SpectraSim Documentation
Parameter · 05 Boundary Conditions

Acoustic impedance model of a perforated wall

perforate.impedance_model Step 05 · Boundary Conditions

Selects the law that gives the acoustic transfer impedance of THIS perforated wall: the pressure jump across the wall per unit normal particle velocity. Each perforated wall has its own model and parameters. The list follows the wall's perforation type (Perforation type): a micro-perforated wall offers only the two micro-perforate models (circular holes, Maa; slits), a perforated tube or a perpendicular plate only the models for millimetre holes. Without the flow (acoustics only) a wall has no perforation type and the list offers both families, grouped. A wall saved with a model of the other family gets the default of its type when the Boundary Conditions step is opened, and a note says so.

Last updated

Theory

In the acoustic model a perforated wall is a cut through the mesh whose two sides are coupled by a transfer impedance ZpZ_p: the normal particle velocity vnv_n is continuous through the wall, and the pressure jumps by ZpvnZ_p v_n, equation (1). It enters the finite-element system as iωZp\frac{i\omega}{Z_p} times the jump matrix of the cut face (a resistance dissipates sound, a mass reactance adds inertia). The time convention is e+iωte^{+i\omega t}, so a mass reactance is positive imaginary and a resistance non-negative.

Every model computes the normalised impedance of ONE hole, ζh\zeta_h; the wall impedance is the hole impedance divided by the open-area ratio σ\sigma, equation (2) (Lee and Ih, 2003). The models:

  • Elnady 2010 (Elnady et al., 2010): Crandall's oscillating viscous flow in a short tube with an effective viscosity 2.179 μ2.179\,\mu (viscous plus thermal losses), the discharge coefficient CDC_D (effective open area σCD\sigma C_D), a resistive end correction δre\delta_{re}, a reactive end correction 0.5 d0.5\,d reduced by the hole-interaction factor fintf_{int}, the radiation resistance, and terms for grazing flow 0.3 Mg0.3\,M_g and through-hole (bias) flow 1.15 Mb/CD1.15\,M_b/C_D (the paper quotes the last two from Bauer 1977), equations (3)-(5). Recommended: no flow, grazing flow and through-hole flow.
  • Kirby & Cummings 1998 (Kirby and Cummings, 1998): flat plates under fully developed turbulent grazing flow, with the friction velocity u∗u_*; when a packing lies on ONE side of the wall the end correction of that side is made complex (their porous-backing correction), equation (6). For dissipative straight-through silencers.
  • Cross-flow: the whole flow forced THROUGH the holes (perpendicular baffles and cross-flow mufflers): resistance 2.57 Mh2.57\,M_h from the mean through-flow Mach number in the holes, reactance 0.95 k(t+0.75 d)0.95\,k(t + 0.75\,d), equation (7); without flow the resistance 6×10−36\times10^{-3} and the reactance k(t+0.75 d)k(t + 0.75\,d) (Sullivan, 1979) (Munjal, 2014). No grazing term.
  • Lee & Ih 2003 (Lee and Ih, 2003): a regression of measured circular orifices in grazing flow, equation (8), with the reactance clamped to zero below 1/∣b4∣=1511/|b_4| = 151 Hz, where the printed fit changes sign.
  • Physics baseline (thin plate): the exact viscous impedance of the hole as a short tube plus Rayleigh's end correction, equation (9); options Thin plate: use C_D, Hole interaction (thin plate), End-correction factor (thin plate). No flow dependence: an independent no-flow cross-check.
  • Sullivan & Crocker 1978 (legacy) (Sullivan and Crocker, 1978): the measured single-hole resistance of one perforate (Single-hole resistance (Sullivan & Crocker)) and the reactance k(t+0.75 d)k(t + 0.75\,d). Offered only to a wall that already uses it.

With Include bias flow on, the bias resistance 1.15 Mb/CD1.15\,M_b/C_D is added to the hole impedance of every model except Elnady and cross-flow, which carry their own through-flow term. On a micro-perforate this term is borrowed from ordinary perforates and is not validated (a warning is shown).

