Modeling, Simulation and Experimental Validation of Heat Transfer During Selective Laser Melting

Selective Laser Melting (SLM), a laser powder-bed fusion (PBF-L) additive manufacturing method, utilizes a laser to selectively fuse adjacent metal powders. The powders are aligned in a bed that moves vertically to allow for layer-by-layer part construction-Process-related heat transfer and thermal gradients have a strong influence on the microstructural features, and subsequent mechanical properties, of the parts fabricated via SLM. In order to understand and control the heat transfer inherent to SLM, and to ensure high quality parts with targeted microstructures and mechanical properties, comprehensive knowledge of the related energy and mass transport during manufacturing is required. In this study, the transient temperature distribution within and around parts being fabricated via SLM is numerically simulated and the results are provided to aid in quantify the SLM heat transfer. In order to verify simulation output, and to estimate actual thermal gradients and heat transfer, experiments were separately conducted within a SLM machine using a substrate with embedded thermocouples. The experiments focused on characterizing heat fluxes during initial deposition on an initially-cold substrate and during the fabrication of a thin-walled structure built via stainless steel 17-4 powders. Results indicate that it is important to model heat transfer thorough powder bed as well as substrate.

Comparison of Tensile Testing Flow Stress to Orthogonal Plate Machining Flow Stress in FCC, BCC, and HCP Materials at Various Levels of Hardness Using a Videographic Quick Stop Device

Basic and advanced metal cutting research has been an ongoing effort since Cocquilhat’s early work directed towards measuring the work required to remove a given volume of material when drilling in the year 1851. Over the 150+ years since his experiments, one of the persistent issues in metal cutting has been how best to determine the flow stress in a metal undergoing cutting. In all the many models proposed since then, the flow stress of metal flowing in front of a cutting tool has not proven to be the same as the flow stress of metal undergoing a tensile pull. This paper examines the flow stress phenomenon using an improved Videographic Quick Stop equipment at Auburn University. The orthogonal machining plates and tensile specimens were all cut from the same stock. Tensile testing of the stock was performed immediately prior to the machining of the plates in a standard MTS load frame to allow actual metal cutting experiments to be performed and compared to actual load frame data from the same stock. Machining was conducted in a specially modified Cincinnati Horizontal Milling machine using an improved Videographic Quick Stop Device (VQSD) to capture the geometry of the cutting formation simultaneously with the forces in the X, Y and Z-axes using a standard Kistler force plate dynamometer. Utilizing the VQSD greatly increases the number of replicates available for statistical analysis by the metal cutting researcher. This allows for comprehensive multivariate analysis of the data with high confidence (> 95%) in the meaning of the results obtained, along with for powerful regression. The results of the data collection and statistical analysis are then used to populate the various historical models predicting the flow stress in metal cutting. The results indicate that one model is superior to all the other models in predicting the flow stress as predicted by the accompanying tensile test data. Further improvements in this model may lead to instantaneous tensile strength measurement when metal cutting with the need for load frames. This in turn would allow optimization of cutting conditions to match material conditions, resulting in a better product and longer-lived tools.

Thermal Error Modeling and Prediction of a Heavy Floor-Type Milling and Boring Machine Tool

Thermal error modeling and prediction of a heavy floor-type milling and boring machine tool was studied in this paper. An FEA model and a thermal network of the machine tool’s ram was established. The influence of boundary conditions on thermal error was studied to find out the boundary conditions that needn’t to be calculated precisely, reducing the time cost of the work. Superposition principle of heat sources was used in the FEA to get the simulation data of thermal error and temperature. A model based on the simulation data was established to predict the thermal error during the work process. An experiment was performed to verify the accuracy of the model. The result shows that the model accuracy is 87%. The method in this paper is expected to be used in engineering application.

Design of a Personalized Skin Grafting Methodology Using an Additive Biomanufacturing System Guided by 3D Photogrammetry

In this paper, the authors propose a novel method whereby a prescribed simulated skin graft is 3D printed, followed by the realization of a 3D model representation using an open-source software AutoDesk 123D Catch to reconstruct the entire simulated skin area. The methodology is photogrammetry, which measures the 3D model of a real-word object. Specifically, the principal algorithm of the photogrammetry is structure from motion (SfM) which provides a technique to reconstruct a 3D scene from a set of images collected using a digital camera. This is an efficient approach to reconstruct the burn depth compared to other non-intrusive 3D optical imaging modalities (laser scanning, optical coherence tomography). Initially, an artificial human hand with representative dimensions is designed using a CAD design program. Grooves with a step-like depth pattern are then incorporated into the design in order to simulate a skin burn wound depth map. Then, the *.stl format file of the virtually wounded artificial hand is extruded as a thermoplastic material, acrylonitrile butadiene styrene (ABS), using a commercial 3D printer. Next, images of the grooves representing different extents of burned injury are acquired by a digital camera from different directions with respect to the artificial hand. The images stored in a computer are then imported into AutoDesk 123D Catch to process the images, thereby yielding the 3D surface model of the simulated hand with a burn wound depth map. The output of the image processing is a 3D model file that represents the groove on the plastic object and thus the burned tissue area. One dimensional sliced sections of the designed model and reconstructed model are compared to evaluate the accuracy of the reconstruction methodology. Finally, the 3D CAD model is designed with a prescribed internal tissue scaffold structure and sent to the dedicated software of the 3D printing system to print the design of the virtual skin graft with biocompatible material poly-ε-caprolactone (PCL).

