No Federal Rule Says How Quiet a Bedroom Has to Be, and the Office Closed in 1982

Does any law say how quiet your bedroom has to be?

No. Nothing federal reaches into a room you already live in, and that gap surprises everybody we mention it to.

Two in the morning. A motorcycle three streets over, quiet again, and then the fridge kicks on. You start doing arithmetic on how much night is left, which is its own way of staying awake.

Now put yourself at work instead. An inspector can walk in with a meter, put it where your head goes, and write your employer up. OSHA draws the line at 90 dB(A) averaged across an eight-hour shift, with hearing tests and paperwork starting at 85. Louder buys you less time, so 95 dB(A) for four hours, 100 for two, 115 for a quarter of an hour.

Real limits. Somebody enforces them, and your boss can lose money for ignoring one.

Then you drive home and go to bed, and there’s nothing at all. Eight hours at a desk has a rule behind it. Eight hours asleep, in the room where your body is supposed to be repairing yesterday, has none.

We think the split comes down to what counts. Hearing damage builds up, tracks a dose, and shows up on a test a nurse gives you. Sleep damage does none of those three cleanly, so nobody wrote a rule they would later have to defend in front of a judge.

What about the 45 decibels everybody quotes?

Ask how quiet a bedroom should be and one figure comes back at you from everywhere, 45 decibels. That figure exists. It’s federal. It was never a rule.

We went and opened it. The figure lives in a 155-page document, 550/9-74-004, published in March 1974 by EPA‘s Office of Noise Abatement and Control. Its cover says out loud what it is not, that “This document has been approved for general availability. It does not constitute a standard, specification, or regulation.”

Printed at the front, before you reach page one. We had to open a PDF to find it, and almost nobody passing that figure along has.

Table 1 inside it gives three figures and stops. 45 dB inside a home. 55 dB outdoors where people live. And 70 dB averaged across 24 hours anywhere, which is what keeps your hearing intact across a lifetime.

Congress had asked for information, not for limits. The Noise Control Act of 1972 told that office to publish which sound levels protect public health and welfare, with an adequate margin of safety. So it published. Then it warned everybody, in print, not to treat any of it as law.

Which leaves 45 as a benchmark you can measure yourself against. Your bedroom can run well above it and you break nothing, because no law contains it.

That difference is worth knowing. A limit binds somebody. A benchmark only describes a room. Mixing them up is how people end up furious at an agency that never made you a promise.

Why did the federal noise office close?

Seven years. That’s how long it lasted after publishing the thing everybody still quotes.

In 1981 the Administration concluded noise was best handled by state and local government, so the Office of Noise Abatement and Control shut its doors and responsibility moved down a level. We looked for whatever replaced it and came up with nothing. Two laws from that era are still on the books, the Noise Control Act of 1972 and the Quiet Communities Act of 1978. Neither has anybody behind it doing what 1974 did.

So any enforceable night rule covering your address was written by a state, a county or a city, and finding it’s on you.

Brace yourself for that search. Responsibility fell to thousands of local governments at once with no template between them, so two towns a few miles apart end up with two unrelated rules and neither one knows it.

A repealed limit at least tells you somebody weighed it and decided against. A limit left standing with nobody behind it tells you nothing at all, and it keeps getting quoted for decades by people who reasonably assume a federal number means somebody is enforcing it.

The office closed in 1981. Its figures were never repealed, and nobody ever enforced them.

Does 45 decibels apply at 2 a.m.?

You buy a meter, keep it up at two in the morning, and it says 51. Are you over?

Nobody can tell you, and that includes us. The 45 is a day-night average, built by adding 10 decibels to every sound between 10 p.m. and 7 a.m., then averaging all 24 hours together.

Read the penalty for what it is. Somebody in a federal office sat down, agreed with you that noise at night is worse, and baked it into the arithmetic.

So 45 behaves like a budget for a whole day rather than a ceiling for a moment. One loud hour at 3 a.m. eats far more of your budget than the same hour at 3 p.m.

A night figure is buried in that same 1974 document, and it’s far lower. A neighborhood sitting at its outdoor goal of 55 dB would give about 40 dB indoors by day, EPA expected, and about 32 dB after dark, and called the night value “consistent with the limited available sleep criteria.”

So 32 is the closest thing this country has to a federal figure for a room with you asleep in it. An estimate, inside a document from 1974, never restated since.

Whoever wrote it was honest about why, and we would rather show you that than tidy it away. Appendix E says noise “can definitely interfere with sleep”, then says relating exposure to sleep quality is difficult, and that the meaning of what noise does to your sleep “is still an open question.” The bedroom is unregulated in more than one direction. No US agency sets an EMF limit for it either.

That was 1974 describing 1974 evidence. Whoever would have gone back and updated it lost their office seven years later.

