Why a Roof Edge Has Three Jobs That Compete

At an eaves, the published junction record lists what has to happen in one slice of construction: water leaving the underlay has to reach the gutter, roof insulation has to meet wall insulation, air may need to pass into the roof void, and the fascia has to carry the gutter and stay reachable for inspection. It then adds the sentence that matters, which is that none of these is difficult on its own and all of them want the same shallow space.

The shape of the conflict is geometric rather than technical. The depth available narrows exactly as the roof plane approaches the wall, which is to say it is smallest where the two insulation lines need to overlap and where the air path needs to be widest. That is why eaves details exist rather than being obvious.

The ventilated pitched roof record describes the competition and refuses to settle it: carry the insulation right out to the wall to meet the wall insulation and it can block the inlet; hold it back to keep the inlet open and there is a cold wedge over the top of the wall. Resolving both at once is the point of the eaves detail rather than a refinement of it.

The published interface record for this junction agrees that the requirement is genuinely contradictory rather than merely awkward: insulation must be continuous, and an air path must remain open, in the same confined zone. The component that separates the two is a baffle, which the bypass control record describes not as an accessory but as the thing that lets both requirements hold at once.

This is the part that catches people acting in good faith. The eaves record notes that fascia and soffit details that close the edge neatly can close the void with it, which turns a ventilated concept into an unventilated one without any visible change, and that renewing fascia and soffit boards is often treated as decoration although it disturbs the underlay edge, the inlet and the gutter line at the same time.

The cold deck record makes the same observation about insulation: adding more insulation into the joist depth does not improve the roof if it closes the air space the design depends on. And the ventilated pitched record notes that blocking eaves openings to stop draughts in the loft removes the mechanism the roof space depends on to stay dry. In each case the action is an improvement to one duty and a removal of another.

Where the roof does not oversail the wall, the same crowding happens at an upstand. The parapet record describes insulation and drainage pulling the design in opposite directions at the same corner: returning insulation up the inner face reduces the cold construction projecting outward, but it thickens the upstand exactly where the waterproofing has to turn and where an outlet has to pass through.

The published interface record adds the thermal half in its own terms. A parapet is a wall that keeps going after the roof has stopped, so the roof insulation and the wall insulation reach it from two directions and at right angles. If the two layers stop where they arrive instead of joining round it, the wall acts as a fin taking heat out of the building along a line that runs the whole perimeter of the roof, continuously rather than at a point.

The interface record names the signature, and it is the most useful diagnostic sentence in the family. The coldest surface in a room under a flat roof is commonly the line where the ceiling meets the external wall, and mould following that junction round the room, rather than appearing in patches on the wall, is what a thermal layer stopping at the parapet looks like from inside.

The ventilation failure shows differently. The eaves record notes that air entering at the eaves has to leave somewhere higher, so an inlet on its own achieves nothing, and where a high-level outlet is absent or blocked the inlet ventilates only the eaves zone while moist air stays in the void and condenses on the underside of the covering furthest from the edge. That is a long way from the edge that caused it.

The eaves record lists who builds it: a roofer, a carpenter, an insulation installer and often a plasterer, usually on different days. The ventilation record makes the consequence explicit, saying the path is completed by several trades who are each doing something else, and none of them owns the circuit, which is why it is so often broken without anyone deciding to break it.

That makes the edge a coordination problem before it is a technical one. The air barrier record asks which trade carries the line where it passes from the walls into the roof, and notes that where the crossing is left to whichever trade arrives last, it is frequently the point at which the line is lost.

The useful request is a single drawing. The eaves record asks which drawing shows the eaves detail, and whether it shows the insulation, the underlay and the ventilation path together on one section. Where three requirements are drawn on three drawings, they are being reconciled by whoever is standing on the scaffold.

The second question is about survival. The record asks what holds the insulation clear of the ventilation inlet once the soffit is closed and nobody can see it, and who is responsible for keeping that inlet open. Both are questions about the period after the detail is correct, which is when most of these edges stop working.