Codes & standards referenced
Why this matters
The “30” in M30 is not a promise that every 28-day strength result will land on exactly 30 N/mm². It is the characteristic compressive strength, fck: the strength below which not more than 5% of valid test results are expected to fall. Concrete varies because materials, moisture, batching, transport, compaction, curing and testing vary. A grade therefore describes a controlled population, not a single cube.
For proportioning, the producer aims above the grade at the target mean strength fck + 1.65σ. The 1.65σ margin places the specified characteristic strength near the lower 5% point of a normal distribution when the process mean and established standard deviation, σ, represent production. That is why a capable M30 process should centre above 30, while an occasional valid result can lie slightly below 30 without automatically proving that the concrete in the structure is unacceptable.
Specification walkthrough
IS 456 Cl 16.1 treats a sample test result as the average strength of its constituent specimens, subject to the validity check in Cl 15.4. The acceptance decision then uses Table 11 under Cl 16.1 and Cl 16.3. It has two simultaneous gates: the mean of a group of four non-overlapping consecutive sample results and the minimum permitted individual sample result.
For the four-result mean, compare fck + 0.825σ, rounded as stated, against fck + 3 N/mm² and use the greater value. Separately, no individual result may be below fck − 3 N/mm². Passing the mean cannot rescue an individual result below its limit; a strong individual result cannot rescue a deficient group mean.
Note 1 prevents a borrowed or guessed σ from becoming permanent: use Table 8 only until enough project data exist, then establish the actual standard deviation. Note 2 is not an informal relaxation. It is the explicit acceptance regime for concrete quantities up to 30 m³ when fewer than four samples arise from the Cl 15.2.2 frequency. For one sample, the required result is fck + 4 N/mm²; do not apply the ordinary fck − 3 individual limit to that lone sample.
How to apply
Build a valid sample before judging compliance
Under IS 456 Cl 15, one sample consists of three specimens tested at 28 days. The sample result is their average. If an individual specimen varies by more than ±15% from the sample average, investigate the test and treat the result in accordance with the clause rather than quietly discarding the inconvenient cube. Where both 7-day and 28-day information is required, cast six cubes—three for each test age—but Table 11 compliance uses the valid 28-day sample result, not a selected set of the best three from six.
Keep the sequence intact
Enter every valid sample chronologically in the cube register. Form groups of four non-overlapping consecutive results: samples 1–4, then 5–8, then 9–12. Calculate the applicable mean threshold from 0.825σ or +3, whichever is greater, and check all four individual results against fck − 3. Do not cherry-pick four favourable dates or slide past a low result.
Apply Note 2 before forcing a four-result rule
For a quantity up to 30 m³, first determine the number of samples required by Cl 15.2.2. If fewer than four are required, use Note 2: all available sample results average at least fck + 4 and each is at least fck − 2. If only one sample is required, that sample itself must reach fck + 4.
When one sample fails, preserve evidence and investigate
A low result is a trigger, not permission to substitute spare cubes. Quarantine the records; verify identification, casting, curing, transport, machine calibration and calculations; map the sample to the member and batch tickets; and notify the engineer responsible. IS 456 Cl 16.3 sends concrete that does not meet the strength requirement to Cl 17 for further evaluation. The engineer may order core testing or load testing; this is not an automatic “retest the same concrete” option.
Use the code's structural investigation route
For core evaluation under IS 456 Cl 17, take at least three representative cores. Concrete represented by the cores is considered acceptable when the average equivalent cube strength is at least 85% of the specified grade and no individual core is below 75%, subject to the engineer's selection of locations and interpretation. Core extraction, preparation and testing must follow the applicable IS 516:2022 provisions. If evidence remains inadequate, the engineer decides whether load testing, strengthening, restricted use or removal is required.
What goes wrong
Six cubes may have been cast for two ages, but that does not create permission to report the strongest three at 28 days. Maintain specimen identity and average the three designated 28-day specimens for that sample.
A high group mean can hide one seriously low result. Table 11 requires the mean gate and the individual gate. Both must pass; neither substitutes for the other.
Small pours do not wait indefinitely for four results. Up to 30 m³, apply the explicit +4 and −2 limits—or +4 for one sample—instead of inventing a four-sample group.
