Thin Bondline Measurement of Adhesively Bonded Metallic Aircraft Structures using an Ultrasonic Analyzer

Abstract

The use of adhesive bonding for the assembly of aircraft structure is desirable because of the associated benefits of reduced weight, decreased parts count and increased fatigue life. The inability to fully characterize the nature of the bondline has been a limiting factor in applying this technology to the manufacture of primary aircraft structure. One such obstacle is the resolution limitation that has existed for the non-destructive measurement of bondline thicknesses. Up to the present time, ultrasonic methods have been limited in practice to a resolution approximately equal to the wavelength of the energy pulse employed. For a 15 MHz transducer, the highest frequency generally incorporated into off-the-shelf inspection equipment, this equates to a lower measurement limit of about 0.006″. For the FM-73 adhesive system used in this investigation, the acoustic velocity is 90, 600 inches per second [1]. The lower resolution limit will vary somewhat as a function of this speed. This limitation presented a problem because previous research had shown that the lap shear strength of aluminum bonded structure was maximized in bondlines between 0.002″ and 0.006″[2]. Furthermore, for bondlines of less than 0.002″, the lap shear strengths dropped off dramatically as illustrated in figure 1. The data presented is for elevated service temperature for an aircraft application. The trend shown is equally as valid for room temperature and at -67 degrees Fahrenheit. Clearly then, to guarantee the manufacture of a high strength, minimum weight structure while still assuring that a sufficient quantity of adhesive was present, it was necessary to establish a non-destructive inspection process, usable in production, that could quickly and accurately discern bondlines in the 0.002–0.006″ range.KeywordsAcoustic VelocityAircraft StructureBondline ThicknessIncrease Fatigue LifeElevated Service TemperatureThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.