Thursday afternoon, the laboratory report for the third-floor columns lands in the site inbox. The class is f'c = 28 MPa; one line reads 24.05 MPa. Within the hour there are two camps: the client's representative wants the column broken out, and the contractor wants the result set aside because "that laboratory has always been low."
The argument runs three days not because the engineering is disputed, but because almost nobody on site can say from memory whether that figure is a specimen or a test, whether 28 days is the age that carries acceptance status, and who is entitled to declare concrete accepted or rejected.
This post walks the chain in the order a QA/QC engineer meets it: delivery ticket, acceptance criteria, cores, disposition. It begins after the truck discharges.
Strength of a concrete class is acceptable when both hold. First, every arithmetic average of any three consecutive strength tests equals or exceeds f'c. Second, no individual strength test falls below f'c by more than 3.5 MPa where f'c is 35 MPa or less (or by more than 0.10 f'c above 35 MPa). These are the acceptance criteria for concrete strength in ACI 318 and, in Philippine practice, NSCP 2015 Chapter 4 (Structural Concrete). Read the wording in your own printed copy: that chapter follows an earlier ACI vintage than ACI 318-19, so its organisation differs even where the numbers agree.
Get this sentence right. Failing either criterion obliges steps to increase the average of subsequent strength test results — that duty is triggered by either failure, not only by the moving-average one. Failing the individual-test criterion does something additional: it engages the low-strength investigation provisions. So does a second, independent limb — field-cured results indicating deficiencies in protection and curing. Under those provisions coring becomes an available route, not a mandated one (the code permits it), and only once the likelihood of low-strength concrete is confirmed and analysis shows load-carrying capacity is significantly reduced. The remediation duty is not switched off by the investigation duty. When both criteria fail, you owe both.
A composite sample comes from the middle portion of the discharge — never the first material out of the chute, never the tail — remixed before casting, with slump, temperature and moulding started inside the short windows the sampling standard sets. Frequency for each class runs on three parallel bases: per day of placement, per stated volume increment, and per stated increment of slab or wall surface area, whichever governs. Read those three values off your own printed copy; they are edition-dependent. Two exceptions travel with them: a small-quantity exemption that lets the building official waive strength testing below a stated total volume, and a floor requiring enough samples that a class is represented by at least five strength tests.
Standard-cured specimens prove the concrete as delivered. Field-cured specimens, left beside the element under the same weather and curing regime, prove the curing and protection you gave it. A low field-cured result is not evidence against the supplier.
The comparison rule, in words: curing and protection procedures must be improved when field-cured strength at the designated test age is less than 85% of the companion standard-cured strength. That comparison is waived when the field-cured strength exceeds f'c by more than 3.5 MPa.
Five-storey RC building, ready-mix, 150 × 300 mm cylinders, two companions per sample, f'c = 28 MPa.
Cylinder area A = π/4 × 150² = 0.785398 × 22,500 = 17,671.46 mm².
Companion 1: 415,000 N ÷ 17,671.46 mm² = 23.48 MPa. Companion 2: 435,000 N ÷ 17,671.46 mm² = 24.62 MPa. Test 5 = (23.48 + 24.62) ÷ 2 = 48.10 ÷ 2 = 24.05 MPa.
Seven strength tests for the class, in placement order (MPa): T1 32.0, T2 29.8, T3 28.2, T4 27.2, T5 24.05, T6 28.8, T7 30.7.
Criterion 1 — every average of three consecutive tests must reach 28.0 MPa:
One low test poisons three windows, and the last fails by a hair.
Criterion 2 — no test more than 3.5 MPa below f'c: limit = 28.0 − 3.5 = 24.5 MPa. T5 = 24.05, so the deficit is 28.0 − 24.05 = 3.95 MPa > 3.5 → FAILS. Every other test clears; the next closest, T4, is short by 0.8 MPa.
Both criteria failed, so both obligations run: raise the average of subsequent results and investigate the concrete represented by T5.
Field-cured specimen 21.7 MPa against its standard-cured companion 29.8 MPa: 21.7 ÷ 29.8 = 0.7282 = 72.8%, below 85% → curing and protection procedures must be improved. Waiver check: does 21.7 exceed 28.0 + 3.5 = 31.5 MPa? No → no waiver. This is a finding against site curing, not against the delivered concrete.
Measured core diameter 94 mm. A = π/4 × 94² = 0.785398 × 8,836 = 6,939.78 mm². ASTM C42 length-to-diameter corrections apply at L/D of 1.75 or less (1.75→0.98, 1.50→0.96, 1.25→0.93, 1.00→0.87; interpolation permitted); none at L/D = 2.00.
Average = (25.9374 + 24.2083 + 23.7138) ÷ 3 = 73.8595 ÷ 3 = 24.62 MPa. Check 1, average at least 0.85 f'c: 0.85 × 28 = 23.80 MPa; 24.62 ≥ 23.80 → OK, margin +0.82 MPa (87.9% of f'c). Check 2, no single core below 0.75 f'c: 0.75 × 28 = 21.00 MPa; lowest core 23.71 ≥ 21.00 → OK (84.7% of f'c).
The concrete represented by these cores is structurally adequate under the core criteria — 85% on the average of three, 75% on the lowest single core, not 85% on every core. Had either check failed, the cores would still not condemn anything by themselves; the result returns to the engineer of record.
Peso rates below are illustrative; price your own project.
Investigation: 3 cores × ₱4,500 = ₱13,500, plus ₱8,000 mobilisation = ₱21,500. Rework of one 0.45 m × 0.45 m × 3.0 m column: 0.45 × 0.45 = 0.2025 m²; 0.2025 × 3.0 = 0.6075 m³; 0.6075 × ₱4,800/m³ = ₱2,916 of concrete, plus an illustrative ₱65,000 for demolition, rebar, formwork and labour = ₱67,916. Ratio: 67,916 ÷ 21,500 = 3.2 × — demolishing before investigating costs about three times what finding out costs.
Project specifications can be stricter than code; code is only the floor. A capacity check may show the low result does not significantly reduce load-carrying capacity, in which case coring is not automatically required. Core orientation, moisture conditioning and reinforcement intercepted by the core all move the number. Placing-temperature limits commonly quoted on Philippine sites are specification or common-practice values, not a code mandate.
Moving averages across a growing test log, deficit checks and core corrections get retyped into a fresh spreadsheet every project and quietly break on one of them. The free set at RHCES Web Tools covers the routine ones.
No. Acceptance is judged at the age designated for f'c. A low 7-day result is a warning worth acting on — check the ticket, check curing, tighten sampling — but it has no acceptance status.
No, and applying it twice is one of the commonest errors in a core report review. Ask in writing whether the reported values are as-measured or corrected. Confirm the measured diameter too: in Part D, assuming a nominal 100 mm instead of the measured 94 mm would understate every core stress by about 12%.
The structural engineer of record, with the building official — not the laboratory, which reports numbers, and not the contractor, which may propose remedies.