A fourth error is treating any cube below fck as automatic demolition. That ignores the statistical definition and the investigation route. The opposite error—declaring acceptance because the overall project average looks healthy—is equally serious. Compliance is attached to the defined sequence, limits and represented concrete, not to a convenient dashboard average.
References & further reading
- IS 456:2000 — Plain and Reinforced Concrete — Code of Practice (Fourth Revision), Bureau of Indian Standards. Cl 15 sampling and specimen-result validity; Cl 16.1 and 16.3 acceptance; Cl 17 evaluation of results and core-test criteria; Page 30, Table 11 substitute cited above.
- IS 516:2022 — Hardened Concrete — Methods of Test, Bureau of Indian Standards. Applicable parts cover compressive-strength specimens and core extraction, preparation and testing.
- User-provided Table 11 source image — screenshot supplied 03/07/2026, reproducing the Page 30 substitute headed “Amendments No. 1 and 3”. The citation header, three columns, data row and both notes above follow that source.
Use the controlled copies of the standards and the project specification for contractual decisions. The project specification may impose requirements more stringent than the code minimum.
Field-checklist blocks
A short questionnaire, an inspector's checklist, and three worked examples that show how the two gates, the small-pour rule and the specimen-validity check resolve into an accepted group, a rejected group or a core-testing instruction.
Questionnaire — ask before a sample result is entered as compliant
- Is this a valid sample: three specimens tested at 28 days, with the sample result taken as their average, and no specimen varying more than ± 15% from that average?
- Does the established standard deviation exist, or are we still on the Table 8 assumption with fewer than 30 project results behind us?
- Have we formed the group as four non-overlapping consecutive results (samples 1–4, then 5–8, then 9–12) out of the chronological register?
- Which of the two mean branches governs, fck + 0.825σ rounded to the nearest 0.5 or fck + 3? The crossover sits at σ = 3 ÷ 0.825 = 3.64 N/mm², so any established σ above that makes the σ branch the binding one.
- Does the small-pour rule apply here? Is the quantity up to 30 m³, and does the sampling frequency give fewer than four samples, or only one?
- If six cubes were cast for two ages, are we judging the valid 28-day sample result rather than the best three of six?
- Who has been told about this result, and has the engineer responsible been notified in writing?
Inspector's checklist — what to verify on site
- Specimen identity: the three specimens of the sample, their test age of 28 days, and the sample average written down with the ± 15% check against each specimen.
- Register order intact: every valid sample entered chronologically, with the groups of four taken as they fall, no reshuffling to improve a result.
- Mean threshold recorded: the 0.825σ term computed and rounded to the nearest 0.5 N/mm², compared against fck + 3, and the greater of the two written into the register as the limit.
- Individual gate checked on all four results against fck − 3, each one, and not just the lowest.
- Note 2 applied before the four-result rule where the quantity is up to 30 m³: fck + 4 on the mean of all available samples and fck − 2 on each, or fck + 4 on a lone sample.
- Two-age casting accounted for: where six cubes were cast, the three designated 28-day specimens are the ones averaged for Table 11.
- Failure record: records quarantined, identification, casting, curing, transport, machine calibration and calculations all verified, the sample mapped to the member and the batch tickets, and the engineer notified.
- Core route ready: at least three representative cores, with the 85% and 75% thresholds understood, core extraction and testing carried out per IS 516:2022, and the engineer's instruction obtained.
What happens if… three worked examples
What happens. The group mean is (33.0 + 32.5 + 34.0 + 30.5) ÷ 4 = 130.0 ÷ 4 = 32.5 N/mm². Take the fck + 3 branch: 30 + 3 = 33 N/mm², so the mean falls 0.5 N/mm² short and the group fails. Every individual result clears fck − 3 = 27 N/mm², so the individual gate passes. If the established σ is above 3.64 N/mm², the other branch takes over and the shortfall grows.
Likely outcome. The group does not comply. IS 456 Cl 16.3 sends concrete that does not meet the strength requirement to Cl 17 for evaluation, and the engineer may order core testing or load testing. This is not an automatic retest of the same concrete, and spare cubes do not get substituted for the failing group.
Preventive correction. Write both branches into the register every time, so the threshold that governed is visible without reconstructing it six months later.
