Fracture mechanics testing techniques are typically utilized for evaluation of the effects of metallurgical or environmental variables on EAC where the specimen contains a sharp crack. One of the most common and relatively simple techniques for incorporation of fracture mechanics techniques for the evaluation of EAC is through the use of constant load or constant deflection specimens. In the case of constant load specimens, a load is applied to a fracture mechanics specimen using a directly applied dead weight or through a pulley or lever system to magnify the dead weight load. These methods are analogous to those used for constant load tensile specimens discussed previously.

The most common types of specimens utilized for evaluation of EAC are the compact tension (CT), single-edge notched bend (SENB), or double-cantilever beam (DCB) varieties (see Figs. 7-9). Normally, they are fatigue pre-cracked prior to exposure to the environment to produce a sharp crack tip. The fatigue pre-cracking must be performed at a low enough stress intensity to minimize the plastic zone ahead of the crack. This is usually accomplished through load-shedding techniques whereby an initially high peak load is used to initiate the fatigue precrack and the cyclic load is then decreased as the precrack approaches its desired length. In the testing of precracked specimen, it has been found that excessive initial stress intensity of precracked specimens can produce a large plastic zone that can act as a barrier to EAC initiation. In these cases, the effect will be to produce nonconservative data.

The stress intensity at the tip of the crack can be calculated using standard equations as given in ASTM E399 for CT and SENB specimens and in NACE TM0177, Method D for the DCB specimen. As shown for the DCB specimen, side grooves can be utilized to assist in keeping the crack growing in a planar fashion under plane strain conditions. In some cases the crack will tend to grow out of plane resulting in an invalid test. The important consequence of using side grooves is that the equation for the CT or DCB must contain a correction factor that accounts for the geometry and dimension of the side grooves.

Dead weight-loaded specimens are often used to monitor time to crack initiation and can be used to monitor crack growth rate vs. stress intensity K. This is normally performed by measurement of crack opening displacement, which can be related to crack length for a particular specimen geometry using compliance techniques. In the latter case, however, provisions must be made to monitoring crack opening displacement at a rapid rate because the crack growth rate will tend to increase with increasing K as the crack proceeds through the specimen.

An attractive alternative to dead weight-loaded specimens is constant deflection specimens. In this situation, either CT or DCB specimens are loaded to an initial level of crack tip stress intensity by deflection of the arms of the specimen. This deflection is obtained either by inserting the wedge into the specimen or by tightening a bolt arrangement that deflects the arms of the specimen. The initial stress intensity must be above the threshold stress for EAC which allows cracking to initiate. Once cracking initiates, it proceeds while the stress intensity decreases as the crack propagates through the specimen. Thus, this type of test is commonly referred to as a decreasing K test and is extensively utilized for evaluation of EAC in its various forms. Once the stress intensity at the crack tip reaches a value insufficient to sustain crack growth, crack growth will stop. These conditions of load and crack length can be used to define the threshold stress intensity using the appropriate equations for either the CT or DCB specimens.

Sometimes the period required to run a decreasing K test is very long. An alternative is to use a rising load test whereby the fracture mechanics specimen is subjected to an increasing load. In this case, the crack open displacement and load are monitored simullaneously and the results are analyzed in a similar manner to conventional fracture mechanics tests. One of the difficulties in the interpretation of rising load tests is that the threshold stress intensity obtained by this method often differs from that determined by decreasing K tests. The dynamic strain rate in the rising load test can complicate the interpretation of the test result particularly if HE or HEC are the prevalent cracking mechanisms. In these cases, a loading rate is usually selected that results in data that correlate with the results of service experience in terms of resistance to EAC.