M25 Cube Test Result 23 MPa: Pass or Fail? IS 456 Explained

A 28-day concrete cube report arrives. The grade is M25. One result shows 23 N/mm². The site team says the concrete has failed. The contractor disagrees. Who is right?

The important question is not only whether 23 is lower than 25. It is whether the result is a valid concrete sample test result, which acceptance criteria apply, and what the other consecutive results show. One number can raise a concern without being the complete acceptance decision.

Free tool: Calculate the IS 456 acceptance checks yourself

Want to check three-cube specimen consistency and sample-group calculations? Try the free concrete cube calculator from GTSM. It includes worked M25 examples and explains the engineering limitations.

First, understand what the three different numbers mean

1. One cube specimen: the compressive strength of a single tested concrete cube.

2. One sample test result: ordinarily the average strength of three cube specimens cast from the same sample, subject to the variation rule in IS 456, Clause 15.4.

3. A group of four results: four consecutive, non-overlapping valid sample test results used together for the standard acceptance check where the four-result criterion applies.

A report stating that one individual cube crushed at 23 MPa does not automatically mean that the sample test result was 23 MPa. And a sample test result of 23 MPa does not automatically tell you whether the batch or represented concrete satisfies the code.

A worked example: Can 23 MPa be acceptable for M25 concrete?

Take an illustrative site with four valid, consecutive 28-day sample test results of the same M25 concrete grade. Each result below is already the average of three cubes, after specimen validity checks.

Sample 1 = 23 N/mm²; Sample 2 = 30 N/mm²; Sample 3 = 31 N/mm²; Sample 4 = 32 N/mm².

These are deliberately simplified teaching values, not actual GTSM test results and not permission to combine unrelated pours or select convenient samples.

Check 1 — Are the three cubes in each sample valid?

Under IS 456:2000, Clause 15.4, the sample test result is the average of three cube strengths. No individual specimen should differ by more than 15% from that average; otherwise the sample result is invalid under that clause.

For example, cube strengths of 22.8, 23.0 and 23.2 N/mm² average 23.0. They are mutually consistent. But a set with widely differing results needs investigation; taking a convenient average without checking the required variation is not acceptable.

Also verify specimen identification, dimensions, sampling, casting, curing, age at test, machine calibration, loading rate and reporting procedure under the applicable test standard.

Check 2 — Does the group of four satisfy Table 11?

For the general four-result case, IS 456:2000 Table 11 (as amended by Amendment No. 4) requires the mean of four non-overlapping consecutive sample test results to be at least the greater of: (a) fck + 0.825 × established standard deviation, with the prescribed rounding; or (b) fck + 3 N/mm².

Here, characteristic strength fck = 25 N/mm² (M25). If no established site standard deviation exists, assume s = 4 N/mm² for M25 as indicated by IS 456 Table 8, subject to the project's applicable requirements.

0.825 × 4 = 3.3 N/mm². Rounded to the nearest 0.5 N/mm², this becomes 3.5. Therefore, the first group-mean threshold = 25 + 3.5 = 28.5 N/mm². The alternative threshold is 25 + 3 = 28.0 N/mm². The governing threshold is 28.5 N/mm².

Actual group mean = (23 + 30 + 31 + 32) / 4 = 29.0 N/mm². This passes the group-mean check because 29.0 is at least 28.5.

Check 3 — Does every individual sample result satisfy the minimum?

For the standard four-result Table 11 case, each individual sample test result must be at least fck − 3 N/mm². For M25, that is 25 − 3 = 22 N/mm².

Our lowest sample result is 23 N/mm². It passes this individual-result threshold. The other sample results also pass.

So, in this narrowly defined hypothetical example, the four results satisfy both Table 11 checks. A sample result of 23 MPa can therefore occur in a compliant four-result sequence. This is not a blanket declaration that every M25 result of 23 passes.

Two counterexamples show why the full calculation matters

Case A: valid sample results 22, 29, 30 and 32 N/mm². Their group mean is only 28.25 N/mm²: below the required 28.5. The group-mean criterion is not satisfied, even though no individual sample is below 22.

Case B: valid sample results 21, 32, 32 and 32 N/mm². Their group mean is 29.25 N/mm², which is above 28.5. But the 21 MPa sample falls below the minimum individual-result threshold of 22. The individual criterion is not satisfied.

You must meet BOTH checks. A high average cannot rescue a result below the individual minimum, and apparently acceptable individual results cannot rescue a low group average.

