When ultrasonic pulses are used to detect flaws that are near the surface or in thin-layered structures, the natural tendency is to use a narrow pulse. A broadband pulse with a high center frequency and a large bandwidth will allow the flaw echo and interface echoes to be resolved from the front surface echo in an immersion or squirter testing situation. However, there are a number of limitations in using high frequency ultrasound. The major disadvantages are the high attenuation, hence the limited interrogation depth, and the degradation of ultrasonic scan images due to interference effects. Interference fringes can arise from minute thickness changes of the various material layers and can mask the flaw images [1]. In ultrasonic C-scans it is also important that the time gate used for interrogating a certain depth range be properly positioned and that the gate be able to follow the surface contour if the part is not flat. These problems can largely be avoided at lower frequencies, but the correspondingly poorer temporal resolution can impair the detection of near-surface flaws. It is generally accepted that there is a “dead time” associated with the front surface ringdown during which near-surface flaws are undetectable. Authors of this paper have encountered these problems in their studies of (1) foreign object detection in composite laminates, (2) disbond and corrosion detection in adhesively bonded aircraft skins, and (3) defects in the bond between thin composite face-sheet and honeycomb core of sandwich structures.