Equations

p1−p2=Zp vn,Zp=ρc ζpp_1 - p_2 = Z_p\,v_n, \qquad Z_p = \rho c\,\zeta_p
(1)
ζp=ζhσ\zeta_p = \frac{\zeta_h}{\sigma}
(2)
wheremeaningunit
p1,p2p_1, p_2Acoustic pressure on the two sides of the wallPa
vnv_nNormal particle velocity through the wallm/s
ρ,c\rho, cDensity and speed of sound of the local gaskg/m³, m/s
ζh\zeta_hNormalised impedance of one hole-
σ\sigmaOpen-area ratio used by the acoustics-
θ=Re{ikCD[tF(K′d)+δrefintF(Kd)]}+1−2J1(kd)kd+0.3 Mg+1.15 MbCD\theta = \mathrm{Re}\left\{\frac{ik}{C_D}\left[\frac{t}{F(K'd)} + \frac{\delta_{re} f_{int}}{F(Kd)}\right]\right\} + 1 - \frac{2J_1(kd)}{kd} + 0.3\,M_g + \frac{1.15\,M_b}{C_D}
(3)
χ=Im{ikCD[tF(K′d)+0.5 d fintF(Kd)]},ζh=θ+iχ\chi = \mathrm{Im}\left\{\frac{ik}{C_D}\left[\frac{t}{F(K'd)} + \frac{0.5\,d\,f_{int}}{F(Kd)}\right]\right\}, \qquad \zeta_h = \theta + i\chi
(4)
F(x)=1−2J1(x/2)(x/2)J0(x/2),K=−iω/ν,K′=−iωρ/(2.179 μ),δre=0.2d+200d2+16000d3,fint=1−1.47σ+0.47σ3F(x) = 1 - \frac{2J_1(x/2)}{(x/2)J_0(x/2)}, \quad K = \sqrt{-i\omega/\nu}, \quad K' = \sqrt{-i\omega\rho/(2.179\,\mu)}, \quad \delta_{re} = 0.2d + 200d^2 + 16000d^3,\quad f_{int} = 1 - 1.47\sqrt{\sigma} + 0.47\sqrt{\sigma}^{3}
(5)
wheremeaningunit
kkWavenumber ω/c\omega/c1/m
d,td, tHole diameter and wall thickness (d in m in δre\delta_{re})m
CDC_DDischarge coefficient-
MgM_gGrazing Mach number-
MbM_bBias (through-hole) Mach number-
μ,ν\mu, \nuDynamic and kinematic viscosity of the gasPa s, m²/s
J0,J1J_0, J_1Bessel functions of the first kind-
ζh=θf+tc d8νω+ik(t+δδ0 0.85 d)−i 0.425 kd(1−ρeqρ),θf=max⁡(0, [26.16 (t/d)−0.169−20]u∗−4.055 fdc)\zeta_h = \theta_f + \frac{t}{c\,d}\sqrt{8\nu\omega} + ik\left(t + \frac{\delta}{\delta_0}\,0.85\,d\right) - i\,0.425\,kd\left(1 - \frac{\rho_{eq}}{\rho}\right), \qquad \theta_f = \max\left(0,\ \frac{\left[26.16\,(t/d)^{-0.169} - 20\right]u_* - 4.055\,f d}{c}\right)
(6)
ζh=2.57 Mh+0.95 ik (t+0.75 d)\zeta_h = 2.57\,M_h + 0.95\,ik\,(t + 0.75\,d)
(7)
ζh=Rn+imax⁡(In,0),Rn=a0(1+a1∣f−f0∣)(1+a2M)(1+a3d)(1+a4t),In=b0(1+b1d)(1+b2t)(1+b3M)(1+b4f)\zeta_h = R_n + i\max(I_n, 0), \qquad R_n = a_0(1 + a_1|f - f_0|)(1 + a_2 M)(1 + a_3 d)(1 + a_4 t), \qquad I_n = b_0(1 + b_1 d)(1 + b_2 t)(1 + b_3 M)(1 + b_4 f)
(8)
ζh=1ρc[iωρtF(2ksa)+iωρ ϵ fint d],ks2=−iωρ/μ, a=d/2\zeta_h = \frac{1}{\rho c}\left[\frac{i\omega\rho t}{F(2k_s a)} + i\omega\rho\,\epsilon\,f_{int}\,d\right], \qquad k_s^2 = -i\omega\rho/\mu,\ a = d/2
(9)
wheremeaningunit
θf\theta_fFlow-induced resistance of Kirby & Cummings-
u∗u_*Friction velocity of the grazing flowm/s
δ/δ0\delta/\delta_0Reduction of the end correction by the flow (Kirby & Cummings eq. 13)-
ρeq\rho_{eq}Equivalent density of a packing on one side (only then the last term)kg/m³
MhM_hMean through-flow Mach number in the holes-
ai,bi,f0a_i, b_i, f_0Lee & Ih regression constants; f0=412(1+104M)/(1+274d)f_0 = 412(1 + 104M)/(1 + 274d)-
ϵ\epsilonEnd-correction factor (End-correction factor (thin plate), 0.85)-

