Concrete
Concrete crack types, causes, and prevention field guide
Read the crack to find the cause: plastic shrinkage and settlement, drying shrinkage, thermal, restraint, re-entrant corner, structural, ASR, and corrosion cracks, and what each one says.
Direct answer
A concrete crack is the material relieving tension it cannot carry, and almost every slab gets some. The job is controlling where and how concrete cracks, not preventing every line. Most cracks trace to shrinkage fighting restraint; thermal, structural, and corrosion cracks are the rest. Read the timing, pattern, width, and location to find the cause.
Key takeaways
- Concrete is weak in tension, roughly a tenth of its compressive strength, so almost all concrete cracks; you control where and how, not whether.
- Read four things to name a crack's cause: timing, pattern, width, and location.
- Plastic shrinkage cracks open 1 to 6 hours after placement (rapid evaporation); plastic settlement cracks follow the top bars, about 1 mm wide.
- Re-entrant inside corners crack at about 45 degrees within days; prevent with a control joint into the corner plus diagonal reinforcement.
- Structural, growing, leaking, or load-path cracks go to a licensed structural engineer; ACI 224 sets tolerable widths (about 0.016 in interior, 0.007 in for deicing).
Concrete cracks, and what you actually control
A concrete crack is the material relieving a tension it cannot carry. Concrete is strong in compression and weak in tension, on the order of a tenth of its compressive strength, so the moment it gets pulled harder than that, it tears. Almost every slab and most structural members crack somewhere. The honest version of the trade is that you do not prevent cracking. You control where it goes and how wide it gets.
That is the frame for this whole guide. A hairline map across a garage floor and a diagonal crack walking up a loaded beam are both cracks, and treating them the same is how people either panic over nothing or miss the one that matters. The skill is sorting the crack you can shrug at from the crack that ends with an engineer and a repair.
The control-joint layout guide covers the other half of this, putting the shrinkage crack on a line you chose instead of where the slab picks. This guide is about reading the crack you already have: which family it belongs to, what caused it, whether it is still moving, and what to do about it.
How do you read a concrete crack?
Four things tell you what caused a crack: when it showed up, the pattern it makes, how wide it is, and where it sits. Get those four and you can usually name the cause before you reach for a repair manual.
Timing splits the field first. A crack you find within hours of placement, while the concrete is still soft or barely set, is a plastic-stage crack, a different animal from one that opens days or weeks later as the slab dries. A crack that appears months or years out, slowly widening, points at something ongoing: settlement, corrosion, or a chemical reaction in the mix.
Pattern names the mechanism. Short random surface cracks, parallel cracks over the bars, a wide map across a wall, a diagonal line out of a corner, a vertical crack at regular spacing in a beam, each is a fingerprint. Width and location finish it. A hairline that follows the top bar is settlement; a wide crack that traces the load path is structural. Write down all four before you decide anything, because guessing from one of them is how a restraint crack gets blamed on the mix.
The two families: before hardening and after
Every crack lands in one of two families, and the split is whether the concrete had hardened when it cracked. Before hardening, while the concrete is still plastic, you get plastic shrinkage and plastic settlement cracks. After hardening, you get drying shrinkage, thermal, restraint, structural, and the long-term chemical cracks from corrosion and alkali-silica reaction.
The families matter because they point at different people and different fixes. Plastic cracks are a placement and weather problem, owned by the crew and the conditions on the day of the pour. Hardened-state cracks split further: shrinkage and thermal cracks are a mix, jointing, and curing problem, while structural cracks are a design and loading problem that belongs to the engineer, not the finisher.
Keep the two straight and you stop chasing the wrong fix. A plastic shrinkage crack will not be cured by tighter joints, and a drying shrinkage crack will not be cured by fogging the next pour. The rest of this guide walks the families one crack at a time.
