11.6 Concrete Curing, Joints, Testing & Defects
Key Takeaways
- Curing maintains moisture and temperature so hydration continues; specified compressive strength is referenced to 28-day cylinder breaks
- Control joints must be cut to at least one-quarter of the slab depth and are commonly spaced in feet at roughly two to three times the slab thickness in inches
- Standard-cured cylinders test the concrete as delivered; field-cured cylinders test the in-place curing and protection
- A slump test measures consistency, not strength — a passing slump says nothing about compressive strength
- Scaling, crazing, dusting, delamination, and plastic shrinkage cracking each trace to a specific finishing or curing error
Hydration Does Not Stop at the Pour
Quick Answer: Curing is maintaining adequate moisture and temperature so cement hydration continues after placement. Concrete that dries out early never reaches its design strength no matter how good the mix was. Minimum curing is commonly 7 days for normal cement and about 3 days where Type III high-early-strength cement is used.
Concrete does not "dry" — it hydrates. Water is a reactant. Concrete kept continuously moist keeps gaining strength for months; concrete allowed to dry at day two stops permanently at whatever strength it had reached. In Arizona's low humidity, that difference can be 30% or more of design strength, and it is invisible until the cylinders come back.
Curing Methods and Duration
| Method | How it works | Notes |
|---|---|---|
| Ponding / immersion | Standing water on the surface | Best control; practical only on flat, contained slabs |
| Continuous fog or sprinkling | Keeps the surface wet | Water use is high; must not be allowed to cycle wet/dry |
| Wet burlap or cotton mats | Absorbent covering kept saturated | Must never be allowed to dry out |
| Plastic sheeting | Traps moisture | Can cause surface discoloration; seams must be lapped and sealed |
| Curing compound | Membrane sprayed on the surface | Most common on large slabs; must be applied at the specified coverage rate immediately after finishing |
| Insulating blankets | Retains heat | Cold-weather work at higher Arizona elevations |
| Condition | Typical minimum curing |
|---|---|
| Normal Type I/II concrete | 7 days |
| Type III high-early-strength | 3 days |
| High-performance, low w/c mixes | 7 days or more, with tighter moisture control |
Warning: a curing compound sprayed too thin, or applied hours late after the surface has already dried, does not cure — it just makes the slab look wet. Coverage rate and timing are both specification items.
Joints: Three Types, Three Jobs
Concrete shrinks as it dries and moves as it changes temperature. Jointing controls where the resulting cracks go.
| Joint | Purpose | Detail |
|---|---|---|
| Control (contraction) joint | Creates a plane of weakness so shrinkage cracks form in a straight, planned line | Sawcut or tooled to at least ¼ of slab depth (¼ T) |
| Construction joint | Stops a placement where work ends for the day | Bulkhead with keyway or dowels; a planned cold joint |
| Isolation (expansion) joint | Separates the slab from columns, walls, and other fixed elements so they can move independently | Full-depth compressible filler; no reinforcement crossing |
Spacing rules of thumb. Control joint spacing in feet is commonly two to three times the slab thickness in inches: a 4-inch slab gets joints roughly every 8 to 12 feet, a 6-inch slab roughly every 12 to 18 feet. Panels should be kept close to square — an aspect ratio beyond about 1.5 to 1 invites a crack across the middle of the panel regardless of the joints at its edges.
Sawcut timing is the other half. Cut too early and the saw ravels the edge; cut too late and the slab has already cracked on its own. The window for conventional wet sawing is roughly 4 to 12 hours after placement depending on temperature and mix, and early-entry dry saws move that window earlier. In July in Phoenix, the window closes fast — crews that plan to cut "first thing tomorrow" find random cracks waiting for them.
Testing: What Each Test Actually Proves
| Test | Standard | Measures | Does not measure |
|---|---|---|---|
| Slump | ASTM C143 | Consistency / workability | Strength, w/c ratio, air content |
| Air content | ASTM C231 (pressure) / C173 (volumetric) | Entrained plus entrapped air | Strength |
| Temperature | ASTM C1064 | Concrete temperature at delivery | — |
| Unit weight / yield | ASTM C138 | Density and delivered volume | Strength |
| Cylinder casting | ASTM C31 | Specimen preparation and curing | — |
| Compressive strength | ASTM C39 | Compressive strength at test age | Durability, finish quality |
Slump is the one candidates over-read. A 4-inch slump load and a 4-inch slump load can have entirely different strengths, because a superplasticized 0.42 mix and a watered-up 0.58 mix can arrive at the same consistency. Slump verifies that the load matches what was ordered; it is not a strength test.
Cylinders are the acceptance test. Standard practice is a set cast per specified volume or per day's placement, with breaks typically at 7 and 28 days. Two curing regimes answer two different questions:
- Standard-cured (lab-cured) cylinders are stored under controlled moisture and temperature. They test the concrete as delivered and are the basis for accepting or rejecting the mix.
- Field-cured cylinders sit at the jobsite under the same conditions as the structure. They test the adequacy of curing and protection in place, and are used to decide when forms may be stripped or a slab may be loaded.
That distinction is a favorite exam item. If the field-cured cylinders break low but the standard-cured cylinders pass, the mix was fine and the curing failed — which is a contractor problem, not a supplier problem.
Worked example. A specification calls for 4,000 psi at 28 days. Cylinders break at 2,900 psi at 7 days and 4,150 psi at 28 days. The concrete complies: 7-day strength typically runs roughly 65–70% of 28-day strength, so 2,900 psi at 7 days was on track, and the 28-day result is the acceptance criterion.
Reading Defects Backwards to Their Cause
| Defect | Appearance | Usual cause |
|---|---|---|
| Plastic shrinkage cracking | Short, random, parallel cracks appearing within hours | Surface evaporation outrunning bleed water — hot, dry, windy conditions |
| Crazing | Fine, shallow, map-like network | Surface dried too fast; overworking or wetting the surface during finishing |
| Dusting | Powdery surface that will not stop shedding | Finishing while bleed water was still present; working water into the surface; poor curing |
| Scaling | Surface flaking off in patches | Freeze-thaw damage from inadequate air entrainment; deicing chemicals |
| Delamination | Hollow-sounding, separated surface layer | Sealing the surface with a steel trowel too early, trapping bleed water and air beneath |
| Honeycombing | Voids exposing coarse aggregate on formed faces | Inadequate consolidation; segregation; oversized aggregate for the form |
| Random cracking across panels | Cracks ignoring the joints | Joints cut too late, spaced too far apart, or too shallow |
Notice how many of these trace to finishing too early. Bleed water must rise and evaporate before final floating and troweling; a crew that closes the surface while water is still underneath produces dusting or delamination almost every time. This is also why exterior air-entrained flatwork gets a broom or float finish rather than a hard steel trowel.
Exam tip: the exam presents a defect and asks for the cause. Sort by timing. Cracks in the first hours point to evaporation. Surface texture failures point to finishing. Flaking in service points to air entrainment. Cracks ignoring the joint pattern point to jointing depth, spacing, or timing.
A control joint is sawcut 3/4 inch deep in a 6-inch slab, and random cracks appear away from the joints. What is the most likely cause?
Standard-cured cylinders from a placement break above the specified strength, but field-cured cylinders from the same load break well below it. What does this indicate?
An exterior slab develops a hollow-sounding surface layer that separates in sheets. What caused it?
What does a slump test prove about a delivered load?