Spotlight #28 – Acoustics & Music: Standing Waves, Chladni Patterns, Room Modes and the Physics of Sound

Acoustics is the oldest branch of physics. Pythagoras studied string ratios. Galileo dropped pendulums to count frequencies. Chladni scattered sand on violin plates and revealed the shapes of sound modes. Sabine measured reverberation in Harvard lecture halls and founded architectural acoustics. Today the field spans concert-hall design, submarine sonar, medical ultrasound, noise-cancellation headphones and cochlear implants. What unifies all of it is the wave equation — a simple PDE that governs how pressure disturbances propagate and interfere.

The simulations in this wave explore acoustics across six scales: the one-dimensional standing waves in a pipe, the two-dimensional nodal geometry on a vibrating plate, the three-dimensional resonances of a room, reverberation time, and the frequency-to-place mapping inside the cochlea. Each scale reveals something different about how waves behave when they encounter boundaries.

1. Standing Waves in Pipes — Harmonics and Boundary Conditions

When a sinusoidal wave travels down a pipe and reflects off its end, the forward and reflected waves superpose into a standing wave. The spatial pattern is frozen in place: nodes (zero displacement) and antinodes (maximum displacement) occur at fixed positions. The allowed frequencies are determined entirely by the pipe’s length and its boundary conditions at each end.

Wave equation (1D, pressure):
  ∂²p/∂t² = c² ∂²p/∂x²   where c = speed of sound (343 m/s in air at 20°C)

Boundary conditions (for pressure p and displacement u ≡ −∂p/∂x):
  Open end:   pressure node      p = 0 at x = 0 or L
  Closed end: displacement node  ∂p/∂x = 0 at x = 0 or L

Harmonic frequencies:
  Open–Open (both ends open):
    f_n = n · c / (2L),  n = 1, 2, 3, …  (all harmonics present)
    Antinode at each end; fundamental = c/2L

  Open–Closed (one end open, one stopped):
    f_n = (2n−1) · c / (4L),  n = 1, 2, 3, …  (odd harmonics only)
    Fundamental = c/4L, one octave below open–open of same length
    Explains why a stopped organ pipe sounds an octave lower

  Closed–Closed (both ends stopped):
    f_n = n · c / (2L),  n = 1, 2, 3, …  (same frequencies as open–open,
    but node at each end instead of antinode)

Displacement standing wave (open–open, nth mode):
  u(x, t) = A · cos(nπx/L) · cos(2πf_n t)
  Nodes at x = L/(2n), 3L/(2n), …
  Antinodes at x = 0, L/n, 2L/n, …

Real instruments (end correction):
  Open end is not a perfect pressure node; the effective length is L + 0.6r
  where r = pipe radius (Rayleigh end correction).
  Accounts for partial radiation at the open end.

The difference between open–open and open–closed pipes explains the timbral difference between a flute and a clarinet. A flute is open at both ends and produces all harmonics; a clarinet is effectively closed at the reed and open at the bell, so it over-blows to the third harmonic (a twelfth above the fundamental) rather than the octave, and its lowest register is dominated by the fundamental and odd harmonics. The even harmonics, missing from the single-reed spectrum, give the clarinet its characteristic “hollow” timbre.

2. Chladni Patterns — The Shape of a Sound

In 1787 Ernst Chladni scattered sand on a square metal plate, drew a bow across its edge and discovered that the sand migrated away from the vibrating regions and accumulated along lines of zero motion — the nodal lines. Each resonance frequency produced a different, often strikingly beautiful geometric pattern. These figures, now called Chladni patterns, are a direct visual map of the eigenmodes of the vibrating plate.

