Concrete strength cylinders and slab acceptance field guide

The strength test, and why the cylinder is the legal record

A concrete strength test is the compressive break of cylinders cast from a placement, and it proves the one thing the fresh tests cannot: how strong the concrete actually got. Slump, air, and temperature tell you the load was batched and placed right. The cylinder tells you whether the structure has the strength the engineer designed against. Everything else is consistency. This is the number.

The cylinder is also the legal record of the concrete. Long after the forms are stripped and the slab is loaded, the break result is the document that says the placement met its specified strength. When the building is modified, sold, loaded heavier, or questioned after a crack, the strength records are what the structural engineer reaches for. There is no second chance to make them. The concrete in the cylinder is the same concrete in the forms only if the cylinder was made, cured, and broken right, and the whole discipline exists to keep those two things equal.

On a data center the records carry more than the building. The mat under a switchgear lineup, the pad under a generator, and the slab under rows of loaded racks all have to reach their strength and hold it through the design event. The broader special-inspection and turnover picture lives in the data center structural QA overview. This guide is the cylinder itself: how it is made, cured, broken, and accepted, and what to do when it comes back low.

How do you make concrete strength cylinders?

You make strength cylinders to ASTM C31, from a sample taken to ASTM C172, consolidated in equal layers, and identified so the set can never be confused with another. The sample comes first and it controls everything downstream. Take a composite per C172 from the middle of the discharge, sampled at the point of placement for acceptance where you can, because the concrete the chute showed is not the concrete that goes in the forms on a hot pour or a long pump line. The same sample feeds the slump, air, and temperature tests, and the full sampling procedure lives in the concrete slump test guide.

Consolidate by rodding or vibration depending on the slump. ASTM C31 keys the method to consistency: rod the more fluid mixes, vibrate the stiffer ones, and follow the standard for the cutoff. When you rod, fill in equal layers, spread the strokes across the cross section, and drive the rod about 1 in into the layer below so the lift bonds. After rodding each layer, tap the side of the mold ten to fifteen times with the mallet to close the rod holes and release the larger trapped air. Strike the top off flush and finish it without overworking it.

Identify the set the moment it is cast, on the mold or the lid, not on a cap you peel off later. Tie the ID to the batch ticket and the exact placement location, because a cylinder you cannot trace back to a ticket and a spot in the structure is a cylinder nobody can defend when the question comes. Mislabeled and untraceable sets are the quiet failure here. The break can be perfect and still prove nothing if you cannot say which pour it came from.

4x8 or 6x12 molds, and how many per set

The trade casts two cylinder sizes, and they are made differently. The 6 by 12 in cylinder is filled in three equal layers, rodded 25 strokes per layer with the 5/8 in rod. The 4 by 8 in cylinder is filled in two equal layers, rodded 25 strokes per layer with the smaller 3/8 in rod. Use the rod that matches the mold. Rodding a 4 by 8 with the big 5/8 in rod is a common mix-up that tears the smaller specimen instead of consolidating it.

The 4 by 8 has taken over a lot of work because it is lighter to carry and handle, it breaks at a lower load so it fits more presses, and it cures faster in a smaller initial-cure box. The trade-off is that the smaller cylinder shows a little more scatter break to break, so a strength test on 4 by 8 cylinders is the average of three of them, where a 6 by 12 test is the average of two. The project specification and the lab fix which size the job uses, and you do not mix sizes inside one strength test.

A set is more than the cylinders that get averaged for acceptance. A typical set holds the acceptance cylinders for the specified age, plus one or two for an early 7-day read, plus a hold or two in reserve in case a result needs a recheck or a retest. Cast a few extra and you have options when a number looks off. Cast the bare minimum and a single broken or low cylinder leaves you with nothing to fall back on.

What is the initial cure for field cylinders?

Initial curing is the first 24 to 48 hours after casting, on site, and it is where good cylinders go bad. ASTM C31 calls for storing the freshly made specimens for up to 48 hours in a controlled temperature range while preventing moisture loss, before they go to the lab for standard curing. For most mixes that range is commonly 60 to 80 degrees F. For high-strength mixes, the standard tightens the window, so confirm the band against the current edition and the mix strength before you set the box.

The reason the window is so tight is that early temperature sets the early hydration, and the cylinder is supposed to track the curing it would get under ideal handling, not the weather. A cylinder left in the sun on a hot slab cures fast and hot and can read high early and erratic. A cylinder left out overnight in the cold cures slow and reads low, and if it actually freezes before it sets, the specimen is damaged in a way that has nothing to do with the concrete in the structure. Either way the break is about the box, not the pour.