Plastic shrinkage cracking
Plastic shrinkage cracks open in the first one to six hours after placement, while the concrete is still plastic and has almost no tensile strength. They are short, run roughly parallel a foot or two apart, often at a slight angle across the slab, and they do not reach full depth. You see them on hot, dry, windy pours, sometimes opening while the finishers are still on the slab.
The cause is rapid evaporation. When water leaves the surface faster than bleed water rises to replace it, the top skin dries and shrinks against the wetter concrete below, and with no strength to resist, it tears. The surface-defects guide carries the evaporation rate and the action level, around 0.2 lb per square foot per hour, where the precautions kick in.
For crack-typing here, the tell is timing and depth. These open early and stay shallow, which separates them from drying shrinkage cracks that come days later and run deeper. Caught while the concrete is still plastic, they can sometimes be closed by refloating. Once it sets they are permanent, usually a durability and appearance issue rather than a structural one, but a path for water into the surface.
Plastic settlement cracking over the bars
Plastic settlement cracks form in the same first few hours, but from a different mechanism. After placement the solids settle and water rises, and where something blocks that downward settlement, the top bars, a change in section, a void former, the concrete drapes over the obstruction and tears at the surface above it. The crack maps the reinforcement.
The pattern is the giveaway. Settlement cracks run in lines directly over the top bars, often spaced to match the bar spacing, commonly around 1 mm wide, and you can sometimes find a crescent-shaped void under the bar where the concrete hung up. On a slab with a top mat you get a grid of cracks tracing the steel. Deep sections, wide bar spacing, and a wet, high-slump mix that settles a long way all make it worse.
The fixes are mix and placement. Revibration after initial settling closes them on some jobs, and a lower-slump, more cohesive mix with adequate cover settles less. The rookie read is to call these drying shrinkage, but drying cracks ignore the bar layout and settlement cracks follow it exactly. The bars in the pattern are the diagnosis.
What is drying shrinkage cracking?
Drying shrinkage is the big one, the crack most concrete eventually shows. As the concrete dries over days, weeks, and months, it loses moisture and shrinks, on the order of a few hundredths of an inch per foot, up to roughly 600 millionths of strain for typical mixes. If the concrete could shrink freely, it would just get a little smaller. It cannot, so it cracks.
Restraint turns shrinkage into a crack. The subgrade drags on the bottom of a slab, columns and walls hold it at fixed points, reinforcement resists the movement, and the concrete's own mass fights it. The shrinkage goes into tension, concrete is weak in tension, and when the tension beats the tensile strength, it tears. This is exactly why control joints exist, and the control-joint layout guide covers the spacing, depth, and timing that put this crack on a line you chose.
The tell is timing and pattern. Drying shrinkage cracks show up days to weeks out, not in the first hours, and they run between restraint points, across the narrow dimension of a panel, or from a re-entrant corner. They are usually narrow and stable once the slab finishes drying, which puts most of them in the cosmetic-to-manageable bucket. The way you fight them is less water in the mix, real curing, sensible joint spacing, and reinforcement to hold the inevitable cracks tight, covered in the prevention section below.
Thermal cracking
Concrete expands when it warms and contracts when it cools, and when something restrains that movement, the contraction goes into tension and cracks. Thermal cracking shows up in three flavors on real jobs: early thermal from the heat of hydration, in-service thermal from daily and seasonal swings, and the freeze damage that breaks concrete from the inside.
Early thermal is the mass-concrete case. Cement hydration is exothermic, so a thick pour heats up in the core while the surface cools to the air, and the core can run far hotter than the edge. Two things crack it: the temperature difference across the section, and the restrained contraction as the whole mass cools back down a day or two later while it is still weak. Specs cap the allowable difference to control this, commonly cited around 35 degrees F between core and surface, with the exact limit set by the project and the mix. Confirm the number against the spec, because it drives the cooling pipes, the insulation, and the placement temperature.
