First Year Engineering is coming soon

We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.

Acoustics of buildings

Principles, formulas, and design steps for room acoustics: reverberation, absorption, insulation, echoes, and auditorium design.

Drafted with Aria, reviewed by the AiCanCode.org team · Updated . How we write and check our notes. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

Acoustics of buildings tells us how to make classrooms, auditoriums, meeting rooms, cinemas, studios, and hospitals sound clear and comfortable. Engineers use it to control reverberation, prevent echoes, reduce noise transmission through walls/ceilings, and to design spaces so speech and music are easy to hear.

Key ideas

  • Decibel (dB): a logarithmic unit expressing a ratio. Sound pressure level Lp = 20·log₁₀(p/p₀) dB; sound intensity level LI = 10·log₁₀(I/I₀) dB.
  • Absorption coefficient (α): fraction of incident acoustic energy absorbed (not reflected) by a surface at a given frequency and incidence, dimensionless (0 to 1 ideally). α = 0 means perfectly reflective; α = 1 means perfectly absorbing. Note: measured α in labs may slightly exceed 1 due to edge effects; treat as ≈1 in design.
  • Equivalent absorption area (A, metric sabins): the total absorption of a room expressed as the area of a perfectly absorbing surface in m². For a surface of area S and coefficient α, its contribution is α·S. The room’s total A = Σ(αᵢ·Sᵢ). One metric sabin = 1 m² of perfect absorber.
  • Diffuse field assumption: sound energy is uniformly distributed with random incidence at surfaces. Many room-acoustics formulas assume this.
  • Mean free path (ℓ̄): average distance between reflections in a room with volume V and surface area S, given by ℓ̄ = 4V/S under diffuse field.
  • Reverberation: persistence of sound after the source stops, due to multiple reflections.
  • Reverberation time (T₆₀): time for the sound level in a room to decay by 60 dB after the source stops. It is the central design parameter for room acoustics.
  • Sabine’s formula (assumes low average absorption and diffuse field): Exact statement: In a diffuse field, for rooms with small average absorption coefficient, the reverberation time is T₆₀ = 0.161·V/A, where V is the room volume (m³) and A is the equivalent absorption area (m² sabin). Assumptions: diffuse field, ᾱ small (typically ᾱ ≲ 0.2–0.3), uniform distribution of absorbers, negligible air attenuation. Common derivation (steps to reproduce in exam):
    1. Let E be acoustic energy in the room. Under diffuse field, the rate of wall collisions is c·S/(4V). With average absorption ᾱ, the fractional energy loss per collision is ᾱ.
    2. So dE/dt = −(c·S·ᾱ/4V)·E, giving exponential decay E(t) = E₀·exp[−(cSᾱ/4V)t].
    3. A 60 dB drop means energy (and intensity) falls by 10⁻⁶. Set E(T₆₀)/E₀ = 10⁻⁶ and solve: T₆₀ = (6·ln 10)·(4V)/(cSᾱ).
    4. Using c ≈ 343 m·s⁻¹ at 20°C, the constant becomes 0.161, and ᾱS = A gives T₆₀ = 0.161·V/A.
  • Eyring (Norris–Eyring) formula (better when absorption is high): Exact statement: For a diffuse field, the reverberation time is T₆₀ = 0.161·V / [−S·ln(1 − ᾱ)], where ᾱ = (ΣαᵢSᵢ)/S, S is total surface area. Assumptions: diffuse field, independent reflections as a Markov process, valid for higher ᾱ. For low ᾱ, −ln(1−ᾱ) ≈ ᾱ and it reduces to Sabine.
  • Air absorption correction: At high frequencies or very large volumes, add air attenuation as an equivalent absorption 4·m·V to A, where m is the attenuation coefficient (m⁻¹) depending on temperature, humidity, and frequency. Then A_eff = Σ(αᵢSᵢ) + 4mV.
  • Optimum reverberation time: The range of T₆₀ that gives best clarity for the use of the room.
    • Speech (classroom/lecture hall): roughly 0.8–1.2 s (smaller volumes toward 0.7–0.9 s).
    • Orchestral music (concert hall): roughly 1.6–2.0 s (depending on volume and style).
    • Multipurpose auditoriums compromise, e.g., 1.2–1.6 s, possibly with variable acoustics.
  • Echo: a distinct, delayed repetition of sound. For speech, delays ≥ 0.1 s are perceived as separate echoes. For a single reflecting wall, this implies source–wall distance ≥ ≈17 m (because the extra path for the round trip is ≈34 m at 343 m·s⁻¹ for 0.1 s).
  • Flutter echo: a rapid “ping-pong” between two parallel hard surfaces due to repeated specular reflections; treated by breaking parallelism or adding diffusion/absorption.
  • Focusing and dead spots: concave surfaces can focus sound, causing hot spots and quiet areas. Avoid large concave domes or use diffusers.
  • Sound insulation (airborne): preventing sound transmission from one room to another through partitions.
    • Transmission coefficient (τ): ratio of transmitted to incident sound power.
    • Transmission loss (TL): TL = 10·log₁₀(1/τ) dB. Higher TL means better insulation.
    • Mass law: For a limp, non-resonant, infinite panel away from coincidence, TL ≈ 20·log₁₀(m′·f) − 47 (m′ in kg·m⁻², f in Hz) in SI. Shows +6 dB per octave slope.
    • Noise reduction (NR): the difference in sound levels between source and receiving rooms due to partition and room absorption: NR = L₁ − L₂. In diffuse-field rooms without significant flanking, NR ≈ TL + 10·log₁₀(S/A₂), where S is the common partition area (m²) and A₂ is the equivalent absorption area of the receiving room (m² sabin).
  • Absorbers (room treatment):
    • Porous absorbers (mineral wool, fiberglass, acoustic foam, heavy curtains): convert sound energy to heat via viscous/thermal losses; best at mid-high frequencies; thickness ≳ 1/4 wavelength improves low-frequency absorption; spacing off the wall helps.
    • Resonant absorbers (panel or Helmholtz resonators): tuned to absorb low frequencies near their resonance; narrow bandwidth.
    • Diffusers (irregular surfaces, QRD/PRD types): scatter sound to improve diffusion and reduce flutter/focusing without over-deadening.
  • Background noise criteria: In quiet lecture rooms, aim for background noise around 30–35 dBA; in auditoriums, 25–30 dBA for high quality. HVAC design and isolation must meet these.
  • Basic design procedure for a lecture hall/auditorium:
    1. Determine room volume V and intended use (target T₆₀ range by use and volume).
    2. Estimate existing equivalent absorption A₀ from surfaces and seats (occupied seating has significant absorption per person).
    3. Use T₆₀ target to compute required A_req = 0.161·V / T₆₀ (Sabine) or Eyring.
    4. Add/select materials to supply ΔA = A_req − A₀, distributed on appropriate surfaces to avoid echoes and flutter. Check low-frequency issues (consider resonant absorbers if needed).
    5. Shape: avoid large parallel walls (or add diffusers), avoid concave focusing; ensure good coverage from loudspeakers if using reinforcement.
    6. Control noise: treat ducts, use silencers, select low-noise HVAC, ensure sufficient transmission loss of partitions.

