Fatigue tests were performed on specimens machined from an angle iron of ~7 mm in thickness, which was detracted from a transmission tower. The material has been in service for more than 20 years. The location of the tower was in central Bohemia in a rural countryside.
The chemical composition of Atmofix 52A steel Two types of flat specimens for fatigue tests machined
according to Fig. 1 were used. The first type was machined in such a way that the corrosion layer covered the two frontal sides of the specimen gauge length having 10 mm width. The remaining two sides were milled. The sharp edges of the gauge length after machining were made round with a radius of 0.5 mm. The thickness of the specimen was determined by the thickness of the angle iron. The cross-section was about 70 mm2. specimen type was machined in such a way that the corroded layer was milled away. Thus, the second type of specimen was characteristic for the base material. Its thickness was about 5.5 mm. An example of the specimen with corroded layer is shown . The specimen heads were slightly reground by reason of good fastening in the grips of the fatigue machine.
Fatigue tests were performed in load control on a resonant fatigue machine Schenck PVQO at frequency of about 40 Hz. Tests with the highest stress amplitudes were conducted on a servohydraulic testing machine Zwick/Roell Amsler MC25 at a frequency of 10 Hz or 3 Hz with the aim to prevent the heating of specimens.
Fatigue fracture surfaces were observed in a scanning electron microscope (SEM) JEOL 6460. The standard
metallographic observation of material structure and corrosion layer was performed by means of standard light microscopy.
The experimentally determined S-N data for loading with the stress ratio R = –1 (R is defined as the ratio of the minimum to maximum stress in a loading cycle) are shown. Arrows in the figure indicate run-out specimens, i. e. specimens with fatigue life higher than 107 cycles. The fatigue life of specimens, which have two sides of the gauge length with a corrosion layer, is substantially lower than that of base material. The full lines describe the best fit of all data corresponding to the failed specimens. The fatigue limit, depicted by the horizontal dashed line, was determined as the stress amplitude at which three
specimens remained unbroken after application of 107 cycles. S-N curves describing the fatigue strength under cyclic loading with the stress ratio R = 0 are shown . Again, the detrimental effect of the corrosion layer is obvious.