Approved Document E requires evidence of sound insulation performance of party walls and floors for: 

• Terraced and semi-detached houses – party walls 

• Flats – party walls and party floors 
• Rooms for residential purposes i.e. rooms that are slept/lived in but share facilities with other rooms e.g. hotels, care homes, student accommodation, hostels and houses of multiple occupation (HMOs). 

These could be newly built or created through a material change of use, for example, a shop or office converted into flats. 

There are 2 routes to compliance: 

1. Subscribe to use Robust Details (accredited wall and floor details that have been proven to achieve more than the requirements of Approved Document E. A fee is paid every time the details are used.
2. Pre-completion sound insulation tests on the walls and/or floors to ISO 140-4/7 

There are pros and cons to each approach, so select the route that is best suited to you. Robust Details mean that little to no site testing will be required, but not every eventuality can be covered by them. They are designed to comfortably exceed the requirements to allow some leeway for variation in site installation. This can mean they are more expensive to construct than what is necessary to meet the requirements when tested on-site. 

The arrangement of rooms along party walls may mean that it is not practicable to undertake sound testing. The requirements relate to party walls between ‘habitable’ spaces – these are rooms used for living, sleeping and eating, not bathrooms, toilets, circulation spaces or corridors. 

Approved Document E includes a sampling regime for houses, flats and rooms for residential purposes. The samples are based on a ‘set of tests’ for every 1 in 10 units: 

For houses, a set is 2 airborne sound insulation tests on party walls – so for every 10 units, 2 tests would be required (this also means 2 tests would be required for 2 units) 

For flats, a set is made up of 6 tests: 

• 2 airborne sound insulation tests on party walls 
• 2 airborne sound insulation tests on party floors 
• 2 impact sound insulation tests on party floors 

So for every 10 units, 6 tests would be required, depending upon the layout of the flats it may not be possible to complete all 6 tests on smaller developments. An example is if the flats are separated by a stairwell, so there are no party walls between the flats. 

If the flats are above/adjacent to commercial spaces such as shops, additional testing is required between the commercial areas and flats to confirm that the airborne sound insulation is sufficient.  

For rooms for residential purposes, a set is made up of 3 tests: 

• 1 airborne sound insulation test on party walls 

• 1 airborne sound insulation test on party floors 
• 1 impact sound insulation test on party floors 

So, for every 10 units, 3 tests would be required. 

The requirements are as follows: 

Sound is measured in decibels (dB). A whisper is about 30 dB, normal conversation is about 60 dB.  

The targets are based on setting a ‘reasonable’ level of sound insulation. For a normal conversation in a neighbouring property, sound insulation of 45 dB or more would reduce this to an imperceptible level. If the neighbours were watching a film at 80 dB, the noise is likely to be audible but drowned out (masked) by day-to-day noises in a property.  

The regulatory levels do not mean the wall or floor is soundproof. If the neighbours are having a loud party and you are trying to sleep then it will be disruptive, even if the wall/floor passed the test. 

Decibels are on a logarithmic scale which means that a 6 dB difference in sound pressure level is equivalent to a halving/doubling of the sound energy. This means that being 6 dB or more under or over the requirements represents a significant failure. 

The airborne test is based on making a loud noise (around 100 dB) using a loudspeaker emitting pink noise (a bit like TV static) on one side of the wall/floor. The sound level in the source room (where the loudspeaker is placed) is measured. With the loudspeaker in the same position at the same level, the sound level is measured in an adjacent room (the receive room) on the other side of the party wall/floor. Measurements are taken in different positions in both rooms and averaged to get an accurate picture of the sound fields. The difference (D) between sound pressure levels in the source and receive rooms is the reported sound insulation in decibels. This needs to be ABOVE the required level in Part E. 

The impact test is based on a ‘tapping machine’ which drops small weights onto the uncovered floor to represent footfall and stimulate structure-borne noise. In this case, the sound is only measured in the room below (the receive room). Again, multiple positions and readings are taken and averaged. The sound insulation level (L) in decibels needs to be BELOW the required level in Part E. 

Measurements are taken for reverberation (echo) and background which are factored into the calculation. Readings are weighted to reflect the human hearing range and perception of loudness. In addition, the airborne result includes corrections (Ctr) to reflect sensitivities to lower frequency noise. 

Unfortunately, sound insulation issues are unlikely to be a quick fix. The sound insulation properties of a wall or floor are based on a combination of the combined mass, stiffness, damping and critical/coincidence frequencies of the products/materials that make up the floor and wall. Sound insulation is affected by their relative arrangement, along with voids, insulation and isolation from each other and the surrounding structure. 

In instances of a specific problem, the test engineer may be able to hear (using their ear and the sound level meter) whether the noise transmission is greater in a part of the room. This may relate to an obvious feature e.g. a structural beam or column.  

Sometimes, it is clear that noise levels are greater close to the wall than in the middle of the room. This suggests that sound is flanking around the floor via the walls. If the noise is evenly spread it suggests there is an issue with the makeup of the floor or wall. The frequency of the noise can indicate whether the problem is linked to a lack of mass or another cause.  

Based on observations on site and the shape and levels of the graph in the report, it is often possible for an acoustic consultant to determine the issue and suggest remedial solutions. Where the problem is less evident, a separate visit to the site for further investigation may be required before proposing solutions. 

Typically, solutions are invasive and expensive and could mean adding ‘sacrificial’ walls or ceilings, which eat into valuable space. In some cases, it is necessary to remove existing walls and floors to replace the materials or reconfigure the arrangement. For this reason, we strongly advise that you consult your architect or an acoustic consultant to review the proposed wall/floor details against the suggested solutions in Approved Document E before starting work. This will minimise the chances of issues during sound testing.