Date of Award

Winter 2019

Project Type


Program or Major

Mechanical Engineering

Degree Name

Master of Science

First Advisor

Todd Gross

Second Advisor

Igor Tsukrov

Third Advisor

Nikhil Padhye


The goal of this effort was to determine how accurately finite element models of 3D woven composites constructed using Dynamic Fabric Mechanics Analyzer predict the intrinsic residual stress field that results from cooling going from the curing temperature to room temperature. Hole drilling methods were used to estimate and measure the local stresses at selected locations in 3D woven composites. Blind holes were virtually drilled in one orthogonal and multiple ply-to-ply models and the resulting in-plane surface displacement fields were compared to the surface displacement fields measured using electronic speckle pattern interferometry.

The shape of the experimental and predicted displacement fields were similar in both the orthogonal and ply-to-ply structures, except along the orthogonal warp tow, where the fields were opposite. The magnitude of the measured displacement fields was approximately 1/5th that of the predicted value both parallel and perpendicular to the tows in the orthogonal structure, except along the orthogonal warp tow, which was -1/5th. This is attributed to microcracking below the warp tow relieving the residual stress. The magnitude of the experimental displacements was between 1/3rd to two times larger than the prediction transverse to the tow axis in the ply-to-ply structure. Along the warp tow axis, on one side the experimental result matches the prediction while on the other side the experimental displacement was two times larger. Along the weft tow in the ply-to-ply the experimental displacement matched the predicted displacement.

The effect of drill depth on surface displacement was studied by drilling holes in 0.5mm depth increments and comparing experimental results to finite element models of an orthogonal and a 12x10 picks per inch ply-to-ply structure. The experimental displacements match the predicted displacements in shape and magnitude. The model does a good job of predicting the effect of drill depth on surface displacement.

The effect of pick spacing and volume fraction on surface displacements from hole drilling was studied for composites with a ply-to-ply weave with 12x12 (warp/weft) picks per inch (ppi), 10x12 ppi, and 10x8 ppi. The finite element models used in this section contained a resin overburden on top of the tows, affecting the accuracy of the predicted displacements. The experimental displacement fields were a good qualitative match to the predictions. The experimental transverse displacements were approximately 1/4 of the predicted displacements. The model increasingly under predicts the displacement along the tow axis as the number of picks-per-inch decreases.