Concrete Strength Acceptance Calculator
Switches between statistical and non-statistical criteria automatically.
Inputs
Result
Formula
| Statistical · mean | mfcu ≥ fcu,k + λ1·Sfcu |
| Statistical · minimum | fcu,min ≥ λ2·fcu,k |
| λ1 / λ2 | 10-14: 1.15 / 0.90; 15-24: 1.05 / 0.85; ≥25: 0.95 / 0.85 |
| Non-statistical | mfcu ≥ 1.15·fcu,k (1.10 for ≥C60), fcu,min ≥ 0.95·fcu,k |
C30, 10 specimens, mean 35.0, standard deviation 1.5, minimum 33.0: statistical method with λ1 = 1.15 and λ2 = 0.90. Mean must be at least 30 + 1.15×1.5 = 31.73 and is 35.0; minimum must be at least 27.0 and is 33.0 - accepted.
FAQ
Why switch to a stricter method below 10 specimens?
A standard deviation from a handful of results is unreliable and usually understates scatter, which would let bad batches pass. The code therefore drops the deviation and compensates with a higher mean requirement (1.15×). It is also why important structures require at least one set per 100 m³ or per shift - sample size is itself evidence.
Can a batch pass if the mean is fine but the minimum is not?
No - both conditions must hold. A low minimum points to a weak spot, perhaps poor curing or a faulty specimen. Test the actual structure (rebound hammer or cores) to find out which, then follow the GB 50204 procedure. Making new specimens is not an acceptable answer.
Which data should the standard deviation come from?
All specimens in the same batch - never a selected subset. Where a grade has several batches, the code allows pooling three or more consecutive batches, but they must share grade, mix, production process and age. Cherry-picking is the most common form of manipulation and the first thing auditors check.
Related calculators
Results are estimating aids only and are not a design basis. Construction and final accounts must follow the drawings, the applicable standards and an approved method statement. Anything affecting structural safety, shoring or falsework must be checked by a suitably qualified engineer.