Why cube testing matters — and what the 28-day break really decides
A concrete cube test is how a structure proves it is as strong as it was designed to be. Fresh concrete is sampled at the pour, cast into standard 150 mm cubes, cured in a controlled water bath, and crushed in a compression machine to measure the stress it carries before it fails. That crushing figure — in N/mm² — is compared against the specified grade (M20, M25, and so on), and it is the only direct evidence that the concrete actually in the building meets its grade.
The 28-day strength is the acceptance criterion. Concrete gains strength over time as cement hydrates, and IS 456 defines the characteristic strength at 28 days — that is the number the design assumed and the number acceptance is judged against. The 7-day break is an early warning only: at 7 days a normal OPC mix has typically reached roughly two-thirds of its 28-day strength, so a low 7-day result flags a problem — wrong mix, bad materials, poor curing — while there is still time to react. A good 7-day result is reassuring, but it never substitutes for the 28-day test. That is exactly why each sample is six cubes: three for the early 7-day indicator, three for the 28-day decision.
What happens if a cube test fails
A failed test — 28-day results below the acceptance limits of IS 456 Cl 16 — does not automatically condemn the structure, but it does trigger a defined escalation rather than a shrug. In order:
- Check the test, then the concrete. First confirm the result is genuine (correct sampling, curing, and machine calibration — not a cube handling error). If the concrete is genuinely suspect, the engineer investigates that specific pour.
- Core testing. Cut cores from the hardened member and test them, to measure the in-situ strength of the actual concrete rather than a cube cast beside it. This is the primary way IS 456 lets you assess concrete already in place.
- Load testing. If cores are inconclusive, the member can be subjected to a load test per IS 456 to check it carries its design load with acceptable deflection and recovery.
- Rectification or demolition. If the element still cannot be shown to be adequate, it is strengthened, or in the worst case removed and recast. The decision rests with the structural engineer against IS 456 — the cube result is what forces the question to be asked.
None of this is possible without traceable records: which sample came from which pour, its 7-day and 28-day figures, and the location it represents. A missing or muddled Cube Test Register is how a marginal result becomes an argument instead of a decision.
Worked example — a 45 m³ slab pour
Say you place 45 m³ of M25 for a floor slab in a single shift. Read the frequency table: 45 m³ falls in the 31–50 m³ band, so it needs 4 samples.
Volume = 45 m³ (31–50 band)
Samples = 4
Cubes/sample = 6 (3 at 7 days · 3 at 28 days)
─────────────────────
Total cubes = 4 × 6 = 24 (12 broken at 7 days, 12 at 28 days)
So the site casts 24 cubes from that slab — and because it all went in one shift, the one-sample-per-shift rule is already satisfied by the four samples. Push the same pour to 120 m³ and the volume rule gives 4 + floor((120−50)/50) = 5 samples = 30 cubes. Enter any volume above and the calculator returns the samples, the total cubes, and the 7-day / 28-day split.
For estimation and site-planning reference only. Sampling frequency is the IS 456:2000 Cl 15.2.2 minimum; each sample is 6 cubes tested per IS 516 (3 at 7 days, 3 at 28 days), and at least one sample must be taken from every shift regardless of volume. Acceptance, core testing, load testing and any rectification are decisions for the structural engineer against IS 456. Not a substitute for the project’s specified quality-assurance plan.