The fix is an initial-cure box with temperature control and a minimum-maximum thermometer logging the range, kept out of the sun and the wind, with the cylinders protected from drying and from being bumped. Record the initial-cure temperatures, because the first question on a disputed low break is always whether the cylinders were held in range. Cylinders rattled across a rough site in a truck bed, or stacked next to a heater, or left on the ground in February, fail an investigation before the concrete is ever suspected.

Standard-cured vs field-cured cylinders

Standard-cured and field-cured cylinders come from the same concrete and answer two different questions. Standard-cured cylinders get the controlled initial cure and then lab curing under ideal moisture and temperature, so they measure the potential strength of the concrete as delivered. They are the acceptance specimens. When ACI 318 acceptance is judged, it is judged on standard-cured cylinders, because the point is the concrete, not the jobsite weather.

Field-cured cylinders are left to cure alongside the structure, in the same conditions the slab or member actually sees, so they measure the in-place strength under real curing. Their job is operational: deciding when the concrete has gained enough strength to strip forms, remove shores, post-tension, or load the member. A field-cured cylinder that has sat through the same cold nights as the slab tells you what the slab has, where the standard-cured cylinder tells you what the slab could have had if everything went right.

The two read differently on purpose, and the gap between them is itself a check. ACI 318 treats field-cured strength that falls below about 85 percent of the companion standard-cured strength at the same age as a flag that the in-place protection and curing need to improve. That comparison is field-cured against its own companion lab-cured cylinder, not against the specified strength. A job that casts only standard-cured cylinders has no way to make a defensible stripping or loading call, which is the rookie gap on a fast schedule: the acceptance cylinders look fine in the lab while nobody can prove the slab they are walking equipment onto has the strength yet.

Lab curing, capping, and the break

Once the standard-cured cylinders reach the lab, they go into standard moist curing until the test age. That means a moist room at controlled temperature with free water on the surface, or a saturated lime-water bath, so the specimens stay continuously wet. The lime in the bath keeps the water from leaching calcium out of the cylinder surface. A cylinder that dried out on a shelf in the lab reads off the same way a poorly field-cured one does, so the curing discipline does not end when the box leaves the site.

Before the break, the ends have to be plane and square so the load goes in evenly. Capping handles that. Bonded sulfur-mortar caps under ASTM C617 are melted and poured onto the ends, and unbonded caps under ASTM C1231, the neoprene pads in steel retainer rings, sit on the ends without bonding. The pads are common now because they are fast and clean, but each pad system is qualified to a strength range and a number of reuses, so a worn or out-of-range pad gives a bad break. A rough, soft, or out-of-plane cap concentrates load and breaks the cylinder low for a reason that is pure technique.

The break itself is ASTM C39, the compressive strength test. The capped cylinder is centered in the testing machine and loaded at a controlled rate until it fails, and the maximum load divided by the cross-sectional area is the compressive strength in psi. The technician records the load, the strength, and the fracture pattern, because the break type tells you something: a clean cone is a sound specimen, while a columnar or one-sided shear can point to a bad cap, an off-center load, or a defect. The number on the report is only as good as the cap and the centering that produced it.

What age are concrete cylinders tested?

Concrete strength is referenced to an age, and unless the specification says otherwise, that age is 28 days. The 28-day basis is convention, not physics. Concrete keeps gaining strength well past 28 days, but the gain slows enough by then that 28 days became the standard age the specified strength is defined against and the age acceptance is judged at. When a drawing says 4000 psi, it means 4000 psi at 28 days unless it names another age.

The 7-day break is the early read. Most ordinary mixes reach roughly two thirds to three quarters of their 28-day strength by 7 days, so a 7-day result is a leading indicator. A 7-day that lands where the mix history says it should is reassurance the 28-day will pass. A 7-day that comes in well under the expected ratio is an early warning to watch the set and have the holds ready, not a failure on its own.

Some mixes are specified to a later age, commonly 56 days, and a few to 90. Mixes with high replacement of cement by slag or fly ash gain strength slower and keep gaining longer, so judging them at 28 days would penalize a concrete that is fine by its design age. Mass-concrete mixes for heavy mat foundations often use a later acceptance age for the same reason. When the spec sets a later age, that age is the acceptance age, and the 28-day break becomes just another information point. Read the specified age off the spec before you schedule the breaks, because breaking and judging at the wrong age is a paperwork failure that fails good concrete.

What is the concrete strength acceptance criteria?

