Fatigue life of a component is composed of two stages: crack initiation and crack propagation. Fracture mechanics is the method used to predict crack growth life. It is particularly useful for components which contain defects or a crack. The main applications of fracture mechanics based fatigue analysis include determination of the maximum tolerable initial flaw size (often involving welds) for design life, calculation of crack growth life with a known or assumed crack size and planning of inspection intervals. To predict crack growth accurately using this method, it is essential to conduct tests to obtain the fatigue crack growth rate (FCGR) of the material under appropriate environmental conditions.
FCGR is often expressed as a function of stress intensity factor range, ΔK, in terms of the Paris power law:
da/dN = A ΔKm [1]
where da/dN is crack growth per cycle, ΔK is a function of stress range, crack size and specimen geometry, A and m are material constants which need to be determined by carrying out fatigue crack growth tests.
There are two well recognised standards for conducting FCGR tests: ASTM E647 (1) and ISO 12108 (2). They are largely in accord, but there are some small differences between the two. Examples of such differences include the specification of the parameter C, which is the magnitude of load drop during a decreasing ΔK FCGR test (-0.08mm-1 in ASTM E647 while -0.1mm-1 in ISO 12108) and the definition for the crack growth threshold, ΔKth: 10-7mm/cycle in ASTM E647 while 10-8mm/cycle in ISO 12108.
FCGR tests require use of standard specimen geometries which include compact tension (CT), single edge notch bend (SENB), single edge notch tension (SENT) and centre crack tension (CCT) specimens. Both ASTM E647 and ISO 12108 accept these specimen geometries and provide the K solutions. Figure 1 shows the geometries of SENB and CT specimens which are more often used than the other two. Compared to the SENB specimen, the CT specimen has the advantage of being more economical in material which can be important when the material sample is limited. However, machining of such a specimen is more expensive. Another advantage with the SENB specimen geometry is that it is easier to set up a test in a corrosive environmental chamber.
Standards (1, 2) specify the size requirements for specimen, notch and pre-crack. Notch is often produced by electrical discharge machining (EDM) and pre-crack is introduced by cyclic loading. It is important to ensure that the last maximum K (Kmax) in pre-cracking is less than the Kmax used at the start of a FCGR test. To achieve this, pre-cracking is often conducted at a stress ratio lower than that used in the subsequent FCGR test.