How Loud Is 40 Decibels? Which Documented Figures Sit There

The most-cited government figure in this region is the EPA's, published in 1974 in a document whose full title is worth reading slowly: Information on Levels of Environmental Noise Requisite to Protect Public Health and Welfare with an Adequate Margin of Safety. It identifies 45 decibels indoors and 55 decibels outdoors as levels that prevent activity interference and annoyance.

The EPA was explicit that these are not single-event or peak values. They are averages of acoustic energy over periods such as 8 or 24 hours, and over long periods such as years. A quiet room that briefly rises above 45 dB is not in breach of anything; the identified level describes a long-run average environment.

The other documented number nearby comes from product labelling rather than health policy. Under EU Regulation 2019/2017, a dishwasher whose declared airborne noise emission is below 39 dB(A) re 1 pW is placed in noise class A. That declaration is a laboratory sound power figure for the eco programme.

Sound power describes the total acoustic energy a source radiates. It belongs to the machine and does not change when you walk away from it. Sound pressure is what a microphone responds to at a particular point in a particular room, and it falls as you move away from the source.

Because they are different quantities with different reference values, a declared 39 dB(A) re 1 pW and an environmental 45 dB are not two points on the same scale. Putting them in one comparison table — which this page previously did — creates a false impression of precision. Neither figure is wrong; the comparison is.

This is the single most common error in decibel content, and it is why every figure on this site now carries its quantity in the same sentence as its number.

People usually arrive at this number while trying to answer a practical question: is my bedroom quiet enough, or is that appliance quiet enough to run at night. The published record answers that less directly than most articles imply.

Environmental night-noise guidance published by health authorities is generally expressed as an outdoor level at the dwelling façade, as a long-term average, and for populations rather than individuals. It is not a bedroom target you can check with a phone in one evening. We did not find a primary regulatory text defining an indoor bedroom limit in absolute terms.

What you can do reliably is compare. The difference between your bedroom with the window open and closed, or between the fridge running and not running, is a comparison your own device can make honestly, because a comparison does not depend on absolute calibration.

Quiet environments are where consumer microphones behave worst. Their own electrical self-noise sits within the range you are trying to measure, and browsers commonly apply automatic gain control and noise suppression that raise or reshape low-level signals in ways the page cannot see.

Uncalibrated, our meter reports an A-weighted digital signal level in dBFS. That is a relative figure, not sound pressure, and in a genuinely quiet room it says more about the microphone than about the room. After a one-time local calibration against a trusted reference reading it reports an estimated dBA value, still on consumer hardware.

If you need a defensible number for a quiet space — a noise complaint, a building acceptance test — that requires a calibrated instrument and a stated measurement procedure, not a browser.