Under ACI 318, concrete strength is accepted on a pattern of tests, and a strength test is the average of at least two 6 by 12 cylinders, or three 4 by 8 cylinders, from the same sample broken at the specified age. Acceptance is satisfied when both of two conditions hold: every arithmetic average of any three consecutive strength tests equals or exceeds the specified strength, and no single strength test falls below the specified strength by more than 500 psi, or by more than 0.10 times the specified strength when that strength is over 5000 psi.

Read those two conditions together, because each catches something the other misses. The three-consecutive-test average is the durability check on the overall production: it allows the normal scatter while making sure the mix is centered above the specified strength on average. The single-test floor is the safety net under any one result: it allows a low test but not a collapse, and the 500 psi or 0.10 f prime c shortfall is the line a single test cannot cross. A run of tests that averages fine but has one test deep below the floor still fails, and a single result a little low while the average holds does not, by itself, condemn the concrete.

Confirm the exact provisions against the adopted code edition and the project specification, which can be stricter and usually is on a mission-critical building. The specification and the engineer of record control the criteria. The two conditions are stable across recent cycles, but the section numbering and the surrounding evaluation rules move, so cite the requirement by what it says, not by a remembered clause.

Specified strength, required average strength, and why the mix is over-designed

The specified strength, f prime c, is the strength the structure was designed for, and the number on the drawings. It is not the strength the mix is proportioned to hit on average. If a producer targeted exactly f prime c, half the tests would land below it by the normal scatter of concrete, and the acceptance criteria would fail constantly. So the mix is proportioned to a higher target, the required average strength, f prime cr, set so the statistics pass.

The required average strength is calculated from f prime c and the variability of the producer's concrete, using the standard deviation of their test history under ACI 318 and ACI 301. The tighter and more consistent the producer's record, the smaller the margin they need above f prime c. A producer with no reliable history gets a larger default margin, because uncertainty has to be covered by overdesign. This is why two plants can both supply a 5000 psi mix and proportion to different actual targets: the better-controlled plant earns a leaner, cheaper margin with the same acceptance.

The practical takeaway on the inspection side is that breaks coming in comfortably above f prime c are not waste or a richer mix than needed. That headroom is the overdesign doing its job, absorbing the scatter so the three-test average stays above the line. When the breaks start landing close to f prime c instead of up near the required average strength, the mix has lost its margin, and the next normal low result is the one that fails. Watch where the breaks cluster, not just whether each one passes.

What do I do if a cylinder breaks low?

A single low cylinder is the start of an investigation, not a verdict. ACI 318 accepts concrete on a pattern of tests, so one number under the line triggers a defined sequence, and condemning a placement on one low cylinder is the classic overreaction that costs everyone money for nothing. Work it in order.

First, separate the cylinder from the concrete. Check the break: the cap, the fracture pattern, the loading rate, whether the companion cylinder in the same test broke consistent with it. Then check the cylinder handling: the initial-cure temperatures, whether it was protected from drying and freezing, whether it got rattled or dropped. Then look at the companion fresh tests for that load, the slump, the air, and especially the unit weight under ASTM C138, which catches a load that was watered down. If a single cylinder reads low while its companion and the fresh tests are clean, the cylinder is the most likely problem, not the pour.

If the in-place concrete is genuinely in question, the strength of the structure is evaluated on the concrete itself, not the cylinders. The standard-cured cylinders are held and retested to confirm the result. Field-cured cylinders, if cast, are checked for what the in-place curing delivered. When that is not enough, the engineer of record directs cores under ASTM C42, drilled from the hardened concrete in the suspect area. Cores are evaluated against an acceptance that allows for coring damage and the difference between a core and a lab cylinder: the concrete is considered adequate when the average strength of three cores is at least 85 percent of the specified strength and no single core is below 75 percent. Cite those factors as the core acceptance, and confirm them against the adopted edition before you put them on a report.

Why a low break is usually the testing, not the concrete

Most low breaks are made in the testing, not in the truck, and a tech who knows that saves a lot of good concrete from getting cored out. The fresh concrete was sampled, batched, and placed under more eyes than the cylinder ever gets. The cylinder, meanwhile, passes through a half-dozen steps where a single careless move costs strength on the report with no change to the concrete in the forms.

