Use of the Direct-Current Electric Potential Method to Monitor Large Amounts of Crack Growth in Highly Ductile Metals

Abstract

The direct-current electric potential method is receiving increasing attention for monitoring crack extension in J-resistance curve testing. Among its advantages over the unloading-compliance method are: (1) no time-consuming unloadings are required, (2) a continuous record of crack extension versus displacement can be obtained, and (3) the method can be used at higher strain rates where unloading compliance cannot be used. Despite the advantages of the direct-current electric potential method, questions persist regarding its ability to monitor large amounts of crack growth in highly ductile materials where large displacements and large amounts of plastic strain occur. This paper presents details of an experiment conducted on a 3T planform-size compact specimen of 25.4 mm thickness to assess the ability of the direct-current electric potential method to accurately measure crack extension in a highly ductile material. The material selected was Type 304 austenitic stainless steel. It was found that the Johnson expression, often used to calculate crack extension from direct-current electric potential data, significantly underestimated the actual amount of crack extension. However, a simple modification of the Johnson expression resulted in excellent agreement between calculated and measured crack extensions.

Numerical investigation of compliance equations used in the R‐curve testing for clamped SE(T) specimens

“…For example, the values of |e 2 | of Eqs (6a) and (7) are 10.4% and 9.4%, respectively, for a/W = 0.1 and B/W = 4. It is interesting to note that the trend in |e 1 | associated with Eqs (8) and (9) is opposite to that in |e 1 | associated with Eqs (5a)- (7), in that for a given a/W, |e 1 | associated with Eqs (8) and (9) (11), (12a), (12b) and (12c) lead to much more accurate predictions of C with values of |e 2 | generally being smaller than 10%. For a given a/W, the values of |e 2 | associated with Eq.…”

“…The results indicate that using ρ (a/W) and ρ (BCE) combined with Eqs (5a)-(9) and (11) lead to highly accurate predictions of a/W and C for both the PS and SG specimens with all the a/W and B/W ratios considered. The corresponding errors of predictions are less than 2% for most of the specimens, and the maximum errors of prediction error associated with Eqs (5a), (6a), (6b), (7), (8), (9) and (11) are 3.7, 4.8, 4.2, 4.2, 6.3, 4.6 and 4.2%, respectively. It is noted that the highest errors of prediction for Eqs (5a)- (9) and (11) are corresponding to the SG specimens with a/W < 0.2, which can be attributed to the relatively poor fitting of the data by Eqs (18) and (22) corresponding to large ω and small γ values (i.e.…”

“…(7) Eq. (8) Eq where f(BCE) is the right hand side of Eqs (5a), (6a), (6b), (7), (8) or (9). The modification factor can also be interpreted as the model error associated with f(BCE).…”

“…C decreases). Figure 4 indicates that ρ (a/ W) generally increases as ω increases for Eqs (5a)- (7), which implies that these 2D-based equations become less and less adequate under the 3D condition as the specimen thickness increases and/or the crack length decreases. The value of ρ (a/W) is relatively insensitive to ω for Eqs (8) and (9).…”

“…): where P is the applied load, and C can be directly measured if the UC method is adopted. For test methods (such as the potential drop method) that directly measure the crack length instead of the compliance, the following equation is needed to evaluate C in Eq. based on the measured crack length: where g (•) represents the function that relates BCE 2D to a / W .…”