This action might not be possible to undo. Are you sure you want to continue?

**Haluk Erola) and Cem Meriçb)
**

(Received: 5 January 2009; Revised: 27 April 2009; Accepted: 28 April 2009)

One way of attenuating broad band noise that propagates through ducts or pipe systems is to use one or more reactive-type acoustic components, each of which is speciﬁcally designed for optimal performance at a particular frequency range. This paper uses the transfer matrix method for predicting the transmission loss (TL) of a system with resonators connected to the main duct and a side branch duct modiﬁed with a zero-mean-ﬂow expansion chamber. The method is applied to different conﬁgurations, and the numerical predictions are compared with the results obtained using ﬁnite element methods. Our results indicate that broadband resonators and a side branch duct with an expansion chamber, ﬁtted as a countermeasure for broadband noise, can offer signiﬁcant beneﬁts. © 2009 Institute of Noise Control Engineering.

Primary subject classiﬁcation: 37.6; Secondary subject classiﬁcation: 34

1

INTRODUCTION

In intake systems of internal combustion engines, side branch-type mufﬂers are often used, for example, for turbocharger noise control. Two types of noise generally occur in turbochargers—blowing or whooshing noise, and pulsation noise at frequencies of about 1000– 3000 Hz. The resulting noise can be very audible both inside and outside the vehicle. Recent advances in modeling and accurate performance prediction have led to the development of simulation methods for practical mufﬂer components in commercial design. Mufﬂer designers need simple, fast and accurate modeling tools, especially in the preliminary design evaluation stages. Finite element and boundary element methods are often used to provide valid results over a wide range of frequencies. However, these methods are time-consuming, they require highly trained personnel, and the commercial software is usually expensive. Therefore, plane wave-based models, such as the transfer matrix method, can quickly offer initial prototype solutions for mufﬂer designers. However, we note that the transfer matrix method has a limited frequency range. A resonator can be an effective acoustic attenuation device at low frequencies. Recently, Selamet and co-workers employed several approaches to examine

a)

b)

Mechanical Engineering, Istanbul Technical University, Gümüşsuyu, 34439, Istanbul TURKEY; email: erolha@itu.edu.tr. Mechanical Engineering, Istanbul Technical University, Gümüşsuyu, 34439, Istanbul TURKEY.

Noise Control Eng. J. 57 (5), Sept-Oct 2009

Helmholtz resonator applications1–4. The Helmholtz resonator is considered one of the most efﬁcient tuning methods and is widely used in vehicle exhaust and intake systems. Modiﬁcations have been made to improve the acoustic attenuation of the Helmholtz resonator. Birdsong and Radcliff proposed a smart Helmholtz resonator for use in the vehicle intake system5. In addition to a Helmholtz resonator, the Herschel-Quincke (HQ) tube is a simple implementation of the interference principle in an acoustic attenuation device. Considering the application of HQ tubes to systems such as compressors and the intake and exhaust systems of internal combustion engines, the effect of a mean ﬂow is not negligible. Selamet et al. modeled the HQ tube, removed many of the geometric restrictions associated with previous work and examined innovative new approaches6. Using a threedimensional model that took into account the curvature of the HQ tube, they determined that, with just a plane wave, the one-dimensional model (which approximates the tube as being straight) is sufﬁcient. Selamet and Radavich then examined a side branch expansion chamber in the context of an HQ tube7. They investigated how geometrical variation of the side branch and HQ tube inﬂuenced the resonant frequencies of the system. Another innovative use of the HQ tube by Selamet and Easwaran was to extend the previous analysis of the two-duct HQ tube without ﬂow8. They found closed form solutions for the TL and resonance locations for the n-duct conﬁgurations. They also found that the additional branches increased the frequency band and the amount of noise attenuation. The effects of ﬂow on the attenuation characteristics of different

476

B.11. with no mean ﬂow. were the ﬁrst to study the potential of HQ tubes for attenuating higher order modes in two-dimensional ducts12. If the duct diameter is less than half the wavelength. b1. t͒ and ͑L . lc2 and lc3. Thus. with zero mean ﬂow. t͒ p͑L. the system shown in Fig. Resonators. k is the wave number or propagation 477 . the TL of a system with two resonators connected to the main duct and a side branch duct modiﬁed with an expansion chamber.e. for optimal acoustic attenuation performance in the 1000– 3000 Hz frequency range. t͒ = p+ei͑t−kx͒ + p−ei͑t+kx͒ . t͒ = C D ͑0. The neck diameters and lengths are hn1. c1. t is the time. The diameter of each element is hhq. used the transfer matrix method to predict the TL of a mufﬂer system16. such publications only examined the effects of plane waves on the HQ tubes. One of the most popular references on the acoustics of ducts and mufﬂers is the work of Munjal22. t͒. There are two resonators connected to the main duct. 57 (5). however. e2 and f. and the numerical predictions are compared with results from ﬁnite element methods. e1. be considered plane waves. We discuss the limitations of the plane wave approach. we demonstrate a mufﬂer tuned for the intake system of an internal combustion engine that features a turbocharger. The ﬁrst suggestion for the combined use of HQ tubes and quarter-wave resonators appeared in work by Trochon17. p͑L . All calculations assume that a 1D model is valid. this means that the 1D model for wave propagation is valid. The rigid straight main duct has a constant cross section. 1 is divided into 10 elements: a. d. in this study. c2. This requires that the maximum duct diameters are sufﬁciently small for the investigated frequency range. The sound pressure ﬁeld inside the tube is p͑x. and A.10 and Zhichi et al. Plane wave propagation in a rigid straight pipe with length L can be described by its transmission (or four-pole) matrix. as follows: ͫ ͬ ͫ ͬͫ ͬ A B p͑0. the waves in the duct can Noise Control Eng. Torregrosa et al. ͑2͒ Figure 1 shows a system diagram. 1—System with two resonators connected to the main duct and a side branch duct modiﬁed with an expansion chamber. the TL of the system. The length of the main duct is lmd. The modiﬁed side branch duct connected to the main duct is also divided into 3 elements of lengths lc1. Chen and Hastings used a variation of the HQ tube to examine its ability to reduce ﬂuid-borne noise in power steering hydraulic transmission lines15. Hallez et al. are connected to the main duct. Peaks of high attenuation were seemingly reduced and shifted in the presence of ﬂow. The resonator cavity lengths and diameters are lhr1. ͑L. Furthermore. In more complex systems the effects of higher order modes on the HQ tube system become very important. ln1 and ln2. t͒ ͑1͒ 2 THEORY where p͑0 . The transfer matrix method is applied to different conﬁgurations. Strunk investigated a variation of the HQ tube for hydraulic systems used in off-road earth-moving vehicle ﬂuid power systems14. t͒ are the sound pressures and volume velocities at the input and the output. in our analytical model. ͑0 . and its diameter is hmd. extended the two-dimensional analysis of higher order modes to three dimensions.. is the angular frequency in radians per second. hhr1 and hhr2. hn2. This paper uses the transfer matrix method to predict. Another interesting implementation of the HQ concept is described by Graefenstein and Wenzel18. This study deals with the transfer matrix method for predicting. as shown in Fig. hard-walled circular ducts13. Fig. J. 1. i. hec and hhq. Sept-Oct 2009 where i = ͱ−1. The basic theory of the HQ tube and recent contributions to the understanding of the acoustical behavior of HQ ducts is available elsewhere19–21.HQ tube conﬁgurations have been investigated by Fuller and Beis9. However. Designs based on HQ tubes have been awarded many patents for reducing the exhaust noise of internal combustion engines and have been used in various practical problems. b2. the analytical model can be converted for direct connection scenarios. c3. t͒ . t͒. by setting the diameter and length of the neck. lhr2. Gerges et al. C and D are four-pole parameters. Brady et al.