In-service thermal cracks come from a slab or wall that heats and cools against a restraint, a long unjointed run in the sun, a tank wall, a bridge deck. Freeze damage is its own thing: water in under-aired concrete expands when it freezes and breaks the paste, which the surface-defects guide covers as scaling. The shared lesson is that temperature swings need somewhere to go, through joints, air entrainment, or a mix and a placement plan built for the heat.
Restraint cracking: the common thread
Restraint is the reason shrinkage and thermal movement become cracks instead of harmless size changes. Anything that stops concrete from moving freely while it shrinks or contracts builds tension into it, and the crack relieves that tension. Name the restraint and you have usually named the crack.
The usual restraints are predictable. Subgrade friction drags on the bottom of every slab on grade, more on a rough or stabilized base than a slip sheet. Edges, returns, and changes in thickness concentrate stress. Embedded pipes, footings, and the slab being cast tight against a wall or column all pin the concrete at a point it wants to move away from. Re-entrant corners, covered next, are the sharpest version. Even reinforcement and adjacent older concrete restrain a new pour.
The practical move is to cut the restraint where you can and relieve the stress where you cannot. Isolate slabs from columns and walls so they can shrink free, use a slip sheet to cut subgrade drag, and put joints where the restraint forces the crack anyway. Blaming a restraint crack on a bad mix is one of the most common wrong calls in the trade, because the mix was fine and the detail held the slab where it could not move.
What is a re-entrant corner crack?
A re-entrant corner is an inside corner, the notch where an L-shaped or T-shaped slab turns back on itself, or the inside corner of a blockout, a pit, a column pocket, or a door opening. Shrinkage stress concentrates hard at that inside angle, and a crack tends to shoot out from the corner at roughly 45 degrees into the slab, often within days of the pour. Leave the corner bare and it cracks, close to every time.
The cause is geometry, not the mix. The corner is a stress riser. The slab on each side shrinks, and the two pulls meet at the inside angle with nowhere to go. This is one of the most reliable cracks in concrete, which is also why it is one of the most preventable.
Two standard moves, and you usually do both: put a control joint into the corner so the crack has a planned line, and add diagonal reinforcement across the corner to catch any crack that still forms. The control-joint layout guide carries the detail, commonly a couple of short bars set diagonally near the surface. The mistake worth calling out is leaving inside corners with no joint and no diagonal steel, then being surprised by the one crack that was guaranteed off the drawing.
Structural and flexural cracks: the serious ones
Structural cracks come from load and movement, not from drying, and they are the family that gets an engineer involved. They show up where the structure is being pulled or sheared harder than it was built for: overload, foundation settlement, an undersized member, or a design that did not account for the real loads.
Pattern points at the mechanism. Flexural cracks are vertical and form on the tension face of a beam or slab, often evenly spaced where the bending is highest, in the bottom at midspan or the top over a support. Shear cracks run diagonally, around 45 degrees, near the supports of a beam or out of the corners of openings. Settlement cracks come from one part of a foundation moving relative to another, often diagonal and wider at one end, tracking the part that dropped. A crack that follows the load path or grows over time is the one to take seriously.
This is where you stop diagnosing and call a licensed structural engineer, a PE. Width is one signal, and a crack you can fit a quarter into, or one near a column or a load, gets looked at, but width alone does not make a crack structural and a hairline can still be one. The engineer decides whether the member is overstressed, whether settlement is ongoing, and whether the fix is a structural repair or replacement. Guessing at a structural crack is how a cheap conversation becomes a liability.
Crazing and map cracking
Crazing is a network of fine, shallow hairline cracks in a random pattern, like a dried lakebed or a cracked windshield, rarely deeper than about 1/8 in. It is a surface phenomenon, caused by the top skin drying and shrinking faster than the concrete below, and it is cosmetic, covered in full in the surface-defects guide. The reason it earns a mention here is that it gets confused with two cracks that are not cosmetic.
Map cracking, a larger network across a wall or slab, can be crazing scaled up, or it can be the surface sign of alkali-silica reaction or corrosion underneath. Depth and timing separate them. Crazing is shallow and shows up early as the surface dries. A deeper map that grows over months, especially with gel or staining at the cracks, is a chemical problem, not a finishing one.
