Where this starts
Curling is one of the few slab defects with a mechanism simple enough to state in a sentence: concrete shrinks as it dries, the top surface dries first because it is the exposed one, and a slab whose top has shrunk more than its bottom becomes a shallow dish. The edges and corners lift because they are free to.
That immediately explains where it appears. It is worst at FREE EDGES and at JOINTS, because those are the only places the slab can move upward; it is largest at corners, where two free edges meet; and it is worse on thin slabs than thick ones, because a given difference in shrinkage between top and bottom bends a thin section more.
It also explains why the causes below are almost all about DRYING rather than about the mix. Anything that makes the top dry faster relative to the bottom increases the gradient: a dry slab base, no vapour barrier under it, heat or draught above it, curing stopped early, or a mix with more water to lose.
Curling matters for three reasons. A lifted edge at a joint is unsupported, so traffic across it breaks the edge down. A curled slab telegraphs through a floor finish laid over it. And a slab that has curled has also cracked in many cases, because the restraint that stops it lifting has to go somewhere.
What it could be, and what tells them apart
These are candidates to check against your own building, in the order they most often turn out to be the answer. None of them is a diagnosis of your situation.
The top dried much faster than the bottom
Most commonLifting concentrated at free edges and corners, appearing in the weeks after the pour rather than immediately, and largest where the slab was most exposed to sun, heat or draught.
How to check: Ask what the conditions were in the first days and whether the slab was covered. A slab left open in drying weather is the usual answer.
Curing stopped too early or never started
Most commonCurling together with a weak, dusty surface, since both come from the same cause — the surface losing its water before hydration finished.
How to check: Scratch the surface in a sound area. A surface that powders under a coin lost its water, which is the same thing that curled it.
No vapour barrier, or one placed wrongly
CommonThe bottom stayed wet while the top dried, so the gradient persisted for months. Curling that is still increasing long after the pour points here.
How to check: Establish what is under the slab. A slab cast directly on a dry granular base behaves differently from one cast on a membrane.
Joints too far apart, or sawn too late
CommonCurling concentrated at a small number of joints with large panels between them, often with random cracking mid-panel as well.
How to check: Measure the panel sizes and look for cracks that are not at joints. A cracked panel means the joints did not do their job.
A mix with more water than it needed
CommonWidespread curling with a generally weak surface, and often bleed marks or a laitance layer — the whole slab shrank more because it had more water to lose.
How to check: Whether the surface is weak everywhere rather than in patches. That points at the mix rather than at local drying conditions.
Slab too thin for its panel size
Less commonPronounced lifting on a comparatively thin slab, with the amount of curl obviously larger than on thicker slabs nearby.
How to check: Measure the actual thickness at an edge. Curl for a given moisture gradient falls sharply as thickness rises.
What to check, in order
The order matters — each step is here because of what it rules out for the ones after it.
- 1
It turns an impression into a number and shows the pattern — corners worst, then edges, then joints — which is the signature that distinguishes curling from settlement.
- 2
Curl that is still increasing months later means the moisture gradient is still there, which points at what is under the slab rather than at how it was cured.
- 3
A weak, dusty surface and curling usually share one cause — water lost from the top too early — so finding both together identifies it without any other test.
- 4
It separates a curling problem from a jointing problem. A slab that has cracked mid-panel was going to move whatever the curing was.
- 5
A membrane under the slab keeps the bottom wet and prolongs the gradient; a dry base lets both faces dry and reduces it. It changes the diagnosis and the remedy.
The calculations behind this
Several of the causes above are a quantity that was specified or assumed. These compute them, with the formula and the standard behind each.
When to stop and get somebody in
- Where a lifted edge at a joint is carrying traffic and breaking down, since the repair has to restore support beneath the edge rather than just fill the gap.
- Where the slab has cracked as well as curled, particularly if the cracks have a difference in level across them, which is a different and more serious mechanism.
- Before laying a rigid floor finish over a curled slab — the curl telegraphs through, and grinding or re-levelling is a specified operation rather than a cosmetic one.
- Where the slab is a structural element rather than a ground-bearing floor, in which case the deformation is a design question rather than a finishing one.
This page lists candidates to check, not a diagnosis — nobody has seen your building. If the symptom involves anything structural, read this first.