The cube test is the only direct proof that the concrete in your slab reached the strength the design assumed. It only works if cubes are sampled properly, cured properly, and judged against the right criteria.
How Many Samples (IS 456, Clause 15.2.2)
| Quantity of concrete in the work (m³) | Number of samples |
|---|---|
| 1 – 5 | 1 |
| 6 – 15 | 2 |
| 16 – 30 | 3 |
| 31 – 50 | 4 |
| 51 and above | 4 + 1 for each additional 50 m³ or part |
One sample is three cubes. The sample’s strength is the average of the three. If any individual cube differs from that average by more than ±15%, the sample is invalid (clause 15.4). Extra cubes are usually cast for 7-day testing.
Casting and Testing
- 150 mm cube moulds, filled in layers and compacted, following IS 516
- Cubes marked with date, location and grade, kept undisturbed on site, then demoulded and cured in water at 27 ± 2 °C until testing
- Tested in a calibrated compression testing machine at 7 and 28 days
- As a site rule of thumb, 7-day strength is often around two-thirds of 28-day strength. Acceptance is judged on 28-day results only.
Acceptance Criteria (IS 456, Table 11, as amended)
| Grade | Mean of 4 non-overlapping consecutive results (N/mm²) | Any individual result (N/mm²) |
|---|---|---|
| M15 | ≥ fck + 0.825 × established SD, or fck + 3, whichever is greater | ≥ fck − 3 |
| M20 and above | ≥ fck + 0.825 × established SD, or fck + 4, whichever is greater | ≥ fck − 4 |
- Until a standard deviation is established from site results, use the assumed values in IS 456 Table 8.
- For small pours of up to 30 m³, where fewer than four samples are taken, the mean of all samples must be at least fck + 4 N/mm² (M20 and above), and each individual result at least fck − 4.
Example: an 18 m³ M25 slab needs 3 samples (9 cubes). Mean of the samples ≥ 29 N/mm², each sample ≥ 21 N/mm².
When Results Fail
Don’t panic and don’t hide it. The usual sequence is to check the sampling and testing records, take cores from the member (IS 456 clause 17.4), and if needed do a load test or a structural review for strengthening. The cube register must record which member every sample came from. Without that, a failed result can’t be traced to a member.
Reference
- IS 456:2000, clauses 15, 16, 17 and Tables 8, 11 (with amendments)
- IS 516 (Part 1/Sec 1):2021, Testing of strength of hardened concrete (formerly IS 516:1959)
Summary for site use. Always work from the current edition of the standard and your project specification. Where they differ, the structural engineer’s specification governs.