Cc and Dc p4 p5 = Tc . for “element f. which can be expressed as the multiplication of the transfer matrices of each of the elements c1. t͒ = ͑p − p ͒e c + − it .” the relations between the sound pressures and the volume velocities are represented by four-pole parameters as is the transfer matrix of the rigid straight pipe.constant. 3 and 4 are the volume velocities. p2.” For “element a. For “element c.” the relations between the sound pressures and the volume velocities are represented by four-pole parameters as ͫͬ ͫͬ p8 p9 = Tf . 57 (5). 8 9 ͑14͒ T= ͫ ͬ A B = C D ΄ cos͑kL͒ S i sin͑kL͒ c i c sin͑kL͒ S cos͑kL͒ ΅ ͑9͒ where junction points 8 and 9 are the input and the output of “element f. Similarly. where p1. The description of the subsystems in terms of their four-pole parameters is very convenient. 9 ͑10͒ ͑0. at the second junction point. ͑3͒ ͑4͒ ͫͬͫ Aa Ba p0 = 0 Ca Da ͬͫ A18 B18 C18 D18 ͬͫ ͬͫ ͬ Af Bf Cf Df p9 . and the output pressure and volume velocity are p͑L. t͒ = ͑p+e−ikL + p−eikL͒eit . 3 6 ͑15͒ where junction points 3 and 6 are the input and the output of “element d. (3)–(6) where junction points 0 and 1 are the input and the output of “element a.” the relations between the sound pressures and the volume velocities are represented by four-pole parameters as Ac. Then the inlet pressure and volume velocity are p͑0. ͑L͒ ͑8͒ Then. Thus. because the output of one system is the input of the next. Bc.” respectively. Conditions of pressure equilibrium and conservation of volume velocity at the ﬁrst junction point yield p1 = p2 = p3 = p4 . Furthermore.” respectively. ͑16͒ ͑17͒ ͫͬ or 478 p9 p0 = TaT18Tf 0 9 ͫͬ 1 = 2 + 3 + 4 . as T c = T c1T c2T c3 = Equation (7) can be rearranged as ͫ ͬ and then p͑0͒ = ͑0͒ ΄ cos͑kL͒ S i sin͑kL͒ c i c sin͑kL͒ S cos͑kL͒ ΅ͫ ͫ ͬ Ac Bc . Here. from which it follows that the transmission matrix of a system that features multiple cascaded subsystems can be formulated in matrix form as ͫͬ ͫͬ p3 p6 = Td . and 1. J. S is the cross sectional area of the acoustic domains.” respectively. 2. ͑0͒ ͑7͒ where Tc is the transfer matrix of the side branch duct modiﬁed with an expansion chamber. 0 1 ͑11͒ where is the density. Sept-Oct 2009 . t͒ = . t͒ = ͑p+ + p−͒eit . for “element d. ͑18͒ Noise Control Eng.” the relations between the sound pressures and the volume velocities are represented by four-pole parameters as ͫͬ ͫͬ ͫͬ ͫͬ p0 p1 = Ta . 4 5 ͑12͒ ͫ ͬ p͑L͒ = ͑L͒ ΄ cos͑kL͒ S − i sin͑kL͒ c c − i sin͑kL͒ S − cos͑kL͒ ΅ ͫ ͬ ͬ p͑0͒ . Cc Dc ͑13͒ p͑L͒ . and x is the plane wave propagation axis. c2. c ͑5͒ ͑6͒ where T18 is the transfer matrix of the system between the inlet “element a” and the outlet “element f. from Eqns. p3 and p4 are the sound pressures. p5 = p6 = p7 = p8 . ͑p+e−ikL − p−eikL͒eit ͑L. and c3. and c is the speed of sound.

Finally. at junction point 5 is 2 = Chr1p10 . 7 and 8 are the volume velocities. Bc + Bd ͑42͒ Examining “element d. Thus. (38) and rearranging results in A dp 8 − A cp 8 + B d 6 = B c 5 . (12) the sound pressure and volume velocity at the junction point 4 can be written as p 4 = A cp 5 + B c 5 . ͑29͒ ͑30͒ Thus. From Eqn. and Thr2 = Tb2Te2 . Bc Bc Ahr2 giving ͑41͒ 6 = 4 = C cp 5 + D c 5 . p2 = Ahr1p10 . ͩ BcChr2 − Ahr2͑Ad − Ac͒ ͑Bc + Bd͒Ahr2 ͪ ͩ ͪ p8 + Bc 8 . ͑19͒ 7 = Chr2p11 . ͑32͒ ͑33͒ where p5. Ahr2 ͑34͒ 0 = C ap 1 + D a 1 . (25) and (29) give Ahr1p10 = Adp6 + Bd6 ͑43͒ 3 = C dp 6 + D d 6 . Then. e1 and e2 as Thr1 = Tb1Te1 . ͑36͒ ͑37͒ ͑38͒ Ahr Bhr . Noise Control Eng. (40) into Eqn.5 + 6 = 7 + 8 . the pressure and velocity relationships between junction point 7 (the upper of the second resonator) and junction point 11 (the end of the second resonator) are p7 = Ahr2p11 . ͑24͒ Eqns. Bc Bc ͑39͒ Thus. 57 (5). ͑27͒ ͑28͒ Substituting Eqn. The transfer matrix corresponding to resonators connected to the main duct can be expressed as the product of the transfer matrices of the elements b1. Substituting p7 = p8 into Eqn. b2. ͑25͒ ͑26͒ Substituting p5 = p6 = p8 into Eqn. Eqns. ͑35͒ Junction points 8 and 9 correspond to the input and output of “element f. Similarly. p 8 = A fp 9 + B f 9 . At junction point 0. Thus. (28) and (29) give A dp 6 + B d 6 = A cp 5 + B c 5 . (31) yields p8 = Ahr2p11 . J. ͑23͒ ͑22͒ The volume velocity 7 can be expressed in terms of the sound pressure p8 as 7 = Chr2 Ahr2 p8 . (30) yields 479 . 5. 5 = ͑40͒ Junction points 4 and 5 are the input and the output of the side branch duct modiﬁed with an expansion chamber.” respectively. ͑20͒ ͑21͒ The sound pressure at the end of the second resonator can be written as p11 = p8 . the sound pressure and volume velocity at junction point 2 (the upper part of the ﬁrst resonator) may be obtained from the sound pressure and volume velocity at junction point 10 (the end of the ﬁrst resonator). Ad − Ac Bd p8 + 6 . the volume velocity. the sound pressure and the volume velocity from Eqn. (26) and (28) into Eqn. 6. (11) give p 0 = A ap 1 + B a 1 . the sound pressure at the end of the ﬁrst resonator is p10 = Ad Bd p8 + 6 . p7 and p8 are the sound pressures. Sept-Oct 2009 Equation (43) may be rearranged further by substituting p8 = p6.” the pressure and velocity relationships between junction points 3 and 6 are p 3 = A dp 6 + B d 6 . (19) results in Chr2 Ad − Ac Bd p8 + 6 + 6 = p8 + 8 . Ahr1 Ahr1 ͑44͒ ͑31͒ Substituting Eqns. p6. and 5. Thr = Chr Dhr ͫ ͬ 8 = C fp 9 + D f 9 .

