (active subsystem), which is characterized in that the electroacoustic active subsystem acts on the combined passive absorber in such a way that in the hybrid absorbers thus formed, the acoustic impedance of the channel-side surface of the absorber lining results in an absorber function U, which in a desired frequency range reaches or approximates the value of the absorber function U _{wp of} the winding point between the first and the second silencer mode (main claim).

Advantageous embodiments are in the subclaims specified.

According to the invention, the absorber function U should therefore come sufficiently close to the value U _{wp of} the absorber function in the upper winding point of the lowest damped mode, ie between the first and second damper modes, over a wide frequency range, especially at low frequencies. "Sufficient" should be measured by a significant improvement in sound absorption compared to dampers of known design.

The following explanations are intended to implement this Illustrate the inventive concept by way of example. It will become clear that it is different implementation options. In it and in the justified Possibility of inexpensive solutions is seen as an advantage of the invention. Because the idea of the invention because of the flexibility of its implementation also through a larger number of examples cannot be fully exhausted more emphasis is placed on the executions, the single-minded Describe the realization in individual steps. The examples given here are not intended to limit the scope of the invention.

6. Realization of the idea of the invention 6.1 Passive subsystem

The realization of the inventive idea is demonstrated by way of example on a rectangular gap channel (see Figs. 1 and 2). It is not difficult for the person skilled in the art to transfer to other channel forms.

6.1.1 Series connection of the passive and active subsystem

Furthermore, the implementation of the hybrid absorber lining according to the invention in the series connection according to Fig. 4 is first discussed.

For the dimensioning of a muffler according to the invention, it is initially helpful and important that the acoustic effectiveness of the active subsystem can be described simply by an acoustic impedance Z _b on the surface of the sounder, with which the passive subsystem ( 1 ) in Fig. 4 on the other Back is complete.

The passive subsystem is a linear acoustic four-pole. According to the rules of the four-pole calculation (see also [1, Eq. (4.7)], the terminating impedance Z _b , with which the absorber function U _{wp is} achieved on the front of the hybrid absorber, results from the formula

Z _{h is} the front impedance of the passive subsystem alone if it is sound-proof on its rear side (idle case), and Z _w is the front impedance of the passive subsystem alone if it is soundproofed on its rear side (short circuit -Case).

As an example, we consider a passive subsystem consisting of a layer of thickness d of a porous absorber material, which is covered on its front by a thin cover layer. The porous absorber material is described by the flow resistance R = Ξ · d / Z₀ (standardized with Z₀) with the length-related flow resistance Ξ of the porous absorber material. It is known how the characteristic propagation constant Γ _an and the characteristic (standardized with Z₀) characteristic impedance Z _on the porous absorber material can be calculated from this (see for example [2] or [17]).

The cover layer with the series impedance Z _s consists of an acoustic parallel connection of a friction resistance R _s with a mass reactance of a mass m _s related to the area. Then the series impedance Z _{s of} the cover layer follows:

with the air density ρ₀. If you choose the resistance term R _s large against the reactance term (2nd term in the denominator), then Z _{s describes} the series impedance of a cover layer consisting of a (mass) foil. Conversely, if one chooses the reactance term large against R _s , then one has a pure friction resistance, as is represented, for example, by woven and non-woven fabrics. If the cover layer is applied tightly to the porous absorber layer, the reactance of the elastic cushioning through the porous layer must still be connected in parallel. Since this is only a demonstration of the procedure, this further complication is dispensed with.

The in Eq. (7) required impedances Z _h and Z _w then result from:

The terminating impedance Z _b required according to the invention is thus known for the selected example of the passive subsystem.

Fig. 14 shows an example of the so-called "locus" of the terminating impedance Z _b in the complex plane with running frequency variables f · h in Hz · m. The points drawn in the curve lie at thirds intervals of these variables. This illustration applies to a simple porous absorber layer (without top layer) with the parameters d / h = 0.5 and R = 0.5. The curve is composed of arcs, which are known to be simulated by resonance systems (see below under "Active subsystem"). At the circular arc to the right of the imaginary axis, sound energy goes into the active subsystem; here it basically represents an electronically controlled passive absorber. However, since this circular arc is run counterclockwise with increasing frequency, it cannot be realized by passive components, because according to a fundamental theorem of circuit theory, the loci curves of passive circuits are always run clockwise. This should be pointed out because the word "active" in the active subsystem is not necessarily (albeit usually) to be equated with acoustic energy output. On the curve parts to the left of the imaginary axis, the active subsystem must actually feed sound energy into the back of the passive subsystem.

For a different structure of the passive subsystem from a porous absorber layer with d / h = 0.5 and R = 1.0, which is covered with a dense film with the mass parameter m _p = m _s / (ρ₀d) = 1, 0, Fig. 15 shows the associated locus of the termination impedance Z _b . The locus here consists of a circular arc of a single-circuit resonance system at low frequencies and approximately the vertical straight line of a spring reactance at high frequencies. The active subsystem must deliver energy at all frequencies.

In the further example of Fig. 16, the locus of the termination impedance for a passive subsystem is plotted from a porous absorber layer with the parameters d / h = 0.5 and R = 3.5, which in Fig. 10 is a passive absorber lining with broadband -Character is shown. The shape corresponds to an attenuated resonance circuit, whose impedance contracts at high frequencies to a convergence point near Z _b = -1.

In general, the examples shown have in common with all differences in shape that they start at low frequencies at the value Z _b = - (R + R _s ) and end near Z _b = -1 at high frequencies.

The use of a porous absorber layer in the passive absorber is not inevitable (although, as will be shown below, it is advantageous) for the realization of the inventive idea. In the calculation, a porous absorber layer can be replaced by an air layer of the same thickness by using for it in the formulas Γ _an → j and Z _an → 1. Fig. 17 with the locus of the terminating impedance Z _b according to the invention is an example of an air layer with d / h = 0.5, which has a dense fleece on its front with the series resistance R _s = 2 and the mass parameter m _p = 1 is covered. As you can see, this locus has a rather complicated shape. However, this is not intended to express the fact that, for technical reasons, it may not also be expedient to construct the passive subsystem of the absorber lining according to the invention without a porous absorber layer, only from foils, membranes and frictional resistances.