Default and rule

Default
Elnady 2010 (new walls); old case files keep the global model they stored
Rule

New perforated walls use Elnady 2010; a micro-perforated wall uses the circular-hole micro-perforate model (Maa 1998). A case file written before the model became a per-wall setting has no entry for the wall and uses the global perforate model it stored.

Source

The equations and constants are those of the original papers (Elnady et al., 2010) (Kirby and Cummings, 1998) (Lee and Ih, 2003) (Sullivan and Crocker, 1978). The program lists Elnady 2010 first as the recommended model; Sullivan & Crocker is no longer offered for new walls. Cross-flow: Sullivan's resistance θ0=2.57 U0/c\theta_0 = 2.57\,U_0/c with U0U_0 the mean hole speed (his eqs. 16-18) and Munjal's reactance 0.95 k(t+0.75 dh)0.95\,k(t + 0.75\,d_h). Munjal writes the resistance as 0.514 d1M/(lσ)0.514\,d_1 M/(l\sigma), equation (3.128) of the 2nd edition = (3.153) of the 1st, with MM the mid-stream Mach number of the pipe (0.514=2.57×0.8/40.514 = 2.57 \times 0.8/4; 0.8 = mean over mid-stream speed). Before 2026-11-07 the program used 4×0.514 Mh4 \times 0.514\,M_h with the MEAN hole Mach number (resistance 20 % low) and no factor 0.95. Validation of the jump formulation: a flat perforated sheet in a duct agrees with the exact TL=20lg⁡∣1+ζ/2∣TL = 20\lg|1 + \zeta/2| within 0.016 dB.

Valid range Perforated (mm holes): Elnady 2010, Kirby & Cummings 1998, cross-flow, Lee & Ih 2003, physics baseline (thin plate), Sullivan & Crocker 1978 only on a wall that already uses it. Micro-perforated (sub-mm holes): circular holes (Maa 1998), slits (Guo, Allam & Abom 2008)

Micro-perforated plates

Sub-millimetre holes or slits with an open-area ratio of about 0.2-5 %: circular holes after Maa (1998) with the edge factors and grazing-flow terms of Allam and Abom (2011) (Hole edges of a micro-perforate); slits after Guo et al. (2008) as given by Kabral et al. (2014) (Slit width, Grazing factor of a slit micro-perforate). The flow solution of such a wall uses the viscous resistance of the same papers (Micro-perforate flow resistance).

Micro-perforates: flow regime

Both micro-perforate models are for no flow or GRAZING flow along the wall, the regime they were derived and measured in: the flow term is KM/σK M/\sigma (circular holes, K=0.15K = 0.15, with the end correction times FF) or βM/σ\beta M/\sigma (slits). Allam and Abom (2011) (sec. 3.1.3) state that, the steady pressure-drop coefficient of a micro-perforate being high, only grazing flow is of interest. They are not models of flow THROUGH the plate (a micro-perforated baffle across a duct, bias flow).

