Dominique GUEUGNAUT, Adil BOUJLAL, Aymeric LOPITAUX, Romuald BOUAFFRE
Papers # 2016 Berlin
A specific methodology has been designed by ENGIE’ R&T Division to allow an accurate determination of the initiation (incubation) time of the NCBT data. This methodology is based on two different data processing procedures. The first data processing proceeds from searching the minimum of the second derivative of both a 3rd order and a 6th order polynomial fitting of the curve. Such approach is not well suited to very flat curves or to unsteady curves on the one hand, and needs to define a low limit of the strain for data processing, in order not to emphasize the curvature at the beginning of the test on the other hand. The second data processing proceeds from searching the minimum of the strain rate all along the test. Such an approach is pertinent and does not require any low strain limit to process the data. Moreover, processing the data in real time allows one to stop the test a short time after the minimum strain rate has been detected. This preferred procedure gives way to an accurate determination of the initiation (incubation) time, hence reducing drastically the NCB Test duration for a quick evaluation of a PE grade.
In the last five years some new and innovative tests have been designed to assess for Slow Crack Growth (SCG) behavior of PE100 and PE100 RC with the objective of reducing drastically the test time duration. Three tests have been focused on by the WG20 of the ISO TC138-SC5 committee. These tests are the Strain Hardening Test (SHT), the Cyclic Crack Round Bar Test (CCRBT) and the Notched Cylindrical Bars under constant load Test (NCBT). These three tests allow to assess for SCG resistance of PE100 and PE100 RC in a reasonable test duration less than 1,000h with the major advantage to avoid the utilization of any surfactant suspected of modifying the SCG process. The two former tests were the subject of an intensive evaluation through round robin tests managed by the WG20 and finally became ISO standards in 2015. The NCBT was first evaluated in 2013on PE100 and PE100 RC by ENGIE’R&T Division and then proposed to the WG20. The promising results presented during the Plastic Pipes XVII opened the way to the standardization process and to a CD document in 2015. One of the advantages of the NCBT is the possibility to detect in a very accurate way the crack initiation (or incubation) time and thus to shorten the test time duration in a very effective proportion. The crack initiation time corresponds to the “trigger” for crack to progess and can be modelled by means of an elasto-visco-plastic theoretical approach as presented formerly as, for instance, during the Plastic Pipes XVI. A first method was designed by R&T Division in order to detect the initiation time. This method, based on the fitting of the NCBT curve by means of polynomial laws, required approximately 2/3 of the up-to-failure curve and was very sensitive to the flatness of the curve. In order to avoid this, a refined method is proposed which focuses on the strain rate. The method was successfully applied on four PE100 and five PE100 RC. Then the “live” detection of the minimum strain rate allows a very accurate determination of the crack initiation time and a drastic test time reduction for a quick evaluation of a PE grade.Furthermore, the true stress applied to the specimen can be calculated – for comparison purposes with the target one -once the initiation time has been determined by provoking the failure of the material in liquid nitrogen.