Fatigue phenomenon consists of two parts, fatigue initiation and crack propagation until failure. When using optical devices, definition of initiation depends on optical resolution of the equipment. In our tests, we detect fatigue initiation by acoustic emission which is more sensitive than optical method. Acoustic emission is registered during test by two sensors. Fatigue initiation is easily detected by an energy burst. This method is also reliable and gives supplementary information on fatigue mechanism. Stress range versus number of cycles to initiation is fitted by a power law similar to Basquin’s one. (9) where ’i is fatigue initiation resistance and an exponent. is the gross stress range. Figure 5: Evolution of ratio Ni/Nr versus number of cycles to failure for X52 steel. Table 8: Fatigue initiation resistance parameters with and without hydrogen charging for X 52and X70 steels Steel, environment ’i fatigue initiation resistance (MPa) Exponent R2 X70 air 395 -0.016 0.96 X70 Hydrogen 368 -0.010 0.97 X52 air 325 -0.018 0.90 X52 Hydrogen 296 -0.011 0.95 Evolution of ratio Ni/Nr versus number of cycles to failure is plotted in Figure 6. We note that fatigue crack propagation is faster in presence of hydrogen because the difference Nr-Ni is strongly reduced. Barsom and McNicol [11], Jack and Price [12], Clark [13] and Truchon [14] have used the parameter K to express fatigue resistance on notched specimens. Boukharouba et Al [15] have compare fatigue initiation criteria with a criterion based on the elasto plastic notch stress intensity factor K on welded specimen made in low strength steel. The same parameter has been used to plot a fatigue initiation curve K (= f(Ni). One notes that this curve are not sensitive to pipe steels but sensitive to environment. 0,40 0,45 0,50 0,55 0,60 0,65 0,70 0,75 0,80 0,85 0,90 10000 100000 1000000 Number of cycle to failure Ratio Ni/Nr Without hydrogen With hydrogen absorption i i N '
RkJQdWJsaXNoZXIy MjM0NDE=