Predicting and testing the tensile relaxation of concrete

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Tensile relaxation is an important property in structural concrete members such as concrete overlays and patch repairs under sustained imposed restraint. Tensile relaxation helps in reducing tensile stresses in concrete which leads to mitigation of cracking in restrained concrete members. Normally, it is assumed that tensile relaxation is influenced in a similar way by the same parameters that influence creep despite potential differences between the two. Moreover, relaxation is predicted from functions that link it to creep. Whereas it is easy to find published literature on creep of concrete, little can be found on the relaxation behaviour of concrete. More research is therefore needed to understand the factors that influence relaxation and its prediction. This study aimed at comparing the influence of selected factors on creep and relaxation to establish a correlation if any between the two. The parameters investigated include: w/c ratio, age of loading, initial stress-strength ratio and aggregate content. Tensile relaxation tests are difficult to perform and hence prediction models are often relied upon to predict relaxation. Two simplified approximate methods are suggested in the fib Model Code to obtain the relaxation function from the creep function. The Age Adjusted Effective Modulus method (AAEM) given in the MC2010 and the Approximate Relaxation Function (ARF) in MC90-93 are mentioned. Values from the two were compared with experimental relaxation results to verify their accuracy in predicting the relaxation potential of selected concretes. Results show that tensile relaxation of concrete, similar to creep is affected by parameters such as w/c ratio, age of loading and aggregate content. It is observed that the magnitude of tensile relaxation, with other factors held constant, was found to reduce with a reduction in w/c ratio, increase in age and increase in aggregate content. The initial stress-strength ratio does not seem to have a significant influence on relaxation. However, an increase in the initial stress leads to a small and un- proportional increase in relaxation. The use of models linking the creep function and relaxation function to predict low-age relaxation is probably valid. Although both the AAEM and ARF methods yield fairly good predictions of relaxation, the simplicity of the AAEM makes it a better option.

Includes abstract.

Includes bibliographical references.