Open Collections

Battilana, Martin Gordon Anselmo

University of British Columbia

Carbon fiber composites are increasingly used in aerospace applications with tight tolerances on dimensional control, requiring strict management of process induced deformation (PID). From the extensive modelling that has been done to predict PID, one of the key uncertainties is the effect of moisture absorption as water molecules from the air diffuse into the polymer matrix of a composite part, leading to swelling due to moisture expansion. While current literature work has only focused on the effects of uniform and symmetric moisture absorption, this work takes it one step further by analyzing the effect of both symmetric and asymmetric moisture absorption. To accomplish this, initial experiments were conducted to measure the coefficients of moisture expansion which relate changes in concentration to moisture swelling strains. Then, experiments subjecting L-shape samples to a moisturizing and drying cycle were conducted for both symmetric and asymmetric moisture absorption (boundary conditions) to measure the spring-in and warpage of the samples. These experiments revealed that there was almost an order of magnitude more spring-in for asymmetric moisture absorption compared to symmetric moisture absorption. Numerical models were developed for the diffusion and deformation processes to allow for the calculation of spring in and warpage, which showed good agreement with the experimental results. These models provide additional insights about the concentration profiles within the sample as well as being able to split the total spring-in response into in-plane and through-thickness components to understand the individual effects. By doing this, it was found that the in-plane effects, which scale with 𝐿/ℎ¹⋅⁵ (where 𝐿 is flange length and ℎ is thickness), were responsible for the significantly large spring-in values observed for the asymmetric moisture absorption. Therefore, dimensional change due to moisture absorption is not a concern for thick parts during manufacturing and assembly even for asymmetric moisture absorption while for thin samples, which are often used for material characterization, very small amounts of asymmetric moisture become critical for controlling PID. Future work is recommended to be focused on improving the diffusion model as spring-in and warpage are very sensitive to the concentration profiles.

Text

2025-06-19

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/91374

Materials Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/