학술논문
Effect of Interlaminar Normal Stresses on the Uniaxial Zero-to-Tension Fatigue Behavior of Graphite/Epoxy Tubes
Document Type
stp-paper
Source
Composite Materials: Fatigue and Fracture (Third Volume), Jan 1991, Vol. 1991, No. 1110, pp. 659-666.
Subject
Language
English
Abstract
During the past several years, the Mechanics of Materials Laboratory of Rensselaer Polytechnic Institute (RPI) has developed a method of obtaining biaxial fatigue data under axial/torsion loading. A thin-walled tubular specimen can be made from prepregs by a layup procedure and tested in an MTS servohydraulic axial/torsion testing machine with computer control. We have provided completely reversed load-controlled fatigue data on graphite/epoxy materials under uniaxial and combined loadings using [± 45]s and [0, ± 45]s layups. The edgeless specimen eliminates suspected end effects and can be used for tests involving significant compressive loading.
Interlaminar normal stresses were thought to influence fatigue performance by enhancing delamination. To check on this hypothesis, zero-to-tension fatigue tests were run on graphite/epoxy [± 45]s tubes with and without internal pressurization. The pressure levels were chosen so as to compensate the suspected interlaminar tensile stresses. Fatigue test results in the range from 104 to 106 cycles with and without pressurization were within the same reasonable scatter band. In the course of testing, it was discovered that performance could be considerably improved by providing a restraint in the hoop direction by either inserting a mandrel or by including a 90° layer on the inside and outside. Fatigue tests with [90,±45]s specimens under zero-to-tension loading showed a significantly improved fatigue performance.
Interlaminar normal stresses were thought to influence fatigue performance by enhancing delamination. To check on this hypothesis, zero-to-tension fatigue tests were run on graphite/epoxy [± 45]s tubes with and without internal pressurization. The pressure levels were chosen so as to compensate the suspected interlaminar tensile stresses. Fatigue test results in the range from 104 to 106 cycles with and without pressurization were within the same reasonable scatter band. In the course of testing, it was discovered that performance could be considerably improved by providing a restraint in the hoop direction by either inserting a mandrel or by including a 90° layer on the inside and outside. Fatigue tests with [90,±45]s specimens under zero-to-tension loading showed a significantly improved fatigue performance.