Noise Exposure Dose Calculator (OSHA, NIOSH, EU)

Occupational noise dose under all three regimes at once — OSHA’s 5 dB exchange rate, NIOSH’s 3 dB recommendation and the EU’s L_EX,8h against its 80, 85 and 87 dB(A) values — with the formulas quoted from 29 CFR and the hearing-protector qualifier that everyone drops.

Last updated 24 Sep 2026

Noise dose under OSHA, NIOSH and the EU

Levels and hours → dose in three regimes

How many parts does the day have?Each part is a steady A-weighted level for a number of hours. Unused parts are locked out.

Part 1 — level, dB(A)An A-weighted sound level, the quantity every one of these regimes is written in.

Part 1 — hours

Part 2 — level, dB(A)

Part 2 — hours

Part 3 — level, dB(A)

Part 3 — hours

Part 4 — level, dB(A)

Part 4 — hours

A scale, not a circuit, and the picture the chart below cannot give: every threshold the three regimes set, on one A-weighted axis, with your own daily exposure level marked. The three rules are Directive 2003/10/EC, 29 CFR 1910.95 and the NIOSH recommendation; the short marks in line with each other are your own L_EX,8h to the nearest decibel, and the exact figure is above. Read down them and you can see at once which limits you are past and which you are not, and how differently the three regimes are spaced. LAV and UAV are the EU's lower and upper exposure action values, ELV its exposure limit value, PEL OSHA's permissible exposure limit and REL the NIOSH recommended exposure limit. The EU row and the OSHA row are LAW where they apply; the NIOSH row is a recommendation and is not law anywhere. None of these figures accounts for hearing protection: the EU's two action values must be judged without it, and its 87 dB(A) limit value is the one figure here that is compared against the protected exposure instead. Note also that the OSHA row is not directly comparable with the other two, because the eight-hour average beneath it is computed on a 5 dB exchange rate rather than an energy basis.

90.53dB(A)Example

two hours at 95 dB(A), four at 88 and two at 80 — an eight-hour day

Three regimes, three sums

OSHA: T = 8 / 2^(L−90)/5, D = 100 Σ(Cᵢ/Tᵢ), TWA = 16.61 log₁₀(D/100) + 90 · NIOSH: T = 8 / 2^(L−85)/3 · EU / ISO 1999: L_EX,8h = 10 log₁₀(Σ Tᵢ · 10^Lᵢ/10 / 8 h)

Lᵢ: the A-weighted sound level of part i of the day, in dB(A)
Cᵢ: hours actually spent at that level
Tᵢ: hours PERMITTED at that level by the regime in question
D: dose, as a percentage. 100% is the whole of that regime’s allowance for one day
TWA: the equivalent steady level for eight hours that would give the same dose
L_EX,8h: the EU’s daily noise exposure level: the day’s energy spread over eight hours
16.61: not arbitrary. It is 5 / log₁₀ 2 = 16.6096, the 5 dB exchange rate inverted

Worked example

two hours at 95 dB(A), four at 88 and two at 80 — an eight-hour day
OSHA permits 4.00 h at 95 dB(A), 10.56 h at 88 and 32.00 h at 80, so the dose is 100 × (2/4 + 4/10.56 + 2/32.00) = 94.1%
That is an eight-hour TWA of 16.61 log₁₀(94.1/100) + 90 = 89.56 dBA — under OSHA's 90 dBA permissible limit, and over its 85 dBA action level, so a hearing conservation programme is required
NIOSH permits 47.6 min at 95 dB(A), 4.00 h at 88 and 25.40 h at 80, so the same day is a 360% dose — 3.6 times the recommended limit
L_EX,8h = 10 log₁₀((2×10^9.5 + 4×10^8.8 + 2×10^8.0) / 8) = 90.53 dB(A)
So this shift is inside OSHA's permissible limit and 3.5 dB over the EU exposure limit value, from the same three measurements. The exchange rate did all of it
Removing only the two hours at 95 dB(A) would drop L_EX,8h to 85.32 dB(A) — the loudest quarter of the day is carrying nearly all of the exposure