The structure of the passive subsystem according to the invention is not limited to the exemplary layer made of a porous absorber with a cover layer. [16, par. 4.4] describes a method for determining the front-side impedance of an M-layered absorber, whereby individual layers can also be air layers and the layers can also have cover layers. This very general procedure can be used for the determination of Z _h and Z _w in Eq. (7) by setting the reflection factor r _{M of} the layering on the back to r _M = 1 (sound-hard) or r _M = -1 (sound-soft). This shows how the task of determining the termination impedance Z _{b is} solved as a subtask in the implementation of the invention for practically every structure of the passive subsystem.

The determination of the termination impedance according to the invention is also not limited to the calculation methods exemplified here of an analytical-numerical description of the passive subsystem using dimensions and material data, although this route is preferable since it is the best for dimensioning the active subsystem (see below) Provides clues. One can also build a passive absorber intended for the application of the invention directly and measure its input impedance (once with a sound-proof termination for Z _h and then with a sound-proof termination for Z _w ) as a function of frequency using acoustic impedance measurement methods which are known in the literature have been described several times and are also standardized. To create a formula-dependent frequency dependency, one can then use the measured values according to Eq. (7) subject the formed impedance Z _{b to} a numerical regression using known methods. When selecting the functions according to which the regression is carried out, the structure of the active subsystem can expediently be taken into account (see below).

This shows that all passive subsystems whose input impedance can be calculated and / or measured can be included in the determination of the terminating impedance Z _b according to the invention.

In the construction of an absorber lining according to the invention, the determination of the terminating impedance Z _b according to the invention is an intermediate step for the design of the active subsystem. Z _b is the acoustic input impedance of this active subsystem connected in series. The essential part of the active subsystem is the acoustic impedance Z _{m of} the loudspeaker diaphragm, since only it can be influenced by electronic control. In technical applications of the invention, the terminating impedance Z _b generally still contains acoustic components which cannot be influenced electronically. Such a portion is provided, for example, by the "loudspeaker box" with which the loudspeaker is bordered on the back. Such a loudspeaker box is usually necessary, firstly to avoid that the sound of the loudspeaker vibration is emitted to the outside, but also to avoid that the pressure of the channel flow presses on the membrane.

Fig. 18 shows schematically a hybrid absorber lining according to the invention including a loudspeaker box. In many cases, the loudspeaker box is partially filled with sound absorption material in order to avoid disturbing resonances of the membrane mass with the spring stiffness of the air cushion in the box (however, it can also be advantageous to use this resonance in a targeted manner to create one of the resonance loops of the local curve from Z _b acoustically!). In these cases, the loudspeaker box is again to be described as a layered absorber with a reverberant termination. Now, almost without exception, speaker membranes are incompressible and thin compared to the sound wavelength (at least in the frequency range of interest here). The terminating impedance Z _b therefore arises from a series connection of the sought membrane impedance Z _m with the input impedance Z _{e of} the box, which is measured behind the membrane: Z _m = Z _b -Z _e . The termination impedance Z _b is still according to Eq. (7) determined. The input impedance Z _e is obtained by using the calculation method for the input impedance of stratified absorbers described in [16, Section 4.4]. This subtask can thus be solved in the same generality as the determination of Z _b .

Fig. 19 shows the locus curve of the membrane impedance Z _m to be set according to the invention of a structure according to Fig. 18, the air gap in front of the loudspeaker (for the meaning see below) here being 10% of the layer thickness in front of the loudspeaker and the loudspeaker box 80% with absorber material is filled. The entire absorber lining is covered on its front side with a film with mass parameters m _p = 2. In this figure, the variable f · h (points on the curve again at intervals of thirds) runs through the value range from approx. 4 [Hz · m] to 100 [Hz · m].

Both for answering the question with which approximations one can work with in order to "approximate the turn point" according to the invention, and for the design of the active subsystem, it is extremely helpful to use the low frequencies of interest in the [16, Eq. (4.16a), (4.16b) and Eq. (6.78)] given approximations for the absorber parameters Γ _an , Z _an and for their product Γ _an · Z _an and their quotient Γ _an / Z can be derived _from analytical approximations for the membrane impedance Z _m according to the invention:

If the last break is omitted, the terminating impedance Z _b arises. It is assumed that the passive absorber is layered in front of the loudspeaker with M absorber layers (including possible air layers) of thicknesses d _i with length-related flow resistances Ξ _i and cover layers with series impedances Z _{s, i} . Likewise, in the loudspeaker _box with the total depth d _{box there is} a number of M _box absorber layers (including any air layers) with the thicknesses d _i . Since the mass reactance of cover layers is usually negligible at low frequencies, Z _{s, i} → R _{s, i go} against the frictional resistance of the cover layers. Then Z _b becomes a linear polynomial in k₀h and in Z _m there is a reactance with the sign of a mass reactance and the frequency dependence of a spring reactance. The electronic synthesis of this approximation in the active subsystem is very simple. This approximation is used below for examples of damping curves for silencers according to the invention.

6.1.2 Parallel connection of the passive and active subsystem

In the case of the hybrid absorber lining in series connection according to Fig. 4, the active subsystem supports the passive absorber by generating an appropriate terminating impedance Z _b .

However, the primary task of the invention is to generate an absorber function U = k₀h · G = U _wp on the front of the absorber lining. This is proportional to the effective wall admittance G. It is now part of the basic experience of acoustics that with flat structured absorber surfaces the acoustically effective admittance is the area mean of the admittances of the individual partial areas, as long as the dimensions of the partial areas remain small to the sound wavelength (typically less than about a quarter sound wavelength). This condition creates no problems at the low frequencies that are of interest here.

So if F _{p is} the area of the passive part of the absorber lining with a wall admittance G _p on the surface and F _{a is} the area of the active part with an admittance G _a on the surface, then the invention must be:

The admittance G _{a of} the active part is thus according to the invention

certainly.