Important exception: Small pours do not use the same rule

Table 11 includes a specific note for concrete quantities up to 30 m³ where fewer than four samples are required under the sampling-frequency rules. For these cases, the mean of the available sample test results must be at least fck + 4 N/mm², with the specified minimum individual result of fck − 2 N/mm². Where only one sample is prescribed, that single sample must meet fck + 4 N/mm².

For M25, this means a group-mean minimum of 29 N/mm² for the applicable small-pour exception, and a minimum individual result of 23 N/mm² where more than one sample is available; for a one-sample case, the sample must be at least 29 N/mm².

Therefore, never copy an example involving four test results and apply it directly to a small single-sample pour. First identify the actual quantity, required sampling frequency, grade, and contractual quality requirements.

How many samples are normally required?

Under the nominal sampling schedule in IS 456, Clause 15.2.2: 1–5 m³: 1 sample; 6–15 m³: 2 samples; 16–30 m³: 3 samples; 31–50 m³: 4 samples; above 50 m³: four plus an additional sample for every further 50 m³ or part thereof. Additional sampling may be required by project specifications, site conditions and quality-control arrangements.

One sample for the 28-day strength result normally means three test specimens. Do not confuse three cubes from one sample with three independent samples.

What should an engineer do when results really are low?

Step 1 — Preserve the records: concrete grade, RMC batch details, sampling location, casting time, cube numbers, curing history and test dates.

Step 2 — Validate the testing: check specimen dimensions and condition, the three-cube variation, the machine and test procedure, and whether the report uses N/mm² correctly.

Step 3 — Apply the correct acceptance provisions to the complete relevant set of results. Do not select or rearrange results to make the average look better.

Step 4 — Inform the responsible site, QA/QC and structural-engineering team. Treat a genuine nonconformity as requiring investigation, not as an automatic verdict on the safety of the entire structure.

Step 5 — When justified, undertake further assessment under the applicable provisions of IS 456 and IS 516, potentially including carefully specified core testing and complementary site investigations. The responsible engineer must decide the appropriate investigation and remedial action.

Rebound hammer and ultrasonic pulse velocity tests can provide useful supporting information, but their readings should not simply be substituted for properly interpreted cube or core strength values.

Five mistakes in 28-day cube test discussions

Mistake 1: Calling a single cube specimen value the complete test result for a sample.

Mistake 2: Declaring a whole concrete grade rejected because one valid sample average is below fck, without checking applicable acceptance criteria.

Mistake 3: Looking only at the four-result average and ignoring the individual-result limit.

Mistake 4: Applying the four-result rule to a small pour where the Table 11 note governs.

Mistake 5: Assuming that meeting a cube-strength acceptance test automatically proves the constructed member is safe, properly placed, cured, detailed and free of defects.

Save this checklist before deciding pass or fail

□ Confirm the specified grade and the age of test.

□ Identify whether the number is one cube strength or the average of three specimens.

□ Check individual specimen variation and sample validity.

□ Confirm the concrete quantity and sampling schedule.

□ For the relevant rule, check the group mean AND the individual-result limit.

□ Establish the correct standard deviation or documented assumed value.

□ Check the current edition, amendments and project specifications.

□ Escalate nonconforming results for formal engineering assessment instead of making an unsupported site-level safety declaration.

Indian Standards and reference documents

IS 456:2000 — Plain and Reinforced Concrete, Clauses 15, 16 and 17, including applicable amendments. BIS Standards Catalogue

IS 456:2000 — Amendment No. 6 (2024), official BIS publication. Amendment No. 4 is relevant to the Table 11 acceptance criteria used in the example. BIS Amendment No. 6 PDF

IS 516 (Part 1/Sec 1):2021 — Hardened Concrete: Strength Testing. Official BIS standard preview

IS 1199 (Part 1):2018 — Sampling of Fresh Concrete. IS 1199 (Part 5):2018 — Making and Curing of Test Specimens. IS 516 (Part 4):2018 — Sampling, Preparing and Testing of Concrete Cores.

For an introduction to specimen preparation and cube testing, see: Why Concrete Cube Testing Matters — GTSM

The takeaway

A cube strength value is test evidence. Concrete acceptance is a documented engineering decision made under the applicable code and project requirements. Numbers should be interpreted, not merely compared with the grade name.

GTSM Testing Labs LLP, Ahmedabad, provides construction-material testing, concrete cube testing and selected structural and geotechnical testing services for project quality assurance. For test requirements, share your specified grade, sample identification, age at test and location.

Learn about GTSM's testing services: GTSM Testing Labs | Phone: +91 98240 49466

Disclaimer: All numerical examples are fictional and for education only. Confirm the standard and its applicable amendments, actual data validity, and contractual requirements before applying acceptance criteria. This post is not certification of a particular batch or a structural safety assessment.