An extractor fan is the least glamorous thing in a bathroom and the one that decides whether the rest of it survives. Black mould in the grout, paint peeling above the shower, a ceiling that has gone soft, a door swollen in its frame — those are ventilation failures wearing a decorating problem's clothes. And the depressing part is that most of the bathrooms we see with mould already have a fan. It is running. It is simply not extracting anything.
This guide is about why that happens, and about the question that governs the whole job in Central London and gets asked last instead of first: where is the air actually going?
What the fan is there to do
A shower puts a remarkable amount of water into the air in a few minutes. That moisture goes somewhere. If it is not removed, it condenses on whatever is coldest — an external wall, a window reveal, the ceiling, the back of a cupboard — and stays there long enough to grow things. Mould is not a cleaning failure. It is a moisture-removal failure, and no amount of bleach fixes the cause.
The point of a fan is to get that air out of the building before it condenses. Bathrooms engage Building Regulations Part F for ventilation, alongside Part G for sanitation and hot water and Part P for the electrics, so this is a regulated matter rather than a preference. Our guide to building regulations for bathrooms covers how those parts fit together across a refit.
Note the phrase "out of the building". That is where Central London gets difficult.
The real question: where does the air go?
In a house, a bathroom usually has an external wall or a loft above it, so the duct run is short and obvious. In a Central London flat it very often has neither. The bathroom is internal — carved out of the middle of a plan when a large flat was subdivided — with no window, no external wall, a neighbour above and a neighbour below. So the air has to travel, and the route is where the job is won or lost.
The routes that actually exist, roughly in order of how well they tend to work:
- Straight out through an external wall. The best outcome and the least available. It needs a wall you are permitted to put a grille through — and on a listed elevation or in a conservation area, that permission is a real question, not a formality.
- Into an existing communal duct or riser. Common in purpose-built mansion blocks, which frequently have shared extract shafts. When one exists and is working, it is the natural answer. But it belongs to the building, not to you, connecting into it needs the freeholder's or managing agent's agreement, and older shafts are sometimes blocked, sometimes shared in ways that move smells between flats, and sometimes fitted with equipment that fights an individual fan.
- A long duct run through a ceiling void to an outside face. Workable, and the most common answer in conversions. Also the one most often done badly.
- Into a void, a loft, a ceiling space or a "somewhere". Not a route. It is a way of relocating your moisture problem into the structure, where it will do more expensive damage more quietly. If a previous installation terminates into a void, that is a defect regardless of how neat the grille looks.
Establish the route before choosing the fan. Every disappointing installation we are called to started with a fan chosen from a box and a route improvised afterwards.
Why long ducts kill fans
Here is the thing people find counter-intuitive. A fan's stated performance is measured in ideal conditions. Put it on a long duct with several bends, a flexible hose that sags between joists, and a grille with a shutter on the end, and its actual output collapses. It still spins. It still makes a noise. It moves almost nothing. Homeowners then conclude the fan is too weak and buy a louder one, which moves almost nothing more loudly.
What fixes it is duct design, not fan power: rigid duct rather than sagging flexible, the shortest sensible route, as few bends as possible, correct diameter maintained all the way — no squeezing a round duct into a smaller hole because a joist was in the way — and a proper termination. Where a run is genuinely long, an inline fan mounted along the duct rather than a small unit at the ceiling grille is generally the better engineering answer, because it is designed to work against duct resistance, and the motor noise ends up remote from the room. In a mansion block with a bathroom in the middle of the plan, that is frequently the only design that performs.
Sagging flexible duct also collects condensate, which pools at the low point and eventually finds a ceiling. A duct that dips is a leak waiting for a schedule.
Air has to get in as well
The other half of ventilation gets ignored. A fan cannot remove air from a sealed box. If the bathroom door is tight to the floor and the fan is running against a closed room, it will achieve very little. Provision for air to enter the room — typically an undercut at the door — is part of the design, and it is why an expensive fan in a room with a heavy draught-sealed door underperforms.
The electrics: zones, isolation and control
Bathroom electrics are more prescriptive than the rest of the flat, for the obvious reason. Installations are designed and tested to BS 7671, which sets out zones within a bathroom and what may be installed where, and the work falls under Building Regulations Part P and must be certified. Practically, this affects the equipment selection — its suitability for the position it occupies — the protective measures on the circuit, and how the fan is isolated for maintenance.
That last point matters and gets skipped. A fan needs a means of isolation so that someone can work on it safely, and in a bathroom that usually sits outside the room. If a fan has been wired with no accessible way of isolating it, servicing it becomes an unnecessarily invasive job.
Control is a design choice with practical consequences. A fan tied to the light switch runs only while someone is in the room, which is exactly when the moisture has not yet condensed — hence overrun timers, which keep it running afterwards, which is when the work actually gets done. Humidity-sensing units respond to conditions rather than to the light switch, which suits an internal bathroom used by several people. Continuous low-rate units run quietly all the time and boost when needed, which suits a flat with a persistent damp problem. None is universally right. What is universally wrong is a fan that only runs for the two minutes somebody is drying their hair.
Consents: check before you cut a hole in anything
Two permissions bear on this, and they are separate.
If the building is listed, works affecting its special character typically require listed building consent, and that includes the interior as well as the elevation. A new grille on a protected façade is exactly the kind of intervention that needs asking about — and the answer is sometimes no, which changes the design rather than ending it. In a conservation area, additional planning controls may apply to anything visible externally. Confirm the position with your local authority before you commit to a route. Our guide to listed building consent for bathrooms works through the wider version of that question.
If you are a leaseholder, most leases require the freeholder's or managing agent's consent for alterations, and ventilation touches this more than most bathroom work does: a new grille is a change to the external face of the building, a duct through a ceiling void may pass through space that is not part of your demise, and connecting into a communal shaft is by definition a change to something shared. Many freeholds in this part of London sit with large estates and their licence-to-alter requirements vary — read your own lease and ask. Our guide to a licence to alter in a mansion block explains that process.
Access, timing and the block
The Central London practicalities apply as they always do. Working in a bathroom in the middle of a flat means carrying materials in through a communal entrance and a lift that will need protecting and often booking with the porter. Many leases restrict working hours, which matters here because ducting through a ceiling can be a noisy hour or two. And if the route involves the communal riser, the managing agent decides when that is opened, not you.
One more: if the fan is being replaced as part of a wider bathroom refurbishment, the duct route has to be designed at the beginning, before the ceiling is closed and before the tiling starts. A duct is easy while the ceiling is open and close to impossible afterwards. Sequencing it late is the most expensive small mistake in a bathroom refit.
How we do it
We start with the route, then the duct, then the fan, then the control — in that order, because reversing it produces a fan that does nothing. We will tell you if the honest answer is that the existing route is a dead end and the fix is a different one. Our electrical work starts from £80, with set prices published on the electrical page, and bathroom work is quoted as a fixed price against a written scope before it starts. Full bathroom refurbishment starts from £2,800.
Where mould has already taken hold, the fan is only part of the answer and the making-good is the rest. For landlords and managed portfolios, recurring mould complaints are cheaper to solve properly once than to repaint every year, and they sit naturally within planned maintenance. Local work across the W9 mansion blocks is covered by our Maida Vale electrician page.
If your bathroom has a fan and still has mould, send us a photo of the fan and, if you can, of where the duct goes. That second photo is the one that answers the question.