Transmission loss involves neither the source nor the radiation impedance. (44) into Eqn. Sept-Oct 2009 D18 = + are obtained. 1 is obtained as T18. These parameters are depending on the system geometry. C18 is the ratio of the velocity at the inlet to the outlet pressure for the hypothetical case of the outlet being rigidly ﬁxed ͑Zf → ϱ͒. (27) results in p 1 = A cp 8 + B c 5 . 57 (5). The transfer matrix of the system shown in Fig. (42) and p10 from Eqn. D c͑ A d − A c͒ Bc ͑50͒ By substituting 5 from Eqn. Equilibrium of volume velocity at the second junction point yields 480 Noise Control Eng. + ͑ B c + B d͒ ͑ B c + B d͒ B c ͑48͒ are obtained. in this study. and B18 is the ratio of the inlet pressure to the velocity at the outlet for the hypothetical case of the outlet being totally open or unconstrained ͑Zf → 0͒22. and D18 is the ratio of the velocity at the inlet to the velocity at the outlet for the hypothetical case of the outlet being totally open or unconstrained ͑Zf → 0͒22.1 = Chr1p10 + Cdp8 + Dd6 + Ccp5 + Dc5 . These expressions for the elements C18 and D18 imply certain physical signiﬁcance. (45) results in 5 = 7 + 8 − 6 = − Chr2 Ahr2 p8 + 8 1 = ͩ + + ͩ BcChr2 − ͑Ad − Ac͒Ahr2 ͑Bc + Bd͒Ahr2 ͪ ͩ ͪ p8 + Bc 8 . and they are not affected by connections to elements upstream or downstream. (50). A18 is the ratio between the inlet pressure and the outlet pressure for the hypothetical case of the outlet being rigidly ﬁxed ͑Zf → ϱ͒. (40). The elements A18 and B18 have individual physical meaning. Bc + Bd Bc + Bd ͑47͒ ͑53͒ BdChr1 BcChr2 − Ahr2͑Ad − Ac͒ ͑Bc + Bd͒Ahr2 ͑Bc + Bd͒Ahr2 Bc ͑Bc + Bd͒Ahr2 + Dd͑BcChr2 − Ahr2͑Ad − Ac͒͒ + DcBd ͑BcChr2 − Ahr2͑Ad − Ac͒͒ + ͩ ͪ p8 BdChr1Bc ͑Bc + Bd͒Ahr1 D bB c B dD cB c + 8 . where C18 and D18 are two elements of the transfer matrix T18 as deﬁned by Eqn. A18. It is an invariant property of the . (10). + ͑ B c + B d͒ ͑ B c + B d͒ B c ͑46͒ ͪ − B2 c Chr2 − Bc͑Ad − Ac͒Ahr2 ͑Bc + Bd͒Ahr2 p8 − Then. where A18 and B18 are the other two elements of the transfer matrix T18 as deﬁned by Eqn. (49) into the Eqn. in terms of p8 and 8. B18. On the other hand. C18 and D18 are frequency-dependent complex quantities embodying the acoustical properties of the system. 6 from Eqn. the sound pressure at the output of “element a” is found. as long as the system elements can be assumed to be linear and passive16. Bc ͑Bc + Bd͒Ahr2 and BdChr1Bc ͑Bc + Bd͒Ahr1 D bB c B dD cB c . Bc + Bd ͑49͒ AdChr1 Ahr1 Ahr1 + Cd + Cc Substituting p1 = p4 and p5 = p8 into Eqn. ͑45͒ Substituting 5 from Eqn. Za and Zf are the impedances of “element a” and “element f. Bc + Bd ͑51͒ Then. AdChr1 Ahr1 + BdChr1 BcChr2 − Ahr2͑Ad − Ac͒ Ahr1 + ͑Bc + Bd͒Ahr2 D c͑ A d − A c͒ Bc C18 = + Cd + Cc + Dd͑BcChr2 − Ahr2͑Ad − Ac͒͒ ͑Bc + Bd͒Ahr2 + DcBd ͑BcChr2 − Ahr2͑Ad − Ac͒͒ . the characteristic impedances are assumed to be equal at the inlet and the outlet: Za = Zf = Z. A18 = Ac + BcChr2 Ahr2 − B2 c Chr2 − Bc͑Ad − Ac͒Ahr2 ͑Bc + Bd͒Ahr2 ͑52͒ and B2 B cB d c B18 = Bc − = . as p 1 = A cp 8 + BcChr2 Ahr2 p 8 + B c 8 B2 c 8 . J.” respectively. (10).

