10.033 Concrete Shrinkage

Q Our project specifications indicate that a specific class of concrete mix shall not exceed a shrinkage limit of 0.040% at 21 days drying. The concrete supplier does not have lab test data to support the performance of the mix, but has some field test results from a recent project. Should we use this information to accept the mix?

— San Francisco Structural Engineer

A 8/07 – Response prepared by William Wahbeh, a responsible engineer at Signet Testing Laboratories and a registered engineer in California

The typical standard used to reference shrinkage limits for a project is ASTM C157 “Length Change of Hardened Hydraulic-Cement Mortar and Concrete”. This test method is a very sensitive laboratory test based on specific criteria for mixing, sampling, curing and measuring. Some criteria may vary, such as storage, allowing for either water or air, which can have significant influence on the test results. SEAONC developed a modified procedure to C157 in the 1960’s commonly referenced in project specifications that make several changes that include sample size (4x4x11 vs. 3x3x11), initial curing (7 days vs. 28 days wet cure) and air drying (50% RH). It is this modified procedure that most project specification limits are based. As with interpreting and analyzing any test results, it is critical to make sure you are comparing apples to apples.

Although every project would like to limit shrinkage to the least possible amount, it is important that limits be specified only when necessary and if the proper quality control, including laboratory testing, can be established. Local materials or mix proportions may not be able to meet shrinkage requirements without the addition of costly admixtures that can affect other properties of the mix. Lab test values should be used as a basis to determine the acceptability of materials and proportions and should not be used categorically. Furthermore, as with concrete compression testing, the results from shrinkage testing are not necessarily representative of the performance of the mix in-place because of the complexity of the factors that influence shrinkage. Similarly, field cast shrinkage samples are typically found to be greater than lab cast samples. Some specifications allow for 15% to 25% higher tolerances, while the SEAONC “Supplementary Recommendations for Control of Shrinkage in Concrete” gives maximum ranges for differenct classes of concrete at 21 days drying for lab from .036 to .060, while field cast specimens are in the range of .048 to .080.

Given the factors noted above, field-testing data can give some indication of the quality of the shrinkage characteristics, however reliance on this information for material acceptance should be avoided. The bottom line is that, in the absence of a new ASTM for field-testing or modified specifications, there is no substitute for laboratory trial batching to determine the shrinkage limits of a specific mix.