What closed with it was the only body in the country whose job was the sound where people live. Everything since has come from agencies with another remit. Transport looks at roads, labor at shifts, housing at what a building must be built to.

Which is a stranger outcome than a rule being struck down. A repealed limit tells you somebody decided against it.

Has anybody written a figure for a sleeping room?

One exists, and nobody enforces that one either. We found it well outside government, with building biology testers, people who turn up at your house with meters and walk into your bedroom.

Standard Point A8 of their guidelines, ninth edition, out in August 2024, bands your sleeping area four ways.

Far tighter than anything we found in federal files, and their own closing section calls those bands recommendations rather than binding limits.

Standard Point A8 opens with a principle instead of a figure, that there should be no individual noises or vibrations perceived as disturbing. Your room can measure low and still fail that sentence, which is roughly how one dripping gutter takes a whole night off you.

Those bands and every federal figure above don’t belong on one scale. Building biology values are a plain average, with no night penalty inside them. A day-night average is a different animal. Set one against the other and your bedroom at 45 looks like it belongs in the severe band. That comparison was never available to anybody.

One pair does survive, and we think it’s worth more than the rest of this section. That 32 is itself a nighttime average with no penalty added, so it belongs beside the table above, and it falls inside slight anomaly. Code sets a floor and caution aims at a ceiling. Those two agree far more closely than their headline figures suggest.

Building biology also wants two weightings rather than one. It asks for dB(C) alongside dB(A), and dB(Z) for anything under 50 Hz, which otherwise barely shows up. Highway rumble and your neighbor’s heat pump both live down there.

Why is the road measured and not the room?

Nobody measures your bedroom. We went looking, and every federal figure near roads and airports gets taken outdoors, roughly 6.5 feet from a wall, and exists to judge a project rather than a house.

Highway rules give homes like yours 67 dB(A) for an hour, measured outside. The same table carries an indoor figure, 52 dB(A), and look at what it covers: auditoriums, hospitals, libraries, schools and places of worship. Your house appears once, in the outdoor row, and never again.

Read that twice, because we did. Somebody listed every indoor place worth protecting from highway noise, and left out where you sleep.

Airports work from a day-night average of 65 dB. A project counts as a noise problem when it pushes that up by 1.5 dB or more somewhere already at or above 65. Your house does qualify as noise sensitive. And the agency adds a caveat about its own figure: “The FAA recognizes that there are settings where the DNL 65 dB standard may not apply.”

Federal housing rules draw the line in that same spot. A site at 65 dB or below is acceptable, no special approvals needed. They ask for 5 decibels of extra soundproofing between 65 and 70, and 10 decibels between 70 and 75.

Now watch what two of those figures do together. We ran this arithmetic ourselves, so treat it as ours rather than as an official finding. That 1974 document assumes an average 15 dB drop between outdoors and indoors with windows partly open. Apply it to a site called acceptable at 65 and you arrive near 50 inside a bedroom. HUD’s own interior goal is 45.

Acceptable outside. Five decibels past their own goal inside. Nobody checks yours.

Somebody did eventually go and ask people how all this feels. The Neighborhood Environmental Survey mailed a questionnaire to more than 10,000 residents living near 20 airports. Replies came back showing a substantial rise in people highly annoyed, across the entire range, quiet end included. A follow-up call found most of the people likely to be bothered reporting they had been startled, frightened or awakened by aircraft at home.

That last word is the whole point. Awakened. And we can’t find anybody who set a limit on it.

What makes a bedroom quieter tonight?

Start with your window, because every federal figure on this page already assumes one. That 15 dB drop between outside and inside assumes windows partly open, so shutting yours is the cheapest move we can offer you and it happens tonight.

Then think about position before you think about treatment. A bedroom on the far side of a house from a road starts ahead of any window you can buy. Federal housing rules say as much: quieting a site comes before soundproofing a building. So if you’re choosing between two rooms, you already own the strongest tool on this page, and it costs nothing.

Measure before you spend, and we’d do it in two ways. Standard Point A8 asks for a C-weighted figure alongside an A-weighted one, and that pair separates a rumble from a rattle. They want different fixes, and A-weighting alone talks you out of hearing the first.

Then borrow that standard’s own guiding principle while you decide: “Any risk reduction is worth it.” Feasibility comes first. Crossing a band tells you a room is louder than a cautious standard prefers. It doesn’t tell you anybody sleeping in it has been harmed, and the standard says as much about its own figures.

One last thing, plainly. Your real question was never whether you’re above a legal ceiling, because nothing federal comes anywhere near your bedroom. It’s what your room does at two in the morning, how far that’s from 32, and which fix you can afford this year. We’d rather hand you those three than a limit nobody ever wrote.