Free tool
How Loud Is It, Really?
Sound is the most common reason somebody is unhappy with an otherwise correct installation, and it is the specification figure customers are least equipped to read. 19 dB(A) and 25 dB(A) look almost the same. In a bedroom at two in the morning they are not. This page explains what those numbers mean, using everyday comparisons to give them a sense of scale.
What this tool can and cannot tell you
The everyday comparisons here are rough anchors, not measurements — how loud a real library or a real conversation is depends entirely on the room and the moment. Manufacturers measure sound under their own test conditions, so figures from two brands are a guide to relative quietness rather than numbers you can compare to the decibel. A quoted figure is usually the quietest fan speed, not the level while a system is pulling a room down to temperature. And correct sizing affects how a system sounds in a room more than its specification does.
The five levels you will see quoted
dB(A) is a measurement of sound, weighted for human hearing. The scale is not a straight line, so a small numerical change can be very noticeable — which is why a handful of decibels matters more than it looks. Select a level to read what a system at it is actually like to live with.
Every figure here is sound pressure at a stated distance, which is the measurement closest to what you would actually hear. It is not the same as sound power, and the two should never be compared directly — there is more on both below.
- Roughly like
- For illustration only: about the level of a very quiet bedroom at night, with the window shut and nothing running.
- What it is like in a room
- At the very bottom of what a person notices in a still room. Most people have to stop and listen for it, and in a room with any background sound at all it disappears entirely.
- Typical of
- The lowest fan speed on a well-specified domestic wall-mounted unit. Not every unit will go this low, and it is worth checking the figure rather than assuming.
- Where it suits
- This is the level worth paying attention to for a bedroom or a nursery, where the system runs while somebody is trying to sleep.
The everyday comparisons above are for illustration only. They are rough anchors to give the numbers a sense of scale, not measurements and not exact equivalents — how loud a real library or a real conversation is varies widely, and depends entirely on the room, the distance and the moment. Use them to get a feel for the difference between one figure and another, not to predict what a particular system will sound like in a particular room.
Put two side by side
The most useful thing to know about two figures on two quotations is whether the gap between them is one you would ever notice.
16 dB(A) apart — a large difference. A gap of ten decibels is conventionally described as roughly twice as loud, so this is not the small step the numbers make it look.
At 19 dB(A): At the very bottom of what a person notices in a still room. Most people have to stop and listen for it, and in a room with any background sound at all it disappears entirely. At 35 dB(A): Clearly present in a quiet room, and unremarkable in a busy one. You would notice it stopping. It sits under conversation and under a television without competing with either.
Why a few decibels is a bigger difference than it looks
3 dB(A)
3 dB(A) apart
About the smallest difference most people reliably notice. Two units three decibels apart sound similar side by side, and you would struggle to tell them apart from memory.
6 dB(A)
6 dB(A) apart
A difference you would notice immediately if one replaced the other in the same room. This is the gap that separates a unit somebody is happy with from one they mention.
10 dB(A)
10 dB(A) apart
Conventionally described as roughly twice as loud, even though the numbers only differ by ten. This is the point of the whole tool: the scale is not a straight line.
What to know before you compare two figures
- What dB(A) actually is
- A decibel is a measurement of sound. The (A) means it has been weighted for human hearing — adjusted so that the frequencies people hear well count for more than the ones they barely hear at all. It is the figure that best reflects how loud something seems, rather than how much sound energy is physically present.
- The scale is not a straight line
- Decibels are logarithmic. That is why ten decibels is conventionally described as roughly twice as loud rather than a fifth louder, and why a small numerical change can be very noticeable — the difference between 25 and 32 dB(A) is far bigger than seven out of twenty-five suggests.
- Published figures are measured under controlled conditions
- A quoted sound level is taken in a test environment designed to make the measurement repeatable, at a stated distance from the unit. Your room is not that environment, so treat the figure as a fair basis for comparing units rather than as a prediction of what you will hear.
- The room changes what you actually hear
- How far you are from the unit, what the walls, floor and furnishings are made of, where the unit is mounted and how much background noise there already is all change the result. A hard-surfaced room with a unit above a desk sounds quite different from a carpeted room with one above a doorway, at the same quoted figure.
- Manufacturers measure differently
- Each manufacturer quotes under its own test conditions, so two figures from two brands are a guide to relative quietness rather than numbers you can put side by side and trust to the decibel.
- Sound pressure and sound power are different measurements
- Every figure on this page is sound pressure — how loud it is at a stated distance, which is what you would hear. Manufacturers also publish sound power, the total sound produced regardless of distance, which is always the higher number. The two cannot be compared with each other.
- The quoted figure is usually the quietest setting
- A headline sound level is normally the lowest fan speed. The unit will be louder while it pulls a room down to temperature, and quieter once it is holding it.
- Sizing affects noise more than the specification does
- A correctly sized system settles to a low, steady output and stays there. An oversized one cycles up and down, and it is the changes people notice and complain about — not the level.
- Where it goes matters as much as what it is
- The same unit is a different experience above a bed and above a doorway. Position, and the position of the outdoor unit relative to your windows and your neighbours', are decided at survey.
Related services
What this tool is a first step towards.
Want this checked properly?
A survey measures the things this page had to ask you to estimate. It is free, there is no obligation, and you get an answer based on your building rather than on a general rule.
Before you rely on this
- This is indicative, not a specification. It is a guide to help you plan and ask better questions. It is not a design, not a quotation and not a substitute for a survey.
- A survey is what decides. An engineer measures the things this page can only ask you to estimate, and confirms where equipment can physically go, how it will be run and what the electrical supply allows.
- Installation and refrigerant work are regulated. Air-conditioning systems contain refrigerant under pressure and are connected to a fixed electrical supply. Installing, moving, opening, recharging or decommissioning one is work for a qualified, F-Gas registered engineer — never a DIY job.
- We do not publish a saving. Nothing on this page claims what you would save against any other way of heating or cooling a room. What you would save depends on what you have now, how you use it and what you pay for energy, and we will not put a number on that without knowing all three.