Do not tear out a sound slab over crazing, and do not write off a growing map pattern as crazing without looking deeper. The next two sections cover the chemical cracks that wear a map-cracking disguise.
Alkali-silica reaction (ASR)
Alkali-silica reaction is a slow chemical crack from inside the concrete. Reactive silica in certain aggregates reacts with the alkalis in the cement paste to form a gel, and that gel swells when it takes on water, pushing the concrete apart from within. It needs three things at once: reactive aggregate, high alkalinity, and moisture. Take any one away and it does not run.
ASR is a long-term problem, years not days, and it shows as map cracking, often with gel weeping or staining at the cracks and sometimes popouts over reactive particles. In unreinforced concrete it makes the classic map pattern; where reinforcement restrains it, the cracking aligns with the steel. It is sometimes called concrete cancer because it is progressive and, once established, hard to stop. Severe cases lose strength and have ended in demolition.
The defense is in the mix, before the pour. Use non-reactive aggregate, or where a reactive aggregate has to be used, suppress the reaction with a low-alkali cement and supplementary cementitious materials such as fly ash or slag. Confirming ASR takes a petrographic examination of cores, not a field guess, so a suspected case goes to a lab and a materials engineer, not to a patch.
Corrosion cracking: rust along the bar
Corrosion cracking starts inside the concrete, at the reinforcement, and works its way out. Steel in sound, high-alkalinity concrete is protected by a passive film. When chlorides reach the bar, from deicing salt, seawater, or a chloride-bearing mix, or when carbonation drops the pH at the steel, that film breaks down and the bar starts to rust. Rust takes up several times the volume of the steel it came from, commonly cited around two to six times, and that expansion splits the cover.
The pattern is unmistakable once you know it: cracks that run straight along the line of the reinforcement, parallel to the bars, often with rust staining bleeding out of them. Left alone, the cracks widen, the cover spalls off in sheets, and the exposed steel corrodes faster with nothing protecting it, so the damage accelerates. Thin cover and a chloride-loaded environment are the setup, which is why parking decks, marine structures, and salted bridges are where you see it most.
Prevention is cover, mix, and keeping chlorides out: adequate cover for the exposure, a dense low-permeability mix, and no chloride-bearing admixtures where steel is present. Once it is cracking and spalling, repair means removing the contaminated concrete back past the bar, cleaning or replacing the steel, and patching, often with corrosion-inhibiting measures. That is a specialty repair, not a cosmetic one.
When should you worry about a concrete crack?
Width is the first thing people ask about and the most over-read. A crack is worth worrying about when it is structural, active, leaking where it cannot leak, or letting water to reinforcement, and width is only one clue to any of those. A wide dormant shrinkage crack in a garage floor can be fine, and a thin crack walking up a loaded beam is not.
For appearance and corrosion protection, ACI 224 has historically published guide values for tolerable crack width by exposure, on the order of 0.016 in for dry interior conditions, down to roughly 0.007 in where deicing chemicals are involved and tighter still for water-retaining structures. Treat those as guidance, not a pass-fail line. They are about service and durability, they vary by edition, and they were never meant to certify that a crack is structurally safe. The applicable code edition and a structural engineer control that call.
A field rule that holds up: a clean hairline that is not growing, not leaking, and not on a load path is usually a monitor-and-move-on. A crack that is wide, growing, staining, leaking, or near a column or a load gets measured, dated, and routed to an engineer. The number that matters is not today's width, it is whether the width is changing.
Active vs dormant cracks
Before you repair a crack, find out whether it is still moving, because the answer decides the repair. An active crack is still working, opening and closing with temperature, ongoing settlement, deflection under live load, or a reaction still running. A dormant crack has stabilized and is not going anywhere.
It matters because a rigid repair on an active crack fails. Glue a moving crack shut with epoxy and it simply cracks again, next to the repair, the first time the thing that moves it moves. You either treat an active crack with something flexible that can move with it, or you find and fix the cause of the movement first, then repair.
