Cure Cycle In-process Monitoring of Advanced Composites via Reusable Flexible Ultrasonic Sensing Film

Abstract

Environmental and cost-saving advantages derived from the use of composites attract the aerospace and automotive industries as these materials offer significant structural and aerodynamic advantages over traditional metal structures. The composites industry, however, is concerned with the manufacturing processes as they cannot provide fast enough cycle time to match metal alloy processes. This research aims to develop a sensing technology in the form of a reusable in-situ cure monitoring and assessment system that can predict the formation of manufacturing defects and monitor the degree of cure.

First, a thin-film material is chosen from various PTFE-based material by prioritizing the debonding effect and signal transmission through the composite part. Then, the film is used to sandwich piezoelectric actuators and sensors to monitor out-of-autoclave carbon fiber composite plates using ultrasonic Lamb waves by temporarily adhering to the manufactured part creating an effective electromechanical coupling between the sensing film and the laminate. Initial results, through the analysis of the fundamental antisymmetric A0 mode at low frequencies, indicate that analyzing the velocity and amplitude of these waves over cure time determines minimum viscosity, gelation, and vitrification points. Experimental results also prove the feasibility of using such a reusable film for different post-curing cycles, always determining certain cure parameters.

Since some of the concerns this industry faces are the energy and time spent on long curing cycles to achieve permanent bonding between the matrix and fibers, the reusable PTFE thin sensing film is then used to monitor the same cure parameters for a shorter curing cycle than that suggested by the CFRP manufacturer. The results show that the determined three cure parameters are offset by the same time deducted from the cycle, highlighting the practicability of using such technology. To verify the viability of this approach, tensile testing and dynamic mechanical analysis (DMA) are performed on these composites. Tensile testing results show that the average tensile modulus for the shortened cycle is of similar values if not slightly higher than that of the normal cycle. DMA results verify both previous conclusions: time shift of cure parameters and enhanced mechanical and thermal properties of the shortened cycle.

With the help of DMA results, a computational model for the CFRP plate is developed to imitate the experimental in-process monitoring. The storage modulus for the used CFRP is extracted throughout the curing cycle and its trend is implemented into COMSOL combined structural and electrostatics multiphysics to simulate the same mechanical fluctuations of the CFRP during curing. Then, Lamb waves are excited and sensed via sandwiched piezoelectric transducers in a reusable PTFE sensing film to monitor the structural health of the structure. The three cure parameters are determined for both curing cycles, validating the efficiency of this numerical method. Also, while closely analyzing these computationally generated curves, an additional cure parameter defined as “gelation initiation” is proposed. Additionally, the decomposition of different wavefield modes is scrutinized to describe their scattering throughout the layered structure. A new entrapped antisymmetric mode appears inside the CFRP laminate at the start of the cure, which suggests that the previously studied A0 mode had been initially converted from the CFRP S0 mode.

Finally, the curing of adhesives used for structural bonding is monitored. By joining two fully cured CFRP plates with a prepreg epoxy film to be cured in the oven, the same previous methods are used to analyse the data extracted from the ultrasound monitoring of this adhesive. Additional S0A0 mode that is converted at the overlap is also analyzed. Then, post-cure monitoring on the CFRP plates is performed to remove the effect of heat and determine more accurate cure parameters. In the numerical model, sole A0 mode is actuated to enhance the scrutiny of mode conversion at the overlap. The numerical results, although heavily dependent on the adhesive film DMA curing results input, still highlight the desired cure points. Lastly, cocuring of both adhesive film and non-cured CFRP prepreg laminates is also tested experimentally where the A0 amplitude curve show more sensitivity towards the added epoxy cure parameters.