The admittance G _p for any layered passive absorber can then be determined using the method described above. Furthermore, the terminating impedance Z _b and / or the membrane impedance Z _{m of} the active part, which in turn can contain a layered absorber between the channel-side surface and the loudspeaker membrane or can be terminated with a loudspeaker box, can be determined by the methods there.

6.1.3. Mixed types of series connection and parallel connection of the passive and the active subsystem

In a further embodiment of the invention, the hydride Absorber lining made of mixed types of series connection and parallel connection of the passive subsystem with the active subsystem.

The two basic types are spatially arranged next to each other in a simple but complex apparatus as shown in Fig. 20. Of greater interest is the embodiment of the invention shown in Fig. 21, where an active subsystem works both in series and in parallel with a passive subsystem. This makes it possible to implement favorable frequency curves for the transfer function of the signal former.

A preferred mixed form of the hydride absorber according to the invention is shown schematically in Fig. 22. With her, the front of the loudspeaker works in series on a passive absorber, and the back creates a parallel admittance G _a in parallel. This is possible because the parallel admittance G _a must have the character of a passive admittance, but the terminating impedance Z _b , as the examples above show, often has the character of a negative passive impedance. This change of sign occurs automatically when you use the front and back of the speaker diaphragm as an active signal generator. This embodiment of the invention can also save the space required for a loudspeaker box, and the pressure equalization of the flow pressure via the loudspeaker is also given.

In the mixed forms, the control signal for the active subsystem can vary as appropriate either in the part of the series connection or in the part of the Parallel connection can be recorded by the signal pick-up (microphone).

7. Active subsystem

The active subsystem with the schematic structure of a possible embodiment of the invention according to Fig. 3 has the task of generating a prescribed acoustic impedance. This impedance is the ratio of the sound pressure on the vibrating surface (membrane) of the signal generator to the speed of this surface averaged over the surface.

Here, an important advantage of the invention compared to the task of active sound insulation has already been introduced, namely that the precise vibration distribution of the membrane is not important. Differences in the oscillation amplitude and phase can namely compensate for each other in the upstream passive absorber (in series connection) or in the upstream chamber (in parallel connection) across the cross-section, so that constant wall impedances arise on the channel-side surface of the hybrid absorber. An air gap between the membrane and the passive absorber, as indicated in FIGS. 4, 18, 20, 21 and 22, is particularly beneficial for such a transverse compensation of the sound field.

Another, compared to the task of active noise control, significant advantage of the invention is that the required control signal detected by the signal pickup (No. (1) in Fig. 3) is not the sound pressure without the contribution of the active subsystem, but the actual sound pressure, which is a superposition of the sound pressure of the noise to be reduced, which penetrates from the front through the passive absorber, and the sound pressure of the loudspeaker of the active system. The above-described difficulties of control in the case of active noise protection, namely having to first eliminate the contribution of the active system from the control signal, are completely eliminated in the present invention from the approach of the inventive concept.

Fig. 23 shows a basic diagram of the active subsystem in a basic form for realizing the invention . A microphone ( 1 ) picks up the sound pressure p in front of the loudspeaker ( 5 ). Together with a microphone amplifier ( 2 ) that may be required, the pressure p is transmitted into an electrical voltage u 1. At the other end of the signal chain, a loudspeaker ( 5 ) produces a sound velocity v of its membrane by driving the loudspeaker via a power amplifier ( 4 ) with the electrical voltage u₂. The signal former ( 3 ) then serves to transform the microphone signal according to the invention in order to achieve the required acoustic termination impedance Z _b = p / v. The subdivision of the passive subsystem shown is a functional subdivision rather than an enumeration of necessary components. The required amplifications can also be easily integrated into the signal former. The function of the signal shaper according to the invention becomes particularly clear if one assumes that both on the microphone side the relationship between p and u 1 is linear and regardless of the frequency and on the speaker side there is such a relationship between v and u 2. Then the transmission factor u 1 / u 2 to be achieved by the signal shaper is, apart from a frequency-independent factor, equal to the terminating impedance Z _b according to the invention or the membrane impedance Z _m .

The frequency-independent linear relationship assumed here between p and u ₁ is no problem with the usual microphone technology here frequency and amplitude range of interest. In contrast, speakers have usually a frequency response of the acoustic vibration at a constant electrical control. The representation and calculation of this frequency response from the constructive data of the loudspeaker with the standard methods Electroacoustics possible (see for example [18,19]). Methods also belong to reduce this frequency response to the prior art. It must therefore will not be discussed here in more detail. In connection with the invention it is sufficient if one considers the remaining frequency dependency of the speaker the frequency dependence of its to be achieved by the signal former Includes transfer function (as multiplication of transfer factors or by adding transfer dimensions, which everyone knows for themselves are); it is then the "combined" transmission factor of the Signal converter spoken.

This inclusion of the speaker frequency response in the combined Transfer factor of the signal shaper according to the invention makes the application the invention largely independent of the selected type of speaker (electrodynamic, electrostatic or magnetodynamic loudspeaker). The selection can be made according to aspects of the transmission range of the  Loudspeaker, by cost, by operational security and by simple Realization of this combined transmission factor.

The realization of this combined electrical transmission factor is then the last subtask to apply the invention. The function of the Signalformers is in the considered basic form of implementation of the invention in essentially that of a frequency-dependent electrical filter. The filter characteristic required according to the invention (frequency curve of the Transfer factor by amount and phase) is by applying the known solution methods for the subtasks known.

If one assumes that with today's well-known digital-electronic Methods practically every transfer function can be realized at reasonable costs is which can be described mathematically (with somewhat continuous functions), then this last sub-task for the application of the invention is also due to look at known techniques in a relaxed manner.

Nevertheless, there is a cost interest in keeping the frequency response of the combined transmission factor as simple as possible by suitable selection and combination of the passive subsystem and the loudspeaker. As the local curves of the terminating impedance Z _b and the membrane impedance Z _m shown above as an example, it is an advantage of the invention that there is scope for design here. Such clever combinations make it possible to also implement the combined transmission factor according to the invention in electronic analog technology. A powerful electronic circuit technology is available for this purpose, so that there is no need to go into this here either.

