Home › Field Notes › When to Worry About Concrete Cracks

When Should You Worry About a Crack in Concrete?

Why does concrete crack in the first place?

Before you can decide whether a crack matters, it helps to accept the thing nobody selling concrete likes to say out loud: concrete cracks. It is not a defect waiting to happen, it is a property of the material. Cracking is what concrete does. The job is not to prevent it, which you cannot fully do, but to control where and how it happens.

There are a handful of reasons a slab or wall cracks, and they are worth knowing because the cause usually tells you how much to care.

• Shrinkage. This is the big one for the cracks people actually notice. As fresh concrete cures it gives up water, and as it dries it shrinks a little. If that shrinkage is restrained, and it almost always is, the stress gets relieved by a crack. This is why crews cut control joints into a slab: to give the shrinkage a planned line to follow instead of a random one.
• Thermal movement. Concrete expands when it warms and contracts when it cools. Big temperature swings, especially early on, move the slab, and movement that is held back finds relief the same way shrinkage does.
• Settlement. If the ground underneath moves, the concrete moves with it. A slab is only as stable as what it sits on, and a soft spot or a washout in the subgrade shows up later as a crack.
• Structural. True overload, a load the element was not designed to carry, is the least common cause and the most serious. This is the category you never want to wave off.

Hold onto that list, because when you look at a crack, you are really trying to work backward to which of these caused it. Shrinkage and thermal are usually the material behaving normally. Settlement and structural are the ones that earn a phone call.

Which cracks are cosmetic and which deserve attention?

Here is the honest version, and I will hedge before I land it. There is no single measurement that draws a clean line between fine and not-fine, and your local spec or engineer is always the authority on a given job. That said, in the field you learn to read four things together, and the pattern they form usually tells you more than any one number.

The single most useful habit is to watch a crack over time. A hairline crack that shows up, sits there, and never changes is almost always shrinkage doing its job. A crack that keeps opening, keeps growing, or develops a vertical offset where one side is higher than the other is telling you something is still moving, and movement is the thing you follow.

So the rule of thumb, softened and then landed: narrow, stable, joint-following, flush cracks in flatwork are almost always cosmetic. Wide, growing, offset, oddly-patterned, or structural-location cracks are the ones you stop and investigate.

What does a crack tell you about the mix, the curing, or the subgrade?

A crack is information, not automatically a verdict. When somebody points at a crack and says "bad concrete," I try to slow that down, because the crack is usually pointing somewhere other than the mix.

Read the crack as a clue and ask three questions in order.

1. Curing. Was the concrete kept moist and protected while it gained strength, or did it dry out too fast? Rapid early drying, often in heat or wind, drives shrinkage cracking. A lot of what looks like a material problem is really a curing problem.
2. Joints. Were control joints cut, and cut deep enough and soon enough, and spaced right for the slab? If shrinkage cracks are wandering across a slab instead of following joints, the joint plan, not the mix, is usually the story.
3. Subgrade. What is under the slab? Uneven or poorly compacted ground concentrates stress and settles unevenly, and the crack shows you where.

Only after those three does it make sense to look hard at the mix itself, and even then the honest framing is "consider" rather than "blame." The point is that a crack tells you to investigate the whole system, how the concrete was made, placed, cured, jointed, and supported, not just to indict the material. Read it that way and you fix the real cause instead of the convenient one.

What are the general ways to reduce the cracks you do not want?

Since you cannot abolish cracking, the goal is to control it, and there are a few general categories to consider. I am going to stay in categories here on purpose. This is the "consider X if you are dealing with Y" way of talking about it, not a product list.

• Joint design. If you are dealing with random shrinkage cracking across flatwork, the first thing to consider is whether the control joints are spaced, timed, and cut deep enough to steer the cracking. Good joint design is the cheapest crack control there is.
• Curing practice. If cracks are showing up early and fine across the surface, consider whether the curing kept moisture in long enough. Protecting the slab from drying too fast, especially in heat and wind, prevents a lot of shrinkage cracking before it starts. Follow the general hot-weather practice your local spec and industry guidance lay out.
• Reinforcement options. If you are dealing with a slab where you want to control crack width and hold cracks tight, reinforcement is the category to consider. That can mean conventional reinforcement, and it can mean fiber options, including macrofibers as a category, which are used to help manage cracking in certain slab applications. The right choice depends on the job, the loads, and the spec, so treat this as a conversation with your engineer, not a default.
• Mix adjustments. If shrinkage is your recurring problem, consider whether the mix itself can be adjusted for the conditions. Mixes can be designed with shrinkage and the placement environment in mind. This is where you lean on your producer and the mix design rather than a field fix.

Notice that none of those is a silver bullet, and none of them is a brand. Crack control is a system: joints, curing, reinforcement, and mix working together. Pull one lever in isolation and you usually just move the problem.

When should you actually call an engineer?

You do not need an engineer for a stable hairline crack in a driveway. You do need one when the crack crosses from cosmetic into a question about how the structure is behaving. Call in a qualified engineer when:

• The crack is wide rather than hairline, or is clearly growing over time.
• There is a vertical offset, one side of the crack higher than the other.
• The pattern is diagonal, stepped, or map-like, rather than following a joint.
• The crack is in a structural element, a footing, beam, column, or load-bearing wall.
• Water is passing through where it should not, or the crack sits somewhere failure would be costly.

Any one of those is a fair reason to get eyes on it. Several together is a strong one. The point of calling early is not alarm, it is that a crack is easier and cheaper to understand while it is small, and an engineer can tell you whether you are watching normal movement or the start of a real problem. When in doubt, the professional call is the cheap insurance.

The takeaway, and a question to ask

Concrete cracks. That is the starting truth, not the failure. Most cracks are the material doing exactly what it does as it cures and moves, and reading them well is a matter of four things: width, pattern, location, and movement. Narrow, stable, joint-following, flush cracks in flatwork are almost always fine. Wide, growing, offset, oddly-patterned, or structural-location cracks are the ones you stop for.

Next time a crack shows up, before anyone blames the concrete, ask one question of whoever placed it: how was this cured, how were the joints planned, and what is the subgrade like? The answer to that will tell you far more than the crack itself, and it points you at the real fix instead of the easy scapegoat.