Ordinary Alterations That Create a New Cavity Drainage Requirement

The masonry cavity wall record states the general case once and then leaves it. Wherever the void stops or is bridged, whether by a lintel, a change of construction, a floor bearing or an abutting roof, the drainage arrangement has to be re-made. The tray record adds that a tray without an exit does not drain and that outlets without a tray above them drain nothing, so the pair works only together.

What makes this an alteration question rather than a construction one is that all of these interruptions can be created after the wall exists. Nothing about the original wall changes; a new line is introduced across the void, and the water that used to run past that line now arrives at something.

This is the most common of them, and the records describe why it is also the most awkward. An abutment needs a tray inside the wall as well as flashings outside it, and the tray belongs to the masonry trade while the flashing belongs to the roofer. The junction between their two pieces of work is inside the wall where neither can see it.

The abutment record adds a timing problem on top of the ownership one. Provision for the concealed tray has to be made while the wall is being built, by a trade that will have left before the roof is weathered, and where it is missed the remedy involves taking part of the wall apart again. Where the wall already exists, as it does on an extension, that remedy is the starting point rather than the fallback.

The record also asks a question an owner can put directly: has a tray been provided at this abutment, and can its discharge points be seen from outside the building? The weeps have to discharge above the cover flashing, which means two trades have to agree a level that neither of them controls alone.

Cutting an opening into a cavity wall interrupts the void on every side at once. The cavity wall record describes the head as needing interception and outlets, the jambs as needing a closer that carries no water across, and the sill as having to throw water clear of the leaf below. The frame is normally fixed to the inner leaf, so the seal between frame and closer is also where the wall's air line passes.

The tray record then names the failure that follows from doing half of it. Without stop ends, a tray discharges into the wall beyond the ends of the opening, so damp appears to one side of a window rather than above it and is diagnosed as something else entirely. The window is usually blamed, and the window is usually not involved.

Any line where the outer leaf stops carrying itself and hands its weight back to the structure is also a line where the void is closed. The veneer record makes the point generally: every place the leaf is carried is a place the void is closed, so support and drainage are the same detail, and support provided without interception and open outlets below concentrates water at exactly the point where the wall is least able to shed it.

The infill wall record adds a setting-out consequence. Every support angle, nib or floor edge crossing a cavity is an interruption, and water travelling down inside the wall arrives at it, so where trays are provided their stop ends and drainage openings have to coincide with the support rather than with the drawing grid. A tray drawn to a grid and a support built to a structure are two different lines.

An insulation upgrade is not usually thought of as a drainage alteration, and the records treat it as one. Trays over lintels, at abutments and at the base were detailed to discharge into an open void; after filling they discharge into a material. That is why the presence, continuity and stop ends of those trays are among the first things a survey looks for.

The retrofit record is also clear about what a survey finds. Existing walls frequently have interruptions with no tray at all, which is a survey question rather than an assumption, and introducing fill both changes what any existing trays are dealing with and reveals whether they were ever installed. It adds that vents, flue ways, ducts, boxes and tray ends each become either a place fill escapes into or a designed air path that is blocked, and that none of these is drawn on an ordinary existing house.

Each of these alterations is commissioned as something else. An extension is a room; an opening is a view; a support is a structural item; a fill is an energy measure. The interruption is a by-product, and the record's observation is that because no single package owns cavity interruptions, the tray is frequently the item that nobody realised was needed.

That is why the published design questions are framed as an audit rather than as a specification. Where in this wall is the cavity interrupted, and has each of those interruptions been reviewed? Where two separate runs meet, which one laps over the other and in which direction does the water travel? Both are questions to ask of the whole wall, including the parts nobody is altering.

It is worth stating the boundary explicitly, because the neighbouring failure looks similar from inside the room. This guide is about a requirement being created: a new line across the void that water now arrives at, needing interception and a way out. It is not about a connection being made by accident across a void that is otherwise intact, which is the subject of the companion guide on accidental bridges.

The distinction matters for diagnosis. If the void was interrupted and nothing was provided, the answer is an arrangement that was never built. If the void is uninterrupted and something is carrying water across it, the answer is a connection that should not exist. Those lead to different investigations and different remedies, and an alteration can produce both at once.