Concrete strength testing with cylinders field guide

What strength testing proves, and why you wait on it

Concrete compressive strength testing is how you prove the material poured into the forms actually reaches the strength the structure was designed around. The mix design sets the target on paper. The cylinder break is the evidence the delivered concrete met it. Until a set of cylinders comes back at or above the specified strength, every load the structure carries is carried on faith.

The mechanism is direct. The structural engineer sized the columns, beams, and slabs against a single number, f'c, the specified compressive strength. Pour concrete weaker than that and the safety margin the engineer built in shrinks toward nothing, quietly, with no symptom you can see from the surface. A cylinder crushed in a machine is the only routine check that puts a real number against the design assumption.

The companion mix design guide covers how the supplier proportions a mix to hit f'c and why the water-cement ratio runs the strength. This guide picks up after the truck arrives. How you sample the concrete, cast the cylinders, cure them, break them, and read the result against the acceptance rule. Two different jobs. The supplier owns the recipe. The testing proves what showed up.

The specified strength f'c, and why the mix is designed above it

f'c is the specified compressive strength the structure was designed for, the number on the drawings, almost always called out at 28 days. Common values are 3000, 4000, and 5000 psi for buildings, higher for columns and special work. That is the floor the concrete has to reach, not the average it should land on.

The supplier does not aim the mix at exactly f'c. Concrete strength varies from batch to batch, so a mix that averaged f'c would fail close to half its tests by chance alone. ACI handles this with overdesign. The required average strength, f'cr, is set above f'c by an amount that grows with the producer's strength variability, measured as standard deviation from past data. A plant with tight, consistent results carries a smaller cushion. A plant with scattered data carries a bigger one.

Carry that in your head before anyone panics over a single number. The mix was built with margin on purpose, which is exactly why one cylinder a little below f'c is not automatically a failure, and why the acceptance rule later is statistical rather than pass-fail on one break.

How are concrete test cylinders made?

A field technician casts the cylinders from a sample of the fresh concrete taken at the point of delivery per ASTM C172, and makes them per ASTM C31. Standard sizes are 6 by 12 in and the now-common 4 by 8 in. A set usually runs two to four cylinders or more, so there are specimens to break at each age plus a hold cylinder in case a result is questioned.

Filling and consolidation follow the standard exactly, because consolidation that varies makes strength that varies. A 6 by 12 cylinder fills in 3 equal layers, each rodded 25 times with the standard tamping rod, the rod reaching about an inch into the layer below. A 4 by 8 cylinder fills in 2 layers, each rodded 25 times. After rodding each layer, tap the sides of the mold with a mallet to close the rod holes, then strike off the top flush. Same rod, same count, every cylinder.

When in the pour you sample matters. Take it from the middle portion of the discharge, not the first surge or the last dregs, within the time window the standard allows after the start of discharge. Make all the cylinders for one test from one sample of one load. The smaller 4 by 8 takes three cylinders to make a single strength test where the 6 by 12 takes two, because the smaller specimen scatters more and needs more breaks to average out.

Curing and handling: from the cure box to the lab

The most common reason a good pour fails a cylinder break is the cylinder, not the concrete, and it usually traces to the first day or two on site. ASTM C31 calls for initial curing in a controlled temperature range, commonly 60 to 80°F, held tighter for high-strength mixes (around 68 to 78°F for 6000 psi and up). Keep the specimens moist so they do not dry out, and protect them from freezing, from the sun, and from getting jostled.

The mechanism is straightforward. A cylinder that freezes, dries, or bakes on a truck dashboard the first night gains strength differently than the structure and breaks low. A cylinder rolled around the bed of a pickup gets micro-cracked and breaks low. Either way the concrete takes the blame for the handling. On a real site you hold the range with an insulated curing box, often thermostatically heated with water inside, that keeps the cylinders in spec no matter what the weather does. Cold pour, you heat the box. Hot pour, you shade and cool it.

Leave a set in the sun on a slab or in a freezing gangbox overnight and you have thrown the test away. The break will read low and there is no clean way to prove it was the handling rather than the mix.

