Approved Document E Acoustics — What Part E Actually Requires for New Buildings

The Regulation That Generates 200,000 Sound Tests Per Year

Every year, approximately 200,000 pre-completion sound insulation tests are carried out in England and Wales under the requirements of Approved Document E (ADE) of the Building Regulations. Each test costs £400–800 per element, adding £2,000–5,000 to the completion cost of a typical residential project. When a separating wall or floor fails, the remedial works average £3,000–8,000 per element — and the property cannot be signed off for occupation until the retest passes.

ADE was first introduced in 1985 with relatively modest requirements. The 2003 revision (effective July 2004) substantially tightened the performance standards and introduced mandatory pre-completion testing (PCT) for the first time. The 2015 amendment updated references and clarified certain construction details but left the performance thresholds unchanged. Today, ADE 2003 (amended 2015) remains the governing document for sound insulation in residential buildings in England and Wales.

This guide covers every requirement in ADE, explains the two compliance routes (pre-completion testing and Robust Standard Details), and includes a worked example for the most common test scenario: a party wall in a new-build terraced house.

What ADE Actually Covers

ADE addresses four types of sound transmission in dwellings:

1. Airborne sound insulation through separating walls and floors between dwellings
2. Impact sound insulation through separating floors between dwellings
3. Internal sound insulation within a dwelling (walls and floors between rooms)
4. Reverberation in common areas (corridors, stairways, hallways)

Airborne Sound Insulation: The Numbers

ADE Table 0.1 specifies the minimum airborne sound insulation for separating elements between dwellings. The metric is DnT,w + Ctr — the standardised level difference (weighted) with the spectrum adaptation term for low-frequency performance. The Ctr term penalises constructions that perform poorly at low frequencies (below 200 Hz), which is where most subjective noise complaints originate (bass from music, television, footsteps).

New-Build Requirements

Conversion Requirements

The conversion requirements are 2 dB more lenient than new-build, recognising the practical constraints of upgrading existing structures. However, 43 dB is still demanding — many Victorian and Edwardian conversions struggle to achieve this without significant intervention (independent leaf wall linings, floating floors, upgraded ceiling treatments).

Why DnT,w + Ctr, Not Just Rw

The choice of DnT,w + Ctr as the performance metric (rather than the simpler Rw laboratory rating) is deliberate and important:

DnT,w is the standardised level difference, measured in situ per ISO 16283-1. It captures the actual performance of the separating element as installed, including flanking transmission through junctions, bypassing through services, and the effect of room size and absorption. This is always lower than the laboratory Rw of the wall or floor construction.

Ctr is the spectrum adaptation term per ISO 717-1, calculated from the difference between the reference curve and a pink noise spectrum weighted by a low-frequency traffic noise spectrum. Constructions with poor low-frequency performance (lightweight frames, thin plasterboard leaves) receive a negative Ctr value of -3 to -8 dB, while heavyweight constructions (dense blockwork, concrete) typically have Ctr values of -1 to -3 dB.

The practical consequence: a lightweight steel-frame party wall might achieve DnT,w = 52 dB but DnT,w + Ctr = 44 dB — a 1 dB fail. The same wall with a dense blockwork core might achieve DnT,w = 50 dB but DnT,w + Ctr = 47 dB — a clear pass. The Ctr term systematically favours mass-based construction at low frequencies.

Impact Sound Insulation

Impact sound insulation applies only to separating floors (not walls). The metric is L'nT,w — the standardised impact sound pressure level (weighted) per ISO 16283-2 and ISO 717-2.

Requirements

Note that impact sound is measured as a level in the receiving room, so lower is better (the opposite convention from airborne insulation where higher is better). An L'nT,w of 62 dB means that the standardised impact noise level in the room below, when the standard tapping machine operates on the floor above, does not exceed 62 dB.

The standard tapping machine per ISO 10140-3 consists of five steel hammers of 500 g each, dropping from a height of 40 mm at a rate of 10 impacts per second. It is a deliberately harsh test that simulates hard-heeled footfall on an uncarpeted floor. Real-world impact noise from residents walking in socks or soft shoes is typically 10–15 dB lower than the tapping machine.

