Vibroacoustic Coupled Analysis
Theory: mass reflects, matched wave speed lets sound through
Overview
Do panels of the same weight insulate equally?
Nearly, at low frequency. Sound striking a panel tries to shake it, but heavier panels move less and reflect the sound, so low-frequency insulation depends mostly on surface density (mass per m²). At higher frequency stiffness matters. Bending waves in the panel speed up with frequency and at some frequency match the speed of sound in air. At this coincidence frequency sound and panel waves move in step and sound passes through easily. Stiffer, thicker panels coincide lower, so among equal-weight panels, one whose coincidence lies in an audible band insulates worse in practice.
Mass law
f is frequency (Hz) and m″ surface density (kg/m²). A field-incidence approximation valid well below the coincidence frequency.
Coincidence frequency
c is the speed of sound in air (343 m/s), D bending stiffness, h thickness and c_B bending-wave speed; c_B = c at f_c.
Laminated glass and double glazing
Window glass tends to coincide in the speech and traffic band (1–3 kHz), a weak point for insulation. Laminated glass bonds two panes with a soft resin interlayer that damps bending vibration and shallows the coincidence dip. In double glazing the air between panes acts like a spring, creating a low-frequency mass–air–mass resonance, so air gap and pane thicknesses are varied to shift resonance and coincidence to less audible frequencies.
Worked examples
Example: surface density, coincidence and mass-law transmission loss
| Panel | Surface density | Coincidence frequency | TL at 500 Hz | TL at 2 kHz |
|---|---|---|---|---|
| Steel 1 mm | 7.9 kg/m² | 12.0 kHz | 24.9 dB | 36.9 dB |
| Steel 2 mm | 15.7 kg/m² | 6.0 kHz | 30.9 dB | 42.9 dB |
| Aluminium 2 mm | 5.4 kg/m² | 6.0 kHz | 21.6 dB | 33.7 dB |
| Glass 6 mm | 15.0 kg/m² | 2.0 kHz | 30.5 dB | (near coincidence) |
| Gypsum board 12.5 mm | 10.0 kg/m² | 2.9 kHz | 27.0 dB | 39.0 dB |
Mass-law TL rises 6.0 dB per doubling of surface density or frequency. For 6 mm glass the mass law would give 42.5 dB at 2 kHz, but coincidence pulls the real value well below.
6 mm glass and 2 mm steel weigh about the same, yet differ at 2 kHz.
Glass is thick and stiff in bending, so it coincides low, at 2 kHz, where its bending waves move in step with sound in air and sound passes through easily, dropping well below the mass-law 42 dB. Thin steel is compliant in bending and coincides high at 6 kHz, so at 2 kHz it performs close to the mass law. 2 kHz matters for speech and traffic noise, so this dip governs window insulation. The dip depth depends on panel damping — the less damping, the deeper. Vibroacoustic analysis therefore includes structural modes, damping and coupling to the acoustic field to obtain transmission loss by frequency.
Analysis workflow
- Use the mass law and coincidence frequency to judge problem bands and material suitability.
- Get structural modes and damping from a finite-element model.
- Model the acoustic field with FEM or BEM, coupled to the structure at the boundary.
- Compute transmission loss or radiated sound by frequency, checking coincidence and resonances.
- Calibrate damping and boundary conditions against measurements such as reverberation-room tests.
“Thicker glass didn't make it quieter”
In one building, window glass was thickened from 5 to 10 mm to cut road noise, but it didn't get as quiet as expected. Doubled mass should give 6 dB by the mass law, but 10 mm glass coincides near 1.2 kHz, actually worsening insulation in a key traffic-noise band. Switching to laminated glass of the same weight to damp vibration shallowed that dip and reached the target insulation.
Common mistakes
Mistakes and fixes
| Mistake | Effect | Fix |
|---|---|---|
| Mass law at all frequencies | High-frequency TL overestimated | Check coincidence |
| Assuming thicker is always better | Coincidence moves lower | Damping or laminated build-ups |
| No damping | Exaggerated peaks and dips | Measure damping |
| Ignoring leaks through gaps | Real insulation lower | Assess gaps |
| Solving structure and air separately | Coupling effects missed | Solve coupled |
I'd like to learn related topics.
Related pages include acoustic–structure interaction, Lighthill's acoustic analogy, noise prediction, muffler design and underwater acoustics.
Related Topics
Experience the theory firsthand with the interactive simulator for this field
All Simulators