Level Thresholds: What Replaces the Step You Removed

An upstand at a doorway works without any moving part, without maintenance and without anyone noticing it. It raises the point at which water can enter above the level water reaches on the surface outside. The record describes a flush junction as removing that passive protection and shifting the duty onto drainage and drying, which is a different kind of arrangement with different failure modes.

That is why the records treat a flush junction as a design decision rather than a finish decision. Whether it is appropriate at all depends on exposure, sheltering, the receiving drainage and the use of the building, and the record declines to state that it suits any particular doorway.

The record is precise about this, and it is the point most often lost. A channel intercepts water travelling across the surface towards the opening. A pan catches what is blown past the door seals and lands behind the frame. They deal with water arriving by different routes and neither covers the other.

The two ways of getting it wrong are therefore symmetrical. Provide only the channel and driven rain arrives inboard of the line it was defending. Provide only the pan and it becomes the collection point for everything the paving has delivered, which is far more water than it was formed to handle.

A threshold channel is a small inlet on a much larger arrangement, and its behaviour depends on what happens downstream of it. The record notes that where the receiving system backs up, the channel at the door is one of the lowest inlets on the site, which is precisely the condition in which it is most needed.

The published question to put is therefore not whether interception exists but where it connects and whether the consequence of that connection surcharging has been considered. The bound surface outside matters too: a surface that sheds rather than absorbs delivers what lands on it towards the lowest point available, and where such a surface has been laid to fall towards a building, no threshold detail is being asked a fair question.

At ground level the threshold is where the horizontal ground moisture barrier and the wall base barrier have to be lapped together around a frame. The damp-proof barrier record calls it the shortest length of that junction on the building and the hardest to reach once the threshold plate is fixed, and notes that the desire for a low or level threshold and the need for an upstand pull in opposite directions.

The same record describes the barrier at a threshold as compressed into the least available height and having to connect to something made of a different material that arrives at a different time, with the sealing components used there depending on adhesion to surfaces that are frequently damp, dusty or freshly cast. That is the ordinary condition at that line rather than an unlucky one.

The same level access question behaves differently where the external surface is a roof, because there is habitable space underneath. The trafficked deck record describes the threshold as the point where the assembly argues with the building: a doorway that has to be level for use and an upstand that has to be measured up from the concealed waterproofing are competing for the same vertical distance, and neither the surfacing nor the waterproofing can concede it alone.

Two consequences follow that do not exist at ground level. The record states that the usual resolution is drainage placed in front of the door and a change in the build-up around it, which is a design decision rather than an adjustment made on site. And it warns that a level threshold is not achieved by reducing the upstand, because treating it that way moves a design problem onto the waterproofing, with the consequences of an unresolved junction moving from an inconvenience at a front door to a leak into a room.

Where the surfacing stands on adjustable supports the arrangement changes again. A pedestal terrace offers a drained void beneath the walking surface, so the level at the door and the level of the waterproofing beneath it are different things, and that separation is what makes a flush junction achievable. It depends entirely on the void staying clear, and the record calls level access the most constrained junction on the whole terrace.

One published case turns the threshold into a limit on the whole roof. Where a roof is arranged to store rainwater and release it slowly, the storage volume is not set by the void but by the lowest point at which water could leave, and a door threshold is one of the points that can cap it.

The record puts this in two directions. A threshold has to be resolved with the storage concept rather than detailed as an ordinary entrance, and altering a threshold later alters what the roof does in a storm. It also notes that where no exceedance route is provided the roof still finds one, and it will be the lowest point of the perimeter, which is usually a door.

The threshold is simultaneously the wettest part of the elevation and the place where the floor insulation, the wall insulation and the frame all stop. The record draws the conclusion: cutting the thermal layer to make room for a channel or a fixing puts the coldest internal surface exactly where the most water is, so a thermal decision here is also a condensation decision.

Hardware clearance is competing for the same vertical room. Raise the external surface to reduce a step and the door seal compresses differently or the leaf begins to catch; lower it to keep the channel draining and the step returns. Because the door is manufactured to its own tolerances and the paving is laid to site tolerances, the setting out has to be agreed before either is ordered rather than reconciled afterwards.

Interception at a doorway silts with grit, leaves and the material walked over it. The record is direct about the consequence: a channel that cannot be lifted and cleared silts up and stops being a channel, becoming instead a shallow reservoir held against the door.

The practical questions are therefore ordinary ones. Does the grating lift? Can the channel be reached with a brush? Can the outfall be rodded? And can an owner do those things without specialist help, which is what decides whether they happen at all.