Fundamentals of Thermoforming Processes of Carbon Fiber Reinforced Plastic (CFRP) Parts

Zhang, Weizhao

doi:https://doi.org/10.21985/n2-etnz-cf61

Work

Fundamentals of Thermoforming Processes of Carbon Fiber Reinforced Plastic (CFRP) Parts

Public

Download PDF

Download All Files (.zip)

Using carbon fiber reinforced plastics (CFRPs) to reduce vehicle weight has received growing attentions because it can effectively curtail greenhouse gas emissions and slow down global warming. The thermoforming process is among the most promising techniques for mass production of CFRP parts because it is highly automated and has relatively low production cost. This process consists of two steps: in a preforming step, thermoplastic CFRPs or prepregs are stacked, heated, and deformed on a press; then, in a curing step, resins solidify either from cooling-down or chemical reaction to obtain the permanent shape of the final part. Basic concepts, tools and procedures for the thermoforming process are quite mature these days, but the process has not been widely utilized. The major reason is that it involves ample design freedom in terms of material design, raw material preparation and process parameters. All these factors need to be considered to maximize performance-to-weight ratios of parts and minimize material waste in production. Existing optimization methods for these factors, however, require complex and expensive experimental trials, leading to high cost for product development. To address process design challenges and enable mass production of CFRP parts via thermoforming, the fundamentals of this process are researched by a numerical method, which is elaborated in this thesis. Between the two steps in thermoforming, preforming determines most of the parts’ geometry and fiber directions, which are crucial to the final parts’ performance. As a result, this thesis focuses on the theory, realization and applications of preforming modeling and its integration into part performance modeling. Specifically, the thesis contains: 1) Experimental characterization of prepregs to obtain the material parameters needed in the numerical simulation for preforming processes, i.e.: (a) uniaxial tension, bias-extension, and bending tests to measure, and (b) friction tests that are designed and performed to reveal the interlayer properties. As a potential low-cost replacement for the experimental friction tests, a hydro-lubricant model is also developed to study the relative motion mechanism of the fabric-resin-fabric system. 2) Part-level modeling for preforming processes simulation. An improved non-orthogonal material model that can capture the fiber-direction-dependent anisotropy of CFRP raw materials under large shear deformation is developed to simulate the preforming process. The material properties in the model are calibrated using the experimental methods. The prediction by this non-orthogonal material model is then validated by benchmark tests. 3) Multiscale modeling for preforming based on numerical characterization. A numerical material characterization approach is developed using the representative volume element (RVE) method at the mesoscale to account for the tension-shear coupling effect of the prepregs. A geometric modeling technique is established to generate the RVE structure of the closely packed prepregs. A modular Bayesian approach is then applied to calibrate the yarn properties and fit the stress emulator for simulation at the macroscale. Upon development, this multiscale model is utilized for numerical prepreg blank geometry design. 4) Integration of the preforming and performance models. An integrated preforming-performance simulation method is developed to include local fiber orientations and part geometries resulting from preforming in performance analysis. This method significantly increases the prediction accuracy of the performance simulation. Moreover, the performance analysis can serve as feedback for preforming parameter selection. The capabilities and potential applications of the integrated simulation method are demonstrated via numerical and experimental validation. In summary, a numerical method has been established to research the fundamentals of thermoforming processes of CFRP parts. This method includes material characterization, preforming modeling and performance modeling. The benchmark tests indicate that given the properties of prepregs characterized either by experiments or by mesoscopic RVE simulation, the preforming model can accurately predict the process effects. These prediction results are then mapped to the performance model to predict elastic behavior of final CFRP parts and provide feedback to preforming parameter design. Prediction accuracy of the performance model and design optimization capabilities of the integrated simulation method are validated by experimental and numerical comparisons. In the future, this relatively inexpensive numerical method could replace the existing experiment-based design methods for thermoforming processes to reduce process development cost and broaden applications of CFRPs in industry to reduce greenhouse gas emissions and control global warming.

Creator