Identification of Distributed Fatigue Cracking by Dynamic Crack-Closure

Abstract

Early fatigue damage is typically a distributed phenomenon affecting the structure at many locations. Multiple-site crack initiation and growth occur at numerous points of stress concentration, e.g., at rivet holes in a wing panel. Coalescence of such relatively large cracks can lead to sudden fracture, and may occur at loads significantly below what would be expected from considering the fracture resistance of single cracks [1,2]. Under laboratory conditions, a great variety of NDE techniques are available for early fatigue damage detection and characterization. These techniques include acoustic emission, linear and nonlinear ultrasonics, vibration analysis, eddy current inspection, magnetics, thermal imaging, X-ray, etc. Unfortunately, most of these methods cannot be directly adapted to infield inspection of aging aircraft. One reason for this is that the most sensitive ultrasonic and eddy current methods are essentially short-range, localized measurements. On a large airframe structure, their use is limited to the inspection of a few suspected areas of high stress concentration, but they tend to loose sensitivity very rapidly when applied to distributed and multiple-site cracking. Large-scale, overall inspection has inevitably lower sensitivity to distributed multiple-site cracking, especially in the presence of artifacts causing false alarms. Strong artifact signals are caused by uncertainties in the inspection procedure (e.g., coupling in ultrasonic testing or lift-off in eddy current testing), inherent geometrical features (e.g., nearby holes and edges), and additional irregularities (e.g., uneven machining, mechanical wear, corrosion, etc.). There seems to be a need for specialized long-range ultrasonic and eddy current techniques which do a better job at distinguishing real fatigue cracks from inherent structural and material variations. These techniques need to be fine-tuned to characteristic features exhibited by fatigue cracks and only fatigue cracks. First, we shall demonstrate that crack-closure under external deformation can be exploited to achieve significant improvements in the threshold sensitivity of long-range ultrasonic and eddy current inspection methods. Second, we shall demonstrate the feasibility of producing dynamic crack-closure by localized cooling with a commercial freezing spray.