Testing and Quality Control of Concrete (Part One)


Concrete is unique among construction materials in that it requires various degrees of processing between arriving on site and being placed, compacted, finished and cured. This is true whether the concrete is site batched or arrives ready mixed. It is therefore necessary to sample the concrete and conduct concrete testing to ensure that it complies with the requirements of the project specification.

The South African Bureau of Standards (SABS) publishes the relevant standard methods, among which are:

  • SANS 5861-2: Concrete testing – Sampling of freshly mixed concrete
  • SANS 5861-3: Concrete testing – making and curing of test specimens
  • SANS 5862-1: Concrete testing – Consistence of freshly mixed concrete (slump test)
  • SANS 5863: Concrete testing – Compressive strength of hardened concrete
  • SANS 5865: Concrete testing – The drilling, preparation, and testing for compressive strength of cores taken from hardened concrete

In practice the most commonly specified tests are the “slump test” and the “cube test”.

Concrete Quality Control

No test can be valid unless the sample is representative of the concrete to be tested. If there is any deviation from the sampling method described in SANS 5861-2, doubt is cast upon the subsequent test results. In other words if the sampling is suspect, don’t even waste time making and testing the specimens.

Specified sampling frequency varies but is commonly one sample per 50m3 of concrete, or part thereof, per day for each grade of concrete being produced.

The “Slump Test”

This test measures the consistence (wetness) of the concrete in the medium “workability” range. (Workability is a qualitative term describing the amount of work required to place and compact fresh concrete).

The test is used to check that the water content of the concrete does not deviate significantly from the design value. The allowable deviation from the specified slump is plus or minus 25 mm or 1/3 of the specified slump - whichever is the greater.

Large variability in slump test results, indicate one or more of:

  • Poor control of amount of water added
  • Unacceptable variations in aggregate batching
  • Change in aggregate water requirement
  • Poor testing procedure

Making and Curing of Test Specimens

The three crucial factors in making and curing the specimens (normally cubes) are:

  • The condition of the cube moulds - steel moulds are supplied in matched sets of pieces and the sides and bases from different moulds must not be mixed
  • The compaction of the concrete – a concrete cube of 150 mm side should weigh approximately 8.10 kg, while a 100 mm cube should weigh approximately 2.40 kg
  • The curing temperature – this should be between 22 and 25ºC

It is important that the cubes are correctly labelled. Experience has shown the best method is to stick a paper label onto the top surface of the cube while the concrete is still plastic. The cube details are then written on the side of the cube with black lumber crayon when the cube is de-moulded. Scratching the cube number on the top surface of the fresh concrete is not recommended.

The “Cube Test”

This test measures the uniaxial compressive strength of concrete cubes which are made, cured and tested to very specific requirements. It does not measure or predict in any unique way the strength of the concrete in the structure. The test is simply a quality control test which measures the consistency of the concrete in terms of one particular property (“compressive strength”) using an arbitrary test method (SANS 5863).

Testing the same concrete under different conditions, for example specimen size, specimen shape, curing temperature, loading rate, etc. will give different results.

From a quality control point of view, the importance of the cube test result is not just the value of any individual result, but the variability in a series of valid test results.

What Is A Valid Compressive Strength Test Result?

A valid test result is the mean of the results of tests carried out on three specimens which are sampled from the same batch of concrete, and which are made and cured under standard conditions at any particular age (for example, 3 or 7 or 28 days). This is with the condition that the range of strengths between the highest and lowest individual result does not exceed 15% of the mean.

An invalid result, or a series of them, indicates one or more problems with the testing procedure. An invalid result is rejected from a quality control viewpoint and the cause of the invalidity must be investigated.

Typical causes include poor sampling and cube making, mis-labelling of cubes, out of tolerance cube moulds, operator error, and malfunctioning compression machine.

Compression Machine Reference Testing

It is essential that compression test machines undergo regular reference testing at about 3 monthly intervals. The fact that a machine is calibrated does not imply that it is producing reliable results.

The way to carry out reference testing is to make and cure a large batch of cubes and to circulate sets to participating laboratories for testing. The results are then compared. The testing is normally done at three different strength levels, but not necessarily all at the same time.

Tip 7, to follow, will cover acceptance criteria and analysis of strength results.