Run the list of what actually drives a false low. A rushed or skipped initial cure, with the cylinders sitting in the sun, in the cold, or drying in the wind. A cylinder bounced across a rough site or dropped on the way to the box. Sloppy consolidation, the wrong rod for the mold, or the rod holes never tapped out, leaving voids. A rough sulfur cap or a worn neoprene pad out of its qualified range. A specimen that dried in the lab instead of staying in the moist room or the lime bath. An off-center cylinder in the press or a loading rate run too fast. Every one of those breaks the cylinder low while the slab is fine.

This is the part of the job that lives or dies on the technician, and it is why the certifications and the field discipline matter. The concrete cannot defend itself after it is placed. The cylinder is its proxy, and a careless proxy convicts an innocent pour. When a break comes back low, the honest first question is not what went wrong with the concrete. It is what went wrong with how we treated the cylinder, and the answer is usually right there in the initial-cure log and the cap.

High-strength concrete and the data center heavy pours

The heavy concrete on a data center raises the stakes on every step above. The mat or raft foundation under the generator yard, the central utility building, and the switchgear rooms is a thick, high-volume placement carrying concentrated equipment loads, and it is often a mass-concrete pour with its own thermal-control plan. Mass concrete frequently uses high replacement of cement with slag or fly ash to manage the heat of hydration, which is exactly the case where a later acceptance age, 56 or 90 days, is specified, because the mix is designed to keep gaining past 28 days.

Higher specified strengths change the test discipline. On mixes over 5000 psi the single-test acceptance floor switches from a flat 500 psi to 0.10 times the specified strength, and the initial-cure temperature window tightens. High-strength concrete also breaks at much higher loads, so the testing machine has to be rated for it and the caps have to be sound, because a marginal cap that survives a 4000 psi break can fail a 8000 psi one and report a low. The equipment pads, the housekeeping bases under the gear, are a different mix again, and each placement is judged against its own approved mix design and its own acceptance.

The recurring failure across all of it is casting cylinders against the wrong approved mix, or judging a late-age mix at 28 days. On a job running several mixes at once, the mat, the structural slabs, the pads, and the grout, a set tied to the wrong design was never a valid acceptance even if the number looked fine. Tie every set to its ticket, its mix, and its location, and judge it at the age that mix was specified to.

The breaks report and the strength trend

The break report is more than a pass or fail on one set. Read across the sets and it becomes a control chart on the producer, and the trend is where a problem shows itself before a single test ever fails acceptance. A plant whose breaks have been landing near the required average strength and then start drifting down toward the specified strength is losing its margin, and the drift is visible a few sets before a test drops below the line.

Watch three things in the trend. The level, which is where the breaks cluster relative to the specified strength and the required average strength. The scatter, which is how tightly the cylinders within a test and the tests within a class agree, because widening scatter means the production or the testing is getting less consistent. And the 7-day-to-28-day relationship, because a 7-day that suddenly reads low against its usual ratio is an early flag on the set in hand. The moving average of three consecutive tests, the same average the acceptance criteria use, is the line to plot, since it smooths the single-test noise and shows the real direction.

This is statistical quality control, and it is the producer's standard deviation in motion, the same number that set the overdesign margin in the first place. A tightening, well-centered record earns a leaner mix and fewer surprises. A trend going the wrong way is the cue to get the producer, the lab, and the engineer of record talking before a placement fails, not after. The set that fails acceptance is rarely a bolt from the blue. It is the end of a trend somebody should have been reading.

Chain of custody and lab accreditation

A strength result is only as defensible as the chain behind it, from the truck to the broken cylinder. Chain of custody means the set can be traced unbroken: the batch ticket, the sampling time and location, the cast time, the initial-cure record, the transport to the lab, the lab log-in, the curing, and the break. Break any link and the result is arguable. The cylinder that nobody can prove came from this placement, cured in range, and broke at the right age is a cylinder a contractor's lawyer or a producer's engineer can pick apart.

The testing is performed by an independent, accredited testing laboratory, retained so the party reporting the strength is not the party that placed the concrete. The lab accreditation and the technician certification are part of the result, not a formality. Concrete field technicians are commonly ACI-certified for the field tests, and the laboratory itself is accredited and routinely inspected for the strength testing it runs, through programs such as AASHTO accreditation and the Cement and Concrete Reference Laboratory, CCRL, inspections. A break from an unaccredited lab or an uncertified technician is a result the building official or the engineer of record can reject.

Treat the lab as part of the project team from the start. The strength records, the initial-cure logs, and the break reports are documents the owner inherits and the structural engineer reaches for years later when the building is loaded heavier or modified. A test that should have been witnessed and traced and was not cannot be recovered after the concrete is down. The chain is built at the truck, in real time, not reconstructed from memory after the trucks are gone.