͑54͒ expected frequency response of the system. . In these three cases. A sound pressure value of 1 Pa was set for the nodes on the input face of the cavity.element. the input and output faces of the cavity were meshed with ﬁrst order triangular elements. respectively.3. the theoretical transmission loss of the system becomes inﬁnite. 57 (5). . The frequency range examined in this work is 0 – 3000 Hz. respectively. the analytical and numerical results are shown together. The cavity is meshed using ﬁrst order tetrahedral solid elements. and the sound speed was 340 m / s. 1700 and 2. The results were then compared with the analytical analyses. In our ﬁgures.17. The bandwidth of the modiﬁed HQ tube developed by Selamet and Easwaran8 and later used by Trochon17 is dependent on the center frequency of the ﬁlter. J. which is called the cavity in this paper. In view of the many possibilities and variations in terms of the dimensions and number of acoustical elements. These parameters remain constant throughout our analysis. . (55) for lhq = 200 mm. the damping effect of air was not taken into account and was also not considered in our analytical model. the mesh size must be smaller than one sixth of the wavelength (mesh size Ͻ / 6). the ﬁrst. Using the default values in the software.3.113 m. a ﬁnite element method model was developed using the MSc. Unigraphics Nx-4. 1700 and 2. to give the input and output. lhq.2. . we ﬁrst built a solid model of the system using a 3D CAD program. TL ﬁnds favor with researchers who are interested in ﬁnding the acoustic transmission behavior of an element or set of elements in the context of isolated terminations22. This solid model consists of the air in the pipe system.2. and the mesh size used in our numerical analysis is 0. our present study focuses on structures with certain discrete duct lengths and diameters. Two local coordinates with the same x-direction were assigned to these new faces. . Finally. At these frequencies.1. ͑56͒ Setting the lengths of the side branch duct. 2͑lhq − ld͒ p = 0. for geometries (hmd = 18 mm.550 Hz. There are three peaks in the TL curves across several related frequency bandwidths. Figures 2–4 show the effects of the diameters of the main duct and the side branch duct on the acoustic attenuation performance of the system. The density was 1. to 200 mm and 100 mm. hhq = 3 mm). the effect of the diameters of the main duct and the side branch duct on the TL is very clear. In each of the ﬁnite element analysis applications. Then. The starting conﬁguration of the system was suggested based upon results found in the literature7. The transfer matrix method based on the plane wave propagation model is valid when the inﬂuence of higher order modes can be neglected. ͑lhq + ld͒ p = 1. (55).Patran in parasolid format. Actran software package. . hhq = 36 mm). the length of the main duct and the length of the side branch duct are taken into account as ld = 100 mm and lhq = 200 mm.19 and was consistent with our initial resonance calculations. 1700 Hz from Eqns. The resonance frequencies of the system consisting of a side branch duct connected to the main duct can be estimated as f= and f= c p. the TL 481 . second and third resonance frequencies of the duct are approximately 850. Setting the design length of the side branch duct. ͑57͒ c ͑2p + 1͒. The minimum wavelength is = 0. For these cases. . (hmd = 36 mm. the resonance frequencies of the system can be calculated as approximately 1125. the initial resonance calculations show that the resonance frequencies are 850 Hz from Eqn. .8. The duct lengths are signiﬁcant to reach the Noise Control Eng. In numerical acoustic analysis. hhq = 18 mm) and (hmd = 36 mm. respectively. and the main duct. We processed the output data in order to obtain the TL of the system. Our goal was to be able to read the input and output sound power values after the analysis.012 m. (55)–(57) and 2550 Hz from Eqn. The resonance of a duct that is open at both ends occurs when the frequency is f= c p. In order to determine the multi-dimensional effects on the acoustic performance of the system.2.225 kg/ m3. Since it is independent of the terminations. to 200 mm. .3. 2lhq p = 1. lhq. This is related to the ﬁxed ratios of the constituent pipe lengths. Sept-Oct 2009 where c is the speed of sound.250 Hz. the solid model was exported to MSc. From the ﬁgures. we created a material with the default values of air. After meshing the cavity using solid elements. Then. ld. ͑55͒ 3 RESULTS AND DISCUSSION This section is devoted to the numerical evaluations of the expressions given in the preceding section and the ﬁnite element models. For small diameter ratios of the main duct and the side branch duct. the TL of the system can be expressed as TL = 20Log10 ͯͩ B18 1 A18 + + ZC18 + D18 2 Z ͪͯ .

1. 5. The effect of the attachment of a one quarter wave resonator is analyzed in Fig. setting the length of the quarter wave resonator. 2. and the results are in agreement. – – – Numerical.60 55 50 Transmission Loss [dB] 45 40 35 30 25 20 15 10 5 0 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 Frequency [Hz] Fig. hmd = 36 mm.3. lhq = 200 mm. while hmd and hhq increase. (55) for lhq = 200 mm. Furthermore. hhq = 18 mm. and lhq are kept constant according to Fig. ld = 100 mm. 1700 Hz from Eqns. Comparison of Fig. — Analytical. ld. 2—TL of the system consisting of a side branch duct connected to the main duct. frequency bandwidths are narrow. J. both connected to 60 55 50 45 Transmission Loss [dB] 40 35 30 25 20 15 10 5 0 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 Frequency [Hz] Fig. the wider the TL frequency bandwidth. There are three peaks in the TL curves in the related frequency bandwidth. lhq = 200 mm. our results are in satisfactory agreement in the related frequency bandwidth. 57 (5). Furthermore. 5 shows that the quarter wave resonator eliminates any signiﬁcant changes in TL levels. 3 and Fig. separately. ld = 100 mm. The effect of Figures 5 and 6 indicate the TL treatment of the system consisting of a side branch duct and one and two quarter wave resonators. the analytical and numerical results are shown together. . — Analytical. the ﬁrst resonance frequency of the quarter wave resonators can be calculated as approximately 1700 Hz. for geometries. ͑58͒ the main duct. but we note that the greater the ratio of the diameters. lhr = 50 mm. For greater diameter ratios. the initial calculations indicate that the resonance frequencies are 850 Hz from Eqn. We note again that the transfer matrix method can only be used for plane-wave conditions. ld = 100 mm and lhq = 200 mm. . to 50 mm. 482 Noise Control Eng. In the ﬁgures. our analytical and numerical results are in good agreement. hmd = 18 mm. (55)–(57) and 2550 Hz from Eqn. . – – – Numerical. hhq = 18 mm. our comparisons show that given the small diameter ratios. hhq = 3 mm. For these cases.2. hmd = 36 mm. (55). 4lhr p = 0. 3—TL of the system consisting of a side branch duct connected to the main duct. Both analytical and numerical TL curves reveal the same characteristics. In this case. Sept-Oct 2009 . The resonance frequencies of the quarter wave resonator can be calculated as f= c ͑2p + 1͒. lhr.

In this case. the ﬁrst 60 55 50 TL peak moves approximately 100 Hz upwards. ld = 100 mm. lec = 67 mm and lhq = 200 mm. hmd = 36 mm. 57 (5). In this case. the attachment of one more quarter wave resonator is analyzed in Fig. the second peak moves approximately 300 Hz upwards. Figure 8 shows the effect of a side branch duct modiﬁed with an expansion chamber connected to the main duct in terms of the acoustic attenuation performance of the system. The TL frequency bandwidths tend to decrease in the side branch duct and the resonator diameters increase. hmd. A minimum of 10 dB of TL between 1050– 1600 Hz can be obtained with the geometry shown in the ﬁgure. 8 shows that the ﬁrst TL peak moves approximately 300 Hz downwards. hhq = 36 mm. ld. In addition. and there is no change in the Transmission Loss [dB] 45 40 35 30 25 20 15 10 5 0 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 Frequency [Hz] Fig. hmd = 36 mm. ld. 7. and the third TL peak moves approximately 100 Hz downwards at the same time. Comparison of Fig. 5—TL of the system consisting of a side branch duct and a quarter wave resonator connected to the main duct. hhq. and signiﬁcant changes in TL levels and frequency bandwidths are not observed. In order to indicate the effects of the diameters of two quarter wave resonators on the TL characteristic of the system. lhq = 200 mm. The illustration shows results for geometries hmd = 18 mm. 6 indicates that the three TL peaks move approximately 50 Hz downwards. This ﬁgure shows our analytical results. hmd. 2 and Fig. and a quarter wave resonator is connected to the main duct. 6. 4—TL of the system consisting of a side branch duct connected to the main duct. — Analytical. hec = 6 mm. Sept-Oct 2009 483 . and lhq are kept constant according to Fig. 3. — Analytical.Fig. J. 5 and Fig. and lhq are kept constant according to Fig. – – – Numerical. we note that with a greater diameter of the side branch duct. ld = 100 mm. ld = 100 mm. hhq = 3 mm. while only hhq increases. lhr = 50 mm. hhq = 18 mm. lhq = 200 mm. TL behaviors are compared in Fig. 3. Comparison of Fig. – – – Numerical. Both analytical and numerical TL curves show the same trends and the results are in good agreement. Noise Control Eng.