What happens. The quantity is up to 30 m³ and the sampling frequency gives fewer than four samples, so the small-pour rule governs: the mean of all samples must be at least fck + 4 = 34 N/mm², and each result at least fck − 2 = 28 N/mm². The mean here is (35.0 + 34.5 + 33.0) ÷ 3 = 102.5 ÷ 3 = 34.17 N/mm², above 34, and the lowest individual result of 33.0 is well clear of 28. The pour passes.
Likely outcome. Accepted, and it is worth being clear about why. Note 2 is the explicit acceptance regime for small quantities, not an informal relaxation of the ordinary rule, and applying the four-result gate instead would have produced no valid group at all for a pour this size. This pour's samples also stay with this pour; they do not get carried forward to make up four results with the next pour's samples.
Preventive correction. Settle the pour quantity and the sample count at the pour card, so the regime is known before the results arrive.
What happens. The sample average is (36.0 + 34.5 + 27.5) ÷ 3 = 98.0 ÷ 3 = 32.67 N/mm². The 27.5 specimen sits (32.67 − 27.5) ÷ 32.67 = 15.8% away from that average, outside the ± 15% limit, so the result is not quietly accepted or quietly discarded. It goes to the validity check in IS 456 Cl 15.4, and the test and the calculation get investigated. If the sample survives as a valid result of 32.67, it is below the fck + 3 = 33 N/mm² branch and the group fails.
Likely outcome. IS 456 Cl 16.3 sends it to Cl 17, where the engineer may order at least three representative cores. The thresholds there are derived from the grade: 85% of M30 is 25.5 N/mm² for the average equivalent cube strength and 75% is 22.5 N/mm² for any individual core. Note that a single result below fck is not demolition, and an overall project average that looks healthy is not acceptance either.
Preventive correction. Check the specimen spread at the point of calculation, while the moulds, the capping and the machine are still part of the conversation.
Frequently asked questions
M30 identifies the characteristic compressive strength, fck: the strength below which not more than 5% of valid test results are expected to fall. It is not a promise that every 28-day strength result will land on exactly 30 N/mm² — concrete varies because materials, moisture, batching, transport, compaction, curing and testing vary.
The mean of the group of 4 non-overlapping consecutive test results must be ≥ fck + 0.825 × established standard deviation (rounded off to nearest 0.5 N/mm²) or fck + 3 N/mm², whichever is greater.
Every individual test result must be ≥ fck − 3 N/mm². Passing the mean cannot rescue an individual result below its limit; a strong individual result cannot rescue a deficient group mean.
Note 2 applies: for concrete of quantity up to 30 m³ (where the number of samples to be taken is less than four as per the frequency of sampling given in 15.2.2), the mean of test results of all such samples shall be fck + 4 N/mm², minimum and the requirement of minimum individual test results shall be fck − 2 N/mm², minimum. However, when the number of sample is only one as per 15.2.2, the requirement shall be fck + 4 N/mm², minimum.
A low result is a trigger, not permission to substitute spare cubes. Quarantine the records; verify identification, casting, curing, transport, machine calibration and calculations; map the sample to the member and batch tickets; and notify the engineer responsible. IS 456 Cl 16.3 sends concrete that does not meet the strength requirement to Cl 17 for further evaluation. The engineer may order core testing or load testing; this is not an automatic “retest the same concrete” option.
No. A high group mean can hide one seriously low result. Table 11 requires the mean gate and the individual gate. Both must pass; neither substitutes for the other.
Related articles
IS 456 — 15 Clauses Every Site Engineer Must Know
The working clauses behind workmanship, durability, reinforcement and acceptance.
Cube Testing — IS 516
Sampling, moulding, curing and compression testing without losing specimen identity.
IS 10262 Mix Design — Worked M30 Example
How target mean strength and production acceptance fit into one quality loop.
Amit Haridas
Founder & Proprietor, ConcreteInfo. 25+ years of experience in concrete technology, RMC plant operations, construction quality, consulting and technical training across India.
Technical updates, not sales emails
One email when a new article goes out on concrete technology, QA/QC and construction quality. No promotions, and your address is never sold or passed to advertising networks. Subscribing is separate from making an enquiry.