Cracks Above Windows and Doors: The Lintel, and the Wall It Sits On
A diagonal crack out of a window corner is rarely the plaster. Check two things: 150mm of lintel bearing each end, and masonry strong enough to carry it.
A crack climbing diagonally out of the corner of a window or a door is rarely the plaster. Nine times out of ten it is the lintel — or the wall the lintel is sitting on. Two checks settle it: 150mm of bearing at each end (100mm only where the clear span is 1200mm or less), and masonry underneath strong enough to carry the load. The estimator prices structural openings as their own line, and our extensions page covers where these openings get formed.
What a lintel is doing
Above every window and door there is a wall that would otherwise have nothing to stand on. The lintel carries that masonry and delivers the load into the wall at each end.
That delivery is the whole story. A lintel does not fail in the middle; it fails at the ends, or the wall under the ends fails. When it does, the load has to find another route down through the masonry, and it does so along the weakest line — diagonally, up and out from the corner of the opening. That is exactly the crack pattern people photograph and send me.
So when you see one, you are not looking at the crack. You are looking at where the load went instead.
The figures
| The rule | What it requires | Source | |---|---|---| | Lintel bearing | Bearing length 150mm or greater. May reduce to 100mm where the clear span is 1200mm or less. Manufacturers commonly require 200mm on longer spans — follow their data too. | AD A §2C24(a) | | Load at wall base | Combined dead and imposed load should not exceed 70 kN/m at the base of the wall. | AD A §2C24(c) | | Wall ties near openings | Additional ties not more than 300mm apart vertically within 225mm of the vertical edges of all openings. | AD A §2C8 | | Tie spacing generally | 900mm horizontal × 450mm vertical, or not less than 2.5 ties/m². | AD A §2C8 | | Tie embedment | Ties to BS EN 845-1, material reference 1 or 3, embedment not less than 50mm in both leaves. | AD A §2C19, Table 5 | | Cavity tray over openings | Where the tray does not run the full length of the wall — i.e. above every window and door — provide stop ends and at least two weep holes. | AD C §5.5(c) | | Maximum floor span on a wall | 6m, measured centre to centre of bearing. | AD A §2C23 | | Minimum wall thickness | Wall up to 3.5m high and 12m long: 190mm. | AD A Table 3 | | Padstones and beams | Sized by a structural engineer against Requirement A1. There is no standard padstone. | AD A, Requirement A1 |
Approved Documents are statutory guidance to the Building Regulations 2010 for England. Compliance is judged against the Requirements in Schedule 1; your building control body has the final word on your specific building.
The two checks, in order
1. Is there enough bearing? 150mm each end is the figure to hold in your head, dropping to 100mm only on short spans of 1200mm or less. Bearing is easy to lose on site: an opening formed slightly wider than drawn, a lintel cut to fit, a block chased out to let the end sit in. Each of those quietly removes bearing, and nothing looks wrong until it is plastered.
2. Is the masonry under the bearing strong enough? This is the one people miss. A steel beam concentrates a very large load onto a small area, and ordinary lightweight blockwork will crush under it. That is what a padstone is for — a dense concrete or engineering-brick block, sized by the structural engineer for the actual load and the actual masonry, spreading it into the wall below.
Padstones are not a standard item. Substituting whatever dense block happened to be on site is common and is one of the more frequent causes of cracking a year later, because the crushing is slow. The wall does not fail; it just gives, a millimetre at a time, and the plaster above the opening records it.
Cracks after an extension: usually the beam settling
When a large opening is formed and the temporary props come out, the new beam takes up its load for the first time. It settles very slightly into its bearings, and fine hairline cracks can appear at the junction between old and new work.
That is expected behaviour, not bad workmanship — provided the dry-packing was done. Dry-packing is a stiff mortar rammed hard into the gap between the top of the beam and the masonry it is supporting. Without it, the wall above has to drop before it makes contact with the beam, and that drop is exactly the movement you then see in the plaster.
So the useful question after a knock-through is not “why has it cracked?” but “was it dry-packed, and is there a photograph?”. Ask before the props come out, not after the crack appears.
What is not expected: a crack that keeps widening over weeks, a crack you could fit a coin into, a crack that appears on both faces of the same wall, or a door that begins to bind. Those want a structural engineer.
Telling the serious from the cosmetic
Four things to read:
Width. Hairlines are usually shrinkage — plaster, render or a new screed drying. Anything past about 2mm deserves attention.
Direction. Vertical cracks along a plaster joint are usually finish. Diagonal cracks running out of the corner of an opening follow load paths, and load paths are structure.
Progression. This is the most informative and costs nothing. Mark each end of the crack with a pencil line and write the date beside it. Photograph it. In six weeks you will know whether you have a defect or a memory.
Company. A crack that arrives alone is usually less serious than one that arrives with a sticking door, a sloping floor, or a matching crack on the outside face of the same wall.
What goes wrong on site
- The opening is formed wider than drawn and the lintel loses bearing at one end.
- No padstone, or a padstone substituted for something weaker.
- The cavity tray above the opening has no stop ends, so water tracks along the tray and comes out somewhere inside instead of through weep holes. The stain arrives months later and gets blamed on the window.
- Extra wall ties around the opening are omitted — they are required within 225mm of the vertical edges, no more than 300mm apart vertically, precisely because the masonry beside an opening has less to hold it.
- Props removed too early, before the beam and its bearings are ready to take load.
Every one of these is invisible after plastering, which puts them in the same category as the fixings and details that separate two quotes for the same job.
The takeaway
Two checks over any opening. The lintel bears 150mm at each end — or 100mm where the clear span is 1200mm or less — and it bears on masonry strong enough to carry it. And after an extension, fine cracks when the props come out are usually the beam settling, not bad work, provided the dry-packing was done.
Get an Itemised Estimate
Or call us on 020 8050 8968. We work across Fulham, Chelsea, Battersea, Kensington, Hammersmith and the surrounding boroughs.
Keep reading
- Why Walls Get Damp: The Damp-Proof Course and the Weep HolesDamp is rarely bad luck. Two numbers keep a wall dry: the DPC sits 150mm above outside ground, and weep holes appear every 900mm. Check both from the garden.
- Why Foundation Depth Changes on Site — and Why That Is HonestThe drawing proposes a depth. The ground decides it and building control confirms it in the open trench — which is why depth is the most common genuine variation.
- Underpinning Explained: Why It Costs More, and When to Avoid ItUnderpinning is ordinary concrete priced by sequence. Short designed bays, poured one at a time — that queue is the cost, and it is what holds the house up.
Get a itemised quotation for your project
Use our instant cost calculator — no calls required, detailed estimate in under 2 minutes.
Get your estimate