The three regimes, side by side

The row that everybody gets wrong is the sixth. Article 3(2) of Directive 2003/10/EC says, in terms: when applying the exposure LIMIT values the worker’s effective exposure shall take account of the attenuation provided by the individual hearing protectors worn, and the exposure ACTION values shall not take account of any such protectors. So an 87 dB(A) reading with no protectors is not automatically a breach of the limit value, and a 79 dB(A) reading achieved only by wearing earplugs is not compliance with the lower action value. The 16.61 in OSHA’s average is not arbitrary either: it is 5 / log₁₀ 2 = 16.6096, the 5 dB exchange rate written the other way round. And the second row hides a third disagreement that even careful sources conflate: a 3 dB exchange rate is not exactly the equal-energy rule, because equal energy halves the time every 3.0103 dB. NIOSH’s published ladder is exactly 3, the EU’s L_EX,8h is exactly energy, and they differ by up to 2.3% of the permitted time at high levels.
	OSHA (United States)	NIOSH (recommendation)	EU Directive 2003/10/EC
Status	Law: 29 CFR 1910.95	A recommendation, not law anywhere	Law in EU and EEA states, through national implementations
Exchange rate	5 dB	3 dB	3.01 dB — a true energy average, as ISO 1999 defines L_EX,8h
Criterion level for 100% dose / the limit	90 dBA for 8 h (permissible exposure limit)	85 dBA for 8 h (recommended exposure limit)	87 dB(A) exposure limit value
Action level	85 dBA 8-hour TWA, a 50% dose — hearing conservation programme required	—	80 dB(A) lower action value and 85 dB(A) upper action value
Peak	140 dB peak sound pressure level (impulsive noise)	140 dBC recommended ceiling	200 Pa (140 dBC) limit value; 140 Pa upper and 112 Pa lower action values
Are hearing protectors taken into account?	No, for the PEL and the action level	No	NO for the 80 and 85 dB(A) action values; YES for the 87 dB(A) limit value, which is compared against the worker’s effective exposure WITH protectors
Formula for permitted time	T = 8 / 2^((L−90)/5)	T = 8 / 2^((L−85)/3)	T = 8 × 10^((87−L)/10)
8-hour average	TWA = 16.61 log₁₀(D/100) + 90	85 + 10 log₁₀(D/100), approximately	L_EX,8h = 10 log₁₀(Σ Tᵢ 10^(Lᵢ/10) / 8 h)

How long you may be exposed, under each regime

Computed from the two published formulas: OSHA’s T = 8 / 2^((L−90)/5) from 29 CFR 1910.95 Appendix A, and NIOSH’s T = 8 / 2^((L−85)/3) from DHHS 98-126. The fourth column is the argument of the page: the two regimes diverge exponentially, from a factor of 2 at 85 dB(A) to a factor of 20 at 110. Two rows correct a figure often quoted wrongly. At 95 dB(A) OSHA permits 4 hours and NIOSH 47.6 MINUTES — not one hour; the 1-hour NIOSH level is 94 dB(A). And OSHA’s own Table G-16 is reproduced exactly by its formula, which is worth checking, because the table stops at 115 dB(A) and the formula does not. Remember which is which: OSHA’s numbers and the EU’s are law where they apply, and NIOSH’s are a recommendation nowhere enacted.
A-weighted level	OSHA permitted time (5 dB)	NIOSH permitted time (3 dB)	OSHA allows this much longer	Time to reach the EU 87 dB(A) limit	Dose per hour, OSHA / NIOSH
80 dB(A)	32.00 h	25.40 h	1.26×	2,405.7 min	3.1% / 3.9%
82 dB(A)	24.25 h	16.00 h	1.52×	1,517.9 min	4.1% / 6.3%
85 dB(A)	16.00 h	8.00 h	2.00×	760.7 min	6.3% / 12.5%
88 dB(A)	10.56 h	4.00 h	2.64×	381.3 min	9.5% / 25.0%
90 dB(A)	8.00 h	2.52 h	3.17×	240.6 min	12.5% / 39.7%
91 dB(A)	6.96 h	2.00 h	3.48×	191.1 min	14.4% / 50.0%
92 dB(A)	6.06 h	1.59 h	3.82×	151.8 min	16.5% / 63.0%
94 dB(A)	4.59 h	1.00 h	4.59×	95.8 min	21.8% / 100.0%
95 dB(A)	4.00 h	47.6 min	5.04×	76.1 min	25.0% / 126.0%
97 dB(A)	3.03 h	30.0 min	6.06×	48.0 min	33.0% / 200.0%
100 dB(A)	2.00 h	15.0 min	8.00×	24.1 min	50.0% / 400.0%
102 dB(A)	1.52 h	9.4 min	9.62×	15.2 min	66.0% / 635.0%
105 dB(A)	1.00 h	4.7 min	12.70×	7.6 min	100.0% / 1,269.9%
110 dB(A)	30.0 min	1.5 min	20.16×	2.4 min	200.0% / 4,031.7%
115 dB(A)	15.0 min	0.5 min	32.00×	0.8 min	400.0% / 12,800.0%

