Concrete Failure Analysis Reveals Multiple Causes
Swimming pool owners noticed that their concrete pool, which had been repaired two years earlier, was once again leaking and cracks were reappearing. Findings of the original investigation concluded that a chemical reaction was responsible for the leaks and cracks but that the reaction had mitigated when the pool was subsequently repaired. The puzzling aspect was that the new cracks were appearing in the same location as where recommended repairs were made. The pool manufacturer wanted to know if the cracks resulted from a construction defect or the concrete used for the repair. They retained RJ Lee Group to identify the root cause of the concrete failure.
We performed petrographic examinations of the concrete samples using reflective light microscopy, transmitted light microscopy and scanning electron microscopy (SEM) with energy dispersive spectrometry (EDS). These analyses confirmed that the cause of the original cracking, alkali silica reaction (ASR), had been mitigated and so was unlikely to have caused the current cracking. However, the analyses did reveal entrapped air voids. This is air that becomes entrapped in the concrete during mixing and when found in large numbers, generally lowers the strength and durability of the concrete.
Visual observations confirmed that the cracks on the concrete floor penetrated through the new plaster and shotcrete because epoxy that had been used to fill the earlier cracks was found in the new cracks and voids. Sample cores showed large variations in mix design and poor uniformity in aggregate. This non-uniformity created excess space that contractors filled in with cement paste which shrank as it cured causing stress.
We found the most intensive cracking on the repaired deep-end walls, pool bench and floor where the concrete walls were thicker. These thicker walls, formed by multiple applications of shotcrete, caused “rebound” and resulted in concrete layers with non-uniform aggregate distribution. Since the aggregate in the concrete of the shallow end of the pool had a more uniform distribution and the walls were thinner, we determined that the deviation in pool design in the deep end of the pool led to stress in the thicker concrete.
The combination of visual observation and analytical techniques in the concrete failure analyses provided our experts with the evidence they needed to weigh each observation’s contribution to the root cause failure. We determined that there was no one single cause for the cracking. Rather a series of errors by both the contractor and the contractor supplier led to the crack development. Although rebound from shotcrete can alter mix constituents and result in a change in mechanical properties, the variations we saw were too great to be attributed solely to rebound. We concluded that the major contributors to the construction defect were a deviation from the pool design evidenced by massive amounts of concrete applied at the deep end of the pool, improper preparation of the concrete mix that resulted in air voids, and a concrete mix design that included poor aggregate uniformity.