At present, PE pipe systems are designed on the basis of experimental regression curves of hoop stress versus time to failure, generated from laboratory hydrostatic pressure tests on whole pipe sections. Such tests, although satisfactory for determining the performance of the parent pipe, are inadequate for assessing the long-term performance of butt fusion welds.
Butt fusion welds are vulnerable to stresses in the axial direction, which may well be generated in service due to bending and thermal contraction. However, in a laboratory hydrostatic pressure test, the stress in the axial direction is only half of that in the hoop direction Therefore, in these tests, failure invariably occurs in the pipe wall before the welded joint, giving no quantifiable data on the long-term integrity of the joint.
A Group Sponsored Project has recently been completed at TWI to develop a new test. This subjects welded lengths of pipe to a constant axial stress at elevated temperature. Results have shown that this test does indeed generate failure in the weld and therefore allows the long-term integrity of the butt fusion joints themselves to be determined.
Results from this work have also suggested that the condition of the externally debeaded surface may have lead to the initiation of slow crack growth through the weld interface.
Based on the results of this initial study, TWI will carry out further work to identify how possible flaws are introduced either during the assembly process or subsequent post fusion operations.
Results have also shown that the inclusion of talc contamination into butt fusion welds in PE100 pipes has a significant effect on the long-term performance of the joint. However, the degrees of contamination studied were also detected during the standard QA examination of the removed external weld bead.
Further work in this area is being carried out as part of a European funded CRAFT project, managed by TWI.
For more information, please contact us.