Telling them apart takes time and a marker, not a guess. Bridge the crack with a tell-tale, a glued glass tab, or a marked and dated line across it, and watch it through a few temperature and load cycles. If it grows or breaks the tell-tale, it is active. If it sits still across the cycles, it is dormant and you can treat it as such. Skipping this step is why a lot of crack repairs come back.
Preventing shrinkage cracks: the toolkit
You cannot stop concrete from cracking, but you can hold shrinkage cracking down to hairlines on planned lines, and the toolkit is short and well proven. Almost all of it comes down to less shrinkage and a place for the shrinkage to go.
Less water is the biggest lever. A wetter mix shrinks more, full stop, so the lowest water content that still places and finishes well is the single best thing you can do for cracking. Good coarse aggregate, well graded and at a sensible maximum size, cuts shrinkage because aggregate does not shrink and paste does. Then cure it. A slab that dries too fast at the surface crazes and cracks; holding the moisture lets it gain strength as it shrinks, covered in the curing side of the surface-defects guide.
Joints and steel handle the shrinkage that is left. Control joints, construction joints, and isolation joints are the primary tool for putting the drying shrinkage crack where you want it, and the control-joint layout guide covers spacing, the quarter-depth cut, timing, and the three joint types in full. Reinforcement does not stop cracks, it holds them tight, distributing shrinkage into many fine cracks instead of a few wide ones. Cut the subgrade drag with a slip sheet, isolate the fixed restraints, and put diagonal steel at re-entrant corners. Do those and the cracks you get are the ones you planned for.
Crack repair: seal it, inject it, or watch it
Match the repair to the crack: what caused it, whether it is structural, and whether it is still moving. Get those wrong and the repair is decoration. ACI 224.1R lays out the methods; the field version is three buckets.
Route and seal is the workhorse for non-structural cracks that need to keep water and debris out. You widen the crack along its face into a small V or U, clean it, and fill it with a sealant, a flexible one if the crack moves and shedding water matters, a rigid filler if it is dormant and takes traffic. It treats the symptom and protects the surface. It does not restore structural continuity.
Epoxy injection is the structural repair. Pressure-injected epoxy bonds the two faces back together and can fill cracks as narrow as a couple of thousandths of an inch, restoring the section, but only on a dormant crack where the cause has been dealt with. Inject an active crack and it reopens beside the cured epoxy. The third bucket is the honest one: monitor. A dormant, hairline, non-structural crack doing no harm gets measured, dated, and left alone, because not every crack is a repair. Structural cracks get the engineer first, the repair method second, and the cause fixed before either.
Crack control on data center and industrial slabs
On a data center or heavy industrial slab, crack control moves to the front of the design, because the floor has to stay flat and tight under racks, hard wheels, and equipment that does not tolerate movement. A crack that would be a cosmetic note on a garage floor is a flatness and a contamination problem under a row of cabinets or a raised floor.
The approach is fewer, better-controlled cracks rather than hoping for none. That means a low-shrinkage mix, generous reinforcement to hold any cracks tight, close attention to subgrade and vapor conditions, and a joint plan built for the traffic, often with dowels and sometimes shrinkage-compensating or post-tensioned designs to push joint spacing way out. The control-joint layout guide covers the heavy-slab jointing and the data center floor in detail.
The crack-typing lesson for these floors is that you diagnose harder and earlier. Random cracking on a superflat floor gets read against a tight spec, mapped, and tied to a cause fast, because the cheap conventional answer, a thin slab and wide joints, is the wrong answer for a floor that runs hard for decades. Spend on the mix, the steel, and the joints up front, because the cracks are where these floors get expensive.
What to document
Most crack disputes turn on whether the crack grew, not on whether it is there, so the record you make the day you find it is what settles the question down the road. Document it the day you find it, with a photo, a scale, a location on the plan, the date, and a measured width, and mark the crack itself with a dated line so the next look can tell you if it moved.
