Experimental and numerical investigation of crack closure measurements with electrical potential drop technique Michael Andersson a, * , Christer Persson a , Solveig Melin b a Division of Materials Engineering, Lund University, Box 118, 221 00 Lund, Sweden b Division of Mechanics, Lund University, Box 118, 221 00 Lund, Sweden Received 21 December 2004; received in revised form 29 August 2005; accepted 23 November 2005 Available online 5 January 2006 Abstract Crack closure measurements during fatigue crack propagation using the electrical potential drop (PD) technique is evaluated using a combination of experiments and simulations. From finite element simulations it is shown that variations in the potential drop value over a fatigue cycle can be explained by crack closure and strain induced conductivity changes. It is also demonstrated that closure measurements made with PD are consistent with closure observations made from in situ SEM observations of crack closure. Finally, it was observed that closure measurements made by a compliance-based measuring technique gave a somewhat lower opening load as compared to the PD-measurement. q 2006 Elsevier Ltd. All rights reserved. Keywords: Crack closure; Potential drop; In situ SEM 1. Introduction The structural integrity of many applications rely on accurate predictions of fatigue lives. This in turn requires fatigue models that are capable of capturing the relevant phenomena occurring. One such phenomenon, which has been found to play a major role in fatigue crack propagation, is crack closure. Crack closure means that the crack remains locally closed at the tip even though a global opening load is applied. Crack closure may be the result of several different mechanisms, such as plastic deformation, crack surface asperities or oxides forming on the surfaces. For a more detailed discussion on different closure mechanisms see for instance [1]. In the late 1960s Elber, see [2,3], discovered that the compliance of his test specimens changed during the load cycle. He interpreted this change as a change in crack length due to a gradual opening of a closed crack and used the compliance curve to measure the amount of closure. To account for crack closure an effective stress intensity range, DK eff , was introduced and defined as DK eff Z K max KK op (1) where K max denotes stress intensity at maximum load and K op stress intensity at crack opening. Elber also found that the load ratio effect during fatigue crack propagation could be eliminated by correlating DK eff against crack propagation rate per cycle. Thus, the load ratio effect could be explained by crack closure. A number of different techniques are used for crack closure measurements, for a detailed review cf. [4,5]. The most common group is the compliance based techniques, used by for instance Elber and described in ASTM-E647 [6]. The principle behind these techniques is that the strain or displacement at some point of the body is measured as a function of applied load. If plasticity is confined to a small region at the crack tip the only non-linearity in the load–strain curve stems from crack closure, and the point of crack opening can be detected as the point after which the compliance remains constant. Although such techniques are often used, it has been found that different placements of the strain or displacement gauge may result in different levels of crack opening, cf. [7]. Also, in some situations, such as after an overload, the compliance based closure measuring techniques have been found to be unreliable, see [8]. Finally, if the material is subjected to high loads with a large amount of plasticity, it is not possible to distinguish closure from inelastic deformation. International Journal of Fatigue 28 (2006) 1059–1068 www.elsevier.com/locate/ijfatigue 0142-1123/$ - see front matter q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijfatigue.2005.11.005 * Corresponding author. Tel.: C46 46 222 7991; fax: C46 46 222 4620. E-mail address: michael.andersson@material.lth.se (M. Andersson).