After initial curing, typically within 48 hours, the cylinders go to the testing lab, still protected from rough handling and temperature extremes in transit. At the lab they are stripped from the molds and moved to standard curing: a moist room or a lime-saturated water bath held near 73°F at above 95 percent relative humidity until the break age. That standard cure is deliberately ideal, and it is what makes results comparable from one set to the next and one lab to another. Hold onto the fact that it is ideal. It judges the concrete under known conditions, not the curing the slab actually got.

How is the compressive strength measured?

The cylinder is capped and crushed. At the break age the lab measures the specimen, caps the ends so the load spreads evenly, then loads it in a compression machine at a controlled rate until it fails. ASTM C39 governs the test. The compressive strength is the maximum load the cylinder carried divided by its cross-sectional area, reported in psi.

Capping is its own source of error. The ends have to be plane and square, or the cylinder breaks low from a stress concentration at the high spot. Labs use either a sulfur mortar cap per ASTM C617 or unbonded neoprene pads in steel retainers per ASTM C1231. A sloppy cap is one more way a sound cylinder reads low. The machine and the loading rate matter too. Load too fast and the reading runs high; a machine out of calibration reads whatever it reads, right or wrong. The technician records the fracture pattern as well, because a clean cone or columnar break says the test was fair while an odd shear or side fracture flags a capping or alignment problem.

The arithmetic is worth knowing so a report makes sense. A 6 by 12 cylinder has an area of pi over 4 times 6 squared, about 28.3 sq in. A cylinder that fails at 113,000 lb gives 113,000 divided by 28.3, about 4000 psi. A 4 by 8 cylinder has about 12.6 sq in, so the same 4000 psi concrete fails at roughly 50,000 lb. Smaller cylinder, smaller load, same strength. The strength is the load over the area, never the load alone.

The test ages: 7, 28, and 56 days

Acceptance hangs on the 28-day strength for most work, because 28 days is the age f'c is specified at and the age the mix was designed to reach. A standard set is split across ages so you learn something early and confirm it at the age that counts.

The 7-day break is an early indicator. As a rough rule, 7-day strength runs about 60 to 75 percent of the eventual 28-day strength for ordinary portland cement mixes, so a 7-day that lands in that band is a good sign the 28-day will pass. A 7-day well under 60 percent is an early warning to start looking, not a failure on its own. It buys you time to investigate before the acceptance break comes due.

The 56-day age shows up on mixes heavy in supplementary cementitious materials. Fly ash and slag gain strength slower than straight portland cement, so the specification may move acceptance to 56 days to give the pozzolanic reaction time to finish. Break one of those mixes at 28 days, compare it to f'c, and call it low, and you have failed concrete that was never meant to be done yet. Read the spec for the designated test age before you read the result against anything.

What is the acceptance criteria for concrete strength?

Acceptance is statistical, not pass-fail on a single cylinder. Under ACI 318, a strength test is the average of the cylinders in one set, two 6 by 12 or three 4 by 8, broken at the designated age. The concrete is accepted when two conditions both hold: every average of three consecutive strength tests equals or exceeds f'c, and no single strength test falls below f'c by more than 500 psi. That 500 psi floor applies where f'c is 5000 psi or less; above that the floor becomes 0.10 times f'c.

Those exact figures, the 500 psi single-test floor and the three-consecutive-test average, are the ACI 318 framework and what most specifications adopt, but confirm them against the project specification and the edition of ACI 318 the job is built to. A tighter spec can override the code minimum.

The reason the rule reads this way is the same overdesign logic from earlier. One cylinder a little low does not condemn anything, because concrete varies and the mix was built with margin. What the rule catches is a real drop: an average trending below f'c, or a single test more than 500 psi under it. Those are the results that trigger action. Read the rule before you read the cylinder. People reject good concrete and accept marginal concrete by judging one number against f'c instead of running the actual criteria.

Field-cured and lab-cured cylinders answer two different questions

This is the distinction crews and even some inspectors blur, and it costs them. Lab-cured cylinders, also called standard-cured, are kept in ideal moist conditions and judge the concrete and the mix: did the material delivered meet f'c. That is acceptance. Field-cured cylinders are kept right next to the structure, in the same weather and the same curing the slab actually gets, and judge the in-place concrete: is it strong enough yet to strip the forms, pull shoring, stress tendons, or open to traffic.