Capture what family it points to from the timing, pattern, width, and location, the likely cause, and the action you took or recommended. The table below is the same sort the field carries in its head, laid out so a PM, an owner, or an engineer can read it. The action column is the one that matters, because it is the difference between a monitored hairline and a missed structural crack.
| Crack type | Timing | Pattern | Likely cause | Action |
|---|---|---|---|---|
| Plastic shrinkage | 1 to 6 hours | Short, random, roughly parallel, shallow | Rapid surface evaporation | Refloat if plastic; cosmetic once set |
| Plastic settlement | First few hours | Lines over the top bars, about 1 mm | Concrete settling over reinforcement | Revibrate; lower slump, adequate cover |
| Drying shrinkage | Days to weeks | Between restraints, off corners | Restrained drying shrinkage | Joints, steel, less water; usually monitor |
| Thermal (early) | 1 to 3 days | Through-section on mass pours | Hydration heat and restrained cooling | Spec temperature controls; engineer on mass |
| Re-entrant corner | Days | 45 degrees out of an inside corner | Stress concentration at the notch | Joint into the corner plus diagonal steel |
| Structural / flexural | Any time, often grows | Vertical on tension face, or diagonal shear | Overload, settlement, undersized member | Call a structural engineer |
| Corrosion | Months to years | Along the bar line, rust staining | Rebar corrosion, chloride or carbonation | Remove cover, treat steel; specialty repair |
| Alkali-silica reaction | Years | Map cracking, gel or staining | Reactive aggregate plus alkali and moisture | Petrographic test; materials engineer |
Common mistakes
- Treating every crack the same, panicking over a cosmetic hairline or shrugging off a structural one.
- Blaming the mix for a restraint crack, when the detail held the slab where it could not move.
- Calling a settlement crack drying shrinkage, when the crack follows the top bars exactly.
- No joints, joints too far apart, or joints cut too late, so the slab cracks where it wants.
- Too much water in the mix, which raises shrinkage and the cracking that comes with it.
- No cure, so the surface dries before it hydrates and crazes and cracks.
- No joint and no diagonal steel at a re-entrant or inside corner, the most preventable crack there is.
- Epoxy-injecting an active crack without fixing the movement first, so it reopens beside the repair.
- Ignoring a crack that follows the load path or grows over time instead of routing it to an engineer.
Field checklist
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
Standards and references
The project structural drawings, the specification, and for any structural crack the engineer govern. Everything below is the framework those sit on, and where the contract documents are stricter, they win.
ACI 224R, control of cracking in concrete structures, is the main reference for why concrete cracks and the tolerable crack widths by exposure, and ACI 224.1R covers the causes, evaluation, and repair of cracks, including route and seal and epoxy injection. ACI 302, the guide for concrete floor and slab construction, and ACI 360, for slabs on ground, carry the jointing and slab detailing. ACI 318, the structural concrete code, governs the design side that structural cracks point back to, and ACI 301 is the specification, including the mass-concrete temperature limit. The exact document numbers and provisions shift between editions, so confirm the edition the project adopted before citing one.
On materials and diagnosis, ASTM covers the test methods, a petrographic examination of cores diagnoses ASR, and the mix supplier and a materials engineer own the aggregate-reactivity and chloride questions. Name the standard that controls the point, verify the edition, and let the project specification and the engineer override any rule of thumb.
Units, terms, and conversions
Crack widths read in inches and millimeters, and the same crack gets described both ways across a spec, a report, and a field note, so keep the conversion handy. Crack width is the number that travels; depth, length, and movement are the rest of the picture.
A hairline crack is commonly taken as under about 1/16 in, roughly 1.5 mm, though the term is loose and width alone does not classify a crack. The ACI guide values run small, on the order of 0.016 in (0.4 mm) down to 0.007 in (0.18 mm) by exposure. Drying shrinkage runs around a few hundredths of an inch per foot, up to roughly 600 millionths of strain. Keep the units straight between the drawing, the report, and the crack you are looking at.
