The potential use of composites for weight savings is now widely recognised
and they are extensively used in both aircraft control surfaces and doors.
However, if major weight savings are to be realised it is essential that c
omposites be used in 'primary' structural components, i.e. wing and fuselag
e skins. Before this can be achieved it must be shown that means of transmi
tting the load from the composite structure to the underlying, or adjacent
structure, are both safe and meet the FAA damage tolerance certification re
quirements.
For stiffened composite panels, one potential failure mechanism is the sepa
ration of the skin from the stiffeners; resulting from excessive 'through t
he thickness' stresses. This failure mechanism is also present in bonded co
mposite joints and composite repairs. Currently, failure prediction due to
in-plane loading appears to be relatively well handled. Unfortunately, this
is not yet true for matrix dominated failures. Consequently, it is essenti
al that a valid certification methodology, which addresses all of the possi
ble failure mechanisms, including failure due to interlaminar failure, be d
eveloped. To aid in the understanding of such failures the present paper ou
tlines the results of a series of experimental, analytical and numerical st
udies into the matrix dominated failures of bonded joints, composite repair
s and rib stiffened post buckled structures. (C) 1999 Elsevier Science Ltd.
All rights reserved.