Failure analysis of bonded composite joints is essential to the design of modern aerospace vehicles where adhesive joints are widely used. Recently, methods for stress and failure analysis of composite bonded joints were implemented within the HyperSizer® stiffened panel design/analysis/optimization software package. The capability uses efficient, non-finite element based methods to calculate the adhesive stresses and detailed in-plane and out-of-plane interlaminar shear and peel stresses in the adherends for predicting damage initiation and failure. Previously, HyperSizer predicted failure using 19 different leading ply and interlaminar stress/strain-based fracture and delamination bonded joint failure theories. Now, the joint failure analysis capability has been extended to include damage tolerance residual strength by using the virtual crack closure technique (VCCT), which predicts the growth of an existing crack by comparing calculated strain energy release rates to critical values determined from tests. Implementation of VCCT, specific to the HyperSizer joint analysis method, is discussed and results are compared to finite element analysis and experimental results from the literature for composite and isotropic joint adherends.

No abstract

No abstract

Due to weight and cost goals, a grid-stiffened panel concept is being used for redesign of a structural component on the Minotaur OSP space launch vehicle. By designing the structural panels to carry operational loads past the point of initial buckling (local postbuckling), the resulting grid stiffened panel concept is lighter and 30% less costly to manufacture than other design candidates such as the existing honeycomb sandwich panel concept flown today. During June 2001 in Seattle, Boeing performed a structural certification experiment of a composite, grid stiffened, cylindrical panel loaded in axial compression. Pretest predictions were made for linear elastic (bifurcation) buckling, and non-linear post buckling. The tools used for pretest analysis were HyperSizer, and the FEM based tools MSC/NASTRAN and STAGS. Local buckling of the facesheet triangular shaped skin pocket occurred at a load of around 230 (lb/in). The test panel was able to sustain considerable additional loading, with post buckling failure occurring at 1320 (lb/in). The HyperSizer post buckling pretest prediction was 1300 (lb/in), the STAGS pretest prediction was 1250 (lb/in), and the MSC/NASTRAN pretest prediction ranged from 1425 to 2000 (lb/in). HyperSizer's implementation of local post buckling based on an effective width approach is presented.

No abstract