SAMPLING AND CONDITION ASSESSMENT OF DUCTILE IRON PIPES

About 300 ft (91.4 m) of DI pipes were exhumed in each of four North American water utilities in an effort to gain a thorough understanding of the geometry of external corrosion pits in ductile iron (DI) pipes, which would lead to a better ability to assess the remaining life of these pipes. The exhumed pipes were cut into sections, sandblasted and tagged. Soil samples extracted along the exhumed pipe were also obtained. Pipe sections were scanned, using a laser scanner that was specially developed at the National Research Council of Canada (NRC) for this purpose and the scanned data were processed using special software developed for this purpose. The pipes were virtually sliced into rings of equal lengths, where each ring was characterized by three geometrical attributes, namely maximum pit depth, pit area and pit volume. Statistical analyses were performed on the geometrical attributes of the corrosion pits found on these rings. Soil characteristics were investigated for their impact on the geometric properties of the corrosion pits and were found not to have a substantial and consistent impact. Based on the results of the statistical investigation, methods were proposed to discern the condition of a ductile iron pipe based on a set of random samples. In this paper we describe the development of these methods, including the sampling scheme, the probabilistic inference on the pipe condition and the confidence bounds for the discerned results.