lhr = 50 mm. hec = 18 mm. 57 (5). 7—TL of the system consisting of a side branch duct and two quarter wave resonators connected to the main duct. hhr1. hhq = 3 mm. J. Comparisons show that for small diameters. and lhq are kept constant according to Fig. hhr1. Speciﬁcally. there 40 35 30 25 20 15 10 5 0 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 Frequency [Hz] Fig. the dimensions that are held constant are lhr = 50 mm. 8 and Fig. in addition. Figures 9 and 10 show the TL behaviors of the system consisting of two quarter wave resonators and a side branch duct modiﬁed with an expansion chamber 60 55 50 45 Transmission Loss [dB] and all connected to the main duct. for geometries (hmd = 18 mm. 9 demonstrates that when two quarter wave resonators are connected to the main duct a new resonance peak appears at approximately 1700 Hz. Sept-Oct 2009 . for the larger diameters. 484 Noise Control Eng. In this case. Comparison of Fig. analytical and numerical results are in good agreement. lhq = 200 mm. both the analytical and numerical results are shown together. hec = 6 mm. and two quarter wave resonators are connected to the main duct. In these two cases. — hhr = 27 mm. hhq = 18 mm. there is a new peak at 3000 Hz. hmd. In this case. The TL levels are higher than the ﬁrst case. hhq = 9 mm. Calculations indicate that the new resonance peak results from the quarter wave resonators given hhr = 3 mm. 3. results are in satisfactory agreement across related frequency bandwidths because of the limitations of the plane wave approach. ld = 100 mm.Fig.hr2 = 9 mm). lhr = 50 mm. ld. In the ﬁgures. lhr = 50 mm. lec = 67 mm and lhq = 200 mm. lhq = 200 mm. – – – hhr = 18 mm. hmd = 36 mm. — Analytical. 6—TL of the system consisting of a side branch duct and two quarter wave resonators connected to the main duct. – – – Numerical. ld = 100 mm. hmd = 36 mm. hhq. respectively.hr2 = 3 mm) and (hmd = 18 mm. frequency of the third peak. ld = 100 mm.

— Analytical. Furthermore. 45 40 35 30 25 20 15 10 5 0 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 Frequency [Hz] Fig. In these two cases. but the fourth TL peak moves approximately 100 Hz downwards. lhr = 50 mm.60 55 50 45 Trasmission Loss [dB] 40 35 30 25 20 15 10 5 0 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 Frequency [Hz] Fig. but we note that the greater the diameters.hr2 = 3 mm. In this case. hhr1.hr2 = 9 mm. 8—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber connected to main duct. the TL frequency bandwidths are narrow. respectively. hmd. In this case. 9 and Fig. hec = 18 mm. ld = 100 mm. we analyze the effects of the position and the length of the expansion chamber on the TL behavior of a system consisting of two quarter wave resonators and a side branch duct modiﬁed with an 60 55 50 Transmission Loss [dB] expansion chamber. hmd = 18 mm. lec = 60 mm. hhr1. lc1 = 80 mm. lec = 67 mm. ld. lc1 = 60 mm. no systematic or noticeable change occurs in TL levels. lec = 67 mm. hmd = 18 mm. are ﬁve peaks in the TL curves for a related frequency bandwidth given the geometries taken into consideration. – – – Numerical. the dimensions that are held constant are hmd = 18 mm. Comparisons between Fig. the TL levels are higher. Noise Control Eng. The major effect of the length of the expansion chamber lies in shifting the TL peak frequencies of the system. hhq = 3 mm. the wider the TL frequency bandwidths. – – – Numerical. lhr = 50 mm. 10 show that for small diameters of the resonators and the side branch duct. hhq = 3 mm. hhq = 9 mm. lhq = 200 mm. and an expansion chamber is located at the side brunch duct. there are no signiﬁcant changes in the frequencies of the ﬁrst three peaks. — Analytical. lc1 = 100 mm). hec = 6 mm. 2. Figures 11 and 12 compare the TL behaviors of the system in order to show the effects of the geometric parameters on the TL characteristic. hec = 6 mm. lec = 67 mm and lhq = 200 mm. J. Sept-Oct 2009 485 . Additionally. ld = 100 mm. all connected to the main duct. hhq. ld = 100 mm. 57 (5). and lhq are kept constant according to Fig. With the change of the position of the expansion chamber. lec = 80 mm) and (lc1 = 20 mm. lhq = 200 mm. 9—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two quarter wave resonators connected to the main duct. for geometries (lec = 40 mm. lc1 = 40 mm.

For the case in Fig. hhr1.n2 = 12 mm. we analyze the diameter and length of the neck on TL behavior of the system consisting of two Helmholtz resonators connected to main duct and a Fig.hr2 = 36 mm. — lec = 60 mm. the major effect of the position of the expansion chamber to the TL curve is to shift the resonant frequencies.hr2 = 50 mm).hr2 = 45 mm) and (hmd = 18 mm. respectively. ln1. hhr1. (56) and (57) and 2250 Hz from Eqn. lec = 67 mm. – – – lec = 40 mm. hhq = 9 mm. ͱ h2 n 2 hhr lhrln . hmd = 18 mm.hr2 = 9 mm. However. hec = 18 mm. Figures 13 and 14 present the TL behaviors of the system for geometries (hmd = 36 mm. Sept-Oct 2009 .n2 = 15 mm. lec = 67 mm. 57 (5). ld = 100 mm. which after some rearrangement is given by c f= 2 side branch duct. 1125 Hz from Eqn.n2 = 12 mm. 13. The resonance frequencies of the Helmholtz resonators were based on the volume of the cavity and the radius of the opening. — – — lec = 80 mm.hr2 = 9 mm. 486 Noise Control Eng. — Analytical. ld = 100 mm. calculations show that the resonance frequencies are 730 Hz from Eqn. lhr1. hn1. ld = 100 mm. 11—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two quarter wave resonators connected to the main duct. – – – Numerical. hhq = 36 mm. (55). lhq = 200 mm. hec = 24 mm. initial calculations indicate that the resonance frequencies are 690 Hz from Eqn. ld = 100 mm. lhq = 200 mm. 10—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two quarter wave resonators connected to the main duct. J. lhq = 200 mm. lhr = 50 mm. (55). 1700 Hz from Eqns. ͑59͒ In the next case. hmd = 18 mm. hhq = 9 mm. hhq = 18 mm. lhq = 200 mm. lhr = 50 mm. For the case in Fig. (55). (59). ln1. 14. hhr1. hhr1. lhr1. (59).n2 = 12 mm. 1125 Hz from Eqn.Fig. hn1.hr2 = 36 mm. hec = 18 mm.