Achieving L'nT,w ≤ 62 dB requires a resilient layer in the floor construction. For concrete floors, a floating screed on a 25 mm resilient layer (mineral wool, recycled rubber, or foam) is standard. For timber floors, a combination of resilient bars on the ceiling below, mineral wool infill between joists, and a resilient overlay or floating platform on the floor above is typically needed.

The Two Compliance Routes

ADE provides two distinct routes to demonstrate compliance: pre-completion testing (PCT) and Robust Standard Details (RSDs). Most projects use one or the other; some use both.

Route 1: Pre-Completion Testing (PCT)

PCT is the default compliance route. For each group of dwellings of the same construction type, a sample of separating elements must be tested by a registered testing body (a member of the Association of Noise Consultants or equivalent).

The sampling rate is specified in ADE §0.2:

• 1 set of tests for the first pair of dwellings of each construction type
• 1 additional set for every 10 further pairs of the same type
• Minimum 1 test per element type (wall, floor) per construction type

If any test fails, ADE requires remedial works and retesting. There is no averaging — every individual test must pass. This is the provision that causes the most cost and delay in residential construction.

Route 2: Robust Standard Details (RSDs)

RSDs are pre-tested construction details registered with Robust Details Ltd (RDL). If a developer uses an approved RSD, the project is exempt from pre-completion testing, provided:

1. The developer registers the site with RDL before construction begins
2. The construction follows the registered detail exactly (no substitutions)
3. An RDL inspector visits the site during construction to verify compliance
4. The developer pays a registration fee per plot (currently £90 per separating element)

The registration fee per element is significantly cheaper than the cost of PCT (£400–800 per test), making RSDs the preferred route for volume housebuilders. However, RSDs require strict adherence to the specified construction detail — any deviation (different blockwork density, different plasterboard thickness, missing cavity sock, etc.) voids the RSD exemption and triggers the need for PCT.

Worked Example: Party Wall in New-Build Terraced House

The Project

A terrace of 12 houses in south-east England. Three-storey construction. Each house is 5.5 m wide × 9.0 m deep. Party walls are full-height separating walls between each pair of dwellings. The developer has chosen PCT as the compliance route.

Party Wall Construction

The specified party wall is a masonry cavity wall:

• Two leaves of 100 mm dense aggregate blockwork (density ≥ 1,850 kg/m³)
• 75 mm cavity (minimum) with 25 mm wire-tied butterfly ties at 900 mm horizontal / 450 mm vertical centres
• 13 mm wet plaster finish to each face (or 12.5 mm plasterboard on dabs with all joints sealed)

Predicted Performance

The laboratory Rw for this construction (per BS EN ISO 10140) is approximately 55–58 dB, depending on blockwork density and plaster finish. The Ctr term for dense masonry is typically -2 to -3 dB.

In situ performance is reduced by flanking transmission. For a masonry terraced house, the primary flanking paths are:

• Through the external wall at the junction with the party wall
• Through the floor slab (if continuous across the party wall)
• Through the internal leaf at party wall junctions

DnT,w = 52 dB, Ctr = -3 dB → DnT,w + Ctr = 49 dB

This exceeds the ADE requirement of 45 dB by 4 dB — an adequate margin.

Critical Construction Details

The predicted performance depends on correct construction of several key junctions:

1. Party wall to external wall junction: The cavity in the party wall must be closed with a flexible closer (mineral wool cavity sock, not rigid blockwork) to prevent sound flanking around the end of the party wall through the external cavity. This single detail accounts for approximately 40% of party wall test failures.

2. Party wall to floor junction: If the floor structure is concrete, it should be discontinuous across the party wall (separate pour on each side, or a structural break at the wall line). If continuous, the flanking path through the slab reduces the effective DnT,w by 2–4 dB.

3. Services penetrations: Any pipe, cable, or duct passing through the party wall must be sealed with intumescent mastic or acoustic sealant. Back-to-back socket outlets on the party wall must be offset by at least 150 mm horizontally and the boxes sealed with acoustic putty pads.

