Why a 33 Decibel Fan Is Louder Than a 30 Decibel One

This page contains affiliate links. If you buy through one we may earn a commission, at no cost to you.

It does not have one. That sounds like pedantry and it is the entire practical answer, because a decibel figure for a fan is a pressure reading taken at some distance, under a weighting curve that deliberately ignores part of what you can hear, and both of those decisions are usually unstated. Two fans whose numbers are eleven decibels apart on the box can swap places at your ear, and the arithmetic that shows it takes one logarithm.

We have quoted these figures ourselves without explaining them. Our piece on why a fan's CFM rating is not the breeze you feel cites 33 dB(A) on one fan, below 40 dBA on another and at or below 60 dB on a third, and the most it does is note that the last one is a ceiling rather than a figure. This page is the part we owed that one.

Pressure falls with distance, by six decibels a doubling

Sound spreads out. The power leaving the fan is fixed, the area it is spread over grows with the square of distance, so the pressure at any point falls with the square of distance too. In decibels that works out at about six fewer every time you double your distance, or six more every time you halve it.

The convention for a published figure is one metre, and you should know that it is a convention rather than a disclosure. Not one of the listings below states a measuring distance. So run the conversion yourself and compare three fans at the distance each is actually used from.

A VZXAB 120mm RGB case fan publishes about 21.6 dBA at about 1650 RPM, with eight anti-vibration pads, a PBT frame, glowing RGB blades and an eight-effect loop on a 4-pin connector. That is the quietest number here by a street. But a case fan lives maybe two feet from your head, so at 0.6 metres it is about four and a half decibels up on its own figure, which is 26.

A TALOYA 52 inch ceiling fan with light and remote publishes a stated under 30 dB from a reversible DC motor with six speeds, a 15W LED dimming from 10 to 100 percent across 3000K, 4000K and 6500K, and a flush mount. Thirty is worse than 21.6 on paper. But it is on the ceiling, something like 2.3 metres above a seated head, which takes about seven decibels off: call it 23.

And a VIVOSUN AeroWave E6 Gen2 oscillating clip fan publishes 320 CFM at 33 dB(A) from a dual ball bearing EC motor, with five-level auto oscillation, obstacle auto-retract, a natural wind mode, a local three-cycle timer, an IP54 rating and a patented clamp for round or square poles. Thirty three is the worst figure of the three and it is clamped to the edge of your desk, perhaps 0.4 metres from your head, which adds about eight: call it 41.

So the paper ranking runs case fan, ceiling fan, clip fan, spread over eleven decibels. The ranking at your ear runs ceiling fan, case fan, clip fan, spread over eighteen. The winner changed, and eighteen decibels is roughly four times as loud in the way people actually experience loudness, where ten decibels is about a doubling. Our reviews of all three set out what each publishes: the VZXAB case fan, the TALOYA ceiling fan and the VIVOSUN AeroWave E6.

Then the weighting throws away the part you hate

The A in dBA is a filter. It attenuates low frequencies on purpose, because the ear is less sensitive down there at quiet levels, and the attenuation is severe: about 8.6 decibels at 250 Hz, 16 at 125 Hz, 26 at 63 Hz, 39 at 31.5 Hz, and about 50 by 20 Hz, which is the bottom of where the curve is even defined.

Now work out where each fan puts its energy. The strongest tone a fan makes is its blade passing frequency, which is the rotation rate multiplied by the number of blades. The case fan above runs at about 1650 RPM, and a 120mm fan typically has nine blades, so that is 1650 divided by 60 times 9, around 248 Hz. A-weighting discounts that by under nine decibels, so its rating captures nearly all of what it produces. A 52 inch ceiling fan with three blades runs at maybe 150 RPM, giving a blade passing frequency around 7.5 Hz. That is below the bottom of the A-weighting curve, discounted by fifty decibels or more wherever you place it.

Which means the two numbers are not the same kind of number. The case fan's 21.6 is close to a description of its whole output. The ceiling fan's under 30 describes everything except the slow thump you feel in a quiet bedroom, because that thump has been filtered out by 50 decibels before anybody wrote the figure down. Both are honest measurements. Neither answers the question people are asking, which is whether the thing will annoy them.

