Summary
In short
- State the criterion level, exchange rate and threshold on every survey. Without them the numbers cannot be compared to anything, including your own previous surveys.
- The action level of 85 dBA, not the 90 dBA permissible exposure limit, is what triggers the hearing conservation programme. This is among the most common misunderstandings of 1910.95.
- OSHA's PEL calculation uses an 80 dBA threshold, meaning quieter periods are excluded from the dose, and the action level calculation extends coverage down to 80 dBA.
- A worker who is OSHA-compliant can be well over the NIOSH recommended limit on the same shift, particularly between 88 and 95 dBA. Many organisations adopt 85 dBA as an internal limit for that reason.
- Impulse and impact noise carries a separate ceiling of 140 dB peak sound pressure level, which applies regardless of how few impacts occur.
- Personal dosimetry beats area measurement for mobile workers, because a survey of places tells you little about the exposure of someone who moves between them.
What it is
What it is
What is a noise survey?
Measurement of noise exposure to determine which employees are exposed at or above the action level and the permissible exposure limit, to identify and rank the sources contributing to that exposure, and to inform control decisions and hearing protection selection.
What is the exchange rate?
The relationship between sound level and permitted duration. OSHA's 5 dB exchange rate halves the allowable time for every 5 dB increase, so 90 dBA is permitted for eight hours and 95 dBA for four. NIOSH's 3 dB rate follows the equal-energy principle used internationally and halves the time for every 3 dB, which is considerably more protective.
When to use it
When to use it, and when not to
This measures exposure and identifies sources. The programme it triggers sits separately.
Use it for
- Determining which employees are at or above the action level and the permissible exposure limit
- Identifying and ranking the sources contributing to exposure, to inform control decisions
- Repeat monitoring where production, process, equipment or controls change and may increase exposure
- Supporting hearing protection selection with the actual exposure level
- Establishing the basis for audiometric testing enrolment
Not for
- Audiometric testing, which measures the effect on the person rather than the exposure
- The hearing conservation programme itself, which this survey triggers
- Hearing protection selection and derating, which follows from the measured exposure
- Community and environmental noise assessment, which uses different metrics and limits
- Noise control engineering design, which follows from the ranked source data
Standards
What it is built against
Noise exposure is prescriptively regulated, with a recommendation from NIOSH that most professionals treat as the design target.
| Clause | Requirement | Where it lands |
|---|---|---|
| 29 CFR 1910.95(a) | Permissible exposure limit of 90 dBA as an eight-hour time-weighted average, with a 5 dB exchange rate | Measurements |
| 29 CFR 1910.95(b)(2) | Impulsive or impact noise not to exceed 140 dB peak sound pressure level | Measurements |
| 29 CFR 1910.95(c) | Hearing conservation programme where exposure equals or exceeds the 85 dBA action level | Outcome |
| 29 CFR 1910.95(d) | Monitoring to identify employees for the programme and to enable proper hearing protector selection | Header |
| 29 CFR 1910.95(d)(2) | Representative personal sampling where circumstances make area monitoring inappropriate | Measurements |
| 29 CFR 1910.95(d)(3) | Repeat monitoring where changes in production, process, equipment or controls may increase exposure | Outcome |
| 29 CFR 1910.95(g) | Audiometric testing programme with baseline within six months of first exposure at or above the action level | Outcome |
| NIOSH REL | Recommended exposure limit of 85 dBA with a 3 dB exchange rate, not legally enforceable | Measurements |
What it does not cover
- Audiometric testing, measuring the effect on the person rather than the exposure.
- The hearing conservation programme, which this survey determines the scope of.
- Hearing protector selection and derating, following from the measured exposure.
- Environmental and community noise assessment, using different metrics entirely.
- Noise control engineering design, which the ranked source data informs.
Filling it in
Filling it in well
Record the settings, measure people rather than places, and rank sources so the survey produces engineering work.
Every measurement should carry the configuration that produced it. A time-weighted average of 87 dBA means one thing under OSHA settings and another under NIOSH settings, and a survey comparing this year's figures against a previous survey taken on different settings is comparing nothing.
Area measurements describe places. A maintenance technician, a forklift driver or a supervisor moves between them, and their exposure is the integral of everywhere they went for as long as they were there. Area surveys are useful for identifying and ranking sources; they are a poor basis for deciding who is in the programme.
The loudest machine is not necessarily the largest contributor. A 95 dBA source someone stands beside for six hours contributes more than a 105 dBA source they pass twice. Ranking by contribution to actual dose is what turns a survey into a prioritised list of engineering candidates rather than a list of loud things.
The hearing conservation programme, monitoring, audiometric testing, protection, training and recordkeeping, attaches at 85 dBA. The 90 dBA permissible exposure limit triggers the requirement for engineering and administrative controls. Confusing the two leaves a population between 85 and 90 dBA outside a programme they are legally entitled to.