Realizations of the invention are also of some interest combined transmission factor through electronic "hybrid technology", that is by combining analog with digital components, in particular in the self-adapting active subsystems with which - as below is still set out - changes in operating temperature and / or Flow speed in the silencer can be "tracked".

As has already been repeatedly noted above, an advantage of the hydride absorber according to the invention is that, by suitable selection of the components of the passive absorber, the locus of the terminating impedance Z _b can be given shapes which can be implemented by simple electronic circuits in the active subsystem.

Another useful embodiment of the invention is described with reference to Fig. 24. The Fig. 4, 5, 18, 20, 21, 22 show each an inventive hybrid absorber element. To line a silencer, several of these elements are arranged in the direction of the channel axis (direction of sound propagation) one behind the other in the basic form of the invention until a length of the silencer is produced which provides the required transmission loss. Fig. 24 shows schematically that in this embodiment of the invention, neither the control microphone nor the signal former need to be repeated for each element. This is because there are both the decrease in the sound pressure level between adjacent hybrid absorber elements of width a (namely 8.68 · Γ′a dB) and the phase shift Φ (namely 180 · Γ ′) over the propagation constant Γ = Γ ′ + jΓ ′ ′ in the damper channel. 'A / π degrees) known. In Fig. 24, these level and phase changes in the excitation of successive loudspeakers are simulated in an electrical chain circuit.

In a further embodiment of the invention, use is made of the fact that Speakers are reciprocal converters; that is, they are not just for Sound generation useful, but can also be operated as a sound pickup will. When applying this property to the invention, the Speakers of the active subsystem in "switching breaks" as sound pickups switched, which measures the sound pressure, which from the damper channel through the passive absorber penetrates. Periods for these switching breaks range from one to a few seconds. The control microphone can thus be omitted; it is through replaced a timed switch.

In the previous embodiments of the active subsystem according to the invention, its implementation is essentially an electronic control task, which incidentally is seen as an advantage of the invention over the methods of active noise protection, since these basically represent a control task. In an embodiment of the invention which may be expedient, the function of the active subsystem according to the invention can also be conceived as a control task. The principle is explained using the schematic Fig. 25. In addition to the basic equipment according to Fig. 3, the sound velocity v of the membrane is recorded by a structure-borne noise sensor ( 4 ) placed on the loudspeaker membrane. The signal shaper is thus controlled using control technology methods known per se so that the quotient of the pressure p recorded by the microphone and the sound velocity v recorded by the structure-borne sound sensor forms the terminating impedance Z _b required according to the invention. This configuration as a control task can be expedient to compensate for changes in the properties of the passive subsystem that are difficult to predict, for example due to fluctuations in the production of the passive absorber, whether due to temperature profiles in the latter that are difficult to predict or due to changes during the operation of the silencer, for example due to dirt deposits.

In a further control-technical embodiment of the invention, the primary target variable of the silencer, namely the drop in sound level over a distance Δx of the channel, can also be used as a control variable for optimizing the termination impedance according to the invention. For this purpose, according to the diagram in Fig. 26, the sound pressure is measured at two microphone locations located one behind the other at a suitable distance, and the amount and phase of the loudspeaker oscillation are regulated so that this drop in level is maximized.

8. Side effects and other refinements

For the technical applicability of an invention - and thus for its Invention value - is decisive, as is the application-related deviation from standard situations to the desired performance of the invention - here the achievable sound absorption - impact.

A first tacitly accepted standard situation is the existence of a predominantly symmetrical sound field, which, coming from the noise source, falls on the silencer (symmetrical means: in-phase vibration of the sound field on both sides of the symmetry plane of the damper channel). In fact, this is usually the case. However, special cases of the sound source and / or the duct in front of the silencer are also conceivable, where the sound field vibration is predominantly antisymmetric (in opposite phase on both sides of the symmetry plane). Then the determination equation (6) applies instead of the determination equation (5). This is to be taken into account in the design of the muffler according to the invention with a rectangular cross-section simply by using the value U _wp = 3.71944 + j · 1.89528 for the absorber function U at the winding point according to [5].

A second standard assumption is that the channel-side surface of the Absorber lining is homogeneous, that means without defects and structures. By building the absorber lining according to the invention from hybrid Absorber elements, especially when using the parallel connection or  mixed arrangements, surface structures arise. Now it is known that such imperfections produce higher modes at most. But since this is stronger are damped as the lowest damped mode for which the invention Muffler is designed, can have an adverse impact will.

Silencers in technical systems are often used at high operating temperatures. These change the acoustic properties of the sound-conducting medium in the damper duct. However, it should now be noted that the design steps presented above have been described in dimensionless form. They therefore apply to any gaseous medium, in particular also to a medium that has been changed from normal air by an increased temperature and / or composition (flue gases!). These operating conditions only manifest themselves when you use numerical values for the material constants that occur. These are the speed of sound c₀ in k₀ = 2πf / c₀, the density ρ₀ in Z₀ = ρ₀c₀, furthermore the viscosity and the adiabatic exponent in Ξ and thus in Γ _an and Z _an . The influence of temperature and gas composition on these quantities is known from thermodynamics. It can therefore be included in the determination of the terminating impedance Z _b if one wishes to design a silencer according to the invention for a specific operating temperature and / or gas composition. If operating temperatures change frequently and significantly during operation, it may be necessary to provide a self-adaptive active subsystem rather than operating with a specific operating temperature. As a rule, this will make digital or hybrid electronic control technology necessary for the signal former, in which the consideration of an operating temperature recorded by means of a thermometer does not pose any fundamental problems.