4. Plaster finish: The wet plaster finish (or sealed plasterboard on dabs) is an essential component of the acoustic performance. Bare blockwork without a plaster finish loses approximately 5–6 dB of airborne insulation. This is the second most common cause of party wall test failures — walls that are "not yet plastered" at the time of testing.

Test Results

The acoustic consultant tests one party wall and one party floor per 10 pairs. For the first two houses, the results are:

All elements pass. The building control officer accepts the results and the properties can be occupied.

What If It Fails?

If a party wall test returns DnT,w + Ctr = 43 dB (a 2 dB fail), the remedial options are:

1. Independent wall lining: Install a 12.5 mm plasterboard layer on resilient bars (or proprietary resilient channel) with a 50 mm mineral wool infill on the failing side. This adds approximately 6–10 dB to the in situ performance. Cost: £25–35 per m² of wall area — approximately £1,500–2,500 for a full-height party wall in a typical house.

1. Investigate the flanking path: Before adding mass, identify the specific flanking route. A failed cavity sock at the external wall junction can reduce performance by 4–6 dB and can be repaired for a fraction of the cost of a wall lining — if it is accessible before the external wall is finished.

1. Re-plaster: If the blockwork was tested without plaster finish, applying 13 mm wet plaster and retesting may resolve the failure.

Internal Sound Insulation and Reverberation

ADE §3 addresses sound insulation within dwellings (between bedrooms, between living rooms and bedrooms, etc.) and reverberation in common areas.

Internal Partitions

ADE recommends (but does not mandate testing for) minimum Rw values for internal partitions:

These are design recommendations based on the laboratory Rw of the partition construction. They are not subject to PCT and are not enforced through building control. However, failure to meet these recommendations can result in post-occupancy complaints and claims under NHBC warranties.

Reverberation in Common Areas

ADE §7 specifies that common internal areas (corridors, hallways, stairwells) in buildings containing flats should have absorptive treatment on the ceiling or upper wall surfaces. The standard does not specify an RT60 target but requires that the mid-frequency absorption coefficient of the treatment is at least 0.3 when measured per ISO 354:2003.

In practice, this means an acoustic ceiling tile, a textured plaster finish with demonstrated absorption, or mineral wool boards on the corridor ceiling. The requirement exists because reverberant corridors amplify impact noise from doors closing and footfall on hard floors, which transmits through flat entrance doors into the dwelling.

ADE in Context: Comparison with European Standards

England and Wales have the most lenient residential sound insulation requirements in northern Europe. Scotland's Section 5 requirements are 11 dB stricter for airborne insulation and 6 dB stricter for impact — a significant difference that means constructions acceptable in England may fail in Scotland. The German, French, Dutch, and Swedish standards are all substantially stricter than ADE, reflecting different cultural expectations about acceptable noise between dwellings.

The Ctr term in the English requirement partially compensates for the lower absolute value by penalising lightweight constructions with poor low-frequency performance. However, the net effect is still that ADE is the least demanding residential acoustic standard in the comparison.

Key Takeaways

Approved Document E is the baseline acoustic regulation for residential construction in England and Wales. Its requirements — DnT,w + Ctr ≥ 45 dB airborne, L'nT,w ≤ 62 dB impact — are achievable with standard masonry and concrete construction but require careful attention to junction details, flanking paths, and service penetrations.

The two compliance routes (PCT and RSDs) offer different risk profiles: PCT catches construction errors but costs more and creates schedule risk; RSDs avoid testing but demand strict adherence to registered details with no flexibility for substitution.

For practitioners, the most effective strategy is to treat acoustic performance as a construction quality issue, not a design issue. The separating element constructions that achieve ADE compliance are well-established and reliable — failures almost always result from poor workmanship at junctions, missing cavity closers, unsealed service penetrations, or premature testing before plaster finishes are complete.

Related reading: BB93 school acoustic design guide | Understanding STC and Rw ratings | DIN 18041 vs BS 8233 vs ISO 3382

Check your wall or floor design: Use the AcousPlan sound insulation calculator to predict DnT,w for your separating element construction.