The trap is using one to answer the other. A field-cured cylinder that reads low does not condemn the mix; it tells you the in-place curing was poor or that it is simply too early to load. A lab-cured cylinder that passes does not tell you the slab is ready to strip, because the slab never saw the moist room. Two questions, two specimens, and they are not interchangeable.

Acceptance testing is lab-cured by default. Field-cured sets get added when someone needs to make a stripping or loading decision from strength, and they are cured as identically to the structure as the site allows, sitting alongside the element they represent.

What do you do if a cylinder breaks low?

Do not condemn the structure on a low cylinder. A low break is the start of an investigation, not a verdict. Concrete in place is almost always stronger and more forgiving than a mishandled 4 by 8 suggests, and the test itself is the single most likely culprit.

Work the sequence in order. First, check the test and the handling. Was the sample taken right, the cylinders made and rodded right, the initial curing held in range, the caps plane and square, the machine in calibration. Most low breaks die here, on a handling or lab problem nobody had to pour concrete to fix. Second, if the test holds up, review the structural significance with the engineer of record. One low test on a non-critical element is not a trend on a column line. Third, if the engineer judges it matters, drill cores from the area in question per ASTM C42 and evaluate those instead of the cylinders. ACI 318 gives the core acceptance: the area is considered structurally adequate when the average of three cores reaches at least 85 percent of f'c and no single core falls below 75 percent of f'c.

Only if the cores also fail do you get into load testing, strengthening, or removal, and those are the engineer's calls, not the field's. The fastest way to burn money and credibility is to jackhammer a slab over one bad cylinder that spent the night in a hot truck. Investigate the test first. Call the engineer. Core before you condemn.

Why cylinders break low when the concrete is sound

Most low breaks are the test, not the concrete, and knowing the short list lets you rule causes out fast. A sample pulled from the wrong part of the load reads off. Poor consolidation or under-rodding leaves voids that crush early. A cylinder that dried out, froze, or cooked because the initial cure was not held breaks low. Rough handling that cracked the specimen before it reached the machine breaks low. A cap that was not plane and square breaks low. A compression machine out of calibration, or one loading too fast, throws the number either way. Any one of those reads as weak concrete on the report.

There is a pattern that points the finger. When the in-place concrete looks and sounds solid, the 7-day break was reasonable, and the slump, air, and temperature on that same load were all in spec, but the 28-day cylinders come back low, suspect the specimens before you suspect the structure. That combination almost always means the handling or the lab, not the mix that went in the forms.

The technician, the lab, and the chain of custody

The reliability of the whole result rides on who made the cylinders and who broke them. The field work, making and curing the cylinders along with slump, air, and temperature, is performed by an ACI Concrete Field Testing Technician Grade I, certified by exam to run and record those tests. The lab that breaks the cylinders should be accredited, with its compression machine calibrated on a documented schedule.

The chain is what ties the number to the concrete. Sample taken and logged, cylinders identified to the placement and the truck ticket, initial cure documented, specimens transported within the standard's window, lab-cured and broken at age, and the results tied back to the location they represent. A break with no traceable chain is a number with no home, and it falls apart the moment it is challenged.

When a result is disputed, the chain is what you defend or attack. The certified technician and the accredited lab are not paperwork for its own sake. They are what makes the strength number hold up when real money rides on whether the concrete is good.

How often is concrete sampled for strength tests?

ACI 318 sets a minimum sampling frequency and the specification can demand more. The common minimum is one strength test for each class of concrete placed each day, and at least once for each 150 cubic yards placed, or once for each 5000 sq ft of slab or wall surface area, whichever produces more tests. A separate test is taken for each different mix design used on the project.

The code also carries a small-quantity floor so a minor placement is not left unsampled, and a rule that a class of concrete tested too few times in total still gets a minimum number of tests. Confirm the exact thresholds, the per-day rule, and the small-quantity provision against the project specification and the adopted edition of ACI 318 before you build a testing plan on them.

Treat the frequency as a floor, not a target. On a critical placement the spec usually calls for more sets and more hold cylinders, because the cost of an unanswered question on that element dwarfs the price of a few extra cylinders. The minimum keeps you legal. The consequence of the pour decides how far above it you go.

