Thresholds of fatigue crack growth rates are important characteristics for fatigue crack growth assessment for the integrity of structural components. ASME Code Section XI provides fatigue crack growth thresholds for ferritic steels in air and water environments. The threshold is given as a constant value under a negative stress ratio. However, the thresholds are not clearly defined in the range of negative stress ratios. The definition seems to be maximum stress intensity factors. Besides, the thresholds expressed by the maximum stress intensity factors decrease with decreasing stress ratios. This means that the thresholds under negative stress ratios become unconservative assessments. The objective of this paper is to discuss the definition of fatigue crack growth threshold and to propose the threshold equation for the ASME Code Section XI, based on experimental data obtained from a literature survey.

Fully plastic failure stresses for circumferentially surface-cracked pipes subjected to tensile loading can be estimated by means of limit load criteria based on the net-section stress approach. Limit load criteria of the first type (labeled LLC-1) were derived from the balance of uniaxial forces. Limit load criteria of the second type are given in Section XI of the ASME (American Society of Mechanical Engineering) Code, and were derived from the balance of bending moment and axial force. These are labeled LLC-2. Fully plastic failure stresses estimated by using LLC-1 and LLC-2 were compared. The stresses estimated by LLC-1 are always larger than those estimated by LLC-2. From the literature survey of experimental data, failure stresses obtained by both types of LLC were compared with the experimental data. It can be stated that failure stresses calculated by LLC-1 are better than those calculated by LLC-2 for shallow cracks. On the contrary, for deep cracks, LLC-2 predictions of failure stresses are fairly close to the experimental data. Furthermore, allowable circumferential crack sizes obtained by LLC-1 were compared with the sizes given in Section XI of the ASME Code. The allowable crack sizes obtained by LLC-1 are larger than those obtained by LLC-2. It can be stated that the allowable crack size for tensile stress depends on the condition of constraint of the pipe, and the allowable cracks given in Section XI of the ASME Code are conservative.

A subsurface flaw located near a component surface is transformed to a surface flaw in accordance with a flaw-to-surface proximity rule. The recharacterization process from subsurface to surface flaw is adopted in all fitness-for-service (FFS) codes. However, the criteria of the recharacterizations are different among the FFS codes. In addition, the proximity factors in the rules are generally defined by constant values, irrespective of flaw aspect ratios. This paper describes the stress intensity factor interaction between the subsurface flaw and component free surface and proposes a proximity factor from the point of view of fatigue crack growth rates.

When discrete multiple flaws are in the same plane, and they are close to each other, it can be determined whether they are combined or standalone in accordance with combination rules provided by fitness-for-service (FFS) codes. However, specific criteria of the rules are different among these FFS codes. On the other hand, plastic collapse bending stresses for stainless steel pipes with two circumferential similar flaws were obtained by experiments, and the prediction procedure for collapse stresses for pipes with two similar flaws was developed analytically. Using the experimental data and the analytical procedure, plastic collapse stresses for pipes with two similar flaws are compared with the stresses in compliance with the flaw combination criteria. It is shown that the calculated plastic collapse stresses based on the flaw combination criteria are significantly different from the experimental and analytical stresses.