Experimental investigation of the effects of imperfections on the behavior of composite scarf joints

• Main Author: Jones, Elizabeth A., S.M. Massachusetts Institute of Technology
• Other Authors: Paul A. Lagacé.
• Format: Others
• Language: English
• Published: Massachusetts Institute of Technology 2017
• Subjects: Aeronautics and Astronautics.
• Online Access: http://hdl.handle.net/1721.1/108929

Thesis: S.M., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2017. === Cataloged from PDF version of thesis. === Includes bibliographical references (pages 243-245). === An experimental investigation was conducted in order to characterize the sensitivity of the overall response and failure behavior of composite scarf joints to the details of their geometry. Three parameters, considered to be manufacturing imperfections that could occur during the repair process, are examined: a through-thickness offset between the two adherends, adherends of different relative thicknesses, and adherends with different relative longitudinal stiffnesses. In all cases, the baseline adherend remains the same with a [±15]3s layup, and the manufacturing imperfection is introduced in the opposite adherend. A specimen was designed and utilized for this study. All specimens were made from Toray T800/3900 pre-impregnated carbon fiber/epoxy composite material and 3M Scotch-Weld Structural Adhesive Film AF 555M using controlled processing techniques, some of which were specifically designed to handle the particular details of this work. The achieved geometry of all specimens for each case show successful implementation of the manufacturing imperfections. All specimens were tested in uniaxial tension to failure and the results indicate that the specimens show the desired behavior throughout. The load-displacement responses of all specimens exhibit the same general characteristics, with an initial linear region that gradually transitions to increasingly nonlinear behavior to the point of maximum load. The extensional stress-strain responses for all adherends are linear to the point of maximum load, with moduli equal to those calculated via Classical Laminate Plate Theory. The load-bending strain responses exhibit minimal magnitude of bending for cases with aligned load paths through both adherends, while the specimens with an eccentric load path between the two adherends show bending strain that increases with increasing load, showing membrane stiffening effects at higher loads. The type of failure and location of the failure surfaces throughout the joint region is generally the same for all specimens, with cohesive failure directly in front of the blunt tip of both adherends, crossing via adhesive failure from the corner of the blunt tip to the opposite adherend surface, where cohesive failure occurs until it crosses back via adhesive failure near the midline of the joint to join antisymmetrically with the failure surface from the opposite blunt tip. The value of the maximum loads of the specimens fall into two sets based on the achieved thickness of the film adhesive layer of the specimen: either less than or greater than the nominal uncured film adhesive thickness. However, the type and magnitude of the manufacturing imperfections considered showed no effect on these maximum loads. Voids are observed within the film adhesive layer of all specimens, with differences in the shapes, sizes, and extents of the voids corresponding with the thickness of the film adhesive layer and likely contributing to the maximum loads observed. Recommendations for further work are made. === by Elizabeth A. Jones. === S.M.