Can a rooflight go near a chimney?

What is already happening around a stack

A chimney passing through a pitched slope is weathered on four faces and each is a different detail. The front apron is a single piece of lead dressed down over the covering below, throwing water out onto the slope. The sides are stepped: individual soakers interleaved with each course, covered by a stepped flashing tucked into the mortar joints.

The back is the important one. Water coming down the slope above cannot pass through the stack, so it is collected in a back gutter, a lead-lined trough that carries it sideways around both corners of the chimney and releases it onto the covering at each side.

That release is the reason a rooflight cannot simply be tucked in behind. The stack does not shed water evenly. It concentrates it.

Where the water actually goes

Think of the slope below a chimney as having two fast lanes. Everything the stack intercepts, which on a large Victorian chimney can be a square metre or more of catchment, arrives at the covering in two streams at the corners rather than spread across the width.

Put a rooflight in one of those streams and the head flashing is dealing with several times the volume it was designed and tested for. It may cope. In driving rain with leaf litter in the gutter behind the stack, it may not, and the failure shows as water at the head of the unit during heavy weather only, which is the hardest kind to diagnose.

Directly below the stack is the position to avoid

Of all the possible positions, immediately downslope of a chimney is the weakest, and it is the one people ask for most, because that is often where the room wants light.

Besides the concentrated flow, there is a second issue: the covering below a stack is where debris collects. Soot, mortar dust from the flaunching, moss shed off the brickwork and the general grit that comes off masonry all wash down that line. It builds against anything standing proud of the slope, and a rooflight standing proud of the slope is a very effective collector.

Where the room genuinely needs light on that line, the answer is usually to move the unit sideways out of the stream, or lower down the slope so the flow has spread and rejoined before it arrives.

Beside a stack, and what the coursing will allow

To the side is generally the best position, and the constraint is the stepped flashing. The soakers up the side of the chimney are interleaved course by course, so each course in that zone is already carrying a piece of lead. A rooflight side flashing needs full, undisturbed courses to lap onto.

Give it two clear courses and both details work independently. Squeeze it to nothing and you end up with one piece of lead lapped onto another piece of lead at an angle neither was formed for, sealed at the junction. That junction is a maintenance item forever after, and it will not be reached without access.

Above the stack, and the ridge problem

Upslope of a chimney is usually clean weathering, since the stack is downhill of anything the rooflight sheds. The constraint there is geometry rather than water: chimneys are often close to the ridge, so the space above one may not leave enough covering between the head of the unit and the ridge line.

The head flashing has to tuck under a full course, and there needs to be sound covering above that for the lap to mean anything. Our page on how close to the ridge a rooflight can go deals with that limit in its own right.

Flues, terminals and openable units

If the chimney is live, serving a stove or a boiler, there is a further consideration and it is not about water. An openable rooflight downwind of a working flue terminal can draw combustion products back into the room under certain wind conditions, and it can also affect the pressure regime the appliance is relying on.

This is a matter for whoever commissioned the appliance rather than for us, but it is worth raising before the unit is positioned rather than after. Where the flue is live and the proposed rooflight is close and openable, a fixed unit or a different position is the straightforward way around it.

The state of the stack changes the sequence

Sometimes the survey finds that the chimney’s own weathering is at the end of its life: lead that has been dressed and re-dressed thin, mortar joints that have lost the flashing turn-in, or flaunching breaking up on top.

We will say so, and we will say plainly that it is roofing work and not ours. Our scope is installation and replacement of rooflights, deliberately narrow. What matters practically is the order: that work wants doing while the access is standing and before a new unit is set beside it, not eighteen months later when somebody has to get back up there and disturb the covering again.

How the position is settled

On survey we mark the chimney’s flashing extent on the slope, count the courses out from it in both directions, and set the opening where both sets of lead have full courses to work with. Then we check that against what the room wants, what the rafters allow, and what is visible from the street if the building is designated.

Occasionally those constraints do not overlap and the honest answer is that the position you had in mind is not the one to use. Saying that early is cheaper than proving it later. There is more on how the perimeter is built in our page on lead flashing and weathering detail.

Fifteen years of pitched-roof work around Oxford, where chimneys are close, tall and frequently in the way. The roof is opened and closed the same day, never left open overnight, and the workmanship is guaranteed for ten years. Call 01865 704245 with a photograph of the slope and the stack and we will tell you whether the position works.