Keywords: characteristic length, composite, joint, strength, failure, bearing.

A numerical method is presented to determine the characteristic lengths for the failure analysis of composite joints without testing. In the conventional methods, compressive characteristic length was determined from the result of a combined bearing test and finite element analysis. The present study, however, shows that the same compressive characteristic length can be obtained by numerical calculation without the bearing test. Basic concept to determine the compressive characteristic length is shown in Figure 1(a). In the joint, , , mm. The terms , , , and mean the hole-diameter, width, edge-distance, and thickness of the joint, respectively. Stacking sequence of the graphite/epoxy joint shown is , which is derived from an actual aircraft control rod joint. Idea of the numerical method to determine the characteristic lengths without testing came from the analysis results that the stress distribution in the characteristic length specimens is almost linearly proportional to the applied load although the nonlinear contact analysis is performed. Consequently, as shown in the figure, the application of the bearing failure load is not necessarily required to decide the characteristic length. Although an arbitrary load, not necessarily the tested bearing failure load, is applied, the stress distribution varies in proportion to the applied load and the compressive characteristic length is same

The new tensile characteristic length is defined using the strength of the notched laminate as the reference stress level as shown in Figure 1(b), whereas conventional method uses the strength of the sound laminate. Using the new definition of the tensile characteristic length, it can be calculated numerically without characteristic length test like the compressive characteristic length. As shown in the figure, the tensile characteristic length based on the new method is different from the one by the conventional method because a different definition was used. Therefore, joint failure loads by the two methods should be compared with joint test results for the validation. Figure 2 and Table 1 show the joint shape and the failure loads of the joints, respectively. The failure loads of the joints showing the bearing mode (failure mode, B) are exactly the same as the ones by the conventional method because the compressive characteristic lengths are same. As for the net-tension failure load (failure mode, N), the present method predicts the joint strength more accurately. In conclusion, the finite element results based on the characteristic lengths calculated without testing show the same or better agreement with the joint test results than conventional method using the characteristic length test.

1

Whitney, J.M., Nuismer, R.J., "Stress fracture criteria for laminated composites containing stress concentration", Journal of Composite Materials, 10, 253-265, 1974. doi:10.1177/002199837400800303

2

Chang, F.K., Scott, R.A., "Strength of mechanically fastened composite joints", Journal of Composite Materials, 16, 470-494, 1982. doi:10.1177/002199838201600603

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)