part 2
FAR 2x.573 requires demonstration of damage tolerance for aircraft structures including adhesive bonds. An FAA NPRM issued on 06 January 2010 proposes to extend these requirements to rotary wing aircraft. NPA-2010-04 "Damage Tolerance and Fatigue Evaluation of Composite Rotorcraft Structures"effectively duplicates the process for EASA. These requirements may involve analysis or testing to demonstrate that the structure provides adequate strength in the presence of known and detectable bond defects. Testing generally involves known artificial defects implanted in the bond-line. Production acceptance criteria for tolerable bond defect sizes are established on the basis of static strength tests or analysis on the basis that the bond interface retains full strength. Even service inspection requirements assume that the bond interface retains integrity.
Analysis and testing are meaningless for adhesion or mixed-mode failures because the assumption inherent in these approaches is that the adhesive surrounding the defect maintains an acceptable level of bond strength. In reality, if a disbond occurs in service, it is almost certainly due to degradation of the interface and such degradation is usually not confined to a localised area. The entire interface may be degrading and the bond surrounding the disbond may be weak. It is not possible to predict adhesion disbond growth rates, and defects may grow without any flight loads.
In recent times the FAA has taken steps to address adhesion failure modes. Policy Statement PS-ACE100-2005-10038 and recent amendments to AC 20-107 to “B” status include advice that interfacial failures are unacceptable in the production process and for service defects. These statements are to be commended, but they must eventually be followed up by appropriate NPRM action to amend the FARs to mandate bond durability testing.
Current NDI methods are only generally effective at finding production voids where there is an air-gap. These are the types of defect which cause cohesion failures. The ability of NDI to interrogate interfaces or even to detect weak bonds (kissing disbonds) is extremely limited. Double-sided adhesive tape will pass many NDI inspection methods, especially the tap-test, despite its obvious weakness compared to effective structural bonds. In effect, NDI can only tell that the bond has a defect, it can NOT guarantee the quality of the bond. With reference to NDI for inspection of in-service bonded components, NDI can not detect the onset of insipient disbonding. It is only effective AFTER a disbond has occurred. Hence, because of the ineffectiveness of damage tolerance in assessing the criticality of adhesive bonds with degrading interfaces and the inability of NDI to assess the potential for degradation, there is a risk that a disbond could propagate to a critical size before it is detected.
Therefore, the introduction of damage tolerance will not prevent disbonding from occurring. Worse yet, the assumption that the bond surrounding the defect retains full strength will provide a false sense of security when in fact the bond may be much weaker.
The solution is to mandate bond durability testing using methods which have a demonstrated history of differentiating between good processes and mediocre processes.