Construction Noise Rules: What OSHA and NIOSH Actually Require

OSHA's construction noise standard requires protection when exposures exceed its permissible limits. The general-industry table it draws on, Table G-16, sets a series of duration-level pairs, and the 5 dB exchange rate means each 5 dB increase halves the permitted duration.

The standard also fixes two hard boundaries: continuous noise must not exceed 115 dBA, and impulsive or impact noise must not exceed a 140 dB peak sound pressure level.

NIOSH's recommended exposure limit is 85 dBA as an eight-hour time-weighted average, with a 3 dB exchange rate. The 3 dB rate follows the equal-energy principle: doubling the sound energy halves the permitted time.

NIOSH is a research institute, not a regulator. Its REL carries no legal force in itself; it exists as the health-based benchmark, and it is why so much guidance quotes 85 dBA while the enforceable US number remains 90 dBA.

Directive 2003/10/EC sets a daily noise exposure limit value of 87 dB(A) with a peak limit of 200 Pa, an upper action value of 85 dB(A) with 140 Pa, and a lower action value of 80 dB(A) with 112 Pa.

One structural detail matters: the limit value takes the attenuation of hearing protection into account, while the action values are assessed without it. Crossing an action value triggers duties — protection availability, information, health surveillance — rather than an immediate breach.

None of these frameworks tells you how loud a specific tool is. They describe measured worker exposure over a shift, which is why compliance work uses dosimetry — a badge-mounted instrument logging a whole day — rather than a spot reading.

A spot measurement is still useful for triage: it tells you where the loud work is, which tasks dominate, and whether a quieter method or a barrier changes anything. It cannot substitute for a dose assessment, and it is not a medical judgement about any individual's hearing.

The other thing these frameworks share is a hierarchy of response. Reducing the noise at source — a quieter method, a different tool, a barrier, moving work away from other trades — comes before protective equipment, because protection depends on being worn correctly for the whole exposure. A protector's laboratory attenuation rating is measured under ideal fitting conditions, and real-world attenuation is routinely lower.

None of this requires a precise decibel figure for the tool in your hands. It requires knowing which tasks dominate a shift and how long they run, which is a scheduling and method question as much as an acoustic one.

Occupational rules stop at the site boundary. Noise reaching homes around a project is handled by local ordinances, permit conditions and, for larger works, environmental assessment using propagation models such as the FHWA construction noise model.

Those regimes usually work with time-of-day restrictions and predicted levels at receivers rather than with worker exposure limits, so quoting the 90 dBA occupational figure at a neighbour is a category error.