How to Write an Acoustic Specification — Complete Guide for Architects | AcousPlan

Why Most Acoustic Specifications Fail

Acoustic specifications fail in one of two ways. The first is the brand-only specification: "Acoustic ceiling tiles: Ecophon Focus C, 600×600, installed to manufacturer instructions." This appears specific but provides no performance obligation. If Ecophon Focus C is substituted for a lesser product on site, the specification has not been breached. The performance consequences — higher RT60, poor speech intelligibility, failed compliance testing — become a post-occupancy dispute.

The second failure mode is the performance-without-verification specification: "Reverberation time in all open-plan offices shall not exceed 0.6 seconds." This is a valid performance target, but without a measurement clause, a testing protocol, a pass/fail threshold, and a remediation obligation, it is unenforceable. The contractor can claim the target was met without ever measuring it.

A good acoustic specification is both. It defines performance targets tied to measurable standards, specifies products with reference to test data, and includes a commissioning and verification regime that creates enforceable obligations.

This guide walks through each section of a complete acoustic specification document.

Section 1: Project Description and Scope

The opening section establishes the acoustic brief for the project. It should include:

1.1 Project overview State the building type, intended occupancy, and the acoustic standards that govern the project. For example: "This is a new-build primary school designed to meet Building Bulletin 93:2015 (BB93). All acoustic performance targets in this specification are derived from BB93 unless stated otherwise."

1.2 Scope of acoustic works List the acoustic elements within the contractor's scope: suspended acoustic ceilings, acoustic wall panels, acoustic flooring, partitions with sound insulation requirements, acoustic doors, and mechanical acoustic isolation. Be explicit about what is excluded (structural isolation for mechanical plant, for example, may be a separate specialist contract).

1.3 Reference documents List all standards, codes, and guidelines that the specification draws from. Include version/year. Never reference "current edition" — standards are updated and what is current at tender may not be current at construction.

Typical reference documents for a UK school project:

• BB93:2015 — Acoustic Design of Schools
• BS EN ISO 16283-1:2014+A1:2017 — Field measurement of airborne sound insulation
• BS EN ISO 16283-2:2015 — Field measurement of impact sound insulation
• BS EN ISO 3382-2:2008 — Measurement of room acoustic parameters in ordinary rooms
• BS EN ISO 354:2003 — Measurement of sound absorption in a reverberation room
• BS EN ISO 11654:1997 — Sound absorbers in buildings (αw classification)
• BS EN 13501-1 — Fire classification of construction products

Section 2: Acoustic Performance Targets

This is the core of the specification. For each space type in the building, define the targets. Present them in a table — it is easier to review and impossible to misread in the way that prose can be.

2.1 Reverberation Time (RT60)

RT60 is measured as T20 or T30 per ISO 3382-2 (the 20 dB or 30 dB decay slope extrapolated to 60 dB). State which measurement parameter you require: T20, T30, or EDT. For most compliance purposes, T20 or T30 is used.

Note: RT60 targets should be stated for the unoccupied condition. Occupied RT60 will be lower (people absorb sound) and is generally not tested. Some specifications for concert halls or performance spaces require occupied measurements — state this explicitly.

2.2 Background Noise Level

Background noise level is measured as the A-weighted steady-state level (dBA) with all services operating (HVAC, ventilation, lighting) but without occupancy. It is measured per BB93:2015 Appendix A or BS 8233:2014 Annex A.

2.3 Sound Insulation Between Spaces

Airborne sound insulation is measured as DnT,w (field standardised level difference, weighted) per BS EN ISO 16283-1. Impact sound insulation is measured as L'nT,w per ISO 16283-2.

2.4 Speech Intelligibility (where required)

Speech intelligibility is specified as STI (Speech Transmission Index) per IEC 60268-16:2020, or as STI-PA (STI using the Public Address method) for amplified speech in auditoria and halls.

Section 3: Product Specifications

3.1 Performance-Based Product Specification

Specify by performance criteria, not by brand alone. The specification should read: "Acoustic ceiling tiles shall achieve the following performance requirements: [table of requirements]." Then: "The following products are pre-approved: [list]. Alternatives may be proposed subject to the substitution procedure in Clause 3.3."

