Open Collections

Forghani, Erfan

University of British Columbia

One of the most important factors playing a major role in the final quality of a thermoplastic composite part is the developed degree of crystallinity, which is directly affected by the thermal history (process cycle). The melt and crystallization kinetics model for PEEK developed by Gordnian [1] is a useful tool for predicting this parameter, and features path dependency, integrated melt and crystallization sub-models, crystallization induction time, effect of temperature rate on melt behavior and finally cold and hot crystallization. However, this model needs to be improved by implementing temperature-dependent inert material models such as specific heat capacity and conductivity. To do this, a comprehensive literature review was conducted on available material properties in the open literature for PEEK, both neat and reinforced by carbon fiber. Additionally, modulated differential scanning calorimetry experiments were conducted to measure the specific heat capacity of PEEK/AS4 prepregs. The effect of thermal modulation parameters as well as thermal contact resistance were studied. Best practices, including identification of a range for modulation parameters and methods for capturing thermal contact resistance effects were determined. Using the best data from the literature, material models were developed. Talbot [2] simulated the welding of a PEEK/AS4 lap-shear joint in as a transient heat transfer problem. As a part of her investigation, the weld quality was judged based on temperature distribution at the end of the heat-up cycle, as a proxy for crystallization. In this thesis, this approach is improved by directly using Gordnian’s melt/crystallization model with both constant and temperature-dependent inert material properties. It is shown that melting can be completed faster than previously suggested. On the other hand, the time to reach maximum crystallinity is longer than originally predicted process time for this particular geometry, as cool-down was previously not considered. The effect of constant vs time-dependent inert material properties affects the predicted processing time by about 2.6 seconds longer processing time for the current case. Although this might be inconsequential for a static weld, it may be significant for continuous resistance welding of more complex geometries or larger sizes.

Text

2022-09-01

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/82656

Materials Engineering

University of British Columbia

UBCV

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