Abstract— An analytical model was developed to investigate the choice of failure criterion based on characteristic length method for predicting composite joint strength. The available experimental data for joint strength in literature were used to determine the accuracy of analytically obtained results. The analysis utilized six failure criteria, where it was shown that among the failure criteria investigated no one gave a best fit for all the three plates evaluated for strength, and the fit of individual failure theories is variable from one type of material to another. The results also showed that the accuracy of this method can be improved by computing the average predicted values of more than three failure criteria. Index Terms— Characteristic length, Composite, Failure Criteria, and Joint strength . I. INTRODUCTION Adhesive bonding and mechanical bonding remain the two most common methods of joint design. While an adhesive joint offers the advantage of distributing the load over a larger area than a mechanical joint, its use is very restrictive due to its sensitivity to environmental conditions such as temperature and humidity. Therefore, mechanical joints are widely used as a reliable and practical method to assemble the composite parts that constitute a mechanical structure; they offer the advantage of assembly and disassembly for accessibility and maintenance, and reliability at various environmental conditions. However, a significant drawback when using the mechanical joint is that it always requires joint holes, which can be a source of stress concentration and thereby weakens the intended strength for the structures. Therefore, for the design of composite joints, stress and failure analysis around the fastener hole should first be conducted. During the loading of mechanical joints, the laminate deforms as the applied load increases and the contact surface between the fastener and the plate changes. It should be noted that despite the fact that the distributed stress around the fastener hole could be accurately determined, the failure at any point within the stress concentration region does not necessarily lead to failure of the joints. The strength of the structure will be underestimated if local failure is taken to be the ultimate failure of the joint. It becomes a big task to develop an analytic or computational method that can Manuscript submitted March 2, 2010. The Accuracy of Characteristic Length Method on Failure Load Prediction of Composite Pinned Joints. Olanrewaju Aluko is with the University of Michigan-Flint, 303 E. Kearsley Street, Flint, MI 48502 USA (phone: 810-762-3181; fax: 810-766-6780; e-mail: aluko@ umflint.edu). Quamrul Mazumder., is with University of Michigan-Flint, 303 E. Kearsley Street, Flint, MI 48502, USA. (e-mail: qmazumde@umflint.edu).. adequately model the complicated failure and structural response to loading of composite joints. Because of the complexity of the failure of composite joints coupled with the intension to reduce the cost and materials wastage in practical testing to failure, the analytical determination of bolted joint strength has received the attention of many researchers [1]-[16]. One of the most common and efficient methods of predicting the strength is the characteristic length method. This method was proposed by Whiney and Nuismer [1]-[3], and it has been further developed by Chang et. al [4]. For this improved method, both the characteristic lengths in tension, R t , and compression, R c , must be determined by stress analysis associated with the results of bearing and tensile tests on notched and unnotched plates before employing an appropriate failure theory along the characteristic curve, r c as shown in Figure 1. By definition, the characteristic length (in tension or compression) is the radial distance from the hole boundary over which the plate must be critically stressed to initiate a sufficient flaw that can cause failure. Wang et al. [5] analytically determined failure strength of composite joints in both circular and elliptical holes using stress functions developed by Lehknitskii [6] and Tsai-Wu failure criterion. However, only characteristic length in tension was used around the hole boundary to evaluate the failure load and mode of the composite joint. Whitworth et al. [7], [8] utilized the characteristic curve developed by Chang et. al [4] (which requires both the characteristic length in tension and compression to be determined) along with Yamada failure criterion to evaluate joint strength. Kweon et al. [9] developed a new method for finding the characteristic length in terms of arbitrarily applied loads, as an alternative to the failure load of notched plate in tension and bearing load of the plate when in compression. They evaluated the joint strength along the characteristic curve using Tsai-Wu failure criterion. In spite of the significant effort that has been invested on strength estimation of composite joints by various investigators using characteristic length method, the choice of an effective failure criterion in order to improve the accuracy of the characteristic dimensions remains to be investigated. For this paper, various failure criteria were utilized with the characteristic length method to predict the strength of composite joints with the objective to optimize the prediction accuracy of the characteristic length method. A two-dimensional analysis which utilized stress functions for the computation of stresses at the hole boundary and within the plate was employed. In this investigation of the effective choice of failure criterion on the prediction accuracy of joint strength, six selected failure criteria based on the previous survey by Nahas [10] were utilized. O. Aluko, and Q. Mazumder The Accuracy of Characteristic Length Method on Failure Load Prediction of Composite Pinned Joints Proceedings of the World Congress on Engineering 2010 Vol II WCE 2010, June 30 - July 2, 2010, London, U.K. ISBN: 978-988-18210-7-2 ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online) WCE 2010