Diagrams you may be asked to draw (describe what to draw and label):

  • Sabine’s derivation model: a rectangular room; indicate volume V, total surface S, a small sound ray striking a wall; annotate mean free path ℓ̄ = 4V/S and collision rate cS/4V.
  • Echo path: plan view of a room with source S, listener L, a reflecting back wall; draw direct SL and reflected S–wall–L path; show extra path Δ and time delay Δ/c; mark condition Δ ≥ 34 m for 0.1 s.
  • Flutter echo: two parallel walls with multiple back-and-forth specular rays; show treatment by adding tilted panel or absorber on one wall.
  • Transmission loss concept: incident sound on a wall, small transmitted wave; label incident power W_i, transmitted W_t, τ = W_t/W_i, TL.

Formulas

Lp = 20·log10(p_rms / p0) dB — p_rms: sound pressure (Pa), p0 = 20 µPa; use for pressure-based levels in air. LI = 10·log10(I / I0) dB — I: intensity (W·m⁻²), I0 = 1×10⁻¹² W·m⁻²; use for intensity levels. A = Σ(αi·Si) — A: equivalent absorption area (m² sabin); αi: absorption coefficient (–); Si: area (m²); sum over all surfaces. ℓ̄ = 4·V / S — ℓ̄: mean free path (m); V: volume (m³); S: total surface area (m²); diffuse field. T60 (Sabine) = 0.161·V / A — T60: reverberation time (s); V (m³); A (m² sabin); valid for low average absorption, diffuse field. T60 (Eyring) = 0.161·V / [−S·ln(1 − ᾱ)] — S: total area (m²); ᾱ = (ΣαiSi)/S (–); better when ᾱ is not small. Aeff = Σ(αi·Si) + 4·m·V — air absorption correction; m: air attenuation coefficient (m⁻¹), frequency and humidity dependent. TL = 10·log10(1 / τ) dB — TL: transmission loss (dB); τ: transmission coefficient (–); airborne insulation metric. Mass law: TL ≈ 20·log10(m′·f) − 47 dB — m′: surface density (kg·m⁻²); f: frequency (Hz); valid for limp, large panel away from resonances and coincidence. NR = L1 − L2 dB — noise reduction between source and receiving rooms; L1, L2: sound levels (dB). Echo delay Δt = Δ/c — Δt: delay (s); Δ: extra path length (m); c: speed of sound (≈343 m·s⁻¹ at 20°C). Average absorption ᾱ = A / S — A (m² sabin); S (m²); use to switch between Sabine and Eyring.

