Ernst van der Stok, Jeroen Weller, Frans Scholten
Papers # 2016 Berlin
This paper describes falling-weight tests as a first step towards determining the effect of residual stresses on the fracture behaviour of old PVC-U gas pipes at impact. Results showed that the level of surrounding pipe support, the quality of the material and the diameter of the striker had a major effect on the extent of deformation and the failure mode (brittle or ductile). Due to the large scatter in the data, it could not be concluded that residual stresses had an effect on the fracture behaviour of wellsupported pipes (simulating fine and well-compacted sandy soil).
Third-party damage is the largest cause of the failure of PVC-U pipes. Residual stresses in old PVC-U gas and water pipes were found to vary between 2 MPa and 8 MPa. The effect of these stresses on the impact resistance of PVC pipes is unknown. The resistance of PVC pipes to external blows can be tested in accordance with ISO 3127. In practice, however, multiple external factors, including soil compaction,
This study determined the so-called brittle-ductile transition temperature (TBD) of PVC-U pipes by varying the temperature because failure is generally brittle in cold pipe segments and ductile in warm segments. The TBD was used as a quality parameter to measure the effect of residual stresses on the impact behaviour. Pipe segments, surrounded by a tightly placed jacket pipe to simulate the soil, were impacted in the experiments using a striker.
The results showed that the level of surrounding pipe support has a major effect on the extent of deformation and the failure mechanism. Extensive deformation of the PVC-U pipe prevented failure in some cases but subsequent failure in a brittle, fragmented way also occurred. Where extensive deformation was prevented, only local failure (brittle or ductile) in the pipe segments occurred.
The quality of the material and the diameter of the striker also affected the failure mode (brittle or ductile) of the PVC-U pipe segment. The results demonstrated the difficulties associated with applying laboratory results to practice.
The excavated pipes (110 mm) had a residual stress of 3.5 - 3.8 MPa (measured using the Janson method), which is moderate. Some of these pipe segments were heated for 100 h at 60 °C to reduce the residual stress to 0.3 - 0.6 MPa.
The results showed that residual stresses – in view of the large scatter in the data – had no negative effect on the fracture behaviour of well-supported pipes (simulating fine and well compacted sand soil).