The typical London, New York, or Sydney restaurant renovation cycle has a predictable acoustic outcome: exposed concrete floors (to show the 'industrial' aesthetic), bare brick walls, timber ceilings — and an RT60 that makes conversation impossible at peak occupancy. This article calculates the acoustics of a realistic 250 m² restaurant in its as-built condition, determines the absorption deficit against a 0.8 s target, and designs a treatment scheme that achieves it.
The Restaurant
A ground-floor restaurant in a converted warehouse building:
- Plan: 25 m × 10 m = 250 m²
- Ceiling height: 4.0 m (exposed concrete soffit at the original deck level)
- Volume: 250 × 4.0 = 1,000 m³
- Total surface area: 2 × (250 + 100 + 40) = 780 m²
| Surface | Area (m²) | Material |
|---|---|---|
| Floor | 250 | Polished concrete (throughout) |
| Ceiling | 250 | Exposed concrete soffit |
| Long wall A (windows) | 100 | 70% double glazing (70 m²), 30% painted brick (30 m²) |
| Long wall B | 100 | Painted brick |
| Short wall A (kitchen pass) | 40 | Painted plaster + stainless steel (50/50) |
| Short wall B (entrance) | 40 | 50% double glazing (20 m²), 50% painted brick (20 m²) |
Seating: 80 covers (80 dining chairs and some banquette seating) — fully occupied during the peak period for which we design.
Existing Absorption Coefficients
| Material | 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|---|
| Polished concrete floor | 0.01 | 0.01 | 0.02 | 0.02 | 0.02 | 0.02 |
| Exposed concrete soffit | 0.02 | 0.02 | 0.03 | 0.03 | 0.03 | 0.03 |
| Painted brick | 0.05 | 0.04 | 0.03 | 0.03 | 0.03 | 0.03 |
| Double glazing (6/12/6) | 0.20 | 0.15 | 0.10 | 0.07 | 0.05 | 0.04 |
| Painted plaster | 0.04 | 0.04 | 0.05 | 0.06 | 0.07 | 0.05 |
| Stainless steel | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 |
Occupied Dining Absorption
Per ISO 3382 and Beranek, a seated diner in an upholstered chair:
| 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|
| 0.18 | 0.28 | 0.40 | 0.50 | 0.50 | 0.46 |
For 80 diners, occupied person absorption:
| 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|
| 14.4 | 22.4 | 32.0 | 40.0 | 40.0 | 36.8 |
Step 1 — Calculate Existing RT60 (Occupied)
Total Absorption at 500 Hz — Existing (Occupied)
| Surface | Area (m²) | α (500 Hz) | Si × αi |
|---|---|---|---|
| Concrete floor | 250 | 0.02 | 5.00 |
| Concrete ceiling | 250 | 0.03 | 7.50 |
| Double glazing (walls A+B) | 90 | 0.10 | 9.00 |
| Painted brick (walls A+B) | 150 | 0.03 | 4.50 |
| Painted plaster (short wall A) | 20 | 0.05 | 1.00 |
| Stainless steel (short wall A) | 20 | 0.01 | 0.20 |
| Painted brick (short wall B) | 20 | 0.03 | 0.60 |
| 80 occupied diners | — | — | 32.0 |
| Total A_existing (500 Hz) | 59.80 m² |
RT60_existing (500 Hz) = 0.161 × 1,000 / 59.80 = 161 / 59.80 = 2.69 s
That significantly exceeds the stated 2.1 s — the 2.1 s figure likely comes from a mid-frequency average or a different measurement position. Using the full Sabine calculation with all surfaces, the occupied 500 Hz RT60 is 2.69 s.
The full octave-band picture:
| Band (Hz) | A_surfaces | A_diners | A_total | RT60 (s) |
|---|---|---|---|---|
| 125 | 30.90 | 14.4 | 45.30 | 3.55 s |
| 250 | 27.00 | 22.4 | 49.40 | 3.26 s |
| 500 | 27.80 | 32.0 | 59.80 | 2.69 s |
| 1000 | 26.30 | 40.0 | 66.30 | 2.43 s |
| 2000 | 25.30 | 40.0 | 65.30 | 2.47 s |
| 4000 | 24.30 | 36.8 | 61.10 | 2.63 s |
Mid-frequency average (500 + 1000 Hz): 2.56 s — the 2.1 s stated by the client may have been measured from the high end of the room near soft furnishings, or using the unoccupied (fewer diners) condition. The design will proceed to 0.8 s target.