29 CFR 1910.95 Table G-16, and the formula that generates it

The regulation gives both a table and a formula, and the table is the rounded one: Appendix A’s T = 8 / 2^((L−90)/5) reproduces every row to within 2%, the largest gap being at 102 dBA, where the table says 1.5 hours and the formula says 1.5157. 29 CFR is a work of the United States Government, so both are quoted here directly. The table’s last row reads “0.25 or less” at 115 dBA and the regulation adds that exposure to impulsive or impact noise should not exceed a 140 dB peak sound pressure level; the formula keeps going past 115 dBA and this page lets it, because a dose has to be able to describe a short burst of something very loud.
Sound level, dBA slow response	Table G-16 duration per day	8 / 2^((L−90)/5)	Difference
90 dBA	8.00 h	8.0000 h	0.00%
92 dBA	6.00 h	6.0629 h	1.05%
95 dBA	4.00 h	4.0000 h	0.00%
97 dBA	3.00 h	3.0314 h	1.05%
100 dBA	2.00 h	2.0000 h	0.00%
102 dBA	1.50 h	1.5157 h	1.05%
105 dBA	1.00 h	1.0000 h	0.00%
110 dBA	0.50 h	0.5000 h	0.00%
115 dBA	0.25 h	0.2500 h	0.00%

One working day, three legal answers

Two regulatory regimes cover most of the world’s workplaces and they disagree about the arithmetic, not just the number. That is why a search engine cannot answer “what is my noise dose”. It is not a conversion; it is a question about which rules you are under.

OSHA, in the United States, uses a 5 dB exchange rate. 29 CFR 1910.95 Appendix A gives the permitted time for a level directly: T = 8 / 2^((L−90)/5) hours. Eight hours at 90 dBA is a full dose; every 5 dB louder halves the time you may spend there. The dose over a day is 100 × Σ(Cᵢ/Tᵢ) per cent, and the equivalent eight-hour average comes back out as TWA = 16.61 log₁₀(D/100) + 90. A second figure matters as much as the 90 dBA limit and gets far less attention: section (c)(2) names an 85 dBA eight-hour average — a 50% dose — as the ACTION LEVEL, at which a hearing conservation programme becomes mandatory. Half a dose is not half a problem.

NIOSH recommends a 3 dB exchange rate and an 85 dBA limit, so T = 8 / 2^((L−85)/3). Three decibels is equal energy: doubling the acoustic power halves the time, which is what the evidence on hearing damage supports and what the 1971 5 dB rule’s assumption of recovery during quiet intervals did not survive. NIOSH’s figure is a RECOMMENDATION. It is not law in the United States or anywhere else, and this page never prints it as though it were. What it does do is show you the size of the disagreement, which is larger than most people expect: at 85 dB(A) OSHA allows twice as long as NIOSH, at 95 dB(A) five times as long, at 110 dB(A) twenty times.