Worked examples

  1. Reverberation time of a furnished lecture room Given: Room 15 m × 10 m × 4 m (volume V = 600 m³). Surfaces and absorption at 1 kHz:
  • Floor: carpet on concrete, area S_f = 15×10 = 150 m², α_f = 0.4
  • Ceiling: plaster, S_c = 150 m², α_c = 0.05
  • Long walls: 2 × (15×4) = 120 m², painted brick, α_w1 = 0.03
  • Short walls: 2 × (10×4) = 80 m², painted brick, α_w2 = 0.03
  • Seating: 100 occupied seats, equivalent absorption per occupied seat A_seat = 0.5 m² sabin at 1 kHz Steps:
  1. Compute total absorption A = Σ(αi·Si) + A_seating.

    • Floor: 0.4×150 = 60 m² sabin
    • Ceiling: 0.05×150 = 7.5 m² sabin
    • Long walls: 0.03×120 = 3.6 m² sabin
    • Short walls: 0.03×80 = 2.4 m² sabin
    • Seating: 100×0.5 = 50 m² sabin Total A = 60 + 7.5 + 3.6 + 2.4 + 50 = 123.5 m² sabin
  2. Apply Sabine: T60 = 0.161·V/A = 0.161×600 / 123.5 s

  3. Numbers: 0.161×600 = 96.6; T60 = 96.6 / 123.5 = 0.783 s Answer: The reverberation time at 1 kHz is approximately 0.78 s (speech-friendly). Final: T60 ≈ 0.78 s.

  4. How much extra absorption is needed to meet a target T60? Given: A music rehearsal room with V = 200 m³. Current surfaces give A₀ = 20 m² sabin at mid-frequencies. Target T60 = 1.0 s (for clarity in rehearsals). Steps:

  5. Required total absorption from Sabine: A_req = 0.161·V / T60 = 0.161×200 / 1.0 = 32.2 m² sabin

  6. Additional absorption needed: ΔA = A_req − A₀ = 32.2 − 20 = 12.2 m² sabin

  7. If you use wall panels with α ≈ 0.8 at mid-band, area needed S_panels = ΔA/α = 12.2 / 0.8 = 15.25 m² Answer: Provide about 12.2 m² sabin extra absorption, e.g., roughly 15–16 m² of α ≈ 0.8 panels.

  8. Transmission loss by mass law and expected noise reduction Given: Single-leaf concrete wall, thickness 150 mm, density 2400 kg·m⁻³. Frequency f = 500 Hz. Receiving room and source room have similar absorption (simplify to NR ≈ TL for a first estimate; acceptable for exam unless room corrections are asked). Steps:

  9. Surface density m′ = density × thickness = 2400 kg·m⁻³ × 0.150 m = 360 kg·m⁻²

  10. Mass law TL ≈ 20·log10(m′·f) − 47 dB. Compute m′·f = 360×500 = 180,000

  11. log10(180,000) = log10(1.8×10⁵) = log10(1.8) + 5 ≈ 0.2553 + 5 = 5.2553

  12. 20×5.2553 = 105.106; TL ≈ 105.106 − 47 = 58.106 dB Answer: The wall’s mass-law TL at 500 Hz is about 58 dB (actual NR can be lower due to flanking, stiffness, and room effects).

Common mistakes

  • Mixing units (feet/inches vs meters) or forgetting that A is in m² sabin (not in m² unless α = 1).
  • Assuming TL = NR without checking receiving-room absorption and area; exam problems may require the full partition equation.
  • Ignoring low-frequency problems; porous panels thin compared to wavelength do little below ~250 Hz unless spaced off the wall or combined with resonant absorbers.
  • Not checking for echoes and flutter; parallel hard walls can ruin speech clarity even if T60 is correct.
  • Quoting a single T60 “best value” for all rooms; optimum depends on use and volume.

For your exam

  • Derivations: Sabine’s formula (full steps and assumptions), Eyring formula (state and outline derivation idea with −ln(1−ᾱ)), mean free path ℓ̄ = 4V/S.
  • Numericals: compute T60 from surface data; find added absorption for a target T60; TL by mass law; echo distance/time delay.
  • Diagrams: room ray model for Sabine; echo path diagram; parallel-wall flutter; TL concept with τ and TL.
  • What full-marks answers include: clear statement of the law/formula with assumptions, neat labeled diagrams as asked, correct units and clear steps, and a sentence interpreting the numerical result (e.g., “suitable for speech”).

Quick check

  1. A 12 m × 8 m × 3 m classroom has A = 48 m² sabin. What is T60 by Sabine?
  2. Name two ways to prevent flutter echo in a corridor with parallel hard walls.
  3. Using mass law, how much does TL increase if frequency doubles (all else constant)? Answers: 1) V = 288 m³, T60 = 0.161×288/48 = 0.966 s. 2) Add absorptive panels; make one wall non-parallel or add diffusers. 3) About +6 dB per octave (since TL ∝ 20·log10 f).

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?