As indicated in the initial resonance calculations from Eqn. 13 shows the effect of the Helmholtz resonators connected to main duct. There is no signiﬁcant change in the frequency and the level of the ﬁrst peak. lhq = 200 mm. ld = 100 mm. hhq = 36 mm. the resonance Fig. respectively. hec = 18 mm. hhq = 9 mm. (55). hn1.hr2 = 9 mm. lhq = 210 mm.. — – lc1 = 40 mm. — – – — lc1 = 60 mm. J. which may prove to be useful for noise control. – – – Numerical.n2 = 12 mm.hr2 = 36 mm. The contributions of the necks are also positive in terms of the frequency behavior of the system. In these cases.n2 = 12 mm. 15. The effect of the length of the side branch duct is investigated and shown in Fig.. and two Helmholtz resonators are connected to the main duct. hhr1.lc1 = 80 mm. Figures 15–20 illustrate analytical results from parametric studies on the TL behaviors of the system. Comparison of Fig. hn1. lhr1. Sept-Oct 2009 487 . hhq. ln1. hhr1. 57 (5). The considerable contribution of the length of the side branch duct is seen in the third TL peak. 4 and Fig. ln1. (59). lhq = 204 mm. hmd = 18 mm. and lhq are kept constant according to Fig. 1700 Hz from Eqns. the dimensions that are held constant are hmd = 36 mm. (56) and (57) and 2250 Hz from Eqn. hhq = 36 mm. — Analytical. 3. 12—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two quarter wave resonators connected to the main duct.hr2 = 45 mm.hr2 = 36 mm. In this case.n2 = 15 mm and lhr1. The width of the attenuation bands near resonances is considerable.hr2 = 45 mm. . lhr = 50 mm. lec = 67 mm. — lc1 = 20 mm. Noise Control Eng.Fig. They carry a new TL peak into the related frequency range. lhq = 216 mm and lhq = 222 mm. for geometries lhq = 198 mm. ld = 100 mm. lhq = 200 mm. ld. ld = 100 mm. These conﬁgurations exhibit frequency bands where the transmission loss is high. hmd = 36 mm.n2 = 15 mm. – – – lc1 = 100 mm. but the other peaks shift downwards. 13—TL of the system consisting of a side branch duct and two Helmholtz resonators connected to the main duct. hmd. hhr1.

Fig. hhr1. 16. the resonance frequencies vary between 1200 Hz for ln = 5 mm and 530 Hz for ln = 25 mm. hmd = 18 mm. 14—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two Helmholtz resonators connected to the main duct.hr2 Fig. ln1. hn1. — – — lhq = 216 mm. frequencies change between 1730 Hz for lhq = 198 mm and 1390 Hz for lhq = 222 mm. In these cases. – – – lhq = 210 mm. ln1. ln1.hr2 = 50 mm. as shown in the ﬁgure. At higher frequencies. There is no signiﬁcant change in the TL levels. The main effects of the length of the neck are especially evident in the ﬁrst TL peak. the lower the frequency of the ﬁrst TL peak. 15—TL of the system consisting of a side branch duct and two Helmholtz resonators connected to the main duct.hr2 = 36 mm. lhr1. Sept-Oct 2009 . TL levels increase in the low frequency range. — Analytical.hr2 = 36 mm. ln1. whereas the levels decrease at high frequencies.n2 = 12 mm. lhr1. — – – — lhq = 222 mm. Comparison of the results shows that the longer the length. hn1.n2 = 12 mm. lhr1. 57 (5). The frequency bandwidth of the ﬁrst TL peak narrows while the resonance frequencies decrease. hhq = 36 mm.n2 = 12 mm. lhq = 200 mm. – – – Numerical. especially for the ﬁrst TL peak. for geometries ln1. hhq = 18 mm.n2 = 25 mm. the second and the third TL peaks are affected mainly by the length of the side branch duct. lec = 67 mm. ld = 100 mm.n2 = 20 mm and ln1. — — lhq = 204 mm. we analyze the length of the neck.n2 = 10 mm. J.n2 = 12 mm. 17 for geometries lhr1. hec = 24 mm. — lhq = 198 mm. the ﬁrst TL peak is affected mainly by the length of the neck. 488 Noise Control Eng. the collapse of the TL consistent with the length of the side branch duct is evident. The contribution of the lengths of the Helmholtz resonators is studied in Fig.hr2 = 45 mm.n2 = 15 mm. the dimensions that are held constant are hmd = 36 mm. (59). ln1.n2 = 5 mm. hhr1. ld = 100 mm. hhr1.hr2 = 36 mm. hmd = 36 mm. hhq = 36 mm. as shown in Fig. hn1. From Eqn. In the next case.n2 = 15 mm.hr2 = 45 mm and lhq = 200 mm. ld = 100 mm.

lhr1. In these cases. — — ln1. the dimensions that are held constant are hmd = 36 mm.n2 = 12 mm and ln1. hhq = 36 mm. hn1. There are no signiﬁcant changes in the frequency bandwidth of the ﬁrst TL peak. From Eqn. ln1. ld = 100 mm.hr2 = 40 mm. J. hn1. hhq = 36 mm. the resonance frequencies appear between 790 Hz for lhr = 35 mm and 620 Hz for lhr = 55 mm. 18 for the geometries hn1.n2 = 15 mm. ld = 100 mm. The main effects of the length of the Helmholtz resonators are especially pronounced in the ﬁrst TL peak.hr2 = 40 mm. lhq = 200 mm. hhq = 36 mm. Figure 18 shows the effect of diameters of the necks on the acoustic attenuation performance of the system. lhr1. hn1. (59).hr2 = 50 mm.hr2 = 45 mm.n2 = 15 mm. hmd = 36 mm. hn1. Noise Control Eng. Meaningful changes in the frequencies and TL levels are not observed in the ﬁgure. – – – lhr1.n2 = 20 mm. hhr1. the lower the frequency of the ﬁrst TL peak. ln1. Increasing the neck diameter Fig.n2 = 6 mm.n2 = 15 mm.hr2 = 50 mm and lhr1.n2 = 12 mm.n2 = 27 mm and hn1.n2 = 15 mm.n2 = 5 mm. lhq = 200 mm. — — lhr1.hr2 = 36 mm and lhr1.hr2 = 55 mm. hhr1. — ln1. = 35 mm. the dimensions that are held constant are hmd = 36 mm. — – — lhr1. Sept-Oct 2009 489 .n2 = 18 mm.hr2 = 35 mm. ld = 100 mm. lhr1. hhq = 36 mm. 57 (5). 16—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two Helmholtz resonators connected to the main duct. In these cases.n2 = 36 mm.n2 = 25 mm. hn1. ld = 100 mm. — – — ln1.hr2 = 55 mm. especially for those frequencies higher than 1000 Hz. – – – ln1. — lhr1. the ﬁrst TL peak is affected mainly by the length of the Helmholtz resonators. hmd = 36 mm. 17—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two Helmholtz resonators connected to the main duct. Comparison of our results shows that the longer the length.n2 = 10 mm.hr2 = 45 mm. lhr1.hr2 = 45 mm. lhq = 200 mm. lhq = 200 mm. — – – — ln1.n2 = 12 mm. This ﬁgure shows that the ﬁrst TL peak essentially results from the resonators. hn1. A similar analysis can be carried out using the results depicted in Fig.Fig.n2 = 12 mm.hr2 = 45 mm. — – – — lhr1.hr2 = 36 mm.