The EU works in a third quantity again. Directive 2003/10/EC sets a lower exposure action value at 80 dB(A), an upper action value at 85 dB(A) and an exposure limit value at 87 dB(A), all as L_EX,8h — the daily noise exposure level defined in ISO 1999, which is a true energy average over the day normalised to eight hours. And it carries the qualifier that is dropped more often than any other sentence in occupational noise law: when applying the LIMIT value, the worker’s effective exposure takes the attenuation of the hearing protectors actually worn into account, and when applying the ACTION values it does not. So the same 86 dB(A) measurement triggers the upper action value regardless of what anyone is wearing, and cannot breach the limit value at all if the protectors are working. This page computes the unprotected figure, which is the right basis for the action values and an upper bound for the limit value.

A subtlety worth knowing, because even careful sources run the two together. A “3 dB exchange rate” is not quite the equal-energy rule. Equal energy halves the permitted time every 10 log₁₀ 2 = 3.0103 dB, not every 3.0000. NIOSH’s published ladder — 85 dB(A) for eight hours, 88 for four, 91 for two, 94 for one — is exactly 3 dB, and the EU’s L_EX,8h is exactly energy. The gap between them reaches 2.3% of the permitted time at 115 dB(A). This page computes each regime with its own arithmetic rather than with a shared approximation.

What this page does not do. It does not model hearing protection: derating a protector’s NRR or SNR is a separate calculation and real-world attenuation falls well short of the laboratory rating because of fit. It does not handle impulsive or impact noise, which every one of these regimes treats through a separate peak criterion — 140 dB peak for OSHA, 200 Pa C-weighted for the EU. It does not estimate hearing loss, which is ISO 1999’s other half. And it is not a substitute for measurement: a working day is not three steady blocks, and a dosimeter worn for a full shift is what establishes somebody’s actual exposure. Use this to understand the arithmetic and to compare the regimes. For the units underneath all of it, the sound pressure level converter covers dB SPL, pascals and sound intensity, and explains why dB(A) is a filter rather than a unit.

Frequently asked questions

What is a 100% noise dose?

It depends whose rules you are under, which is the entire point of this page. Under OSHA, 100% is eight hours at 90 dBA, or four at 95, or two at 100 — a 5 dB exchange rate. Under NIOSH’s recommendation, 100% is eight hours at 85 dBA, or two at 91, or one at 94 — a 3 dB exchange rate. The same working day can be 94% of a dose under OSHA and 360% under NIOSH, which is what the default example on this page shows. Neither number is wrong; they are answers to different questions.

Which is right, the 3 dB or the 5 dB exchange rate?

The 3 dB rate is the one with the physics behind it: it is equal energy, so halving the time exactly compensates for doubling the acoustic power. NIOSH adopted it in 1998 and ISO 1999, and through it the EU directive, use it. OSHA’s 5 dB rate dates from 1971 and assumes that quiet intervals let the ear recover, which later evidence did not support. OSHA is still the law in the United States, so US employers must compute the 5 dB dose whatever they think of it. The honest position is the one this page takes: compute both and say which one is enforceable where you are.

How long can I work at 95 dBA?

Four hours under OSHA and 47.6 minutes under NIOSH — not one hour, which is the figure usually quoted. One hour is the NIOSH level for 94 dB(A). Under the EU directive, reaching the 87 dB(A) exposure limit value at 95 dB(A) takes 76.1 minutes of unprotected exposure. Those three answers for one level are why this page exists.

What is L_EX,8h and how is it different from a TWA?

L_EX,8h is the EU’s daily noise exposure level: the total acoustic energy of the working day, averaged as if it had been spread over eight hours. It is defined in ISO 1999 clause 3.6 and adopted by Directive 2003/10/EC. An OSHA 8-hour TWA answers the same question using the 5 dB exchange rate instead, so the two numbers differ for any day that is not steady — for the example on this page, 90.5 dB(A) against 89.6 dBA. Note also that both average over eight hours whatever the shift length, so a twelve-hour day is not diluted by being long.

Does this account for my earplugs?

No, and it must not be read as though it did. Every figure here is the level at the ear with NO hearing protection. That is the correct basis for the EU’s 80 and 85 dB(A) action values and for OSHA’s PEL and action level, all of which are judged without protectors. It is only an upper bound for the EU’s 87 dB(A) exposure limit value, which Article 3(2) says is compared against the worker’s effective exposure with the protectors actually worn. Working that out means derating a protector’s NRR or SNR, which is a different calculation with its own rules and its own large real-world discount for fit.