The consideration of a flow with a is a little more complicated Flow velocity V₀ and a Mach number M₀ = V₀ / c₀. Flow acoustically, the flow superimposition can be completely grasped by wherever k₀ occurs, this is replaced by:

k₀ → k₀w (13)

with the so-called "flow factor"

Although the theoretical-acoustic literature also recommends a change in the boundary condition (from rapid adaptation to elongation adaptation to boundary surfaces), which would lead to the further replacement G _w → G _w · w, comparisons between measurement and calculation show one better agreement if one does not make this additional replacement. With this, Eq. (4) about in:

for z now instead of Eq. (5) the equation of determination applies:

z · tan (z) = jU · w (16)

and, accordingly, U → U · w is replaced in the other determination equations given above. Due to the occurrence of the flow factor w in the determination equations, the winding point U _wp (M₀) is also shifted depending on the Mach number. With the method described in [5], the associated winding point can be numerically determined for a given Mach number M₀. M₀ is set positive if sound and flow in the silencer have the same direction, and M₀ is set negative if they are directed in opposite directions.

When applied to the muffler according to the invention, the consideration of a flow superimposition essentially means an exchange of the constant U _wp . If the flow velocity changes frequently and strongly during the operation of the silencer, an expedient embodiment of the invention consists in introducing this change in the constant U _wp into a self-adapting active subsystem using known control engineering methods, using the display of a flow velocity -Measuring device. With simple frequency responses of the combined transmission factor, this can be accomplished in electrical analog technology, with more complex frequency dependencies, regulation in hybrid or in digital technology can be indicated.

9. Technical progress of the invention

The technical progress of the invention of the hybrid silencer compared to the state of both passive damper technology and (and compared to the active damper technology should be briefly presented.  There will also be further details of a preferred embodiment of the invention.

First, the importance of the task of improving the sound attenuation of silencers in technical systems is explained using a practical example. A flue gas silencer in a link design according to Fig. 2 for a medium-sized thermal power plant with an optimized design according to Fig. 10 had a length L = 7 m with a total width and height of 7 × 7 m². The gate thicknesses were D = 0.2 m, the gap widths H = 0.1 m, i.e. d / h = 2. The absorber filling had a flow resistance r = 1.7. From Fig. 10 one can then read an attenuation D _h = 0.1 dB at f = 100 Hz, i.e. at f · h = 5 Hz · m. At this frequency, the silencer has a total attenuation of D _d = 14 dB. At the frequency f = 50 Hz, at which noise protection requirements were also made, the total attenuation was only D _d = 3.5 dB. A target price per square meter of backdrop area results in a silencer price of around DM 0.5 million. The average flow velocity in the backdrop columns was V₀ = 50 m / s. This results in a drag coefficient (the ratio of pressure loss to dynamic pressure of the flow) of ξ = 2.5. With this information it can be calculated that the flow loss in the silencer, which has to be provided by the blowers in order to push the air through the silencer, is around 3000 kW; with an efficiency of 80% of the blowers, an electrical output of around 3800 kW is applied. With an electricity tariff of 0.10 DM / kWh, the silencer causes annual operating costs of 3.33 million DM / a due to its pressure loss, i.e. many times its purchase value. Here, a normal temperature T₀ was still calculated for a gas flow. At a higher operating temperature of T Kelvin, the power loss increases by a factor of (T / T₀) ^2.5 for the same mass throughput, i.e. at an operating temperature of around 300 degrees Celsius by around a factor of 5.8, resulting in annual operating costs of around 19 .3 million DM / a leads.

Assuming that the overall cross section of the muffler is retained, a muffler according to the invention, which is designed for D _h = 9.5 dB (see damping examples below) with an exhibition ratio d / h = 0.5, could not only dampen at f = 50 Hz to D _d = 15 dB and at f = 100 Hz to D _d = 25 dB, furthermore the flow velocity could be halved to V₀ = 25 m / s and the drag coefficient would decrease to ξ = 0.35. This would reduce the operating costs (at normal temperature) to around 62 TDM / a, and in hot operation to around 360 TDM / a. Even a multi-expensive silencer according to the invention would have (with significantly better sound insulation!) Refinanced its higher investment costs after a short time from the savings in operating costs (apart from the fact that the fans would be cheaper because of the lower power required!).

Fig. 27 shows an example of an application of the invention as a solid curve, the damping for a hybrid absorber consisting of a mineral fiber layer with the parameters d / h = 0.5 and R = 0.5 and an active subsystem, which is a linear approximation the Eq. (7) generated in k₀h for the terminating impedance Z _b , whose exact locus for this example is shown in Fig. 14. For comparison, the damping curve with the passive absorber alone is shown in dashed lines. This example not only shows the high gain in attenuation at low frequencies by the muffler according to the invention, it also illustrates that such gains can already be achieved with simple approximations to the prescribed terminating impedance Z _b , and it suggests a further embodiment of the invention. At frequencies above approximately fh = 70 Hz · m, the approximation used for Z _b obviously no longer applies; here the absorber lining of the passive subsystem alone produces higher damping than the absorber lining with the hybrid absorber in the design used. In such cases, it makes sense to "switch off" the active subsystem at higher frequencies. This means that the termination impedance Z _{b should} then have large amounts. With a hybrid absorber connected in series and with an electrodynamic loudspeaker, this can be achieved simply by the signal former generating an electrical short circuit on the loudspeaker. Then the membrane is known to be "braked" electrodynamically.

Fig. 28 also shows the damping curves of a hybrid absorber (solid line) with a structure and an exact location curve for Z _b as in Fig. 15, whereby an approximation to the exact course was also used here, and (dashed) the damping curve with the passive absorber alone. This example is of interest for practical use because the front of the absorber layer is covered by a film (with m _p = 1), which means that the hybrid absorber can be effectively protected against possible dirt in the sewer flow.

Finally, the last Fig. 29 shows the corresponding damping curves for a passive absorber lining already dimensioned broadly according to Fig. 10, namely that described in Fig. 16. In contrast to the two aforementioned exemplary embodiments, the terminating impedance was not obtained from analytically derived approximations, but rather by interpolation of measured values, as was described above. Since the limitations of an analytical approximation are eliminated in such a procedure, the damping achieved comes closer to the optimum value.

