Ali Yousefpour

Fusion Bonding/Welding of Thermoplastic Composites

Joining of thermoplastic composites is an important step in the manufacturing of aerospace thermoplastic composite structures. Therefore, several joining methods for thermoplastic composite components have been under investigation and development. In general, joining of thermoplastic composites can be categorized into mechanical fastening, adhesive bonding, solvent bonding, co-consolidation, and fusion bonding or welding. Fusion bonding or welding has great potential for the joining, assembly, and repair of thermoplastic composite components and also offers many advantages over other joining techniques. The process of fusion-bonding involves heating and melting the polymer on the bond surfaces of the components and then pressing these surfaces together for polymer solidification and consolidation. The focus of this paper is to review the different fusion-bonding methods for thermoplastic composite components and present recent developments in this area. The various welding techniques and the corresponding manufacturing methodologies, the required equipment, the effects of processing parameters on weld performance and quality, the advantages/disadvantages of each technique, and the applications are described.

Process and performance evaluation of ultrasonic, induction and resistance welding of advanced thermoplastic composites:

The possibility of assembling through welding is one of the major features of thermoplastic composites and it positively contributes to their cost-effectiveness in manufacturing. This article presents a comparative evaluation of ultrasonic, induction and resistance welding of individual carbon fibre-reinforced polyphenylene sulphide (PPS) thermoplastic composite samples that comprises an analysis of the static and dynamic mechanical behaviour of the joints as well as of the main process variables. The induction welding process as used in this research benefitted from the conductive nature of the reinforcing fibres. Hence, no susceptor was placed at the welding interface. Resistance welding used a fine-woven stainless-steel mesh as the heating element and low welding pressures and times were applied to prevent current leakage. Triangular energy directors moulded on a separate tape of PPS resin were used to concentrate ultrasonic heat at the welding interface. The static single-lap shear strength of the joints was found similar for induction and ultrasonic welding. A 15% drop in the static mechanical properties of the resistance welded joints was attributed to incomplete welded overlaps following current leakage prevention. However, the fatigue performance relative to the static one was similar for the three sorts of joints. A comparative analysis of process variables such as welding time, required power and energy was also carried out.

Experimental and Computational Study of APC-2/AS4 Thermoplastic Composite C-Rings

The mechanical performance of APC-2/AS4 thermoplastic composite C-ring samples with different processing conditions was investigated, and experimental results were compared with numerical results using finite element methods (FEMs). Mandrel/substrate preheating was found to be necessary for good-quality parts. Ten sets of samples, with five samples per set, were manufactured using in-situ thermoplastic composite filament winding. For the first five sets, tape preheated to below the glass transition temperature (Tg ) at 110°C was used, while the consolidation pressure for various sets was 5.5, 12.6, 19.4, 26.0, and 32.4 kN/linear-meter. The same pressures were used for the next five sets while the tape was preheated above the Tg at 170°C. Scanning electron microscopy (SEM) was used for quality control. C-ring tests were performed to evaluate failure stress, strain, and deflection of C-rings at room temperature. Samples failed in compression at ring mid-section and inner radius. Samples made with 12.6 and 19.4 kN/linear-meter consolidation pressures yielded the best results. Non-linear FEM was employed to simulate the C-ring experiment using shell, target, and contact elements. The experimental deflection to failure was applied to the model, and the failure stress, strain, and load were determined. The results from non-linear numerical analysis were slightly higher than those determined from available analytical solution.

Metal mesh heating element size effect in resistance welding of thermoplastic composites

The objective of this work is to determine the effects of metal mesh heating element size on resistance welding of thermoplastic composites. The materials to be resistance-welded consisted of carbon fiber/poly-ether-ketone-ketone (CF/PEKK), carbon fiber/poly-ether-imide (CF/PEI) and glass fiber/PEI (GF/PEI). Four different metal mesh sizes were used as heating elements. The samples were welded in a lap shear joint configuration and mechanically tested. Maximum Lap Shear Strengths of 52, 47 and 33 MPa were obtained for the CF/PEKK, CF/PEI and GF/PEI specimens, respectively. The ratio of the heating element’s fraction of open area and wire diameter was shown to be the most important parameter to be considered when selecting an appropriate heating element size.

