Facade sound insulation testing and the as built evidence

Facade sound insulation testing proves that the building you built achieves the internal noise levels the acoustic report promised. The condition usually requires testing of a sample of completed rooms before occupation, with the results compared against the design targets and any shortfall remedied. It is the only part of an acoustic submission that is measured rather than modelled, which is why it is the part that occasionally goes wrong.

Two different tests go by the same name, and the condition should say which one applies. One measures the performance of the facade as a building element. The other measures what a resident would actually hear.

What the verification condition requires

Typical wording obliges the developer to carry out testing in accordance with a method agreed with the local planning authority, to submit the results before first occupation, and to carry out any remedial works needed to achieve the stated internal levels.

The targets are normally expressed as internal ambient noise levels drawn from the guidance in BS 8233, which gives indicative design ranges for dwellings: a daytime level in living rooms, a night time level in bedrooms, and often a maximum level at night to deal with individual pass by events. The condition should state the levels, the periods, the room types and the window condition, which for internal ambient levels means closed.

Three things worth settling before the condition is imposed:

• Which rooms will be tested, expressed as a proportion and a minimum number rather than named units
• Whether the test is an element test with a loudspeaker source or a measurement of actual levels from the real noise source
• What happens on a marginal result, because measurement uncertainty is real and a condition with no tolerance invites argument

Propose the wording yourself in the Noise Impact Assessment. An authority given a workable condition will usually take it.

Rooms selected for testing and why

Sample selection should be worst case rather than representative, because the point of the exercise is to find a failure if one exists.

The rooms to put forward:

1. The most exposed facade, closest to the dominant source, at the floor level where the assessment predicted the highest incident level.
2. Rooms with the largest glazed area relative to their volume, since the glazing usually governs.
3. Rooms with the most ventilator openings, because attenuated ventilators are the commonest weak point.
4. Corner rooms with two exposed elevations.
5. One room on each distinct facade build up, where the scheme uses more than one glazing or ventilator specification.
6. At least one bedroom, because the night time target and the maximum level criterion are harder than the daytime one.

Avoid a sample made up entirely of accessible ground floor units that happen to be finished first. That is how a scheme passes the test and fails in use.

Agree the sample in writing with the environmental health officer before the test date, and record why each room was chosen. A report that explains the selection logic is far harder to dispute than one that lists six room numbers.

Test method and the measured quantity

The element test gives a repeatable number that can be compared against the specified sound reduction performance of the build up, and it is not affected by how busy the road happens to be. The direct approach answers the question the condition usually asks, namely whether the internal level is achieved, but it depends on the external conditions during the test and needs a correction back to the design condition.

Whichever is used, the report has to record the ventilation condition, the state of completion of the room, the background level in the room from any mechanical plant, calibration certificates for the instrumentation, measurement positions, and the measurement uncertainty.

One practical point about timing within the build. Test after glazing, ventilators, internal linings and floor finishes are installed. A room tested before plastering has different reverberation characteristics, and the result will not represent the finished dwelling.

Comparing results to the design assumption

The comparison is only meaningful if the design assumption is written down. The acoustic report should state, for each facade build up, the incident external level used, the required level difference, the glazing and ventilator performance assumed, and the internal level predicted.

Where the test is an element test, compare the measured level difference against the required figure. Where it is a direct measurement, compare the measured internal level against the target, having corrected for the external level measured at the same time.

Two corrections cause most of the disputes. Whether the external level during the test was representative, which matters most on a road with variable flows and is why simultaneous external measurement is needed. And whether the internal level includes contributions from mechanical ventilation, which it should, because a resident cannot separate the two.

Where the original assessment used unattended monitoring to establish the external level, the validity of that baseline becomes relevant again at verification stage, a point we cover in unattended noise monitoring, duration, position and validity. Road traffic sources behave differently from rail, and the design assumptions differ accordingly, as set out in road traffic noise for a new housing site and railway noise and vibration for sites beside a line.

Dealing with a failure

Failures are usually installation rather than specification. Work through the likely causes in this order, because the first three cost very little to fix.

• Perimeter sealing around the window frame, particularly at the cill and at the junction with the reveal
• Ventilator installation: the wrong attenuator fitted, an acoustic ventilator installed without its liner, or a ventilator cut into the frame on site
• Letterplates, cat flaps, extract terminals and service penetrations through the facade
• Glazing supplied to a different specification from the one approved, which a check of the delivery documents will reveal
• Trickle ventilators left in a partly open position during the test
• The wrong glass thickness or an incorrect cavity in a sealed unit

Where the specification itself is inadequate, remediation options include replacing the ventilators with higher performing units, adding secondary glazing, replacing the sealed units with an asymmetric or laminated build up, and in the worst case changing the ventilation strategy to mechanical supply with attenuation.

Retest after remediation, on the same rooms and by the same method. And flag the overheating consequence of any change that reduces openable area, because a scheme that solves its acoustic failure by sealing the facade has created a different problem.

Reporting format the authority expects

A report that gets accepted first time contains:

1. The condition wording and the agreed method statement.
2. The design targets by room type and period, taken from the approved acoustic report.
3. The sample, with the selection rationale and a plan showing each tested room.
4. Instrumentation, calibration certificates and the date of the last laboratory calibration.
5. Measurement positions, including external microphone position and height.
6. Raw results, with the measured quantity named and the correction method shown.
7. Weather and source conditions during the test.
8. A comparison table of measured against target, room by room, with a stated pass or fail.
9. Measurement uncertainty.
10. Any remedial works carried out and the retest results.

Testing by an organisation with recognised accreditation for the measurement carries more weight than testing by the designer, and some authorities require it. Check the condition.

Timing the test within the programme

Build the sequence into the programme rather than discovering it at practical completion.

Allow for the rooms to be substantially complete, for the test itself, which takes a day or two for a sample of six to eight rooms, for the report, for the authority's determination of the discharge application, and for a remediation and retest cycle if it is needed. On a scheme where the acoustic design is marginal, allow eight to twelve weeks in total.

A worthwhile precaution on larger schemes is an early indicative test on the first completed unit on the most exposed elevation, before the remaining units are glazed. It is not a formal verification, and it tells you whether the installation details work while there is still time to change them on 80 flats rather than on four.

Enter your postcode on PlanScope to see which acoustic and amenity documents your scheme is likely to need.

This is general guidance rather than legal advice. Check the position with your local planning authority or a qualified professional before relying on it. Councils usually expect acoustic testing and reporting by a suitably qualified author, and professional review is strongly advised. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.