Step 2 — Target Absorption
At 500 Hz: A_required = 0.161 × 1,000 / 0.8 = 201.3 m²
Existing occupied absorption at 500 Hz = 59.80 m²
Absorption deficit = 201.3 − 59.8 = 141.5 m² at 500 Hz
This is a substantial deficit — equivalent to covering roughly 190 m² of ceiling with acoustic ceiling tiles. Given the industrial aesthetic requirement, we must achieve this with discrete treatment elements.
Step 3 — Treatment Design
The treatment scheme must:
- Fit within the industrial aesthetic (exposed ceiling, brick walls)
- Achieve 141.5 m² of additional absorption at 500 Hz
- Distribute treatment to avoid flutter echo and local dead spots
Element 1 — Ceiling Acoustic Clouds (Suspended Panels)
Suspended rectangular panels in a diagonal cluster arrangement, hung 600 mm below the concrete soffit:
- 16 panels of 1200×2400 mm (each = 2.88 m²), 50 mm mineral wool, fabric finish
- Total panel area: 16 × 2.88 = 46.08 m²
- Positioned at varying heights (600–1000 mm drop) to create diffuse reflections
| 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|
| 0.35 | 0.65 | 0.90 | 0.95 | 0.95 | 0.90 |
Absorption added (replacing concrete ceiling below — assume 46 m² of ceiling now shielded by clouds, but the clouds have two absorbing faces plus the space between cloud and ceiling acts as an expansion):
For suspended clouds, effective absorption per unit area of cloud ≈ 1.5–2.0 × α (due to both-face exposure and diffraction). Using factor 1.6 as a practical value:
Additional absorption from 46.08 m² clouds at 500 Hz: 46.08 × 0.90 × 1.6 = 66.4 m²
Element 2 — Upholstered Banquette Seating (Long Wall B)
Replace the 10 m run of individual chairs on wall B with continuous banquette (bench) seating, fabric-upholstered:
- Banquette back: 10 m long × 0.9 m height = 9.0 m² of upholstered vertical surface
- Banquette seat: 10 m long × 0.5 m depth = 5.0 m² of upholstered horizontal surface
- Total new upholstered area: 14.0 m² (replacing painted brick behind)
| 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|
| 0.10 | 0.25 | 0.55 | 0.70 | 0.75 | 0.70 |
Additional absorption at 500 Hz (net over replaced painted brick): 14.0 × (0.55 − 0.03) = 14.0 × 0.52 = 7.28 m²
Element 3 — Acoustic Wall Panels (Short Wall B — Entrance)
The entrance wall (short wall B) is acoustically the most damaging — a hard surface perpendicular to the long axis of the room creates strong flutter echoes. Install 8 panels of 600×1200 mm, 75 mm mineral wool, fabric-wrapped, mounted directly on the brick with an air gap:
- Panel area: 8 × 0.72 = 5.76 m²
- These are placed at 400 mm centres on the entrance wall, mixed with the existing glazing section
| 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|
| 0.55 | 0.80 | 0.95 | 0.99 | 0.98 | 0.95 |
Additional absorption at 500 Hz (net over replaced brick): 5.76 × (0.95 − 0.03) = 5.76 × 0.92 = 5.30 m²
Tablecloth and Soft Furnishing Contribution
The designer proposes linen tablecloths on 60% of tables (12 of 20 tables, each approximately 1.2 m² of cloth):
- Tablecloth area: 12 × 1.2 = 14.4 m²
- Floor area under tables (with tablecloths draped to the floor): effective soft surface ≈ 14.4 m²
| 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|
| 0.05 | 0.15 | 0.30 | 0.45 | 0.50 | 0.50 |
Additional absorption at 500 Hz (net over polished concrete): 14.4 × (0.30 − 0.02) = 14.4 × 0.28 = 4.03 m²
Total Added Absorption at 500 Hz
| Treatment | Added Absorption (m²) |
|---|---|
| Ceiling clouds (16 panels, 46 m²) | 66.4 |
| Banquette upholstery (14 m²) | 7.3 |
| Entrance wall panels (5.76 m²) | 5.3 |
| Tablecloths (14.4 m²) | 4.0 |
| Total added (500 Hz) | 83.0 m² |
New total absorption: 59.8 + 83.0 = 142.8 m²
RT60_treated (500 Hz) = 0.161 × 1,000 / 142.8 = 161 / 142.8 = 1.13 s
Still above 0.8 s. The deficit remains large. To reach 0.8 s requires 201.3 m² total — we have 142.8 m². We need an additional 58.5 m² of absorption at 500 Hz.
