The ear gives up, the body does not
Habituation to noise is real, and it happens fast. After a few weeks in a flat on a busy road, you no longer notice the cars, you no longer register the bus, and you tell visitors that you have stopped paying attention. All of that is true and none of it is evidence. The mechanism is attentional: the brain stops directing awareness towards a repetitive sound that carries no new information. It does not stop processing it.
During sleep, a single noise event can trigger a micro-arousal, shift the sleeper from a deep stage to a light one, briefly raise heart rate and blood pressure. These reactions are recorded in sleep laboratories in people who report, the next morning, that they heard nothing and slept well. The gap between the account and the recording is the central fact of this subject.
One practical consequence follows. Asking occupants how bothered they are tells you something useful about comfort, and nothing reliable about exposure. People who have lived somewhere for ten years usually report less annoyance than recent arrivals at the same sound level. That is why a sound level meter goes where the sleeper’s head actually is, over several nights, rather than in the middle of the living room during the day.
The 53-decibel threshold
For road traffic, the reference point is 53 decibels Lden, the indicator that averages levels over twenty-four hours with a penalty applied to evening and night. Above that level, the relationship between road traffic exposure and cardiovascular risk is established in the epidemiological literature. The figure deserves to be read correctly: it is a population threshold, derived from large cohorts followed over time. It marks the level at which an effect becomes visible across a group. It says nothing about what will happen to one individual, and it is not a diagnosis.
The guidelines published by the World Health Organization in 2018 set different reference points depending on the source, which is the most instructive part of the document.
| Source | Lden reference | Night reference (Lnight) |
|---|---|---|
| Road traffic | 53 dB | 45 dB |
| Railway traffic | 54 dB | 44 dB |
| Aircraft | 45 dB | 40 dB |
These are outdoor levels, measured at the facade. They do not compare directly with what is recorded inside a bedroom with the window closed, where the expected background level is far lower.
Why aircraft noise is worse at the same decibel level
The table shows an eight-decibel gap between road and aircraft. That is not a typo: at equal sound level, aircraft noise is the most disturbing of the three. Several features contribute. An overflight is a discrete event with a steep rise and fall, whereas road traffic forms a more continuous blanket that the ear settles into. Aircraft noise arrives from above, through the roof and through openings that nobody treats acoustically. It carries a low-frequency component that building elements attenuate poorly. And it is experienced as beyond the resident’s control, which weighs on reported annoyance.
In practice: a home at 50 dB of aircraft noise raises more questions than a home at 50 dB of road noise, and a measurement that does not distinguish the sources misses the point.
The noise you generate yourself
This is the part most often left out of an assessment. Outdoor noise fluctuates, pauses, and can be taken up with the building management. Equipment noise runs all night, with no gaps that would let the sleeper recover.
The usual sources:
- Mechanical ventilation, particularly when the extract vent sits close to the bed head or the unit was installed in a cupboard sharing a wall with a bedroom.
- The outdoor unit of a heat pump or air conditioner, whose compressor produces a low-frequency hum that passes through glazing far more easily than street noise.
- The fridge in a kitchen open to a room where someone sleeps.
- Heating circulators, building booster pumps, and modern lifts whose plant room adjoins a bedroom.
The World Health Organization uses 30 dB(A) as an acceptable background level in a bedroom at night. Plenty of bedrooms fitted with poorly installed balanced ventilation sit above that, and nobody has ever checked. The useful measurement is taken around three in the morning, with equipment running normally, then repeated after shutting each machine down in turn to isolate its contribution.
What glazing changes, and what it does not
Acoustic glazing works well on mid and high frequencies: voices, horns, brakes, tyre roll. It does little for low frequencies, which are exactly those of lorries, buses, compressors and departing aircraft. Standard 4/16/4 double glazing is worth very little acoustically. The gain comes from asymmetric laminated glass with a suitable interlayer.
More importantly, a window is only as good as its weakest point. In most cases that is not the glass but the roller shutter box, the trickle vent in the head member, the seals on the opening leaf, or the facade lining. Replacing the glass while leaving an uninsulated shutter box produces a disappointing result, and that disappointment is common.
Glazing also does nothing about noise transmitted through the building structure: slamming doors, lifts, pipework. It does nothing about noise generated inside the home. And it creates a trade-off worth naming: a window kept shut all night in a home without functioning mechanical ventilation degrades indoor air quality. Treating noise without treating air renewal simply moves the problem.
What we do, in order
The order matters more than the list. We start with a full night at the bed head, with the level history broken down over time, so that continuous background can be separated from events. We then identify sources by switching them off one at a time. We deal with equipment before the building envelope, because it costs less and usually achieves more: isolating a unit, changing a vent, relocating a compressor, fitting an attenuator. The envelope comes next, starting with the weak points rather than the glass. Then we measure again, in the same spot, under the same conditions.
Lif8 measures places and corrects installations. We are not health professionals and these measurements replace no medical opinion: persistent sleep problems are a matter for a doctor, including once the bedroom has been put right.
