Development and Influence of Erosion Voids at Rigid Pipe Joints
MetadataShow full item record
An experiment was designed and conducted to study the effect that erosion voids forming at leaky reinforced concrete pipe joints have on the pipe response to surface loading. The experiment consisted of a jointed system of two 825 mm inner diameter reinforced concrete pipes, buried in a uniform sand. A section of the pipe joint was left unsealed to allow infiltration once the water table was raised to the ground surface. Circumferential strain profiles on the inner and outer wall of the pipe were measured using nylon fibre optics. The strains were then used to calculate the moment and thrust response of the pipe during service load testing. Each stage of the erosion void growth experiment had an initial surge of sand ingress, which then slowed to a low energy flow that carried little to no eroded soil into the pipe. The total dry mass of eroded soil collected through the erosion void growth was approximately 33 kg, which had a volume of around 20 L. The final void geometry was thin, with a thickness ranging from 5 to 20 mm, and widely spread along the pipe system, with maximum vertical and horizontal extents of 0.95 m and 1.12 m, respectively. Gel-encased fibre optics along the outer pipe wall along with a heat source were used to attempt to track the extent of the void contact along the pipe during erosion, but results were inconclusive. Similar to prior analyses and testing conducted with artificial erosion voids, an increase in moment occurred due to the soil support lost in the location of the erosion void. When the service loads were applied over the joint of the pipe system, higher responses to load were recorded in the bell pipe than the spigot pipe. This may be a function of the relative stiffness of the bell compared to the spigot, as well as additional lost bedding support lowering soil pressure on the spigot pipe. More tensile thrusts were also measured in the spigot pipe during post-erosion void load testing. Polymer filler injected into the void was not observed to affect pipe response.