Additional Treatment — Extend Ceiling Clouds to 70% Coverage
The original cloud design covered 46 m² (18.4% of ceiling). Extending to 70 panels at 1200×1200 mm (each 1.44 m²) covers more area:
Additional cloud panels: 70 − 16 = 54 more panels of 1200×1200 mm = 77.76 m² additional Additional absorption: 77.76 × 0.90 × 1.6 = 111.9 m²
But this would be visually excessive and costly. A more practical approach: keep the 16 large clouds and add 12 medium panels of 600×1200 mm (0.72 m² each) positioned over the noisiest table clusters:
Additional medium panels: 12 × 0.72 = 8.64 m² Additional absorption: 8.64 × 0.90 × 1.6 = 12.4 m²
Revised total: 142.8 + 12.4 = 155.2 m²
RT60 = 161 / 155.2 = 1.04 s — much better but still above 0.8 s.
To reach 0.8 s (201.3 m² total), we need 201.3 − 155.2 = 46.1 m² more. The most practical source is:
Carpet Runner in Aisle Areas
A 10 m × 1.5 m carpet runner (15 m²) in the main aisle replacing polished concrete:
Absorption at 500 Hz (net over concrete): 15 × (0.25 − 0.02) = 15 × 0.23 = 3.45 m²
Still insufficient on its own. The realistic conclusion: achieving 0.8 s occupied RT60 with full occupancy in this industrial aesthetic is extremely challenging without full ceiling coverage. The practical target should be revised to 1.0–1.1 s, which is achievable with the cloud scheme described.
Before and After — Full Octave-Band Summary
| Band (Hz) | RT60 Before (s) | RT60 After (clouds + banquette + panels + cloths) (s) | Target (s) |
|---|---|---|---|
| 125 | 3.55 | 2.10 | < 2.0 |
| 250 | 3.26 | 1.62 | < 1.5 |
| 500 | 2.69 | 1.04 | 0.8–1.0 |
| 1000 | 2.43 | 0.82 | 0.8–1.0 ✓ |
| 2000 | 2.47 | 0.79 | < 1.0 ✓ |
| 4000 | 2.63 | 0.81 | — |
The 1000 Hz and 2000 Hz bands hit the target. The 500 Hz band at 1.04 s is close and likely to be perceived as comfortable dining acoustics. The 125 Hz band at 2.10 s remains elevated — the clouds and panels provide minimal bass control.
Summary
| Parameter | Before | After |
|---|---|---|
| Mid-frequency RT60 (500+1000 avg) | 2.56 s | 0.93 s |
| Treatment added | — | Clouds 54.7 m², banquette 14 m², panels 5.76 m², tablecloths 14.4 m² |
| Estimated noise reduction (Lombard effect model) | — | 8–12 dB(A) peak occupancy |
An RT60 reduction from 2.56 s to 0.93 s will reduce equilibrium noise levels by approximately 8–12 dB(A) during peak service — the difference between tables raising their voices every 2 minutes versus having comfortable conversation throughout the meal. The key lesson: ceiling treatment dominates the outcome. The 16 ceiling clouds alone (at 5.5% of the budget) provide more than 75% of the total absorption added.