New Approaches for Improved Efficiency and Flexibility in Process Chains of Press Hardening

In the last decade, press hardening has become a fully established technology in both science and industry for the production of ultra-high-strength structural components, especially in the automotive industry. Beside the improvement of car performance such as safety and lightweight design, the production process is also one focus of trends in technology development in the field of press hardening. This paper presents an overview about alternative approaches for optimized process chains of press hardening, also including pre- and post-processing in addition to the actual forming and quenching process. Investigations on direct contact heating technology show new prospects regarding fast and flexible austenitization of blanks at compact device dimensions. By applying high speed impact cutting (HSIC) for trimming of press hardened parts, an alternative technology is available to substitute the slow and energy-intensive laser trimming in today’s press hardening lines. Combined with stroke-to-stroke control based on measuring of process-relevant parameters, a readjustment of the production line is possible in order to produce each part with individual, optimal process parameters to realize zero defect production of property-graded press hardened components with constant high part quality. Significant research in the field of press hardening was carried out at Fraunhofer Institute for Machine Tools and Forming Technology IWU, in the hot forming model process chain which enables the running of experiments under conditions similar to industrial scales. All practical tests were prepared by design of experiments and assisted by thermo-mechanical FE simulations.

Shell Element Formulation Based Finite Element Modeling, Analysis and Experimental Validation of Incremental Sheet Forming Process

This paper presents the effect of shell element formulations on the response parameters of incremental sheet metal forming process. In this work, computational time, profile prediction and thickness distribution are investigated by both finite element analysis and experimentally. The experimental results show that the thickness distribution is in good agreement with the results obtained with Belytschko-Tsay (BT) and Improved Flanagan-Belytschko (IFB) shell element formulations. These two shell element formulations do trade-off between computational time and accuracy. For more accurate results, the BT shell element formulation is better and for less computational time with good results, the IFB shell element is preferable. Finally, BT shell element formulation has been chosen for FE Analysis of ISF process in HyperWorks, since the results of thickness distribution and profile prediction is in better agreement with the experimental results as well as the computational time is less among the shell elements.

Prediction of Reliability in Friction Welded Dissimilar Joints by Weibull Distribution

The reliability of experimental data obtained in friction welded titanium alloy and stainless steel with copper interlayer by using various interlayer thicknesses and upset time are investigated using the maximum likelihood method for the estimation of the Weibull parameters of the results. The results indicate that among the various process parameters, interlayer thickness was significant. Further the reliability of the tensile strength was estimated using weibull distribution. Using this technique in conjunction with the experimental data, we can predict the output, in this case tensile strength more accurately and minimize their impact. Titanium alloy when directly bonded to stainless steel, improper bonding happens. Hence an interlayer in the form of copper is added to have successful joints.

A Method to Estimate Residual Stress in Metal Parts Made by Selective Laser Melting

Accurate evaluation of residual stresses in structures is very important because they play a crucial role in the mechanical performance of the components. As residual stresses can be introduced into mechanical components during various thermal or mechanical processes such as heat treatment, forming, welding and additive manufacturing. As an additive manufacturing method, selective laser melting (SLM) has become a powerful tool for the direct manufacturing of three dimensional nano-composite components with complex configurations directly from powders using 3D CAD data as a digital information source and energy in the form of a high-power laser beam. Therefore, the application of the SLM technology is necessary to manufacture Inconel 718 superalloy, which has been widely employed in industrial applications due to its remarkable properties. Hence, it is critical to measure and reduce the residual stress in the Inconel 718 parts formed by SLM due to rapid cooling and reheating. In this study, the process-induced residual stress in Inconel 718 parts produced by selective laser melting (SLM) has been investigated using the model established by Carlsson et al., which is an instrumented indentation technique based on the experimental correlation between the indentation characteristic and the residual stress. The samples were sectioned from an Inconel 718 block along its build direction, and subsequently prepared with general metallographic methods for Vickers indentation and measurements by optical microscopy. The residual stress on the scanning surface (Z-plane) and side surface (X-plane) at different build heights have been evaluated in micro-scale with the contact area, indentation hardness and the equai-biaxial residual stress and strain fields. The results show that the residual stress is unevenly distributed in the SLMed parts with some areas have an maximum absolute value around 350 MPa, about 30 percent of the yield strength of Inconel 718. The average residual stresses in the Z-plane and X-plane samples are tensile and compressive, respectively. Besides, the residual stress does not change significantly along the building direction of the part. Moreover, the Vickers hardness of the parts built with the SLM process is comparable to the literature, and the X-plane surface has a higher hardness than the Z-plane surface. The microstructures and texture evolution of the SLM processed Inconel 718 alloy are also investigated. The X-plane shows the columnar structure due to the large temperature gradient while the Z-plane presents the equiaxed structures. The random texture is shown in the SLM processed specimens.

