Open Access
Subscription or Fee Access
On the Quantification of Adhesive Layer Thickness Effect on PWAS Performance and Damage Detection Capability
Abstract
monitoring of aircraft and civil infrastructures, a more accurate knowledge of the system is required to increase prognosis reliability. The adhesive bonding layer between a piezoelectric wafer active sensor (PWAS) and the host structure is often pointed out as a critical factor. Several studies have shown behavioral differences with bond thickness and stiffness, from the so-called “shear-lag†effect to the change in PWAS dynamics. Issues such as the role of adhesive layer on the electromechanical impedance of the PWAS; the identification of bonding defects; or the influence of layer thickness at elevated temperatures on Lamb wave response have been extensively studied. However, there are questions that remain unclear, such as its importance for damage detection capability. The main goal of this work is to search for possible correlations among the layer’s thickness, the PWAS admittance and the response to Lamb waves and, additionally, to evaluate the damage detection capability of PWAS bonded with different thicknesses. On the assumption that a high adhesive shear modulus and a thin bond thickness would improve the performance of the sensor/actuator system, the epoxy DP490 (3M) was selected to glue a matrix of PZT transducers to an aluminum plate in columns of 50, 100 and 200μm adhesive thickness layers. Four techniques were used to analyze the system as illustrates the Figure. Firstly, the impedance and admittance of each sensor were measured. Then, a round-robin test, using one PWAS at a time as actuator, was performed in the pristine and damage structure. After that, the complex aluminum-adhesive-PWAS was encapsulated and cut to conclude the actual thickness of the adhesive layer. Finally, a finite element model was implemented for a better understanding of the physical problem. Understanding and relating the effect of bonding on the electromechanical measurements and Lamb wave signals would allow adjusting and controlling the signal processing for damage detection in a wide range of scenarios. Likewise, knowing the frequency dispersion caused by a specific adhesive thickness would help to determine the time of flight, and therefore, the damage localization with higher precision.
DOI
10.12783/shm2019/32306
10.12783/shm2019/32306