In addition to the improvements to the The invention has further technical advantages over sound absorption the prior art, which are listed below.

The implementation of the active subsystem is an electronic control task instead of - as with the active silencer - a standard task. This will be an advantage seen because experience shows that tax tasks are easier to solve than Control tasks. Nevertheless, the active subsystem according to the invention technical supplements are added, which are a "descendant" of Allow changes to operational parameters.

The control variable in the invention is the total pressure at the measuring location and not - As with the active silencer - the sound pressure of the incident wave, the first recovered from the measured sound pressure using complex algorithms must become.

The application of the silencer according to the invention is - in contrast to the active silencer - regardless of the sound field distribution in the Damper channel. This is important because the sound pressure distribution in technical Investments can neither be reliably predicted nor kept constant. The silencer according to the invention can therefore be independent of the noise source, in which it is to be used, developed and dimensioned, while active silencers are adapted to the respective noise source have to.

The acoustic field impedance to which the loudspeaker of the active Subsystem works by appropriate choice of parameters of the passive Subsystems can be set in a wide range. This has the advantage that relatively "hard" speakers can be used. These are namely in the Usually cheaper and more robust in operation. In addition, when the hybrid absorbers in series connection the sound pressure level at the loudspeaker - and thus the required elongation of the loudspeaker diaphragm - by the upstream passive absorbers reduced.

There is an important advantage of the invention for technical application in that the components of the active subsystem are not - as in the active  Silencers - are directly exposed to the flow. The upstream passive absorber can be used as a thermal barrier in hot gas flows porous fiber absorbers are usually also good thermal insulation have. The upstream passive absorber can also in the invention Take over the function of a dirt filter.

The property is also important for use in flow channels porous passive absorber materials that they respond to turbulent pressure fluctuations have a rectifier effect. That means: the fields of turbulent Pressure fluctuations break down faster in an absorber layer than the fields the sound wave, because at the low frequencies that are mainly of interest here the characteristic diameter of the turbulent pressure fluctuations very much are smaller than the sound wavelengths. The one that occurs with the active silencer Problem of separating acoustic and turbulent pressure fluctuations from each other must largely be omitted in the invention.

Another advantage of the silencer according to the invention in comparison The active silencer is seen in the fact that in the event of a failure of the active Subsystem (during maintenance or malfunction) the passive absorber a certain can ensure remaining sound attenuation.

Another advantage of the muffler according to the invention over the active muffler is that the transition to other channel cross sections can essentially be taken into account by changing the constant U _wp and that - unlike with the active muffler - new algorithms for regulating and new arrangements of the loudspeakers do not develop Need to become.

After all, it's one when planning and sizing one Muffler according to the invention extremely helpful advantage that the individual design steps in standard tasks (such as impedance measurement, Determination of converter constants, design of control circuits etc.) can be disassembled, while with the active silencer practically only the complete System can be experimented.

On the other hand, the invention is not an obvious further development known technical teachings can be recognized only by the functions, which a porous absorber layer in a silencer according to the invention be allocated. Such a layer offers thermal protection for the active Subsystem; it is a dirt and turbulence filter; it lowers the Sound pressure level at the location of the active subsystem due to the internal one  Loss of propagation in the absorber material and thereby makes speakers smaller stroke possible; it takes on the task of a passive damper high frequencies; it ensures residual damping in the event of an active fault Subsystems; as a passive damper, it provides the damping of higher channel modes for sure; As a resistance rectifier, it accomplishes pressure equalization in Transverse direction with non-uniformly vibrating loudspeaker diaphragms.

In summary, the following are possible implementation options Invention:

Regarding the basic acoustic forms:

• - Series connection of the passive and the active element (see Fig. 4)
• - Parallel connection of the passive and the active element (see Fig. 5)
• - Mixed types of these (see Fig. 20-22)

Regarding the active subsystem consisting of signal transducer ( 1 ) signal former ( 2 ) and sound generator ( 3 ) as electro-acoustic variants with appropriate training of the signal transducer:

• - Control via microphone or loudspeaker microphone (see Fig. 3 and 23)
• - Control via structure-borne sound sensor with microphone (see Fig. 25)
• - Control via channel microphones (see Fig. 26)

The electro-acoustic variant of the control via microphone according to Fig. 3 is particularly emphasized, the quotient of the area mean of the speed of the loudspeaker membrane and the sound pressure in front of the latter forming an acoustic terminating impedance Z _b according to the invention, which in the case of the acoustic series connection of the passive absorber and the active subsystem according to equation (7) given above and in the case of acoustic parallel connection of the passive absorber and the active subsystem by a combination of the equations (12) and (7) given above.

10. Description of the pictures

Fig. 1: Rectangular silencer with clear duct width H = 2h and absorber lining of thickness d made of porous absorber material with length-related flow resistance Ξ, duct-side with a sound-permeable cover layer.

Fig. 2: Baffle silencer with baffles of thickness D = 2d and gap widths H = 2h.

Fig. 3: Schematic drawing of the active subsystem with sound pickup ( 1 ), signal former ( 2 ), loudspeaker ( 3 ).

Fig. 4: Series connection of the passive ( 1 ) and the active ( 2 ) subsystem.

Fig. 5: Parallel connection of the passive ( 1 ) and the active ( 2 ) subsystem.

Fig. 6: Solution of Eq. (5); 1. Damper mode.

Fig. 7: Solutions of Eq. (5); 2. Damper mode.

Fig. 8: Sound absorption in a rectangular silencer with a clear duct width H = 2h with an absorber lining, whose absorber function U has the value U = U _wp .

Fig. 9: Sound absorption in a rectangular silencer with a clear duct width H = 2h with optimal absorber lining, with hybrid absorber lining with passive absorber with thickness d made of porous absorber layer with flow resistance R = Ξd / Z₀ and with purely passive lining.

Fig. 10: Approximation for damping curves of rectangular silencers with a clear channel width H = 2h and absorber lining of thickness d made of porous absorber material with length-related flow resistance Ξ, optimized for broadband performance [1].