Thermal Simulations for Cooling Rate Mapping in Electron Beam Additive Manufacturing

The powder-bed electron beam additive manufacturing (EBAM) process is a relatively new AM technology that utilizes a high-energy heat source to fabricate metallic parts in a layer by layer fashion by melting metal powder in selected regions. EBAM can be able to produce full density part and complicated components such as near-net-shape parts for medical implants and internal channels. However, the large variation in mechanical properties of AM build parts is an important issue that impedes the mass production ability of AM technology. It is known that the cooling rate in the melt pool directly related to the build part microstructure, which greatly influences the mechanical properties such as strength and hardness. And the cooling rate is correlated to the basic heat transport process physics in EBAM, which includes a moving heat source and rapid self-cooling process. Therefore, a better understanding of the thermal process of the EBAM process is necessary. In this study, a 3D thermal model, using a finite element method (FEM), was utilized for EBAM heat transport process simulations. The process temperature prediction offers information of the cooling rate during the heating-cooling cycle. The thermal model is applied to evaluate, for the case of Ti-6Al-4V in EBAM, the process parameter effects, such as the beam speed and power, on the temperature profile along the melt scan and the corresponding cooling rate characteristics. The relationship between cooling rates and process parameters is systematically investigated, through multiple simulations, by incorporating different combinations of process parameters into the thermal model. The beam scanning speed vs. beam power curves of constant cooling rates can be obtained from 3D surface plots (cooling rate vs. different process parameters), which may facilitate the process parameters selections and achieve consistent build part quality through controlling the cooling rate.

Tensile Properties of Polylactic Acid (PLA) Additive Manufactured Parts

In the advancement of Additive Manufacturing (AM) technologies, 3D desktop printers have become an accessible solution to address the current manufacturing practices for most industries and the general public. This study explores the effect default build parameters have on the tensile properties of additive manufactured parts by comparing the Young’s Modulus and tensile strength of polylactic acid (PLA) in the elastic region before and after the AM process through experiments and numerical simulations. The build parameters are specified via MakerBot Desktop — the file preparation software for the MakerBot Replicator 3D printer used to create the specimens tested herein. This work presents the tensile mechanical properties for specimens built using low infill rate, low layer resolution, and standard build speed and extrusion temperature to recreate the worst possible part quality attainable using MakerBot 3D Desktop printers. Using these build parameters results in a part with a hollow honeycomb interior structure, and due to its heterogeneous cross sectional area, experimental stress-strain curves do not accurately represent its physical response to tensile loading. Therefore in this case, an experimental-numerical study of the 3D printed specimens is performed, using the load-displacement experimental data acquired from tensile tests to calibrate the ANSYS Structural Mechanics simulations. The goal is to optimize the material properties in our simulation such that the equivalent strain magnitude matches the experiments. This is an approach to determine the experimental Young’s modulus of PLA additive manufactured parts where the AM process, heterogeneous structure, and size greatly influence the part strength. This is completed by studying the worst part quality possible first to better understand this effect. Tensile tests are performed using an ADMET 5603 Universal Test Machine (UTM) synched with a Correlated Solutions 3D Digital Image Correlation (DIC) system. A fine heterogeneous speckle pattern is sprayed on the specimens and used by the DIC system to obtain surface contours of deformation. This data is compared to the displacement fields in the finite element analysis (FEA) simulation of the specimen. When compared to the pre-manufacturing PLA, additive manufactured parts exposed that the post-processed stiffness of the material is increased when tested under this loading condition. The Poisson’s ratio for printed PLA was also noted to decrease when compared to pre-manufactured PLA, due to the larger longitudinal deformation compared to the transverse. Specimens failed by brittle fracture across the hex pattern, showing limited deformation and failing short after. The failure location based on the influence interior geometry has on failure showed that specimens failed by brittle fracture across the hex pattern, initiating fracture in the same region of all specimens.