Governing equation (thin elastic plate, Kirchhoff–Love theory):
  D ∇&sup4;w + ρh ∂²w/∂t² = 0
  w(x,y,t) = plate deflection
  D = Eh³/[12(1−ν²)] = flexural rigidity  (E = Young’s modulus, h = thickness, ν = Poisson’s ratio)
  ρ = density, h = thickness

For a freely-vibrating square plate (side a), separation of variables w = W(x,y)e^(iωt):
  ∇&sup4;W = λ&sup4;W,   where λ&sup4; = ρhω²/D

Simply-supported (SS) rectangular plate — exact solution:
  W_mn(x,y) = sin(mπx/a) · sin(nπy/b)
  f_mn = (π/2) · (D/ρh)^(1/2) · [(m/a)² + (n/b)²]
  Nodal lines: vertical at x = ka/m (k=1..m−1); horizontal at y = lb/n (l=1..n−1)

Chladni’s empirical rule (for square plates, free edges):
  f_mn ∝ (m + 2n)²   ← approximate for free boundary conditions
  The actual free-plate modes require numerical FEM; nodal patterns are more complex
  than the simply-supported case.

Number of nodal lines:
  Mode (m,n): m−1 nodal lines parallel to Y, n−1 parallel to X
  Or m−1 + n−1 curved nodal lines for non-rectangular symmetry.

Degeneracy and mode mixing:
  When two modes have the same frequency (f_mn = f_nm for a square plate),
  any linear combination is also a valid mode — the actual pattern depends
  on initial/boundary conditions. This mixing produces the rotated and curved
  nodal patterns observed in real Chladni experiments.

Chladni patterns are more than beautiful: they are used by violin makers to test the acoustic quality of the top and back plates. A Stradivarius violin top plate has specific resonance frequencies that the maker shapes and tunes by removing wood until the Chladni patterns match the desired templates. Modern loudspeaker designers use Finite Element Analysis of nodal patterns to eliminate unwanted resonances in cones and diaphragms.

3. Room Modes — Resonances in Rectangular Spaces

A concert hall, recording studio, or living room is an acoustic resonator. Sound waves reflecting off the six walls create three-dimensional standing waves — room modes — at specific frequencies determined by the room’s dimensions. At these frequencies, some positions in the room have very high pressure (antinodes) and others have nearly zero pressure (nodes). The listening experience changes dramatically depending on where you sit.

Modal frequencies (rigid-wall rectangular room, L×W×H):
  f_lmn = (c/2) · √[(l/L)² + (m/W)² + (n/H)²]
  l, m, n = non-negative integers (not all zero)
  c = speed of sound

Mode types:
  Axial (one index non-zero):     waves along one axis only
    e.g. f_100 = c/(2L)         ← simplest, strongest modes
  Tangential (two indices):       waves along two axes, reflecting off four walls
  Oblique (three indices):        waves along all three axes, reflecting off all six walls

Mode density (number of modes below frequency f):
  N(f) ≈ (4πV/3c³)f³ + (πS/4c²)f² + (L_total/8c)f   (Weyl formula)
  V = room volume, S = total wall area, L_total = sum of all edge lengths
  Modes are sparse at low frequencies, dense at high frequencies.

Schroeder frequency (transition from modal to statistical regime):
  f_S ≈ 2000 · √(RT60/V)   Hz
  Below f_S: room modes dominate; above f_S: diffuse-field assumption is valid.
  Typical small room (V=40 m³, RT60=0.4 s): f_S ≈ 200 Hz
  Concert hall (V=20 000 m³, RT60=2 s): f_S ≈ 20 Hz

Pressure mode shapes (rigid walls):
  p_lmn(x,y,z) = A · cos(lπx/L) · cos(mπy/W) · cos(nπz/H)
  Nodal planes at positions x = L(2k+1)/(2l), etc.

Problematic intervals (modes too far apart):
  Large spacing between consecutive mode frequencies → uneven bass response.
  Recommended room ratios (Bonello criterion):
    L:W:H ≈ 1.6:1.25:1 (or 1.9:1.4:1) — minimise coincident modal frequencies.

Room modes are the principal reason why small recording studios sound bad in the bass region. Below about 200 Hz the modes are widely spaced; if you are sitting at a pressure antinode for the 80 Hz axial mode, bass sounds boomy; at the node it disappears entirely. Studio designers place absorbers, diffusers and bass traps at the corners (where pressure is highest for the lowest modes) and choose asymmetric room dimensions to spread the modal frequencies.