Is 85 dB a legal limit?

It depends which 85 you mean. In the EU, 85 dB(A) is the upper exposure ACTION value, and reaching it triggers legal duties — protection must be worn, a noise-reduction programme must exist, areas must be marked. It is not the limit value; that is 87 dB(A). In the United States, 85 dBA as an 8-hour TWA is OSHA’s action level, at which a hearing conservation programme becomes mandatory, while the permissible exposure limit is 90 dBA. And NIOSH’s 85 dBA recommended exposure limit is not law anywhere. Three different statuses for one number.

Can I use this instead of a dosimeter?

No. This computes a dose from levels and durations you supply, which means the answer is only as good as your estimate of what those were. A real working day is not three steady blocks; it is thousands of short events, and a personal dosimeter worn for a full shift is what regulators and occupational hygienists actually rely on. Use this to understand the arithmetic, to compare regimes and to see what changing a task would do, and use a dosimeter to establish what somebody’s exposure is.

Related calculators

References

Code of Federal Regulations, 29 CFR 1910.95, Occupational noise exposure, with Table G-16 and Appendix A. A work of the United States Government: the formulas are quoted directly. Appendix A gives the reference duration T = 8 / 2^(L−90)/5, the dose D = 100 (C₁/T₁ + … + Cₙ/Tₙ) and the eight-hour average TWA = 16.61 log₁₀(D/100) + 90. Section (c)(2) names an 8-hour TWA of 85 dB, a 50% dose, as the action level at which a hearing conservation programme is required.
National Institute for Occupational Safety and Health. Criteria for a Recommended Standard: Occupational Noise Exposure, Revised Criteria 1998, DHHS (NIOSH) Publication No. 98-126, and CDC’s Noise and Hearing Loss Prevention — Regulations and Guidance. A work of the United States Government. The REL is 85 dBA as an 8-hour TWA with a 3 dB exchange rate. NIOSH’s own ladder — 85 dBA for 8 hours, 88 for 4, 91 for 2, 94 for 1, 97 for 30 minutes, 100 for 15 — is reproduced exactly by T = 8 / 2^(L−85)/3. It is a RECOMMENDATION and not law anywhere.
Directive 2003/10/EC of the European Parliament and of the Council on the minimum health and safety requirements regarding the exposure of workers to the risks arising from physical agents (noise), Article 3. Exposure limit values L_EX,8h = 87 dB(A) and p_peak = 200 Pa; upper exposure action values 85 dB(A) and 140 Pa; lower exposure action values 80 dB(A) and 112 Pa. Article 3(2) is the sentence everyone drops: “When applying the exposure limit values, the determination of the worker’s effective exposure shall take account of the attenuation provided by the individual hearing protectors worn by the worker. The exposure action values shall not take account of the effect of any such protectors.”
ISO 1999:2013, Acoustics — Estimation of noise-induced hearing loss, clause 3.6, which is the definition Directive 2003/10/EC adopts for L_EX,8h. ISO 1999 is copyrighted and is not reproduced: the numbers on this page come from the directive, which is published in the Official Journal, and the energy-average arithmetic is reconstructed from the directive’s own definition.
IEC 61672-1:2013, Electroacoustics — Sound level meters — Part 1: Specifications, clause 5.4 and Table 3. The frequency weightings are defined in the standard by a table of values with tolerances, not by a formula; the standard is copyrighted and its table is NOT reproduced here. The analytic four-pole function this page uses was checked against all 34 tabulated A-weighting values and reproduces every one to better than 0.05 dB, inside the table’s own 0.1 dB rounding.
National Physical Laboratory. Sound Pressure Level and Intensity Level, NPL acoustics technical guide, resource.npl.co.uk. States both reference values, that they were chosen together in 1932, and that “a 20 µPa acoustic pressure in air gives a 1 pW/m² intensity if the impedance is 400 Rayls” — which is the sentence this page’s temperature input exists because of. NPL quote 409 rayl for air; that is the 25 °C value, and 413 is the 20 °C one.