PSI - Issue 13

Mor Mega et al. / Procedia Structural Integrity 13 (2018) 123–130

125

M. Mega et al. / Structural Integrity Procedia 00 (2018) 000–000

3

Fig. 1: An interface delamination between a transversely isotropic UD fabric and a tetragonal plain balanced weave.

A carbon / epoxy composite laminate plate containing 69 plies was fabricated. As a result of the temperature changes applied during the curing process, residual stresses were induced within the laminate. The maximum temperature measured in the plate during curing was 85 ◦ C. Weight fraction tests were carried out. The results were converted into volume fractions. With respect to the ob tained values, the e ff ective mechanical and thermal properties of each ply within the plate were obtained using the High-Fidelity Generalized Method of Cells (HFGMC) [25]. In addition, the properties were calibrated according to previous mechanical properties obtained from experimental results for each ply. The upper UD fabric, k = 1, is nearly transversely isotropic and may be described by five independent mechanical properties and two thermal properties. The glass fibers were taken into consideration in the matrix in an average sense. The lower woven fabric, k = 2, is a tetragonal material with x 2 = 0 a symmetry plane. This material may be described by six independent mechanical properties and two thermal properties. The measured and calculated volume fractions, as well as the mechanical and thermal properties used for the upper UD ply and for the lower woven ply are presented in Tables 1 and 2, respectively. Strips were cut from the plate using a water jet and glued to aluminum partial disks to form a Brazilian disk specimen which is illustrated in Fig 2a. Its diameter 2 R ≈ 40 mm; an initial delamination of length 2 a ≈ 16 mm along the investigated interface was created by a PTFE strip, 15 µ m thick. During each test, a specimen is placed in the loading frame shown in Fig. 2b with the loading angle ω , presented in Fig 2a. This loading angle is specified for each test and defined as the angle between the vertical load line and the delamination. Eight tests at four di ff erent loading angles were carried out using the BD specimens, in order to determine the fracture toughness for the delamination along the considered interface. The test procedure was based on the mixed mode methodology used in [26, 27, 7, 8]. First, geometric parameters of each specimen were measured, including the composite diameter 2 R and thickness B . In addition, each specimen was photographed with an Olympus Confocal Microscope (model number OLS4100; Tokyo, Japan) using its optical mode. The images were stitched together using ImageJ software [28] to create a high resolution photograph of the composite strip. Using the same ImageJ software [28], the thickness of each stack of plies of the same composite was measured for each specimen, including fourteen above the delamination and thirteen below it. The geometric measurements were used in the FE model. A LaVision system composed of one camera, a programable timing unit (PTU) and computer software are em ployed during the test. The camera used is a monochrome CCD of LaVision (model no. 1101396; Goettingen, Ger many) with a 5 MP Imager Pro SX, resolution 2456 × 2058 pixels, and a Nikon Micro-Nikkor 105 mm f / 2.8 lens. Images of the specimen are acquired at a rate of 5 Hz during the test as the load increases until fracture. The camera was connected to a LaVision external PTU and controlled by DaVis [29] computer software. Using an image of the specimen and frame, obtained by the LaVision system, the loading angle ω is measured with the Vision Assistant [30]

Table 1: Volume fraction, as well as the mechanical and thermal properties for the UD fabric with fibers oriented in the 0 ◦ -direction.

V f

E A

E T

G A

G T

ν A

ν T

α A

α T

( × 10 − 6 / ◦ C ) ( × 10 − 6 / ◦ C ) 65.1 1.46

(GPa) (GPa) (GPa) (GPa)

0.40 77.1 4.8 1.9 1.6 0.26 0.48