Search

This paper investigates the stability of deformable layers produced by material extrusion additive manufacturing. A Computational Fluid Dynamics (CFD) model is developed to predict the deposition flow of viscoplastic materials such as ceramic pastes, thermosets, and concrete. The viscoplastic materials are modelled with the Bingham rheological equations and implemented with a generalized Newtonian fluid model. The developed CFD model applies a scalar approach to differentiate the rheology of two layers in order to capture the deposition of a wet layer onto a semi solidified printed layer (i.e., wet-on-semisolid printing). The semi solidification is modelled by a yield stress buildup. The cross-sectional shapes of the deposited layers are predicted, and the relative deformation of the first layer is studied for different yield stress buildups and processing conditions such as printing- and extrusion-speed, layer height, and nozzle diameter. The results of the CFD model illustrate that the relative deformation of the first layer decreases non-linearly with an increase in yield stress, and that stable prints can be obtained when taking into account the semi solidification. Furthermore, it is found that the deformation is dependent on a non-trivial interplay between the extrusion pressure, the shape of the cross-section, and the contact area between the layers. Finally, the results highlight which process conditions can be changed with benefit in order to limit the requirement on the yield stress buildup and still provide stable prints.

Contains fulltext : 238926.pdf (Publisher’s version ) (Closed access) Soft and electrically active materials are currently being utilized for intelligent systems, including electronic skin, cybernetics, soft robotics, and wearable devices. However, fabricating materials that fulfill the complex requirements of such advanced applications remains a challenge. These attributes include electronic, adhesive, self-healing, flexible, moldable, printable, and strong mechanical properties. Inspired by the recent interest in transforming monofunctional materials into multifunctional ones through nanoreinforcement and mussel-inspired chemistry, we have designed a simple two-step methodology based on halloysite nanotube (HNT) and polydopamine (PDA) to address the grand challenges in the field. In brief, HNTs were coated with PDA and embedded within a poly(vinyl alcohol) (PVA)-based polymeric matrix in combination with ferric ions (Fe(3+)). The final composite displayed a 3-fold increase in electrical conductivity, a 20-fold increase in mechanical stiffness, and a 7-fold increase in energy dissipation in comparison to their nonfunctional counterparts, which arose from a combination of nanotube alignment and mussel-inspired chemistry. Moreover, the developed composite could elongate up to 30000% of its original length, maintain its electrical properties after 600% strain, self-heal within seconds (both electrically and mechanically), and display strain-sensitivity. Finally, it was 3D-printable and thus amenable for engineering of customized wearable electronics.

Advancements in material extrusion additive manufacturing are driven by the need of fast production of high-quality parts. A key innovation was the introduction of short fibers into the built filament, which substantially improve the strength of manufactured components. However, the presence of fibers also affects the viscosity and thermal conductivity of the filament, which are important parameters for the extrusion flow and thus the maximum printing speed. In this paper, we numerically study the effect of fibers on the polymer flow and pressure drop inside the nozzle, which determines the maximum extrusion rate. The thermoplastic polymer flow is simulated with a non-isothermal computational fluid dynamics model and the inclusion of fibers is treated with a continuum approach. The simulations are performed for ABS polymer with short carbon fibers, however, other thermoplastic systems with short fibers (e.g. glass, wood, or nylon) can be integrated into the model. The model provides a virtual window into the process and illustrates that while fibers increase the viscosity of the filament and the pressure drop, they also improve the thermal conductivity and lead to faster melting of the polymer, which has an opposite effect on the pressure drop. Finally, the model quantifies the relationship between the fiber volume fraction and the maximum extrusion rate.

This work presents Computational Fluid Dynamics (CFD) simulations of the polymer flow inside the hot-end during material extrusion additive manufacturing. Two CFD models are investigated: a previously-published one-phase model, where the entire domain is filled with liquid, and a novel model, where the free surface of the polymer inside the channel is resolved. Both models predict a recirculation region between the nozzle wall and the incoming filament. With the free surface resolved, melting of the solid filament and filling of an empty liquefier channel are shown in detail. Moreover, the simulations predict the pressure and temperature distributions inside the channel. The molten polymer (ABS) is simulated as a Generalized Newtonian Fluid with shear- and temperature-dependent viscosity. The numerical results are compared with the experimental measurements of the filament feeding force, which relates to the pressure inside the flow channel. An inverse analysis of the heat transfer coefficient is performed to estimate the thermal resistance at the channel’s wall. It is shown that the model which resolves the free surface is able to predict the feeding force for typical working conditions with a reasonable accuracy. Moreover, it captures the change of the flow regime from stable to unstable extrusion at high feeding rates. A hypothesis that explains the pressure and melt zone oscillations that occur during unstable extrusion is given. The influence of the liquefier temperature, liquefier length and nozzle diameter on the flow are discussed. The predictions of the model become less accurate when different channel geometries are simulated, which is attributed to the simplified material model that does not capture viscoelastic effects and possible buckling of the solid filament.