ln1.hr1. shifts the ﬁrst TL peak upwards and expands the frequency bandwidth. – – – hhq. — – — hhq.hr2 = 24 mm.hr2 = 18 mm.n2 = 18 mm.n2 = 36 mm.hr2 = 45 mm.hr2 = 36 mm. 57 (5). for geometries hhq. The main effects of the size of the diameters are especially pronounced in the ﬁrst TL peak. hhr1. In the middle frequency range.hr1.n2 = 6 mm. In each case. the diameters of both the resonators and the side branch duct are identical. — — hhq. hmd = 36 mm. the resonance frequencies change between 340 Hz for hn = 6 mm and 2100 Hz for hn = 36 mm. 490 Noise Control Eng. lhr1.hr2 = 12 mm.n2 = 12 mm. ld = 100 mm. 19. except for certain narrow frequency bands associated with resonant peaks. the dimensions that are held constant are hmd = 36 mm. – – – hn1. Comparison of the results shows that increasing the diameter shifts the ﬁrst TL peak downwards and decreases the frequency bandwidth. lhq = 200 mm. ld = 100 mm. In these cases. lhr1. (59).hr1.hr1.hr1. hhq. hhq. — – — hn1.hr2 = 18 mm.hr1. the collapse of the TL behavior of the system associated with the size of diameters is evident.hr1.hr2 = 36 mm.hr1. hn1.n2 = 27 mm.hr2 = 30 mm. hhq. lhq = 200 mm. Sept-Oct 2009 . lhq = 200 mm. — — hn1.hr2 = 45 mm and hn1. ln1. ld = 100 mm.hr2 = 36 mm. J. 18—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two Helmholtz resonators connected to the main duct. In the next case.n2 = 15 mm.hr2 = 30 mm and hhq.n2 = 12 mm. especially for the frequencies in between 1000 Hz and 1500 Hz.n2 = 12 mm.n2 = 15 mm. — – – — hn1. In addition. — – – — hhq. hmd = 36 mm.n2 = 15 mm.hr1. ln1. 19—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two Helmholtz resonators connected to the main duct. whereas at the middle frequencies. — hhq. the TL behavior of the system with extended necks is better than in the case of the directly connected resonators. and the ﬁrst TL peak is affected mainly by the diameter of the neck.hr2 = 24 mm.hr2 = 45 mm. we investigate the diameters of the side branch duct and two Helmholtz resonators. — hn1. there Fig. From Eqn. as shown Fig. lhr1. hhq = 36 mm.Fig.hr1.hr2 = 12 mm.

We apply our method to multiple different conﬁgurations.Fig. (59) show that the resonance frequencies are intermediate. in some cases the deviations of the TL peaks seemed to result from the slight differences between the assumed dimensions and geometries. In general. hn1. hhr1. (55)–(57) for ld = 110 mm and ﬁnally. the inﬂuence of the distance between the junction points is examined in Fig. 690 Hz from Eqn. We discuss the limitations of our plane wave approach.n2 = 15 mm. 491 . Figure 20 shows that the change in the distance between the junction points does not affect the character of the TL curves. and we compare our numerical predictions with the results obtained using ﬁnite element methods. 690 Hz from Eqn. lhr1. For the case in Fig. hn1. particularly at higher frequencies. whereas the TL peaks shift downwards together with the increase in the distance. The results are consistent. There is no signiﬁcant change in the TL levels and bandwidths.n2 = 12 mm. ld = 110 mm and ld = 120 mm. the solid models that we Noise Control Eng. The frequency bandwidth of the ﬁrst TL peak narrows while the resonance frequencies decrease. 57 (5). 1550 Hz and 1890 Hz from Eqns. hhq = 36 mm. In general. It is possible to use the transfer matrix method to predict the acoustic behavior of systems in the relatively low frequency range with geometries that are not too complex. 1130 Hz. our results show that increasing the duct diameters leads to more extensive attenuation patterns. 2100 Hz for hhr = 12 mm and 690 Hz for hhr = 36 mm. Helmholtz resonators and side brunch ducts designed to certain optimal dimensions. the calculations show that the resonance frequencies are 690 Hz from Eqn. (55)–(57) for ld = 120 mm. – – – ld = 110 mm. The initial resonance calculations from Eqn. lhq = 200 mm and ln1. especially for the frequencies higher than 2250 Hz. ln1. In these cases. (59).hr2 = 36 mm. J. hhq = 36 mm. (59). lhr1. 1050 Hz. lhq = 200 mm. 4 CONCLUSIONS We wished to construct an engineering design tool for the quick but moderately accurate determination of the TL of a system with two resonators connected to the main duct and the side branch duct modiﬁed with an expansion chamber. Sept-Oct 2009 explored using the ﬁnite element method exhibit more detailed geometries and accurate dimensions than the models that we used with the transfer matrix method.hr2 = 45 mm. Numerical applications show that the 1D analytical approach yields satisfactorily accurate approximations in the 0 – 3000 Hz frequency range. (59). we ﬁnd that the analytical and numerical results are in good agreement. Finally. Our results indicate that it is possible to manufacture a broadband mufﬂer by using reactive type acoustic attenuation devices such as quarter wave resonators. 1100 Hz. some divergences were identiﬁed in regions near the cut-off frequency of the system. — ld = 100 mm.n2 = 15 mm. hhr1. We conclude that the junction point is not a point in the solid models. 1700 Hz and 2250 Hz from Eqns.hr2 = 36 mm. In terms of applications. Therefore.hr2 = 45 mm.n2 = 12 mm. 20. at least in cases where the effects of higher order modes can be neglected. 1410 Hz and 2100 Hz from Eqns. — – — ld = 120 mm. 20 for geometries ld = 100 mm. hmd = 36 mm. We generally found good agreement between the transfer matrix method and the ﬁnite element method results. As expected. 20—TL of the system consisting of a side branch duct modiﬁed with an expansion chamber and two Helmholtz resonators connected to the main duct. (56) and (57) for ld = 100 mm. and the ﬁrst TL peak is affected mainly by the diameter of the neck. is no meaningful change in the TL levels. We derived an analytical model that uses the transfer matrix method for plane wave conditions and with no mean ﬂow. the dimensions that are held constant are hmd = 36 mm.

