Environmental Fatigue Crack Propagation Of Aluminum Alloys At Low Stress Intensity Levels

The environmental fatigue crack propagation in 2024-T3, 7075-T6, and 7178-T6 was studied at low levels of cyclic amplitude of stress intensity, delta K. Both wedge force loading and remote loading techniques were employed to achieve the desired delta K levels, and preliminary experiments were designed to test their compatibility. Testing was carried out in humid air, distilled water, and 3.5% sodium chloride solution, and the observed crack growth rates compared with those in desiccated air. Later studies were also conducted in an inert reference environment with a total water content of less than 2 ppm. When the data are plotted as log delta K versus log d2a/dN, alloy 2024-T3 exhibits a marked slope transition, alloy 7075-T6 a slight slope transition, and alloy 7178-T6 a rectilinear behavior throughout the whole range of delta K studied. The basic shape of these curves is discussed in terms of state-of-stress conditions at the crack tip, frequency effects, environmental effects, strain rate sensitivity, and metallurgical structure. An attempt is also made to correlate the rate of fatigue crack propagation in a particular environment and at a particular delta K level with the fracture topography. (Author).

The effects of environment on fatigue behavior are quite intimately related to such test variables as stress intensity, cyclic frequency, and temperature. The effect of these variables has been shown to be quite complex. However, a consistent pattern of behavior is beginning to emerge which, it is hoped, will lead to a considerably changed understanding of the corrosion-fatigue process. It is considered probable that an increased understanding of the corrosion-fatigue process will also contribute measureably to an improved understanding of general fatigue processes. Material variables such as composition, directionality of properties, and microstructure, have important effects on corrosion-fatigue behavior of aluminum alloys. However, it appears that these variables, at least in gaseous environments, may be important largely because of their effects on the mechanical behavior of aluminum alloys. On the other hand, studies in seawater suggest that these variables become increasingly important as the corrosive nature of the environment increases. This memorandum discusses in detail the effects of both test variables and material variables on corrosion fatigue. In addition, current theories of corrosion-fatigue mechanisms are outlined, and several methods of preventing corrosion fatigue are suggested. (Author).

A series of fatigue tests with specimens subjected to constant amplitude and two-step axial loads were conducted on 12-inch-wide sheet specimens of 2024-T3 and 7075-T6 aluminum alloy to study the effects of a change in stress level on fatigue-crack propagation. Comparison of the results of the tests in which the specimens were tested at first a high and then a low stress level with those of the constant-stress-amplitude tests indicated that crack propagation was generally delayed after the transition to the lower stress level. In the tests in which the specimens were tested at first a low and then a high stress level, crack propagation continued at the expected rate after the change in stress levels.

Results are presented of a series of fatigue tests to study crack propagation and the resulting stress distributions in tension panels. The panels were all of the same general design, and configurations varied mainly in the relative amount of cross-sectional area in the skin, stiffeners, and flanges. The panels were constructed of 2024-T3 and 7075-T6 aluminum alloys. It was found that the average rate of crack growth was slower in panels made of 2024-T3 aluminum alloy than in panels made of 7075-T6 aluminum alloy. All cracks initiated in the skin, and the slowest crack growth was measured in configurations where the highest percentage of cross-sectional area was in the stiffeners. Strain-gage surveys were made to determine the redistribution of stress as the crack grew across the panels. As a crack approached a given point in the skin, the stress at that point increased rapidly. The stress in the stiffeners also increased as the crack approached the stiffeners. During the propagation of the crack the stress was not distributed uniformly in the remaining area.

The delaying effect of a single overload on the 2024 T351 aluminum alloy covering a wide ?K range has been studied in ambient air, secondary vacuum, and high purity nitrogen (N2). It is shown that for a given environment and an overload ratio, the delay behavior can be rationalized in terms of the peak stress intensity factor corresponding to the overload for the studied test conditions. At low ?K values, the overload effect is confined to a grain. The results obtained show that while the size of the overload affected zone is not affected by the environment, the number of delay cycles at an R ratio of 0.5 in vacuum can be ten times as high as that obtained in the two other environments. At an R ratio of 0.1, at low ?K levels, the delaying effects can be similar in vacuum and in air. The observed effects are explained on the basis of mechanisms governing constant-amplitude crack growth behavior of the studied alloy.