Fig. 11: Sound absorption with a passive absorber lining made of so-called "membrane absorbers", a resonator combination tuned to low frequencies.

Fig. 12: Schematic drawing of an active silencer with control microphone ( 1 ), signal shaper ( 2 ), anti-noise speakers ( 3 ).

Fig. 13: Schematic drawing of an active silencer with adaptive control; Control microphone ( 1 ), signal former ( 2 ), anti-noise speaker ( 3 ), correction microphone ( 4 ), controller ( 5 ).

Fig. 14: Locus of the terminating impedance Z _{b of} an absorber layer of thickness d with d / h = 0.5 with normalized flow resistance R = 0.5.

Fig. 15: Locus of the termination impedance Z _{b of} an absorber layer of thickness d with d / h = 0.5 with normalized flow resistance R = 1.0 covered with a foil with m _p = 1.

Fig. 16: Locus of the terminating impedance Z _{b of} an absorber layer of thickness d with d / h = 0.5 with normalized flow resistance R = 3.5 of a passive broadband absorber.

Fig. 17: Locus of the terminating impedance Z _{b of} an air layer of thickness d covered with d / h = 0.5 with a normalized flow resistance R _s = 2.0 and a foil with m _p = 1.0.

Fig. 18: Series connection of the passive ( 1 ) and the active ( 2 ) subsystem, with a loudspeaker box ( 3 ). The terminating impedance Z _b and the input impedance 05147 00070 552 001000280000000200012000285910503600040 0002004027511 00004 05028 of the box Z _{e are} indicated.

Fig. 19: Locus of the membrane impedance Z _m .

Fig. 20: Combination of a series connection ( 1 ) with a parallel connection ( 2 ) of the passive and the active subsystems.

Fig. 21: Combination of a series connection ( 1 ) with a parallel connection ( 2 ) with a common active subsystem.

Fig. 22: Combination of a series connection ( 1 ) with a parallel connection ( 2 ) with a common active subsystem using both diaphragm sides of the loudspeaker.

Fig. 23: Schematic drawing of the functional components of the active subsystem with sound pickup ( 1 ), microphone amplifier ( 2 ), signal former ( 3 ), power amplifier ( 4 ), loudspeaker ( 5 ).

Fig. 24: Hybrid absorbers ( 1 ) arranged one behind the other with an active subsystem consisting of a guide element ( 2 ) and a chain connection of attenuators and phase rotators to simulate the change in level and phase of the sound wave in the damper channel ( 4 ).

Fig. 25: Series connection of a passive ( 1 ) and an active ( 2 ) subsystem, which regulates the ratio of the sound pressure from the microphone ( 3 ) and the speed from the structure-borne noise sensor ( 4 ) to the required ratio of the final impedance.

Fig. 26: Series connection of a passive ( 1 ) and an active ( 2 ) subsystem, which regulates the difference in sound levels at two microphones ( 3 ) arranged in series in the damper channel ( 4 ) to a maximum value.

Fig. 27: Attenuation curves of a silencer with a passive absorber and a termination impedance according to Fig. 14; solid: hybrid absorber; dashed: purely passive absorber.

Fig. 28: Attenuation curves of a silencer with a passive absorber and a terminating impedance according to Fig. 15; solid: hybrid absorber; dashed: purely passive absorber.

Fig. 29: Attenuation curves of a silencer with a passive absorber and a termination impedance according to Fig. 16; solid: hybrid absorber; dashed: purely passive absorber.

11. Literature used

[1] MECHEL, F. P. "Silencer" in "Taschenbuch der Technische Akustik", Edit. M. Heckl, H.A. Müller Springer publishing house, Berlin, 1975

[2] MECHEL, F. P. "Sound absorption" in "Taschenbuch der Technische Akustik", Edit. M. Heckl, H.A. Müller Springer publishing house, Berlin, 1975

[3] MORSE, P.M.J. Acoust. Soc. Amer. 11 (1939) 205-210 "Transmission of sound inside pipes"

[4] CREMER, L. Acustica, Supplements 3 (1953) 249-263 "Theory of airborne noise reduction in a rectangular duct with a swallowing wall and the resulting highest damping measure "

[6] GERBER, O. Acustica, Supplements 2 (1953) 264-270 "Experimental investigations to realize the theoretically possible Maximum attenuation of sound propagation in a rectangular air duct with swallowing walls "

[7] SHORT, U. Acustica 21 (1969) 74-85 "Sound propagation in the canal with periodic wall structure"

[8] KÖLTZSCH, P. "Sound absorbing channels" in "Noise Control", Edit. W. Schirmer Verlag Tribüne, Berlin, 1971  

[9] ESCHE, V. "Ventilation systems and silencers" in "Taschenbuch Akustik", W. Fasold, W. Kraak, W. Schirmer Edit. VEB Verlag Technik, Berlin, 1984

[10] VDI guideline 2567 "Sound insulation through silencers"

[11] FUCHS, H., et al. "Sound absorbing component", DE-Pat. 34 12 432 (1984/1989) "Silencer box", DE-Pat. 35 04 208 (1985/1989)

[12] LUEG, P. US Pat. No. 20 43 416; (1934/1936) "Process of silencing sound oscillations" LUEG, P. DRP No. 6 55 508; (1933/1937) "Method for damping sound vibrations"

[13] GUICKING, D. Progr. Acoustics, DAGA ′89 (1989) 23-36 "Active noise protection - successes, problems and perspectives"

[14] GUICKING, D. "Active Noise and Vibration Control. Annotated Reference Bibliography" 3rd ed. 1988; III. Physics. Inst. Univ. Goettingen, 3400 Goettingen

[15] MUNJAL, M.L., ERIKSSON, L.J. J. Acoust. Soc. Amer. 86: 832-834 (1989) "Analysis of a hybrid noise control system for a duct"

[16] MECHEL, F. P. "Sound absorber", vol. I Hirzel Verl., Stuttgart, 1989

[17] MECHEL, F. P. Progr. Acoustics, DAGA ′84 (1984) 31-52 "Sound absorber"