Am.. 355–361. Dickey. Burdisso and J. A. 9. J. A. 8. P. “The effects of ﬂow on the performance of a reactive acoustic attenuator”. “Circular concentric Helmholtz resonator”. 96(5). R. (2000). Soc. 46. Graefenstein and W.. 1–8. R. 124(2). Sept-Oct 2009 . J. J. 94. Sound Vib.. Genhua and L. 6. Dickey and A. ASME NCA 22. 73– 92. Novak. 284. J. T. A. (1997). Trochon. C. 14. M. Soc. R. Radavich. 723–732. P. Am. 245–255. S. 11. Moya.. Intern. 107. J. Rui. Gerges et al. 261–266. Selamet and P. Hallez. 101. M. C. New York. 27. (1999). “Investigation of the Herschel-Quincke concept for the suppression of higher-order modes in a duct”. S. (1991). (1987).We note that the effect of extending the duct diameters is to widen the frequency bands of attenuation. 132–140. Dickey and J. D. M. “A study of the inﬂuence of mean ﬂow on the acoustic performance of HerschelQuincke tubes”. Y. A. (1998). “Control of higherorder modes in ducts using arrays of Herschel-Quincke waveguides”. InterNoise99. Selamet. Burdisso. N. J. “The HerschelQuincke tube: A theoretical. J. “Silencer for hydraulic piston pump pressure pulsations”. Birdsong and C. 512–517. Active Control of Vibration and Noise. (2008). Hastings. S. A.. Broatch and R. Noise Control Eng. of the Brazilian Society of Mechanical Science & Engineering. “A new type of silencers for turbocharger noise control”. we conclude that the method presented in this paper may be useful in examining the acoustical behavior of these structures in broadband noise control applications. Radavich. 195. 2. D. 492 Noise Control Eng. Ji. 107. WileyInterscience. Selamet and V. 283–298. Am. A.. (1996). Soc.. N. 3177–3185. “Modiﬁed Herschel-Quincke tube: attenuation and resonance for n-duct conﬁgurations”. Soc. “Herschel-Quincke Spiral: A new interference silencer”. Lee. “Helmholtz resonator with extended neck”. 107. 1874–1879. Thus. J. 62. J. 15. 19. 13. “A smart Helmholtz resonator”. Torregrosa. Peizi. C. Stunk. Acoust. A. 2360–2369. J. 22. 5. 127–132. J. Rui.. ASME DE-75. Karlsson. (1997). J. “Helmholtz resonators: onedimensional limit for small cavity length-to-diameter ratios”. A. M. “Effect of expansion chamber on the resonance frequency of side branches and HerschelQuincke tubes”. Bies. “Mufﬂer Modeling by Transfer Matrix Method and Experimental Veriﬁcation”. Novak. R. A. H. SAE Technical Paper Series 911759. 57 (5). G. “Multidimensional effects on silencer performance”. Soc. Acoust. Payri. C. L. Selamet. (1994). Acoust. computational and experimental investigation”. A. Zhichi. P. Selamet and I. J. J. Am. “Application of Quincke tubes to ﬂow ducts as a sound attenuation device”. J. 41–51. (2003). J. E. Smith. J. (1994). (2001). Acoust. N. Chen and M. Off-Highway and Powerplant Congress and Exposition. C. Fuller and D. Smith and R. M. (2000). S. R. J. A. Song. (2000). (2005). Glav and M. “Acoustic performance of a Herschel-Quincke tube modiﬁed with an interconnecting pipe”. (2005). SAE Technical Paper Series 011722. S. Brady. 20. Soc. Acoust. Sound Vib. 10. Munjal. Acoust. Wenzel. Noise Control Eng.. (1979). 16. 7. Acoustics of Ducts and Mufﬂers. J. “The Herschel-Quincke tube: The attenuation conditions and their sensitivity to mean ﬂow”. Åbom. (2003). Selamet. A.. M.. (1996). “Half-wavelength tuning cables for passive noise control in automotive power steering systems”. 102. Z. Am. Radavich and N. “Circular asymmetric Helmholtz resonators”. Am. ASME Forum on Active Noise Control. Radcliffe. ASME Control of Vibration and Noise. M. 18. Am. A. Desantes. Dickey and J. L. Acoust. Selamet. Easwaran. Sound Vib. 164–169. 12. 21. F. 1975–1985. 3. N. Selamet and Z. Torregrosa. (1997). IMECE. B. A. Dallas. (1994). L.. 4. P.. Climent and D. M. A. Soc. P. 17. J. L. SAE Technical Paper Series 01-1436. 5 REFERENCES 1.

- The Power of a Good Sleep
- Get a Good Night's Sleep - 7 Practical Steps.pdf
- X-Ray Diffraction Studies on Electron Density in Organic Crystals
- Stress Analysis using Ccaesar.pdf
- caesarii-141127101848-conversion-gate02.pdf
- Economics of Engine and Compressor Analysis
- Basic Pipe Stress Analysis Tutorial
- Modelling Turbulence Induced Vibration of Pipes
- Pulsation and Vibration Study Of Reciprocating Compressor.pdf
- Advanced Mechatronics Systems - Laboratory - Mechatronics Control Systems - 1st Ed
- 1a_Flanges_Fundamentals.pdf
- Autopipe New User Tips 9a
- A Finite Element Based Study on Stress Intensification Factors - SIF - For Reinforced Tees
- Apostila Instalaoemanutenodemotoreseltricos 130312110408 Phpapp02
- Structural Vibration Modeling and Dynamic Stiffness - Mahmoodi Masih- 2014 - PhD Thesis
- Guidelines Dynamics Vibration
- Pipe Hangers
- API 674 Pumps1
- Reciprocating Pump Selection Guide for Pulsation Studies
- Raised Face Flange Ball Valves
- Flanges Fundamentals
- Ijetae Icmtset 2014 01
- riverin2006rp-1pp (1)
- Guidelines for Transient Analysis
- Water Hammer - Volume 1

by Jose Prado

Application of Resonators and a Side Branch Duct With an Expansion Chamber for Broad Band Noise Control

Application of Resonators and a Side Branch Duct With an Expansion Chamber for Broad Band Noise Control

- Wavesby nguang
- HAARP RESONANCE - CONTROLLED WAVE PARTICLE INTERACTION STUDIES IN THE RADIATION BELTSby THE NIKOLA TESLA INSTITUTE
- Sound - Problems With Answers and Explanationsby Ha Vi
- Figure Shown the Shape of Part of a Long String in Which Transverse Waves Are Produced by Attaching One End of the String to Tuning Fork of Frequency 250 Hzby harshanauoc

- Waves
- HAARP RESONANCE - CONTROLLED WAVE PARTICLE INTERACTION STUDIES IN THE RADIATION BELTS
- Sound - Problems With Answers and Explanations
- Figure Shown the Shape of Part of a Long String in Which Transverse Waves Are Produced by Attaching One End of the String to Tuning Fork of Frequency 250 Hz
- Chladni Plates-CurtisParry (1)
- e16_1
- FEI - 5650A
- magnetotherapy.pdf
- Essay 2 - Creation & Manifestation-factors Affecting the Internal Psychophysiology of m.p.o. - l.e.r.m
- paper1
- Total Population Control
- Paper 11
- Experiment 4
- Solucionario Capitulo 21 Física Serway and Faughn
- 17.07120803
- Physics Waves MCQ
- 10 - Waves
- BLST-CRT
- Microwave
- Purchased a Rife Machine Specifically to Deal With My Mothers Cancer a GB4000
- 0000278
- Experimental Investigation of Vibration Power Flowin Thin Technical Orthotropic Plates by the Method of Vibration Intensity
- lecture23-1233854096875992-3
- 29 JORDAN Bill MineBlastingVibration
- Radio Frequency Identification
- How to Measure Blocking
- 2008 FDTD as a Validation Tool of Transmission Loss Calculation
- R REC v.431!7!200005 Nomenclatura
- Systemic.P8
- Virtual Experimental Modal Analysis (VEMA)
- Application of Resonators and a Side Branch Duct With an Expansion Chamber for Broad Band Noise Control - Haluk Erol

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd