Electric Skylight Installation

Rain Sensors That Trigger Falsely or Fail to Close


A rooflight rain sensor problem comes down to one of three things: the pad is dirty and reading wet when it is dry, the pad is worn or cracked and no longer reading wet when it is raining, or the sensor is fine and the controller behind it has stopped acting on what the sensor tells it. The first is a cleaning job you can do yourself. The second and third are usually a sign the unit is at the end of its service life, and on anything past about fifteen years the honest answer is replacement rather than parts.

We fit new and replacement rooflights across Ipswich, Suffolk and north Essex. We do not repair rooflights, reseal them or patch them, so what follows is diagnosis written to help you work out where you actually stand, not a lead-in to a call-out charge. If the answer turns out to be “clean the pad”, clean the pad and forget about us.

Close-up of a rain sensor pad on the outer frame of an electric rooflight showing two interleaved conductive tracks under flat overcast daylight
Two interleaved conductive tracks. Water bridges them, resistance drops, the controller drives the sash closed.

What the sensor actually is, and why that explains the fault

The sensor is a small plastic pad, usually 40 to 60mm across, mounted on the outer face of the frame or on the head of the sash. Under the surface are two conductive tracks laid out interleaved, like the teeth of two combs pushed together without touching. A low voltage sits across them. Dry, the gap between the tracks is effectively an open circuit. Wet, water bridges the gap, resistance collapses, the controller reads the change and drives the actuator closed.

That is the entire mechanism. There is no moisture gauge, no optical detector, no clever electronics on most domestic units. It is a resistance bridge, and every common failure is a variation on the tracks being bridged when they should not be, or not being bridged when they should.

Two behaviours follow from the design and both get reported to us as faults when they are not.

  • It is not instant. Detection happens in a second or two, but the actuator takes 30 to 60 seconds to run the sash home from full stroke. In a heavy burst you will get some water in during that window. That is normal and it is not a sensor fault.
  • It holds the unit shut afterwards. Most controllers keep the rooflight closed for 10 to 15 minutes after the pad reads dry, so a passing shower does not set the vent flapping. If you press open during that hold period and nothing happens, the system is working correctly.

How the sensor talks to the actuator is covered on how does an electric rooflight open. What follows here is what goes wrong with the sensor itself.

False triggers: the rooflight closes when it is not raining

This is the more common complaint by a wide margin, and near the coast it is the single thing we get asked about most.

Salt film on the pad

Salt is conductive when damp and hygroscopic when dry, which is a bad combination for a device whose entire job is to notice when something conductive lands on it. Within about a mile of open sea at Felixstowe, Harwich, Frinton, Aldeburgh or Southwold, salt-laden air deposits a thin film across the pad over a season. The film then pulls moisture out of humid air on a still, damp morning with no rain at all, bridges the tracks, and the controller does exactly what it was designed to do.

The tell is behaviour, not appearance. A salted pad closes the unit on mornings with heavy dew or sea fret, closes it during a humid evening, and is worse in autumn than in a dry June. If your rooflight is shutting itself at seven in the morning under a grey sky and staying obedient the rest of the day, look at the pad before you look at anything else.

Fresh water and a cloth. Not a sponge with washing-up liquid, because a surfactant residue is itself slightly conductive and you will have solved nothing. Twice a year on the coast, once a year inland. Estuarine sites at Woodbridge, Wivenhoe, Manningtree, Shotley and Pin Mill carry moderate salt loading rather than open-sea conditions, and once a year is usually enough there.

Organic contamination

Inland at Ipswich, Stowmarket, Kesgrave and Bury St Edmunds, the equivalent problem is biological. Algae growth on a north-facing frame, a pollen crust after a heavy May, lime and sap deposits under an overhanging tree, or bird droppings, all of which hold water long after the roof surface has dried. A pad under a mature lime or sycamore is the worst case we see away from the sea.

Physical damage and age

The pad surface degrades under ultraviolet light. On a unit past twelve to fifteen years, the plastic goes chalky, develops fine surface crazing, and those cracks hold water by capillary action for hours after rain has stopped. You get a rooflight that closes in a shower and then refuses to reopen until the afternoon. Cleaning improves it briefly. It comes back within weeks, and that pattern is the useful diagnostic: a fault that returns quickly after a proper clean is a worn pad, not a dirty one.

Rain sensor symptoms and their usual cause
Symptom Most likely cause What it means
Closes on damp mornings with no rain Salt or algae film bridging the tracks Clean with fresh water. Fixable by you.
Closes correctly but reopens hours late Crazed pad surface holding water Pad at end of life. Cleaning buys weeks.
Closes at random, no weather pattern Damp in the connector block or cable entry Corrosion in the wiring. Not a cleaning fix.
Ignores drizzle, closes in heavy rain Normal behaviour on an unheated pad Not a fault. Heated pads reduce it.
Ignores rain entirely, manual control fine Sensor circuit or controller input failed Control system fault. See diagnosis below.
Ignores rain and manual control erratic Controller failing Unit is at end of life.
Closes then immediately reopens Controller logic fault or failing hold timer Control system fault.

No response: the rooflight stays open in the rain

The opposite fault matters more, because the failure costs you a wet floor rather than an inconvenience.

Start with the boring explanation. Fine drizzle and sea fret often will not bridge the tracks. Droplet size is too small, the water beads rather than spreading, and the pad never sees a continuous path. This is a design limitation of every unheated resistive sensor, not a fault, and it is why we specify heated pads on coastal work where sea fret rolls in off the North Sea on spring mornings at Felixstowe and Aldeburgh. A heated pad runs a few degrees above ambient, which encourages fine mist to spread into a film instead of sitting as separate droplets.

Then work through the real possibilities in order. This is all done from inside the house. Never get onto a roof to inspect a sensor, and never onto a wet flat roof under any circumstances.

  1. Confirm the unit drives at all. Open and close it on the handset and on the wall switch. If the actuator runs normally both ways, power and mechanism are fine and the fault is in the sensor circuit. If it does not, you have a different problem and a failed opening mechanism or a motor failure is the place to look.
  2. Check whether the sensor is disabled. Many controllers have a sensor-off mode, sometimes entered accidentally by a long press on the handset. Check the manual for your model. We have surveyed units reported as dead sensors that were switched off two years earlier by a child with the remote.
  3. Test it wet. If the pad is reachable from inside with the sash open, a wet cloth laid across it should trigger a close within a couple of seconds. Do not reach out of an opening at height, and do not lean out over a flat roof. If you cannot reach it safely, do not try.
  4. Note whether anything changed. A sensor that stopped working the week a new handset was paired, or after a power cut, points at the controller rather than the pad.

If it drives on demand but ignores water on the pad, the sensor circuit or the controller input has failed. On a wired sensor that is often corrosion at the connector, which is where the coastal story turns from cleaning into something structural.

Corrosion behind the pad, which is the part nobody sees

The pad is the visible half. Behind it sits a short cable run through the frame into the control gear, and every point where that cable passes through the frame is a route for damp salt air. Connector blocks fur up green, terminal screws lose contact pressure as the metal corrodes underneath, and the sensor circuit starts reading noise instead of rain.

The symptom is a sensor that behaves inconsistently rather than badly. It works in July, misbehaves in November, closes the unit at three in the morning for no reason anyone can identify, then behaves for a fortnight. Owners describe it as the rooflight having a mind of its own, and that description is a reliable indicator of a wiring problem rather than a pad problem.

It is also the fault we cannot make go away with a clean, and the reason we specify the way we do on new coastal installations: sealed cable entries, marine-grade glands where the design allows, 316L stainless rather than 304 at brackets and fixings because the molybdenum content is what resists chloride attack, and the control unit sited inside the warm envelope rather than in a cold roof void where it will condense against its own casing every night of the winter.

A sealed cable entry and marine grade gland where the rain sensor cable passes through a rooflight frame, photographed under flat overcast daylight
The cable entry is where salt air gets in. Sealed glands and a control unit inside the warm envelope.

When the sensor is a symptom, not the problem

Here is the part that usually decides the outcome. On a unit past about fifteen years, a rain sensor fault is rarely alone. It arrives alongside a handset that needs three presses, a sash that stops short of fully closed, or a controller that has started to ignore one of two paired units. Those are not four faults. They are one control system reaching the end of its life and shedding functions in the order of least redundancy first, and the sensor circuit, being the most exposed and the lowest current, tends to go first.

The commercial reality on top of that is availability. Sensor pads are model-specific and rarely stocked beyond about ten years after a range is discontinued, and where the pad is integrated into a control module rather than being a separate part, there is nothing to source at all. That whole picture, discontinued protocols, unavailable modules, pairing that no longer works, is set out on obsolete electric rooflight controls.