Finite Element Method for Active Vibration Suppression of Smart Composite Structures using Piezoelectric Materials

Adaptive or intelligent structures which have the capability for sensing and responding to their environment promise a novel approach to satisfy the stringent performance requirements of future space missions. Analytical, numerical, and experimental results are employed to verify the performance of piezoelectric stacks and patches as well as to determine the natural frequencies of typical strut and panel structures. A strut model with a piezoelectric stack actuator for axial vibration suppression and a composite beam with surface-mounted piezoelectric patch actuator for lateral vibration suppression are considered to model an active composite strut (ACS) and an active composite panel (ACP), respectively. These ACS and ACP are employed to develop an actuator optimum voltage (OV) for active vibration suppression using modal, harmonic, and transient finite element analyses for a range of frequency encompassing a natural frequency. The ACP model demonstrates that the actuator vibration suppression capability depends on the modal shape and location of the actuator. The OV, in this work, is determined by increasing the level of actuator voltage gradually and generating a vibration with same frequencies as the external vibration but 180 out-of-phase, and observing the increasing level of active vibration suppression until an optimum/threshold actuator voltage is reached. Beyond the optimum voltage level, the actuator increased the level of vibration 180 out-of-phase. Modal, harmonic, and transient finite element analyses are performed to verify the results. Selected axial and lateral vibration suppression experiments are also performed to verify the numerical results. The analytical, numerical, and experimental results obtained in this work are in excellent agreements. This work also presents a systematic guideline for the use of piezoelectric stack and monolithic patch smart materials in intelligent structures using the finite element method.

Modelling of mechanical properties of randomly oriented strand thermoplastic composites

There is an emerging interest in the aerospace industry to manufacture components with intricate geometries using discontinuous-fibre carbon/polyether-ether-ketone moulding systems (obtained by cutting unidirectional tape into strands). Great formability and high modulus can be achieved with this type of composites, but the high variability of measured properties can have a detrimental effect on the design allowables. When it comes to prediction of mechanical properties, it is important to capture the average strength and modulus as well as their statistical variability. This article proposes a stochastic finite element technique that uses the concept of randomly oriented strands to model variability, and the application of Hashin’s failure criteria and fracture energies to estimate strength. Overall, the model matches the trends observed during experiments and shows that strength of randomly oriented strand composites is significantly lower than that of continuous-fibre laminates due to the ‘weakest-link’ principle.

Optimization of thermoplastic composites resistance welding parameters based on transient heat transfer finite element modeling

The use of resistance welding technology to join thermoplastic composite aerospace structures is still contingent upon a better understanding of the heat transfer mechanisms occurring during welding, which govern the joint quality and mechanical performance. In this study, two-dimensional (2D) and three-dimensional (3D) transient heat transfer finite element models were developed to simulate resistance welding of thermoplastic composites. The 2D model was used to investigate the effect of the length of the exposed areas of the heating element to air (clamping distance) on the local overheating at the edges and the effects of the input power level on the thermal behavior of the welds. It is shown that controlling the clamping distance improves the thermal uniformity of the weld. The 3D model shows that heat conduction along the length of the laminates influences the thermal uniformity of the weld interface. An optimization chart is developed in order to minimize the undesirable edge effect and to define the conditions required to obtain a complete weld. The results of the 3D model are compared with experimental data.