Micro-perforates: validation

Circular holes against measured transmission loss: Allam and Abom (2011) two-chamber muffler without flow (rms 1.1 dB), three chambers with grazing flow up to Mach 0.1 (the measured trend with flow is reproduced; the level is 5-7 dB high at the maximum; the cause is not settled, a small leak between the chambers reproduces the measurement), Lee et al. (2016) single chamber with a film lining (rms 1.8 dB below 3 kHz), Maa (1998) absorption. Slits: Kabral et al. (2014) without flow (rms 1.8 dB) and with grazing flow at Mach 0.05 (rms 0.6 dB below 2.4 kHz); above the Cremer peak the model is about 5 dB high.

Use when

Elnady: no flow, grazing or through-hole flow; CDC_D measured for d 0.6-2.4 mm and t 0.8-2.4 mm, not for packing touching the holes. Kirby & Cummings: grazing flow with packing on the perforate, fitted for σ 0.20-0.27, t/d 0.29-0.54. Lee & Ih: d 2-9 mm, t 1-5 mm, σ 3-22 %, M ≤ 0.2, 60-4000 Hz. Cross-flow: flow through the holes only. Baseline: no flow.

Validity gate

Frequencies and parameters outside a model's published range are listed by the acoustic QA gate (a warning; the run continues). A negative resistance fails the gate: it would inject energy.

Not modelled

Mean-flow convection in the duct is not part of these impedances; the amplitude (non-linear) dependence of the resistance is not modelled; Kirby & Cummings' porous-backing correction is applied only when a packing touches ONE side of the wall.

References 11

  1. Allam, S. and Abom, M. (2011). A new type of muffler based on microperforated tubes. Journal of Vibration and Acoustics, 133(3), 031005. DOI 10.1115/1.4002956
  2. Elnady, T., Åbom, M. and Allam, S. (2010). Modeling perforates in mufflers using two-ports. Journal of Vibration and Acoustics, 132(6), 061010. DOI 10.1115/1.4001510
  3. Guo, Y., Allam, S. and Abom, M. (2008). Micro-perforated plates for vehicle applications. Proceedings of Inter-Noise 2008, 773-791. Web page
  4. Kabral, R., Du, L., Abom, M. and Knutsson, M. (2014). A compact silencer for the control of compressor noise. SAE International Journal of Engines, 7(3), 1572-1578. DOI 10.4271/2014-01-2060
  5. Kirby, R. and Cummings, A. (1998). The impedance of perforated plates subjected to grazing gas flow and backed by porous media. Journal of Sound and Vibration, 217(4), 619-636. DOI 10.1006/jsvi.1998.1811
  6. Lee, S.-H. and Ih, J.-G. (2003). Empirical model of the acoustic impedance of a circular orifice in grazing mean flow. The Journal of the Acoustical Society of America, 114(1), 98-113. DOI 10.1121/1.1581280
  7. Lee, S., Bolton, J. S. and Martinson, P. A. (2016). Design of multi-chamber cylindrical silencers with microperforated elements. Noise Control Engineering Journal, 64(4), 532-543. DOI 10.3397/1/376399
  8. Maa, D.-Y. (1998). Potential of microperforated panel absorber. Journal of the Acoustical Society of America, 104(5), 2861-2866. DOI 10.1121/1.423870
  9. Munjal, M. L. (2014). Acoustics of ducts and mufflers. 2nd ed., Wiley, Chichester, ISBN 978-1-118-44312-5. Web page
  10. Sullivan, J. W. and Crocker, M. J. (1978). Analysis of concentric-tube resonators having unpartitioned cavities. The Journal of the Acoustical Society of America, 64(1), 207-215. DOI 10.1121/1.381963
  11. Sullivan, J. W. (1979). A method for modeling perforated tube muffler components. II. Applications. The Journal of the Acoustical Society of America, 66(3), 779-788. DOI 10.1121/1.383680