And note what you cannot work out. The clip fan publishes ten speed levels and no RPM at all, so there is no way to calculate its blade passing frequency, which means there is no way to tell what its 33 dB(A) is hiding. That absence is the gap worth noticing in the listing, more than the number itself.

One figure to be sceptical of in the other direction: 21.6 dBA. A tenth of a decibel is below what anyone can hear as a difference, below the repeatability of a practical measurement, and it is quoted with no distance, no speed condition and no standard. The precision is decoration. We would read it as about 22 and treat the first digit as the only one that means anything.

Two listings that tell you something by saying nothing

An iLiving 18 inch wall mounted fan rated at 6360 CFM publishes a great deal: a 168 watt sealed motor, UL listing with an OSHA approved grill, a weatherproof corrosion resistant finish, an adjustable bracket, 20 pounds, compatibility with a separately sold misting kit, and three pull chain speeds with no thermostat, humidistat or timer. What it does not publish anywhere is a noise figure. On a 168 watt motor driving 18 inches of blade that is not an oversight, it is a choice, and it is more informative than a number would have been. Our review of the iLiving wall fan has the full record.

A SEAOURA aquarium chiller fan with timer and temperature control goes one step further and makes the claim in words instead: ultra low noise, quiet operation, alongside a specification that is otherwise unusually complete, with a 3 to 7 Fahrenheit cooling drop, an adjustable threshold range from 60 to 104 Fahrenheit, ten speeds, one to three fan heads, a 24 hour timer and a 5.4 inch gooseneck. Everything else there is a number. Quiet is an adjective. That is worth noticing precisely because the rest of the listing shows they were willing to publish figures where they had them. See our review of the SEAOURA chiller fan.

The free lever, which is bigger than any upgrade

Fan noise rises very steeply with speed. The standard relationship used in fan engineering is that sound power level changes by about fifty times the logarithm of the speed ratio, which gives a result worth memorising: dropping the speed by twenty percent takes roughly five decibels off, for twenty percent less air.

Five decibels is half way to being perceived as half as loud, and it is one click on a dial you already own. Set against it, the gap between two decent fans of the same size is often two or three decibels. So the quietest fan in your house is the one you already have, turned down, and the right way to buy for quiet is to buy a size bigger than you need and run it slow rather than buying a smaller one with a better number and running it flat out.

Then move it, because that is the other free six decibels a doubling. Going from half a metre to two metres is twelve decibels, which is more than the entire published spread between the three fans above. If a fan has to be close, aim it past you rather than at you: the blade noise is loudest straight down the axis.

And decouple it, because structure borne noise sidesteps all of this. A fan screwed hard to a thin panel turns that panel into a loudspeaker, radiating exactly the low frequencies that A-weighting discards, which is why the case fan above ships with eight anti-vibration pads and why they are not a gimmick. A folded towel under a floor fan, or rubber grommets instead of screws, costs nothing and attacks the part no rating describes. We touched on the mounting side in PC cooling and case fans, and room sizing for a ceiling fan in choosing the right ceiling fan for a room.

The honest limit

The six decibels a doubling rule assumes a free field, which your living room is not. In a hard, bare, tiled room the reflections build up and beyond a metre or two the level stops falling much with distance, so moving away helps less than the arithmetic promises. In a carpeted, curtained, furnished room you get most of it. Which means soft furnishings are part of the fan question, oddly enough, and they are already in the room.

The other limit is that you cannot compare across unit systems. Sones are a loudness scale, where one sone is defined against a 1 kHz tone at 40 decibels and doubling the number means doubling the perceived loudness. Decibels are a pressure scale where ten more is roughly a doubling. A bathroom fan at 1.5 sones and a clip fan at 33 dBA cannot be ranked against each other from those two figures alone, and any listing that gives you one while a rival gives the other has made the comparison impossible rather than close.

So the useful version of the original question is not what the noise level of a fan is. It is what level this fan produces, at the distance I will sit, at the speed I will run it, in a room like mine. Three of those four are yours to set, and all three are free.