- Plastic shrinkage crack
- An early surface crack opened in the first hours when evaporation outruns the bleed rate
- Plastic settlement crack
- An early crack over the top bars where fresh concrete settles around the reinforcement
- Drying shrinkage
- The volume loss as hardened concrete dries, which cracks the concrete under restraint
- Restraint
- Anything that stops concrete moving freely as it shrinks or contracts, turning movement into tension
- Re-entrant corner
- An inside corner where stress concentrates and a crack tends to run out at about 45 degrees
- Active vs dormant crack
- Active cracks are still moving; dormant cracks have stabilized, which decides the repair
- ASR
- Alkali-silica reaction, a slow expansive gel from reactive aggregate, alkali, and moisture
- Crack width
- The opening of a crack at the surface, one clue among timing, pattern, and location
FAQ
Why does concrete crack?
Concrete is weak in tension, about a tenth of its compressive strength, so it tears whenever it is pulled harder than that. Most cracks come from shrinkage fighting restraint as the concrete dries, with thermal movement, overload, settlement, and corrosion the other causes. Almost all concrete cracks; the goal is controlling where and how.
What is drying shrinkage in concrete?
Drying shrinkage is the volume loss concrete undergoes as it dries and loses moisture over days, weeks, and months, on the order of a few hundredths of an inch per foot. When restraint stops the concrete from shrinking freely, the shrinkage goes into tension and cracks the concrete. Control joints put that crack where you want it.
When should you worry about a concrete crack?
Worry when a crack is structural, growing, leaking, staining, or on a load path, near a column, or one you can fit a quarter into. Width alone does not classify a crack, and a thin one can still be structural. Measure it, date it, and route anything structural or growing to an engineer.
What is a re-entrant corner crack?
A re-entrant corner crack runs out of an inside corner, the notch of an L-shaped slab or the corner of a blockout, opening, or column pocket, usually at about 45 degrees within days of the pour. Shrinkage stress concentrates at the inside angle. Prevent it with a control joint into the corner plus diagonal reinforcement.
What is the difference between plastic shrinkage and plastic settlement cracks?
Both open in the first hours while the concrete is plastic. Plastic shrinkage cracks are short and random from rapid surface evaporation. Plastic settlement cracks follow the top bars, because fresh concrete settling around the reinforcement tears at the surface above it. The pattern is the tell: random means shrinkage, over the bars means settlement.
What causes cracks along the line of the rebar?
Cracks that run straight along the bars usually mean corrosion. When chlorides or carbonation reach the steel, it rusts, and rust takes several times the volume of the steel, splitting the cover along the bar with rust staining. Early cracks over the bars in fresh concrete are plastic settlement instead. Cover and keeping chlorides out prevent the corrosion case.
How do you tell if a concrete crack is structural?
Read the pattern and the location, not just the width. Flexural cracks are vertical on the tension face of a beam or slab; shear cracks run diagonally near supports; settlement cracks track a foundation that moved. A crack on the load path or one that grows over time is the concern. When in doubt, a structural engineer makes the call.
Can you repair a concrete crack permanently, or will it come back?
It depends on whether the crack is active or dormant. A dormant crack can be epoxy-injected to bond the faces or routed and sealed for good. An active crack reopens beside any rigid repair unless you fix the movement first or use a flexible sealant. Find out if it moves before you choose the repair.
Does every concrete crack need to be repaired?
No. Many cracks are dormant, hairline, and non-structural, and the right move is to measure, date, and monitor them rather than repair. Repair when a crack leaks, lets water reach reinforcement, takes traffic on its edges, or is structural. ACI guidance on tolerable crack width by exposure helps, but the engineer controls the structural call.
People also ask
Codes cited in this guide
This guide is written and reviewed against the published standards below. Always confirm the current adopted edition with the authority having jurisdiction.