In-place and nondestructive methods, and their limits

Cylinders are the acceptance standard, but several methods estimate strength in the actual structure without waiting on a lab break. Each has a job and a hard limit. The rebound hammer, the Schmidt hammer, measures surface hardness and correlates it loosely to strength under ASTM C805. It is good for checking uniformity across a placement and for picking where to drill cores. It cannot accept or reject concrete; the standard itself states rebound numbers are not a basis for acceptance.

The penetration probe, the Windsor probe under ASTM C803, and the pullout test under ASTM C900 drive or pull a fastener and read the resistance. Both correlate to strength better than the hammer, and both are still in-place estimates that need calibration to the specific mix. The maturity method under ASTM C1074 is the strongest of the in-place tools for tracking strength gain, and it gets its own section next.

None of these replace cylinders for acceptance. They answer in-place and timing questions. Use a rebound hammer to argue a slab is uniform enough to investigate or to choose a coring spot, never to overturn an acceptance break.

The maturity method for fast-track form removal

The maturity method estimates in-place strength from the concrete's own temperature history over time, and it is the modern tool for deciding when to strip forms or load early. ASTM C1074 governs it. You calibrate first: cast cylinders from the mix, break them at several ages, and build a strength-versus-maturity curve for that exact mix. Then temperature sensors embedded in the structure log the time-temperature history, the maturity index accumulates, and you read in-place strength off the curve in real time.

The schedule payoff is real. Instead of waiting for field-cured cylinders to break, you know the moment the slab reaches stripping strength as it happens, which can pull days out of a deck cycle on a fast job. The catch is that the calibration is mix-specific. Change the mix and the curve is wrong, so the curve has to be rebuilt for every mix you intend to track this way.

The honest limit: maturity tracks strength gain from temperature, so it does not catch a batching error that made the concrete weak in a way temperature cannot reveal. Pair it with cylinders, do not replace them with it.

The fresh tests that ride with the cylinders

On the same sample the cylinders come from, the technician runs the fresh-concrete tests, because they explain a strength result later. Slump for consistency, air content, concrete temperature, and often unit weight are all taken at the point of delivery alongside making the cylinders.

Each one ties back to strength. Slump well over the ticket can mean water was added at the truck, which raises the water-cement ratio and drops strength; the slump test guide covers that call and the ASTM C94 tolerance. Air content that ran high costs strength too, on the order of a few hundred psi for each added percent of entrained air. Temperature outside the range flags hot or cold weather effects on the set and strength gain. When a 28-day break comes back low, the fresh-test record on that exact load is the first place to look for the why.

A cylinder set with no slump, air, and temperature recorded beside it is a break you cannot interpret. Record them together, on the same load, every time.

Using strength to strip forms, post-tension, and open to load

Strength is more than an acceptance number. It is also the gate for construction operations. You strip forms, remove reshores, stress post-tensioning tendons, and open a slab to traffic or load when the in-place concrete has reached the strength those operations require, a strength set by the engineer and the spec and often well below the 28-day f'c.

The test that answers it is not the lab-cured acceptance cylinder. Use field-cured cylinders cured with the structure, or the maturity method, because both track what the in-place concrete actually did rather than what an ideal cure would have done. Stripping a deck on a lab-cured number is how a slab gets stripped before it is ready in cold weather, when the in-place concrete is colder and slower than the moist room ever was.

Post-tensioning is the blunt case. Stress the tendons before the concrete reaches the specified stressing strength and you can crush the anchorage zone. That strength comes from field-cured cylinders or maturity, verified, not from the calendar and not from the acceptance set sitting in a lime bath across town.

Acceptance on mats, transfer slabs, and data center pours

The large, critical placement is where the testing program gets heavier, and for good reason. A thick mat foundation, a transfer slab, or a data center structural slab carries consequences that make extra cylinders cheap insurance. Expect more sets per placement, extra hold cylinders for later ages, and often field-cured sets plus maturity to manage both the schedule and the heat.