3.2 Required Documentation for All Acoustic Products

For every acoustic product, require:

For sound absorbers (panels, tiles, baffles):

• ISO 354 test report from an accredited laboratory (UKAS or equivalent)
• αw value and sound absorption class (A–E per ISO 11654)
• Absorption coefficients at 125, 250, 500, 1000, 2000, 4000 Hz
• Fire classification certificate (BS EN 13501-1 or BS 476 Part 6/7)
• Installation method statement

• Rw test report from an accredited laboratory (ISO 10140-2 for laboratory measurement)
• Or Robust Details (RD) reference for approved standard constructions
• Fire resistance certificate where required

• Rw test report including door, frame, and seals as a complete assembly
• Brand and series of acoustic seals (do not allow seal substitution without test data)

3.3 Substitution Procedure

Define a formal substitution procedure:

"Any product substitution requires written application to the architect/acoustic consultant at least 28 days before the proposed installation date. The application shall include: (a) ISO 354 or ISO 10140 test report for the proposed alternative, from an UKAS-accredited laboratory; (b) fire classification certificate; (c) statement of equivalence demonstrating the proposed product meets all performance requirements in Clause 2. The architect/acoustic consultant shall respond within 14 days. Approval is valid only for the specified product batch — re-approval is required if the product specification changes."

Section 4: Acoustic Commissioning and Testing

This section is most commonly omitted or inadequate, and most commonly disputed.

4.1 Pre-Completion Testing

Specify the testing required before practical completion sign-off:

• RT60 measurements per ISO 3382-2 in all spaces with RT60 performance requirements
• Background noise measurements with all services operating in all spaces with noise targets
• Airborne sound insulation measurements (DnT,w) at minimum two locations per interface type
• Impact sound insulation measurements (L'nT,w) at minimum one location per interface type
• STI measurements (where specified)

4.2 Testing Protocol

State that testing must be conducted by an independent UKAS-accredited acoustic consultant, not by the main contractor or the acoustic product supplier. Define the measurement equipment requirements: Type 1 sound level meter (IEC 61672), calibrated within the previous 12 months.

4.3 Pass/Fail Criteria and Tolerance

Define explicitly:

"RT60 measurements shall be taken at a minimum of three source positions and three receiver positions per measurement space, with the values averaged. The result shall not exceed the target value in Clause 2.1 by more than 0.1 seconds in any octave band from 250 Hz to 2000 Hz."

"DnT,w results shall be within +1 dB of the target value (i.e., failure is defined as a measurement more than 1 dB below the target)."

The tolerance is important. ISO 16283-1 measurement uncertainty is typically ±1–2 dB. A zero-tolerance spec creates unresolvable disputes. A ±1 dB tolerance on DnT,w is standard professional practice.

4.4 Remediation Obligations

"If any measurement fails the criteria in Clause 4.3, the contractor shall investigate the cause, propose a remediation scheme, and retest within 28 days at the contractor's cost. The architect/acoustic consultant shall confirm in writing that the proposed remediation is acceptable before work commences. A further failed test obliges the contractor to engage an independent acoustic specialist to identify the failure mechanism."

Section 5: Practical Tips for Specification Writers

Use the NBS format: National Building Specification (NBS) provides an acoustic specification template (Section K13 for suspended ceilings, K31 for plasterboard partitions) that is recognised by UK contractors and integrates with BIM workflows.

Include acoustic performance in the BIM execution plan: If the project is delivered with BIM, require that acoustic performance data (RT60 targets, sound insulation requirements) is captured in the BIM model as room data requirements. This creates a live connection between design and specification.

Require acoustic modelling at RIBA Stage 3: Specification writers should not be working from memory. At Stage 3, use acoustic modelling software (or AcousPlan's calculator) to verify that the proposed construction achieves the targets. If it does not, revise the construction specification before tender — not after failed commissioning tests.

Do not mix up Rw and DnT,w: Rw is a laboratory measurement of a construction element in isolation. DnT,w is a field measurement of an installed partition in a real building. Field values are typically 3–8 dB lower than laboratory values due to flanking transmission and workmanship. Specify DnT,w for field verification targets, and Rw for product selection. Do not require DnT,w to equal Rw — it will not.

State measurement conditions: "All RT60 measurements shall be conducted in the unoccupied condition, with heating, ventilation, and air conditioning systems operating in normal mode." Ambiguity about measurement conditions is a common source of disputes.

A well-written acoustic specification is a relatively short document — typically 8–15 pages. Its value is not length but precision: every claim is testable, every product is defined by data, and every obligation has a consequence. That is the standard to write to.