The purpose of this investigation was to gain a better understanding of drying shrinkage of slabs on ground. There were four components to this work:
- background research;
- a site investigation at the Capral Aluminium Warehouse site in Dinmore, in conjunction with Readymix;
- a laboratory investigation at the University of Queensland concrete laboratory, in conjunction with Readymix; and
- the application of codified shrinkage models and their comparison with laboratory results.
The background research involved the consideration of concrete shrinkage. It was seen that good design and construction practices were the most effective way of controlling shrinkage.
The site investigation was carried out at the Capral site and involved documentation of the placement of seven slabs, as well as the general monitoring of other slabs. This site was chosen as there was a high chance that significant cracking would occur, due to its location and the time of year during which pouring began. However, significant cracking did not occur, and may be attributed to the successful design, specification, supply, and construction of the concrete.
The laboratory investigation involved the testing of three standard Readymix concretes. Mix A was the standard Capral Aluminium mix, Mix B was a high shrinkage mix, and Mix C was a low shrinkage mix. Standard specimen tests were carried out at the University of Queensland. The placement and monitoring of a slab and free shrinkage box for each mix was undertaken in attempt to model in situ slabs, and to better understand the effect of restraint on the shrinkage of members. Mix C was disregarded from analysis as test results were inconsistent with design specifications. The effect of base restraint on shrinkage of slabs was clearly identified when comparing the slabs to the free shrinkage boxes. Restraint of the slabs was seen to hide some of the early age shrinkage effects that were observed in the free shrinkage boxes.
Three codified models for shrinkage were considered: AS 3600-2001, AS 3600 revised, and ACI. The variations between the parameters of each model were discussed, and their comparative performance analysed. It was seen that the ACI model performed best, followed by the AS 3600 revised, followed by the AS 3600-2001.