Do electric rooflights leak more?

The mechanism lives inside the weather line

Take a section through an opening rooflight and the order is fixed. Outside, the roof covering laps onto a flashing kit. The flashing laps onto the outer frame. The outer frame carries a drained and vented channel. The sash closes onto a gasket seated in that channel. Only inside all of that does the actuator housing sit, screwed into the head of the frame, in the dry.

Water arriving at the glass runs down the pane, onto the outer frame, onto the flashing and away down the slope. It has to defeat the gasket, then the drainage channel, then the channel’s own outlets before it is anywhere near the mechanism. A unit that gets water to the actuator has a problem several stages earlier than the actuator.

Where water actually enters on a rooflight

In our experience the causes are unglamorous and they are all outside the unit. An upstand built too low, so driven rain and standing snow sit above the level the flashing turns up to. A flashing lap running the wrong way relative to the fall, so water is invited under rather than shed over. A roof covering terminated short of the flashing skirt. A flat roof laid without enough fall, so water ponds on the up-slope side of the kerb and finds the weakest point at leisure.

On a pitched slate roof the classic is a soaker set too tight to the frame, leaving nowhere for the water to run round. On a warm flat roof it is the corner of the kerb, where three planes meet and the covering has to be dressed rather than simply laid. None of those faults care whether there is a motor in the frame.

Gasket compression, and why a motor closes better than a hand

Here is the part that runs the other way. A powered sash tends to seal more consistently than a manual one, not less.

A manual rooflight is closed by a person pulling a handle or winding a pole until it feels shut. Feel is not a measurement. On a cold morning the gasket is stiff, on a hot afternoon it is soft, and the sash gets latched at whatever position the hand stopped at. A chain actuator closes to a fixed end point every single time, at the same force, and holds there through the geared drive without relaxing.

That matters because a gasket only works when it is compressed across its whole perimeter. Uneven compression at one corner is how wind-driven rain finds a route into the drainage channel faster than the channel can shed it.

Travel limits, and a sash that reads closed but is not

The corollary is that commissioning matters. Every powered unit has end-of-travel settings that tell the mechanism where closed is. Set the closed limit a couple of millimetres short and the sash sits on its gasket with almost no compression at the hinge end while looking perfectly shut from the floor. Set it too far and the chain stalls against the frame every cycle, which shortens the life of the drive.

This is a five-minute job at handover and it is the single commonest reason a powered unit underperforms a manual one on weathertightness. We set travel limits at commissioning and check the compression line round all four sides before we leave.

Cable entries: the one genuine addition

To be fair to the question, a mains-powered rooflight does introduce something a manual unit does not have, which is a cable arriving at the unit. Done properly this is a non-event. The cable enters through a grommeted opening in the head of the inner frame, on the warm side, inside the vapour control layer, and never passes through the outer frame or the flashing.

Done badly it is a hole drilled through the frame from outside because nobody planned the route at first fix. That is why the cable is run while the structure is open, before plasterboard, and why first fix wiring is coordinated with the electrician rather than improvised. A solar unit sidesteps the question entirely, having no cable at all.

Condensation is not water entry, and it looks identical

A large share of the wet marks reported on rooflights are not rain. Warm moist air from a kitchen or bathroom meets the coldest surface in the ceiling, which is the glass and the frame, and gives up its moisture there. The water runs down the pane, collects at the bottom of the sash and drips onto the reveal, exactly where you would expect rain to appear.

Two tells separate them. Condensation appears on still cold nights with no rain, and it appears evenly along the bottom edge rather than at one point. Water entry appears during driven rain, usually from a specific corner, and often lags the weather by an hour. Electric units are slightly more prone to condensation reports for a mundane reason: they are the ones people leave shut all winter because the switch is there.

Upstand height, and the number worth arguing about

On a flat roof the figure that decides most of this is the height of the finished upstand above the finished roof surface. We work to 150mm as a working minimum and we will say so at survey even when it costs us the sale, because a kerb built to 75mm to keep the ceiling line generous is the reason a rooflight lets water in three winters later.

On a pitched roof the equivalent figures are the pitch itself against the unit’s minimum, and the distance from the frame to the nearest abutment, which decides whether the flashing can be dressed properly. Both are settled with a tape at survey, and both are more important than which mechanism is inside the frame.

What the ten-year guarantee is actually covering

Four details decide whether a rooflight stays dry, and none of them are visible once the job is finished: upstand height, the fall of the surrounding covering, the direction and size of the flashing laps, and where the covering terminates against the kit. Those are the four we guarantee for ten years, because they are the four the homeowner cannot inspect.

The motor is covered by the manufacturer’s own warranty on the unit, which is a separate document with a separate term. Read both. We install VELUX, Fakro, Keylite and Roto units across Oxford and up to twenty five miles by road, and we quote a range by email then a fixed price after a survey.