Assessment of weld reduction factors through experimental reference curves

Abstract

The 2¼Cr1Mo (P22) is present in most of structural high-temperature applications, as results from a review on materials employed for the construction of both power and chemical plants. So, it fits well for an experimental activity in the laboratory as reference material. ISPESL, the Italian Certification Agency, carried out a low-budget program of creep and low-cycle fatigue (LCF) tests at 550 °C; the specimens (twelve plus two spare) of as-fabricated material came from a thick-wall pipe with a certified circumferential weld. They were base metal and cross-weld fusion-line centered ones. Comparison of the resistance curves obtained from LCF tests with those from codes (ASME and ISPESL) showed for the base metal good correspondence; therefore, showed the reliability of the test proceeding. For the weld it showed instead conservatism reduction at high-strain levels for the ASME design curve; moreover, it revealed an insufficient conservatism of the reduction factor proposed, which is 1/2, for the lower bound ISPESL curve. Comparison of the resistance curves obtained from creep tests with those from codes showed this: for both the base metal and weld there are a good correspondence and acceptable (low) conservatism of the strength reduction factor (SRF) proposed; for the load and temperature levels considered in the tests, it equals one. Still, the creep tests duration was within thirty hundred hours each; therefore the levels of the applied load (150.2–100 MPa) were not as low as those in typical operation conditions. For these cases, the code (ASME) proposes a lower-than-unity SRF: thus, longer tests (lower levels of the applied load) would be useful, also extending the activity to other materials.