Do flat rooflights cause condensation?

Why the moisture shows here first

Air holds water in proportion to its temperature. Cool it far enough and the water comes out of it onto whatever surface did the cooling. That temperature is the dew point, and in an occupied house in winter it is commonly somewhere around 10 to 14 degrees.

A rooflight sits at the highest point of a room, and warm damp air collects there. It contains the largest area of glass in the ceiling, and glass runs cooler than plaster. It is surrounded by a reveal that projects up into the cold roof build-up. Every one of those is a reason for water to appear at the rooflight before it appears anywhere else in the room. The unit is the thermometer, not the cause.

The cold bridge around the kerb

Here is the failure that produces almost all of the complaints, and it is a construction fault rather than a product fault.

A modern flat roof puts insulation above the deck. Cut a hole for a rooflight and build a timber kerb around it, and unless that kerb is insulated on its outside face, you have created a continuous loop of solid timber running from the warm room, up through the insulation layer, to the cold air. Timber conducts heat far better than insulation board does. The inner face of that loop runs several degrees below the ceiling either side of it, and in cold weather it drops below dew point while the rest of the ceiling stays above.

Water then forms in a line all the way around the reveal, and along the bottom of the frame where the two meet. It looks exactly like water tracking in from a failed perimeter, which is why people call a roofer.

Telling the two apart before anybody goes up

Timing and pattern separate them reliably, and you can do this from the floor.

Where the picture is genuinely mixed, both can be present at once. The diagnostic order for actual ingress is on why a flat rooflight is letting water in.

Surface water against water inside the build-up

Two different problems share the word condensation, and only one of them is visible.

Surface condensation is what you can see: water on the glass, the frame or the reveal. It is unsightly and it will mark plaster over time, but it dries and it is fixable. Interstitial condensation happens inside the roof build-up, where warm damp air has got past the vapour control layer and reached a cold surface within the construction. That one you do not see until timber has been wet for a long time. It is the reason the vapour control layer has to be continuous around a rooflight opening rather than simply stopping where the hole starts.

Where the vapour control layer has to go

On a warm roof the vapour control layer sits directly on the deck, below the insulation, keeping room air out of the cold part of the build-up.

At a rooflight it has to be carried up the inner face of the kerb and sealed to the underside of the rooflight frame, so the barrier is unbroken all the way round. That is a taped, deliberate detail, and it takes a few minutes per side to do properly. Left undone, the opening becomes a chimney feeding warm damp air straight into the insulation layer. It is the detail we check first on any survey where a ceiling shows a history of water, because it explains cases that make no sense otherwise.

The room below is producing the water

The kerb decides where the water lands. The room decides whether there is any to land.

A kitchen extension with a hob, a bathroom or wet room, a utility with a vented dryer, or a room where washing is dried indoors all put a great deal of moisture into the air. So does a new extension in its first winter, when the screed, plaster and blockwork are still giving up construction water. A rooflight installed in October into a room finished in September will show water that has nothing to do with the installation and will not return the following year. Rooflights over genuinely wet rooms need their own approach, covered on flat rooflights over a bathroom or wet room.

Moving air across the glass

Even with a properly insulated kerb, a large area of glass in a moist room will occasionally mist at the edges on the coldest nights. Air movement is what clears it.

A trickle ventilator in the frame or the kerb gives a continuous low rate of exchange that keeps the humidity in the room down without opening anything. Mechanical extract in kitchens and bathrooms does the heavy work, and it needs to run for long enough after use rather than switching off with the light. Where the reveal is splayed rather than square, air circulates across the glass instead of stagnating in a shaft, which is a small effect but a real one.

Frame material and warm edge spacers

The unit itself contributes, and it is worth knowing what to look for on a specification.

An aluminium frame without a thermal break conducts heat straight through, and the inner face will run cold. Any current flat rooflight worth fitting has a polyamide thermal break separating the inner and outer sections. The spacer bar between the panes matters too: an aluminium spacer chills the edge of the glass and produces a band of misting 20mm to 30mm in from the frame, where a warm edge spacer in composite or stainless does not. Neither of these will rescue an uninsulated kerb, but both remove a cause you would otherwise be living with.

What we do at the perimeter

Insulation carried up the outer face of the kerb as a continuation of the deck insulation, with no gap at the base. Vapour control layer carried up the inner face and sealed to the frame. Thermal break in the frame and a warm edge spacer in the glass as standard. Reveal splayed where the room benefits from it.

Four details, none of which you can see once the plastering is done, and together they decide whether you spend February wiping a ceiling. That is why the workmanship carries a ten-year guarantee rather than the products alone. We install and replace across Oxford and up to twenty five miles by road, with fifteen years in the trade. Call 01865 704245 or email info@heritageskylightsoxford.co.uk.