If you want to really understand why the current tap test for such disbonds is useless, I suggest you read this:
http://www.adhesionassociates.com/pa...ES%20final.pdf
It actually shows that for short bond lengths where there is a history of adhesion or mixed-mode failure, then NDI may not actually prevent bond failure because the bond strength may decay to unacceptable levels before any detectable disbond can be found. In fact there may not even be a disbond at the time of failure.
Draw your own conclusions about the validity of currently approved methods for assurance of structural integrity of principal structural elements with short overlap bonds.
Regards
Blakmax