4. Concert-Hall Reverberation — RT60 and Room Geometry

Reverberation time RT60 is the time it takes for sound to decay by 60 dB after the source stops. It was defined and first measured by Wallace Clement Sabine in 1900 while studying the acoustics of the Fogg Art Museum in Cambridge. His empirical formula, derived from careful experiments with stopwatch and organ pipe, remains one of the most useful engineering formulas in architectural acoustics.

Sabine formula (1900):
  RT60 = 0.161 · V / A
  V = room volume (m³)
  A = total absorption (m² Sabins) = Σ α_i S_i
  α_i = absorption coefficient of surface i (0 = perfect reflection, 1 = perfect absorption)
  Valid for low average absorption (α_avg < 0.2)

Eyring formula (1930, more accurate for higher absorption):
  RT60 = −0.161 · V / [S · ln(1 − α_avg)]
  S = total surface area
  Reduces to Sabine formula when α_avg → 0

Mean free path (average distance between reflections):
  λ_mfp = 4V/S   (valid for diffuse field)
  Number of reflections per second: c / λ_mfp = cS/(4V)
  Energy decay rate: η = −cSα/(4V) → RT60 = 24 ln(10)/η ≈ 0.161 V/(Sα)

Optimal RT60 by room use:
  Anechoic chamber:     < 0.1 s
  Recording studio:     0.2 – 0.4 s
  Speech (classroom):   0.4 – 0.8 s
  Opera house:          1.2 – 1.4 s
  Symphony hall:        1.8 – 2.2 s
  Cathedral:            4 – 10 s

EDT (Early Decay Time):
  Decay evaluated from 0 to −10 dB (extrapolated to 60 dB).
  Correlates more closely with subjective “liveness” than full RT60.
  A large EDT relative to RT60 suggests late-arriving reflections dominate.

Sabine’s first application of his formula was the Sanders Theatre renovation in 1900: by adding cushioned seats and draping he raised the absorption and brought RT60 from an unintelligible 5.5 seconds down to about 1.2 seconds. The Boston Symphony Hall, designed using Sabine’s formulas in 1900, remains one of the acoustically finest concert halls in the world — a direct legacy of applying physics to architecture.

5. Cochlear Mechanics — The Inner Ear as a Frequency Analyser

The inner ear performs a continuous Fourier transform. The cochlea — a fluid-filled coiled tube about 35 mm long — acts as a mechanical frequency analyser: each position along the basilar membrane resonates at a specific frequency, with high frequencies near the base and low frequencies near the apex. Hair cells at each position translate mechanical motion into neural signals, creating a tonotopic map.

Tonotopic frequency map (Greenwood function, 1990):
  f(x) = A · (10^(a·x) − k)
  x = fractional distance from apex (0 = apex, 1 = base)
  Human cochlea: A = 165.4, a = 2.1, k = 0.88 (gives f in Hz)
  Apex:  ~20 Hz (low frequency)
  Base:  ~20 000 Hz (high frequency)
  ~3.5 mm per octave along basilar membrane

Travelling wave (Georg von Békésy model):
  A stiffness gradient along the basilar membrane (stiff at base, floppy at apex)
  causes incoming pressure waves to travel from base to apex.
  Each wave builds up to a maximum amplitude at the characteristic place for its frequency,
  then rapidly decays — creating a sharp peak that encodes frequency.

Mechanical tuning curve:
  Bandwidth of basilar membrane resonance: Q_mech ~ 3–10 (passive)
  Active amplification (outer hair cells, OHC somatic motility):
    OHCs amplify motion, sharpening tuning to Q ~ 30–100
    This “cochlear amplifier” explains the 40 dB of gain and the fine frequency
    discrimination of the healthy ear.