Numerical modeling is one of the key components in the development of digital twins of Additive Manufacturing (AM) processes, encompassed in the concept of Industry 4.0. Numerical simulations also have a role to play in the improvement of contemporary AM techniques, among which material extrusion is very popular. Computational Fluid Dynamics (CFD) models have recently proven successful for simulating the deposition flow in extrusion AM. Previous modeling works using CFD have been able to predict the influence of processing parameters on the cross-sectional shape of the printed strand, as well as the mesostructure formation resulting from their deposition and fusing. The present study focuses on using CFD simulations to quantify the waviness of a strand extruded on top of a previous layer with an orthogonal print direction, which often occurs when printing a part with a rectilinear infill pattern and alternate 0°/90° raster angles from layer to layer. The variations in the strand width and thickness of the second layer were found to depend on the strand-to-strand gap in the first layer. The CFD model was also used to determine the interlayer penetration depth, which enhances mechanical performances.

A computational fluid dynamics model is used to predict the mesostructure formed by the successive deposition of parallel strands in material extrusion additive manufacturing. The numerical model simulates the extrusion of the material onto the substrate. The model takes into account the effect of the presence of the previously extruded material on the shape of the subsequently deposited strands. The simulated mesostructures are compared to optical micrographs of the mesostructures of 3D-printed samples, and the predictions agree well with the experiments. In addition, the influence of the layer thickness, the strand-to-strand distance, and the deposition configuration (with aligned or skewed layers) on the formation of the mesostructure is investigated. The simulations provide detailed information about the porosity, the inter- and intra-layer bond line densities, and the surface roughness of the mesostructures, which potentially can be used in a model-based slicing software.

The material deposition along a toolpath with a sharp corner is simulated with a computational fluid dynamics model. We investigate the effects of smoothing the toolpath and material over-extrusion on the corner rounding and the corner swelling, for 90° and 30° turns. The toolpath motion is controlled with trapezoidal velocity profiles constrained by a maximal acceleration. The toolpath smoothing of the corner is parametrized by a blending acceleration factor. Analytical solutions for the deviation of the smoothed toolpath from the trajectory of the sharp corner, as well as the additional printing time required by the deceleration and acceleration phases in the vicinity of the turn are provided. Moreover, several scenarios with different blending acceleration factors are simulated, for the cases of a constant extrusion rate and an extrusion rate proportional to the printing head speed. The constant extrusion rate causes material over-extrusion during the deceleration and acceleration phases of the printing head. However, the toolpath smoothing reduces the corner swelling. The amount of underfill and overfill at the inside and outside of the corner are quantified, as compared to an ideal case where the printing head would instantaneously change its speed direction at the corner. The numerical results show that there is an optimal amount of toolpath smoothing where the over-extrusion compensates for the corner rounding; hence improving the quality of the corner. A uniform road width is obtained with the proportional extrusion rate.

Material extrusion additive manufacturing utilizes a thermoplastic polymer in the form of a solid filament as a built material. The polymer melts inside the hot-end channel and flows under the pressure generated by the filament feeding force. The flow of polymer through the hot-end is not fully understood yet, as it involves many complex phenomena, such as phase transition, shear rate and temperature dependent viscosity, as well as viscoelastic effects. In this paper, we investigate experimentally the filament feeding force, as a function of the feeding rate, for different materials (PLA and ABS), liquefier temperatures, nozzle diameters, and lengths of the liquefier. Two extrusion regimes are identified: a linear regime with a stable flow, and a non-linear regime with fluctuations in the feeding force, which are concomitant to unstable extrusion. Increasing the liquefier length and liquefier temperature are found to extend the linear extrusion regime. It is shown that the filament feeding force predicted by the analytical models available in the literature deviates from the measurements, especially at high feeding rates. A model solely based on heat transfer considerations is proposed to estimate the maximum feeding rate before the extrusion becomes unstable. The modelling results agree well with the measurements. The model can be used to select the hot-end design as well as appropriate printing parameters.

We propose a numerical model to simulate the extrusion of a strand of semi-molten material on a moving substrate, within the computation fluid dynamics paradigm. According to the literature, the deposition flow of the strands has an impact on the inter-layer bond formation in extrusion-based additive manufacturing, as well as the surface roughness of the fabricated part. Under the assumptions of an isothermal Newtonian fluid and a creeping laminar flow, the deposition flow is controlled by two parameters: the gap distance between the extrusion nozzle and the substrate, and the velocity ratio of the substrate to the average velocity of the flow inside the nozzle. The numerical simulation fully resolves the deposition flow and provides the cross-section of the printed strand. For the first time, we have quantified the effect of the gap distance and the velocity ratio on the size and the shape of the strand. The cross-section of the strand ranges from being almost cylindrical (for a fast printing and with a large gap) to a flat cuboid with rounded edges (for a slow printing and with a small gap), which substantially differs from the idealized cross-section typically assumed in the literature. Finally, we found that the printing force applied by the extruded material on the substrate has a negative linear relationship with the velocity ratio, for a constant gap.

We investigate experimentally and numerically the influence of the processing conditions on the cross-section of a strand printed by material extrusion additive manufacturing. The parts manufactured by this method generally suffer from a poor surface finish and a low dimensional accuracy, coming from the lack of control over the shape of the printed strands. Using optical microscopy, we have measured the cross-sections of the extruded strands, for different layer heights and printing speeds. Depending on the processing conditions, the cross-section of the strand can vary from being almost circular to an elongated rectangular shape with rounded edges. For the first time, we have compared the measurements of strands’ cross-sections to the numerical results of a three-dimensional computational fluid dynamics model of the deposition flow. The proposed numerical model shows good agreement with the experimental results and is able to capture the changes of the strand morphology observed for the different processing conditions.