Mass concrete adds a wrinkle the cylinders alone do not catch. A thick section holds the heat of hydration, so the core cures hot while the surface cools, and the temperature differential across the section, not just the strength, becomes something the spec controls to keep the placement from cracking. The cylinders judge strength. A separate thermal-control plan, with internal sensors, judges the cracking risk from that differential.

Scale the testing to the consequence. The minimum frequency is sized for ordinary work. On the element the whole building leans on, the spec and the engineer set a heavier program, and the field follows it rather than defaulting to the code floor.

What to document

A strength result with no record behind it cannot be defended when it matters, and the moment it matters is months out, when a load runs and someone asks whether the concrete under it was ever right. Tie every set back to the concrete it represents and the conditions it was made under.

Capture the placement location and element, the truck and ticket number, the mix design and the specified f'c, the sample time, the fresh-test results on that load, the cylinder set ID and the number of cylinders, the initial curing record, the break ages and results, the acceptance rule applied, and whether the set was lab-cured or field-cured and why. If field-cured cylinders drove a stripping or stressing decision, record that decision next to the strength it was based on. The table below sorts the specimen types by what each one actually judges, which is the question people get wrong when they read a report.

Field checklist

Common mistakes

• Mishandling cylinders in the first 24 to 48 hours, letting them freeze, dry, bake, or get jostled, then blaming the concrete for the low break.
• Condemning concrete on one low cylinder without investigating the test, reviewing with the engineer, and coring per ASTM C42.
• Using field-cured cylinders to judge the mix, or lab-cured cylinders to decide when to strip forms.
• Treating a rebound hammer reading as acceptance instead of a uniformity or where-to-core check.
• Poor consolidation or under-rodding, so the cylinder carries voids and reads weak.
• Breaking an SCM-heavy mix at 28 days when the spec designated 56, and calling it a failure.
• Sampling below the ACI 318 minimum frequency, or letting one strength test stand in for two different mixes.

Standards and references

The test methods belong to ASTM and the acceptance belongs to ACI, so cite the one that governs the point. ASTM C172 covers sampling the fresh concrete. ASTM C31 covers making and curing the field specimens. ASTM C39 is the compression test. ASTM C617 and C1231 cover capping. ASTM C42 covers drilled cores. ASTM C805 covers the rebound hammer, C803 the penetration probe, C900 the pullout test, and C1074 the maturity method. ACI 318 carries the strength-acceptance criteria and the core-evaluation thresholds, and ACI 301 is the common reference specification for the work.

The acceptance numbers, the 500 psi single-test floor, the three-consecutive-test average, and the 85 and 75 percent core thresholds, are the ACI 318 framework as it is commonly applied, but the project specification and the adopted edition of ACI 318 control. Confirm the designated test age, the sampling frequency, and the acceptance rule against the contract documents before you apply any of them to a real result.

The field testing is performed by a technician certified under the ACI Concrete Field Testing Technician Grade I program, and the breaking lab should hold accreditation. Those credentials are part of what makes a strength result defensible, not optional polish on it.

Units, terms, and conversions

Concrete strength shows up in a few unit systems and under a few names, so the same idea can read differently across a drawing set, a lab report, and a metric spec.

Strength is reported in psi in the US and in megapascals, MPa, in metric documents. 1000 psi is about 6.9 MPa, so a 4000 psi mix is roughly 28 MPa and a 5000 psi mix is about 34 MPa. Cylinder sizes are 6 by 12 in, which is 150 by 300 mm, and 4 by 8 in, which is 100 by 200 mm. f'c is the specified strength the structure was designed for, while f'cr is the higher average the supplier proportions the mix to actually hit.

f'c

Specified compressive strength the structure is designed for, usually called out at 28 days

f'cr

Required average strength, set above f'c to cover batch-to-batch variation

Strength test

The average of the cylinders in one set, broken at the designated age

Lab-cured (standard-cured)

Cylinders cured in ideal moist conditions to judge the mix for acceptance

Field-cured

Cylinders cured with the structure to judge in-place strength for stripping or loading

psi / MPa

Pounds per square inch and megapascals; 1000 psi is about 6.9 MPa

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is a concrete mix design?
• How do you do a concrete slump test?
• What is the concrete strength acceptance criteria?
• How do you test a concrete slab for moisture before flooring?
• What is a concrete admixture?