Effects of Geometric Optimization of Plug-Supported End-Caps on the Performance of Thick Thermoplastic Composite Pressure Vessels Under External Hydrostatic Pressure

A finite element model was developed to investigate the response of a thick composite pressure vessel under hydrostatic pressure for deep ocean applications. To verify the finite element model, layer-by-layer stress and strain responses at the mid-length of a composite pressure vessel, i.e., free from the end-cap effects, were obtained and compared with an existing analytical solution. Excellent agreement was obtained between the numerical and analytical solutions. The created model was employed to investigate the performance of an APC-2/AS4 thermoplastic composite pressure vessel using plug-supported end-caps with contoured-ends and initial radial clearances. The plug-supported end-caps with contoured-ends and initial radial clearances were modeled as radial simply supported boundary conditions at the ends of the composite cylinder. The pressure vessel has a thickness of 4.3 cm, an inner diameter of 33 cm, an internal effective length of 45.7 cm, and a symmetric sub-laminate configuration of [(90/90/0)s]4 subjected to an external hydrostatic pressure of 71 MPa. The results of finite element analysis revealed that the performance of the pressure vessel greatly depends on the length of the tapered section as well as the tapered radius of the contoured-end plug-supported end-caps. In addition, it is shown that an initial radial clearance of plug-supported end-caps can also affect the performance of the pressure vessel. The optimum performance of the pressure vessel was obtained when the length of the tapered section and tapered radius were 38.1 mm and 3.3 m, respectively. The best initial radial clearance for this pressure vessel was found to be 0.5 mm; however, sealing issues should also be taken into account when selecting the final amount of an initial radial clearance. The comparisons between the performances of the pressure vessels reveal that the stress factor of safety of the pressure vessels using plug-supported end-caps with optimum tapered and initial radial clearance can be 2 and 2.29 times, respectively, greater than that for the pressure vessel with plug-supported end-caps.

Impact of layup rate on the quality of fiber steering/cut-restart in automated fiber placement processes

Abstract Developing reliable processes is one of the key elements in producing high-quality composite components using an automated fiber placement (AFP) process. In this study, both simulation and experimental studies were carried out to investigate fiber steering and cut/restart under different processing parameters, such as layup rate and compaction pressure, during the AFP process. First, fiber paths were designed using curved fiber axes with different radii. Fiber placement trials were then conducted to investigate the quality of the steered fiber paths. Furthermore, a series of sinusoidal fiber paths were fiber placed and investigated. Moreover, a six-ply laminate with cut-outs in it was manufactured in the cut/restart trials. The accuracy of the fiber cut/restart was compared at different layup rates for both one- and bi-directional layups. Experimental results show that it was possible to layup steered fiber paths with small radii of curvature (minimum 114 mm) designed for this study when the proper process condition was used. It was observed from the cut/restart trials that the quality of tow cut was independent of layup speed; however, the accuracy of tow restart was related to the layup speed. The faster the layup speed, the less accurate was the tow restart.

Characterization of resistance-welded thermoplastic composite double-lap joints under static and fatigue loading

An experimental investigation of resistance welding of thermoplastic composite double lap shear (DLS) joints is presented. DLS specimens consisting of unidirectional carbon fibre/polyetheretherketone (CF/PEEK), carbon fibre/polyetherketoneketone (CF/PEKK), carbon fibre/polyetherimide (CF/PEI) and 8-harness satin weave fabric glass fibre/polyetherimide (GF/PEI) composites were resistance welded using a stainless steel mesh heating element. The welded specimens were tested under static and fatigue loadings, and the quality of the welds was examined using optical and scanning electron microscopy. Weld strengths of 53, 49, 45 and 34 MPa were obtained for CF/PEEK, CF/PEKK, CF/PEI and GF/PEI DLS joints, respectively. Indefinite fatigue lives were obtained between 20 and 30% of the ultimate static failure loads of the joints. Performances of the resistance-welded DLS and single lap shear (SLS) joints were compared. It was shown that the effect of joint geometry, that is, DLS versus SLS, on the mechanical performance of the resistance-welded joints is minimal, indicating a good resistance of welded joints to peel stresses.