Audit findings
Common audit findings
Noise survey findings concentrate on configuration, method and what happened afterwards.
| Finding | Clause | What fixes it |
|---|---|---|
| Criterion level, exchange rate and threshold not recorded. | 1910.95(d) | Without them the result cannot be interpreted or compared. |
| Programme scoped from the 90 dBA PEL rather than the 85 dBA action level. | 1910.95(c) | The programme attaches at the action level; this is the most common error. |
| Area monitoring used for mobile workers. | 1910.95(d)(2) | Personal dosimetry where circumstances make area monitoring inappropriate. |
| No repeat monitoring after a process or equipment change. | 1910.95(d)(3) | Change that may increase exposure is an explicit trigger. |
| Impulse noise not assessed against the 140 dB peak ceiling. | 1910.95(b)(2) | The ceiling applies regardless of duration or number of impacts. |
| Sources listed by loudness rather than by contribution to dose. | 1910.95(d) | Rank by dose contribution; the loudest source is often not the largest. |
| Hearing protector selection not derated. | 1910.95(j) | Laboratory ratings overstate real-world attenuation; apply a derating method. |
| Baseline audiograms not established within six months of first exposure. | 1910.95(g)(5) | Baseline is what makes later audiograms interpretable. |
| Survey concluding in hearing protection with no engineering assessment. | 1910.95(b)(1) | Feasible engineering and administrative controls are required above the PEL. |
| Employees not notified of their monitoring results. | 1910.95(e) | Notification of exposure at or above the action level is required. |
Worked case
Case in point: the same shift, two answers
A packing hall was surveyed and the operators returned time-weighted averages of around 87 dBA. The result was reported as compliant with the permissible exposure limit, which it was, and no further action was taken.
Two things followed from that reading that had not been drawn out. At 87 dBA the operators were above the 85 dBA action level, which meant the full hearing conservation programme applied: monitoring, audiometric testing, protection, training and recordkeeping. Being below the PEL is not being below the trigger.
And measured against NIOSH criteria, with an 85 dBA criterion level and a 3 dB exchange rate, the same exposure sits substantially further above the recommended limit than the OSHA figure suggests. Between roughly 88 and 95 dBA the two methods diverge sharply, which is precisely the range most manufacturing noise occupies.
Definitions
Definitions and key terms
- Exchange rate
- How permitted duration changes with level. OSHA uses 5 dB; NIOSH uses 3 dB on the equal-energy principle.
- Criterion level
- The level permitted for a full eight hours, 90 dBA under OSHA and 85 dBA under NIOSH.
- Threshold
- The level below which noise is excluded from the dose calculation, 80 dBA for OSHA hearing conservation purposes.
- Action level
- 85 dBA as an eight-hour TWA, triggering the hearing conservation programme under 1910.95(c).
- Permissible exposure limit
- 90 dBA as an eight-hour TWA, above which feasible engineering and administrative controls are required.
- Noise dose
- Accumulated exposure as a percentage, where 100 percent corresponds to the criterion level for eight hours.
- Peak sound pressure level
- The instantaneous maximum, subject to a 140 dB ceiling for impulsive noise regardless of duration.
- Derating
- Adjusting a hearing protector's laboratory rating to reflect real-world attenuation, which is consistently lower.
FAQ
Frequently asked questions
Why do OSHA and NIOSH give different answers?+
Different criterion levels and exchange rates. OSHA uses a 90 dBA criterion with a 5 dB exchange rate; NIOSH uses 85 dBA with 3 dB, following the equal-energy principle used internationally. Between roughly 88 and 95 dBA the two diverge sharply, and a worker who is comfortably OSHA-compliant can be well above the NIOSH recommendation on the same shift.
Which should we use?+
OSHA's is the enforceable standard in US general industry, so compliance is assessed against it. NIOSH's is a recommendation with no enforcement mechanism, and most industrial hygienists regard it as the better protection target. Many organisations report against OSHA and design against 85 dBA, which is a deliberate choice rather than a compromise.
What triggers the hearing conservation programme?+
The 85 dBA action level, not the 90 dBA permissible exposure limit. Monitoring, audiometric testing, hearing protection, training and recordkeeping all attach at the action level. Workers between 85 and 90 dBA are not legally overexposed and are entitled to the full programme, and scoping from the PEL leaves them outside it.
Is area monitoring acceptable?+
For identifying and ranking sources, yes. For determining who is in the programme, personal dosimetry is far better wherever workers move, because their exposure is the integral of everywhere they went. Maintenance, supervision and materials handling roles are the ones an area survey describes least well.
What about impulse noise?+
It carries a separate ceiling of 140 dB peak sound pressure level under 1910.95(b)(2), which applies regardless of duration or how few impacts occur in a shift. It is not addressed by the dose calculation, and a workplace with impacts, presses or air ejection needs peak measurement as well as time-weighted average.
The agents
What the agents do with it
The survey produces exposure figures. What fails is settings nobody recorded and a population between 85 and 90 nobody enrolled.
Records criterion, exchange rate and threshold with every measurement, and enrols everyone at or above the action level rather than the limit.
Flags process, equipment and control changes as monitoring triggers, and compares surveys only where the settings match.
Ranks sources by contribution to dose so noise control becomes a prioritised engineering list rather than a list of loud machines.
Covers hearing protection fitting and the difference between laboratory ratings and achieved attenuation.
This template lives in KnowHealth — employee wellbeing. Exposure monitoring, health surveillance, case management and return to work.
Meet KnowHealth→Sources
Sources
- 29 CFR 1910.95, occupational noise exposure, OSHA
- NIOSH criteria for a recommended standard: occupational noise exposure
- 29 CFR 1910.95 Appendix A, noise exposure computation
- 29 CFR 1910.95(g), audiometric testing programme, OSHA
- Control of Noise at Work Regulations 2005 (GB), with lower and upper action values