[18] LENK, A. "Electromechanical systems. Vol. 1: Systems with concentrated parameters" 3rd edition, VEB Verl. Technik, Berlin, 1975

[19] REICHARDT, W. "Electroacoustics" Teubner Verlagsges., Leipzig, 1971  

[20] DE 34 25 450 A1

[21] DE 27 12 534 A1

Claims (20)

1. Hybrid silencer, consisting of a passive absorber lining and an active electroacoustic subsystem, characterized in that the electroacoustic active subsystem having a control microphone and a loudspeaker acts on the passive absorber combined with it in such a way that the acoustic impedance in the hybrid absorber thus formed the channel-side surface of the absorber lining results in an absorber function U, which in a desired frequency range reaches or approximates the value of the absorber function U _{wp of} the winding point between the first and the second silencer mode. 2. Hybrid muffler according to claim 1, characterized in that the absorber function U _wp at predominantly symmetrical sound fields in the sound damper channel lies at the winding point between the first and the second symmetrical silencer mode, and that the absorber function U _wp at predominantly antisymmetric sound fields in the sound damper channel Winding point lies between the first and the second antisymmetric silencer mode. 3. Hybrid silencer according to claim 1 or 2, characterized characterized in that the passive absorber from known com components such as porous absorber materials, air layers, membranes, Foils, flexible plates, cover layers with flow resistance such as fleeces, felts, fabrics, perforated sheets, etc., also from acoustic resonators of a known type, such as Helmholtz-Resona gates, plate and foil resonators or combinations of sol components. 4. Hybrid silencer according to at least one of the types Proverbs 1 to 3, characterized in that the hybrid absorber the lining of silencer ducts with a rectangular, round one  or annular cross section or that it is so-called Forms the backdrop in such channels. 5. Hybrid silencer according to at least one of claims 1 to 4, characterized in that the active subsystem from a control microphone, optionally with a microphone amplifier, for measuring the sound pressure in front of the loudspeaker, from this loudspeaker, optionally with a power amplifier, and consists of an interposed electronic signal former which controls the loudspeaker in such a way that the quotient of the area mean value of the speed of the loudspeaker membrane and the sound pressure in front of it forms an acoustic termination impedance Z _b according to the invention. 6. Hybrid silencer according to one of claims 1 to 5, characterized in that the passive absorber and the active Subsystem are arranged in acoustic series connection. 7. Hybrid silencer according to at least one of the types Proverbs 1 to 6, characterized in that the passive absorber and the active subsystem arranged in parallel acoustic connection are. 8. Hybrid silencer according to at least one of the types Proverbs 1 to 7, characterized in that combinations of Series connection and parallel connection of the passive absorber and active subsystem.   9. Hybrid silencer according to at least one of the Claims 1 to 8, characterized in that the rapid on one side of the speaker diaphragm for the termination impedance in Series connection and the fast on the other side of the loudspeaker membrane used for the termination impedance in parallel becomes. 10. Hybrid silencer according to at least one of claims 1 to 9, characterized characterized that reciprocal speakers in the active subsystem be used and that at timed intervals the Loudspeaker for measuring the sound pressure on its membrane. 11. Hybrid silencer according to at least one of claims 1 to 10, characterized characterized in that in addition to the control microphone for the measurement the sound pressure in front of the loudspeaker diaphragm is a structure-borne noise is placed on the speaker diaphragm, which the Rapid measures the membrane, and that with known electronics control loops the quotient of sound pressure and fast on the termination impedance according to the invention is regulated. 12. Hybrid silencer according to at least one of claims 1 to 11, characterized characterized in that with hybrid absorber elements, which in Sound propagation direction are arranged one behind the other the control of the speakers of the following elements through electronic attenuators and phase shifters in the Way is derived that they the level drop and the phase Ge speed of the lowest damped damper mode in the damper duct corresponds. 13. Hybrid silencer according to at least one of claims 1 to 12, characterized  characterized in that in consecutive places in the damper channel with microphones the drop in sound level is measured and that the terminating impedance by known electronic controllers the loudspeaker of the active subsystems used to a maxi the level drop is regulated. 14. Hybrid silencer according to at least one of claims 1 to 13, characterized characterized in that the operating temperature of the muffler is measured sen and that the parameters of the signal former of the active Subsys tems with known control techniques using this Temperature as an input variable can be controlled so that the acoustic impedance the channel-side surface of the absorber lining at different Operating temperatures are maintained. 15. Hybrid silencer according to one of claims 1 to 14, characterized in that the flow rate is measured in the silencer channel and that the parameters of the signal former of the active subsystem using control techniques known per se, using this flow rate as an input variable and the one valid at the respective flow rate Absorber function U _wp are regulated as a system variable in the winding point in such a way that the acoustic impedance of the channel-side surface of the absorber lining is maintained at different flow rates. 16. Hybrid silencer according to at least one of claims 1 to 15, characterized characterized in that the signal former of the active subsystem in elec tronic analog technology or in electronic digital technology or built in hybrid technology from analog and digital components is.   17. Hybrid silencer according to at least one of claims 1 to 16, characterized characterized in that the passive absorber as a heat-insulating layer is trained. 18. Hybrid silencer according to at least one of claims 1 to 17, characterized characterized in that the passive absorber through cover layers its front against the ingress of dirt and / or moisture activity is protected and that these cover layers in the acoustic Characteristics of the passive absorber and in the dimensioning of the signal former of the active subsystem are included. 19. Hybrid silencer according to at least one of claims 1 to 18, characterized characterized in that the length-specific flow resistances porous absorber layers and / or the flow resistances of Cover layers of the passive absorber are chosen so that in one an air gap arranged in front of the loudspeaker diaphragm equal to the sound pressure with an uneven vibration amplitude of the speaker over its surface. 20. Hybrid silencer according to at least one of claims 1 to 19, characterized characterized in that the passive absorber according to known per se Rules are dimensioned such that in the event of a failure of the active Subsystem of the silencer has a certain residual damping.