The non–destructive testing (NDT) of adhesively bonded joints is a specific issue of its own. A typical joint consists of steel or, in some cases, aluminum sheets with thicknesses in the range of 0.7–2 mm. During the manufacturing process, adhesives or sealants are typically applied between these sheets prior to the formation of complex joints by means of spot welds, rivets or clinch flanges.
In order to ensure sufficient joint strength, the nominal thickness of this adhesive layer should be approximately 0.1 – 0.5 mm. Naturally, during the formation process, large forces are applied to these metals, resulting in the deformation of the mating parts. This gives rise to large-scale variations in the thicknesses of the adhesive layers. In fact, in some regions the thicknesses are often found to be less than 0.1 mm, while in others exceed 1 mm. Furthermore, uncured adhesives tend to accumulate in locations where the gap between adherents is increased, thereby leaving voids in neighbouring regions which remain in the joint even after the curing process is complete (the so called “spring back effect”).
To visualize defects in adhesive joints with high spatial resolution, high frequency focused systems have been developed using a pulse-echo technique. Here the variation of the reflection coefficient at the first metal/adhesive interface (closest to the transducer) that results from disbonds is strong enough to obtain sufficient contrast in resulting acoustical images. Unfortunately, the acoustical impedance mismatch between the adhesive or the immersion liquid and the metal – especially for steel – produces prolonged, strong reverberations of the wave in the first metal sheet. In fact, if the time delays of the wave propagation in the adhesive layer and metal sheet are approximately equal (or a multiple of each other) these reverberations sufficiently mask the small echoes returning from the second adhesive/metal or the adhesive/air interface (furthest from the transducer).
To overcome this problem, the IDIR has developed the NDT technique based on comparing the output waveform with a previously recorded reference for the first metal sheet; this is developed to detect disbonds and voids in the adhesively bonded joints with sufficient resolution.
Based on a simple plane wave model, the output waveform is presented as a sum of the three responses associated with the reflection of the ultrasonic wave at the first metal–adhesive interface, second metal–adhesive interface, and second metal–air interface, respectively. The strong response produced by the wave reverberating in the first metal sheet is eliminated through comparison between output and reference waveforms. The reference waveform is obtained through the multiplication of an echo recorded outside of the joint with a damping function. Any disbond that occurs at the second adhesive–metal interface can be detected through phase inversion of the second response. The third received response is associated with the wave reverberating in the second metal sheet. In the case that the thicknesses of the first and second metal sheets are equal, this response is quite strong, and its presence is an indication of a good joint. The developed decomposition algorithm has been applied to the study of steel and aluminum samples having various adhesive layers thicknesses in a range of 0.1-1 mm.