Predicting construction noise under BS 5228

BS 5228 construction noise prediction is an arithmetic exercise with four inputs: the sound power or reference sound pressure level of each item of plant, the proportion of the working period each item is actually running, the distance and screening between the plant and the receptor, and the duration over which the level is averaged. Get those four right and the prediction is defensible. Get the on-time percentages wrong, which is the usual failure, and the result is either an overstated impact that triggers mitigation nobody needs or an understated one that collapses under scrutiny.

The standard is BS 5228-1, the code of practice for noise control on construction and open sites. Part 2 deals with vibration. Councils cite both in conditions, and the environmental health officer reading your assessment will have the tables open.

What the standard provides

Three things, and they are worth separating.

Reference noise data for plant and activities, measured at a stated distance, which saves you measuring a breaker or an excavator yourself. A calculation method for combining plant items, applying distance attenuation and screening, and converting to an equivalent continuous level over the assessment period. And example methods for judging whether a predicted level amounts to an effect that matters.

The standard does not set statutory limits. Limits come from a condition, from a notice under section 60 of the Control of Pollution Act 1974, or from a prior consent under section 61 of the same Act. The distinction matters because an assessment that says "complies with BS 5228" without naming a criterion has not actually said anything.

Building the plant inventory

Start with the programme, not the plant list. For each phase, work out which items are on site at the same time and for how long within a working day.

For each item record the make and type where known, the reference level and the distance it was measured at, the number of units, and the on-time percentage. On-time is the share of the assessment period during which the item is generating noise at that level. An excavator in a 10-hour day might run 70 per cent of the time. A dump truck loading and hauling might be at the receptor-facing part of the site for 20 per cent. A concrete breaker rarely exceeds 30 per cent on-time because the operator repositions, refuels and waits.

Those percentages are the single biggest driver of the answer, and they are also the input a reviewer will question first. Base them on the contractor's method and say where they came from. A table of plant with every item at 100 per cent on-time tells the officer the assessment was not discussed with anyone who will be on site.

Predicting at the nearest receptor

Work out which receptor is nearest in each phase, and remember that the answer changes as the works move across the site.

The calculation for each item combines the reference level with a distance correction, a screening correction where a hoarding or structure intervenes, and the on-time percentage. Distance attenuation for a point source follows 20 times the logarithm of the distance ratio, so doubling the distance from the source takes off about 6 dB. Items are then combined logarithmically and the total expressed as an LAeq over the period.

An illustrative worked example. A breaker with a reference level of 90 dB LAeq at 10 metres, working at 30 per cent on-time, assessed at a facade 40 metres away with no screening, gives roughly 90 minus 12 dB for the distance, then minus about 5 dB for the on-time correction, so in the order of 73 dB LAeq over the period. Add a tracked excavator and a dumper and the combined figure rises by a few decibels rather than doubling, because logarithmic addition of similar levels adds around 3 dB.

Three refinements change results more than people expect. Facade reflection adds in the order of 3 dB where the receptor plane is modelled. Soft ground attenuation applies over grass and bare soil but not over hardstanding. And a solid barrier of adequate height close to either source or receptor is worth several decibels, with the gain depending on the geometry rather than the material.

Significance criteria and the example methods

BS 5228-1 offers example approaches rather than a single test, and the one councils use most is the category method based on existing ambient levels.

Values are threshold levels in dB LAeq over the relevant period, summarised from the standard's example method. Consult the current edition of BS 5228-1 for the full table, the period definitions and the rules for selecting a category, because the category depends on the measured ambient level at the receptor.

The alternative example method treats a change of 5 dB or more above the existing ambient as the trigger. On a quiet rural receptor the two methods can give different answers, and an assessment that reports both is harder to argue with than one that picks whichever is favourable. Where ambient data is thin, strategic noise mapping gives a starting point, as covered in our note on strategic noise mapping for early site appraisal.

Best practicable means and the mitigation hierarchy

Best practicable means is a statutory concept under the Control of Pollution Act 1974, and it bounds what an authority can require. Reasonableness, local circumstances, the current state of technical knowledge and financial implications all come into it.

In practice the hierarchy runs in this order.

1. Choose a quieter method: augered piling instead of driven, hydraulic crunching instead of a breaker, off-site fabrication instead of cutting on site.
2. Site the noisy activity and the compound as far from receptors as the layout allows, and orient loading away from facades.
3. Screen it, with the hoarding line set close to the source and with no gaps at the base.
4. Restrict the hours for the noisiest operations rather than the whole site.
5. Manage it, through plant maintenance, broadband reversing alarms, engine shutdown policies and a named contact for residents.

Price the first three before offering the fourth. Hours restrictions extend the programme, which extends the exposure, and an officer who understands that will accept a shorter, louder phase over a long quiet one in some situations.

Feeding the prediction into a management plan

The prediction is evidence for a document, and the document is what gets conditioned. A Construction Management Plan should carry the predicted levels by phase and receptor, the mitigation assumed in those predictions, the monitoring locations and trigger levels, the complaints procedure and the name of the responsible person.

Set trigger levels slightly below the criterion so there is room to respond before a breach. Say what happens when a trigger is reached: who is told, what stops, and within what period the cause is investigated. Dust and air quality monitoring usually share the same plan and the same receptors, so coordinate them with the Air Quality Assessment rather than running two regimes.

Where the contractor wants certainty, a section 61 prior consent application submitted before works start converts the agreed method and levels into a consent, and protects against a subsequent section 60 notice on the same facts.

Vibration from piling as a separate question

Vibration is not a noise question with different units. It has its own standard, its own metrics and its own receptors, and it belongs in a separate section.

Human response is assessed through vibration dose values under BS 6472-1. Cosmetic and structural damage criteria to buildings come from BS 7385-2 and are expressed as peak particle velocity. The two sets of numbers are orders of magnitude apart, so a level that will annoy a resident is far below one that threatens plaster.

Piling method selection is where most of the mitigation sits, and a pre-works condition survey of neighbouring property protects both sides. Noise assessment of this kind is a job for a suitably qualified acoustician, and councils expect a Noise Impact Assessment to carry an author with relevant competence. A PlanScope draft gives that professional a structured starting point, and professional review is strongly advised before submission.

Enter your postcode on PlanScope to see which environmental documents your site is likely to need.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. For specialist technical reports, qualified professional input may be required.