What we do about it. We do not repair, reseal or patch rooflights. If your sensor fault is contamination, we will tell you to clean it and charge you nothing. If it is a failed circuit on a unit still in support, your options are the manufacturer or an electrician. If the unit is out of support, we quote a replacement, and a like-for-like electric roof window replacement runs £1,600 to £2,300 supply and fit, one day on site, with a 10-year workmanship guarantee.

What a replacement changes

A current unit is not the same product with a newer sticker. The sensor pads on modern ranges sit on a sealed module rather than an exposed terminal, heated versions are readily available, hold-closed timers are adjustable rather than fixed, and the sensor is properly integrated with the controller so a failed pad reports as a fault rather than as silence.

The building fabric moves too. A replacement rooflight in an existing dwelling must achieve a whole-unit U-value of 2.2 W/m²K or better under Part L, and the unit you are replacing, if it is old enough for its sensor to be unobtainable, is almost certainly nowhere near that. You are not just buying a working sensor.

Whether the replacement makes sense yet is a separate question, and when an electric rooflight needs replacing works through the threshold. If your unit is stuck open right now with rain forecast, go to rooflight stuck open in the rain first and deal with the water.

A newly installed electric opening rooflight on a flat extension roof with the rain sensor mounted on the sheltered side of the frame, flat overcast daylight
New unit with the sensor sited on the sheltered side of the frame rather than where the instructions default to.

Common questions about rain sensor faults

Can I just disable the rain sensor?

On most systems, yes, either through a handset sequence or a switch on the control unit. Whether you should is a different matter. A rooflight over a kitchen with a tiled floor is one thing. A rooflight over a stairwell, a carpeted landing or anything with a timber floor below it is worth keeping automatic, because the scenario the sensor exists for is the one where nobody is home. Disabling it as a workaround for a fault also removes the evidence, and you will forget you did it.

How do I clean a rain sensor without going on the roof?

Open the rooflight, and if the pad is on the sash head or the inner face of the frame it will often come within arm’s reach from a step inside the room. Fresh water and a soft cloth, no detergent. If it is not reachable from inside, leave it. Do not lean out of an opening at height, do not step onto a flat roof, and do not go up a ladder to reach it. On many pitched roof windows the sash rotates fully for cleaning, which brings the sensor inboard, and that is worth checking in your model’s manual before you assume it is out of reach.

Does a heated sensor pad fix false triggering?

No. Heating solves the opposite fault. It helps a pad detect fine drizzle and sea fret by encouraging mist to spread into a continuous film rather than beading, so it reduces missed closures. It does nothing about a salt film, because the salt is still there and still conductive when damp. False triggering is a contamination problem, and cleaning is the answer to it.

My sensor works but the rooflight closes slowly. Is that a fault?

Not on its own. A chain actuator takes 30 to 60 seconds to run home from full stroke, and that is normal. What is worth attention is a change: a unit that used to close in 40 seconds and now takes 90, or one that hesitates part way and then continues. That is a mechanical or drive fault rather than a sensor one, and it is covered on our page about a failed opening mechanism.

Get a fixed price for a replacement

If cleaning has not fixed it, if the fault keeps coming back within weeks, or if the sensor is one of several things going wrong at once, tell us the make and model from the data plate on the top edge of the sash, roughly how old the unit is and where the property sits. We will say plainly whether it is worth pursuing parts or whether the unit is finished.

If it is finished, we survey it and give you one written price covering the rooflight, the actuator, the controls and the electrical connection, and that price is fixed. Every installation carries a 10-year workmanship guarantee, and the installers doing the work have 25 years fitting rooflights behind them.

Request a fixed quote, read the wider picture on electric skylight installation, or see how a sensor gets specified on a new unit at adding a rain sensor to a rooflight.

Thinking about this job? We survey across Ipswich, Suffolk and north Essex and give you a fixed price before any work starts, with a 10-year workmanship guarantee.

Get a fixed quote

Get a fixed quote

Tell us about the roof. We will tell you the price.

A survey costs you nothing and there is no obligation attached to it. We look at the roof build-up, the exposure and the opening, and give you one fixed figure, not an estimate that moves once work starts.

If your rooflight is leaking now, say so and we will prioritise the survey.

  • Fixed price, confirmed in writing
  • 10-year workmanship guarantee
  • Installed to current Building Regulations
  • Your roof is never left open overnight
Call Get a fixed quote