Two-tone suppression:
  A tone at f_1 can suppress the response to f_2 if sufficiently close in frequency.
  Reflects the compressive nonlinearity of the OHC amplifier.
  The origin of masking effects used in audio compression (MP3, AAC).

Critical bandwidth (Bark scale):
  The frequency resolution of the cochlea; tones within one critical band mask each other.
  Δf_CB ≈ 25 + 75[1 + 1.4(f/1000)²]^0.69 Hz (Zwicker formula)
  Audio codecs exploit this: discard masked components below the masking threshold.

Georg von Békésy won the 1961 Nobel Prize in Physiology for demonstrating the travelling wave in excised human cochleae. The cochlear amplifier — the active role of outer hair cells — was discovered in the 1980s and explains how the healthy ear achieves both extreme sensitivity (detecting vibrations smaller than the diameter of a hydrogen atom) and fine frequency selectivity. Cochlear implants bypass damaged hair cells by directly stimulating the auditory nerve at multiple electrode positions along the tonotopic map, exploiting the same spatial frequency code that the basilar membrane creates naturally.

6. Acoustic Diffusers — Scattering Sound in Space

A flat wall reflects sound like a mirror: all energy returns in a single specular direction. A quadratic residue diffuser (QRD) — a periodic array of wells of different depths — scatters incoming sound uniformly across many angles. The depths are chosen from quadratic residue sequences, a number-theoretic construction discovered by Manfred Schroeder in 1975. The result: a surface that appears acoustically rough at all frequencies within its design bandwidth, eliminating acoustic glare and flutter echo.

Quadratic residue sequence (period N, N = prime):
  s_n = n² mod N   (n = 0, 1, …, N−1)
  Example N=7: s = [0, 1, 4, 2, 2, 4, 1]
  Each sequence is symmetric: s_n = s_{N-n}

Well depth (maximum depth d_max for lowest design frequency f_min):
  d_n = s_n · λ_min / (2N)   or   d_n = s_n · c / (2N · f_min)
  Total depth span: d_max = (N−1) · c / (2N · f_min)
  For N=7, f_min=500 Hz: d_max = 6 · 343/(14·500) ≈ 294 mm

Far-field scattered pressure (Fraunhofer approximation, M periods):
  p(θ) = Σ_{n=0}^{N−1} r_n · exp(iknd sinθ)   × sin(Mk d_x sinθ/2)/sin(k d_x sinθ/2)
  k = 2πf/c, d_x = well width, θ = scattering angle
  Uniform |p(θ)|² across θ is the design goal (equal energy in all grating lobes)

Why it works (number theory):
  The DFT of the QRD reflection coefficients r_n = exp(iπs_n/N) has constant magnitude:
  |DFT{r_n}| = constant for all frequencies → flat angular power spectrum.
  This is the connection between Gauss sums (number theory) and acoustic diffusion.

Diffusion coefficient d:
  d = [(NΣE_i² − (ΣE_i)²) / ((N−1)(ΣE_i)²)]^(1/2)   (AES standard)
  d = 1 for perfect diffusion, d = 0 for specular reflection.
  QRD typically achieves d > 0.8 over its design bandwidth.

Schroeder diffusers are now standard in recording studios, concert halls and broadcast facilities worldwide. Because their design is rooted in number theory, different prime numbers N produce different period widths and frequency ranges. Real installations combine several periods side by side; the grating lobes from the periodicity add constructively, giving strong diffuse scattering. The connection between pure mathematics (quadratic residues, Gauss sums) and room acoustics is one of the most elegant examples of abstract mathematics finding direct engineering application.

The Wider World of Acoustics

These six simulations cover acoustics from the scale of a pipe bore to a concert hall, sampling some of the most directly perceivable physics in the world. Sound is everywhere, and the wave equation that governs it recurs throughout physics: the same mathematics describes electromagnetic waves, quantum mechanical probability amplitudes, seismic waves in the Earth and gravitational waves in spacetime. Understanding how boundaries, geometry and material properties shape acoustic waves builds intuition that transfers directly to all of these domains.