Strip foundation

Nearly everything is poured 150 thick, and on a wide footing that is wrong

Ask anyone how thick a strip footing is and you will get 150mm. It is such a settled habit that nobody checks it, and on an ordinary narrow footing it is genuinely correct, which is exactly why the habit survives.

Paragraph 2E2c of Approved Document A actually says the minimum thickness is 150mm or the projection, whichever is the greater. The projection is how far the concrete sticks out past the wall on each side. A 600mm footing under a 300mm wall projects 150mm, so 150mm thick is right. An 850mm footing under the same wall projects 275mm, and 150mm thick is then wrong by nearly half.

The reason is load spread. The load leaves the wall and fans outwards through the concrete rather than dropping straight down. If the footing is too thin for how far it reaches, the outer part of that width is not carrying anything, it is an unreinforced lever with a crack waiting at the root of it. Making the thickness at least the projection is what keeps the whole width doing a job.

The ground is identified with your hands, not a lab

Table 10 is unusual and rather good in that it identifies ground by a physical test you can do standing in the trench. A thumb, a finger, a spade and a 50mm square wooden peg:

• Stiff clay can be indented slightly by thumb.
• Firm clay lets your thumb make an impression easily.
• Soft takes a finger up to 10mm, very soft up to 25mm.
• Medium dense gravel or sand needs a pick, and a peg is hard to drive past 150mm.
• Loose sand comes out with a spade and takes a peg easily.

If you are between two rows, take the softer one. The table is a set of minimums and the penalty for being one row conservative is a few hundred millimetres of concrete, which is a great deal cheaper than being one row optimistic.

Where the table stops, and why that matters

The two softest ground types the table covers, loose sands and soft silts, have figures at 20 and 30kN/m and then nothing. That is deliberate, and the note underneath spells it out: foundations on soil types V and VI do not fall within the provisions of this section if the total load exceeds 30kN/m.

It is worth being clear about what that means, because it is easy to read a gap in a table as an invitation to use the last number in the row. It is not. It means the simple approach has stopped applying and the foundation needs designing rather than looking up. This tool says so rather than quietly handing you a wider footing.

Depth has nothing to do with the load

Width comes from the ground and the load together. Depth comes from the ground alone. The 450mm everybody quotes is a frost figure, and it is the minimum on ordinary ground, but on shrinkable clay the governing number is much deeper: 750mm on low shrinkage, 900mm on medium, 1000mm on high. Clay with a Modified Plasticity Index of 10% or more swells and shrinks with the weather, and a footing sitting in the zone that moves will move with it.

Trees are the other half of that. Approved Document A says the depth will commonly need increasing where there are trees nearby, and pointedly does not say by how much, because it depends on the species, the distance and the clay. If there are trees near a clay site, that is a question to put to whoever is designing it rather than one to answer off a table.

One last quantity note. The concrete figure here is the strip itself, width by thickness. If you are trench filling instead, the sum is different: concrete to within a few courses of ground level, so usually two to three times the volume, and almost no muck away to cart off. On a tight site that trade is often cheaper even though the concrete bill looks worse.