What to document

The strength record is judged by whether someone can reconstruct, two years out, what each set proved and tie it to a location in the structure. Capture it at the truck and at the lab, not from memory. The set ID and the batch ticket tie the cylinders to the load and the approved mix. The placement location ties them to the structure. The fresh tests give the context that explains a low break. The initial-cure record answers the first question of any dispute, and the break results against the specified strength are the verdict.

Write down enough that the next person does not have to call you. The table below is the spine of a strength acceptance record for one set.

Common mistakes

• Rushing or skipping the initial cure, so the cylinders never see the controlled temperature range for the first 24 to 48 hours.
• Leaving cylinders in the sun, in the cold, or out where they can freeze before they set.
• Mislabeling sets or casting them with no traceable tie to a ticket, a mix, and a placement location.
• Condemning a whole placement on one low cylinder instead of running the ACI 318 pattern and investigating.
• Casting only standard-cured cylinders and having no field-cured set to make a stripping or loading call.
• Using the wrong rod for the mold, or never tapping the rod holes out, leaving voids that break the cylinder low.
• Breaking on a rough sulfur cap or a worn neoprene pad out of its qualified range.
• Casting against the wrong approved mix, or judging a late-age mix at 28 days instead of its specified age.
• Mixing 4x8 and 6x12 cylinders inside one strength test, or averaging the wrong number for the size.

Field checklist

Standards and references

ASTM C31 is the practice for making and curing strength specimens in the field: the molds, the layers and rodding, the consolidation, and the initial-cure temperature window for the first 24 to 48 hours. ASTM C172 governs the sampling so the cylinders see representative concrete. ASTM C39 is the compressive strength test, the break itself. Capping follows ASTM C617 for bonded sulfur-mortar caps and ASTM C1231 for unbonded neoprene pad caps, each qualified to a strength range. The companion fresh tests, slump under ASTM C143, air under C231 or C173, temperature under C1064, and unit weight under C138, are covered in the concrete slump test guide.

Acceptance and the strength criteria come from ACI 318, the structural concrete code, with ACI 301 as the specification for structural concrete. ACI 318 sets the two acceptance conditions, defines a strength test as the average of the cylinders in a sample, and gives the field-cured-versus-standard-cured comparison and the core-evaluation criteria. Cores follow ASTM C42 for obtaining and testing drilled cores, evaluated against the 85 percent average and 75 percent single-core acceptance in ACI 318. Section numbers and clause references shift between code cycles, so confirm the specific provisions against the adopted edition and any project amendments before citing them on a submittal.

The testing is performed by an accredited laboratory with certified technicians, with accreditation and inspection through programs such as AASHTO accreditation and the CCRL. The structural drawings and specifications control where they are stricter than the floor these documents set, and the engineer of record governs the acceptance and any low-strength evaluation. The broader special-inspection framework, IBC Chapter 17, the statement of special inspections, and the turnover record set, lives in the data center structural QA overview.

Units, terms, and conversions

Concrete strength reads in pounds per square inch on US jobs and megapascals most other places, and the two show up side by side on international work. One MPa is about 145 psi, so a 4000 psi mix is roughly 28 MPa and a 5000 psi mix about 34 MPa. Strength is always referenced to an age, 28 days unless the specification names another, and to the specified strength f prime c the structure was designed for.

The terms below are the ones that travel across the whole make-cure-break-accept chain.

f prime c

Specified compressive strength, the design strength on the drawings, referenced at the specified age and given in psi or MPa

f prime cr

Required average strength, the higher target the mix is proportioned to so the acceptance statistics pass, set from f prime c and the producer's variability

Strength test

The average of at least two 6x12 cylinders or three 4x8 cylinders from the same sample, broken at the specified age

Standard-cured

Cylinders given the controlled initial cure and lab moist curing, used for ACI 318 acceptance of the concrete

Field-cured

Cylinders cured alongside the structure to measure in-place strength for stripping, shoring, post-tensioning, and loading decisions

Initial curing

The first 24 to 48 hours on site, held in a controlled temperature range under ASTM C31 with moisture loss prevented

Capping

Making the cylinder ends plane and square before the break, by bonded sulfur caps (ASTM C617) or unbonded pads (ASTM C1231)

Core

A cylinder drilled from hardened concrete under ASTM C42 to evaluate in-place strength when cylinders are in question

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is a special inspection on a construction project?
• How do you do a concrete slump test?
• How is concrete strength tested?
• Why do you scan concrete before drilling or coring?
• What is an NFPA 110 acceptance test?