Sensors for monitoring early stage fatigue cracking

Abstract

Three sensor systems were evaluated for their ability to detect early stage fatigue cracking in open holes. The sensor systems were (1) a Meandering Winding Magnetometer Array sensor system that induces eddy currents to monitor conductivity changes; (2) a through-transmission ultrasonic technique that monitors energy loss, and (3) the Electrochemical Fatigue Sensor which detects fatigue-induced changes in a metal surface through their effects on the electrochemical double layer. The sensors were mounted individually or in tandem in samples of 7075-T651 aluminum alloy which were then fatigued using a spectrum loading sequence. The samples were also examined at various stages during fatigue cycling using optical or scanning electron microscopy. Detection thresholds of approximately 100 μm were observed, and calibration curves for crack size in terms of the sensor outputs were obtained.

Introduction

Interest in early stage fatigue has continued to increase with each new generation of observational techniques and with the recognition that short fatigue cracks do not always follow the same laws as long cracks [1]. Likewise, the importance of fatigue cracking to the structural safety of aircraft has become an increasingly serious preoccupation of the world’s aircraft fleet managers, as evidenced by the yearly conferences devoted to this issue (e.g. [2], [3]). As part of this interest, the sensing of fatigue cracks for use as a nondestructive inspection technique and for potential use in health monitoring systems is also the subject of much current research [4], [5]. Piotrowski et al. [4] recently tested 20 different sensor systems for their ability to detect subsurface cracks in lap joint structures. They chose a threshold value of 3.8 mm at which to compare the techniques, and found that the probability of detection was “quite low” at this value [4].

In the Safe Life approach to aircraft fatigue, the US Navy considers that a primary aircraft structure with a crack of 250 μm has reached the end of its fatigue life [6]. Thus, the Navy threshold is more than an order of magnitude smaller than the threshold used in the Piotrowski study. This comparison is not strictly valid because the Piotrowski study examined cracks in the subsurface layer of lap joints, which are not usually of primary interest in naval aircraft. However, the comparison serves to highlight the gap between current NDI techniques and the need to detect smaller cracks.

For naval aircraft, fastener holes are often prime locations for fatigue crack initiation, and some fatigue life tracking is focused on the surveillance of several fastener hole “hot spots.” Inspection for 250 μm cracks in fastener holes, particularly holes with the fastener in place, is not possible with current technology. However, next generation health monitoring systems, e.g., “prognosis” approaches, will require detailed information about fatigue cracking at this length scale and below.

In an attempt to address the need for early stage fatigue crack detection at fastener holes, a series of laboratory experiments was undertaken to evaluate the state-of-the-art of currently available sensor systems for cracks in holes. Recent reviews have catalogued most of the available techniques for early stage fatigue crack detection, [7], [8], [9] and several of the most applicable and promising of them were selected for evaluation in this study. It was decided to restrict the work to laboratory testing of open-hole 7075-T651 aluminum coupons that were designed to be representative of current US Navy interest. The idea was to provide an optimum environment for sensor operation in order to establish the current lower limit for crack detection. The sensor systems were also chosen with an eye toward their potential for eventual use in a fielded system. The three chosen techniques were based on measurements of eddy currents, ultrasonic energy and electrochemical behavior.