Modelling of dimensional accuracy and surface roughness in micro-plasma transferred arc additive manufacturing of metallic materials

Abstract

Keywords: Additive manufacturing; Micro-plasma; Theoretical models; Dimensional deviation; Dimensional tolerance; Surface Roughness American Society for Testing and Materials (ASTM) defines additive manufacturing (AM) as a “process of joining materials, usually layer upon layer as opposed to subtractive manufacturing process, to make an object using data of its 3D-CAD model”. AM is one of key enablers of Industry 4.0. Directed energy deposition (DED) is one category of AM processes that uses a focused thermal energy source supplied through a nozzle to fuse the different deposited layers by melting them during their deposition. It manufactures a product by stacking the deposition either vertically or horizontally or both horizontally and vertically which results in multi-layer single-track, single-layer multi-track, and multi-layer multi-track deposition respectively. The DED processes can be classified into two categories according to type of energy source used for melting of the deposition materials: (i) DED processes using high-energy beam such as laser and electron beam, and (ii) DED processes using arc such as gas metal arc (GMA), gas tungsten arc (GTA), and plasma arc (PA). A DED process has unique multi-functional capabilities of freeform manufacturing of a complicated product, coating, cladding, surface texturing, surface alloying, repairing, remanufacturing, rapid prototyping, and rapid tooling for various industrial applications. It is typically used in biomedical, post-injury rehabilitation, aerospace, automobile, marine, sports equipment, power generation, gas turbines, oil, and gas extraction applications. A novel DED process named as micro-plasma transferred arc metal additive manufacturing (µ-PTAMAM) process has been developed at IIT Indore for the metallic materials focussing on their meso-sized AM applications. This process has unique capabilities such as higher energy density, better arc stability even at low current (maximum value of DC current supplied 20 A), and ability to supply deposition material in powder, wire, powder and wire simultaneously, and particulate form enabling its use for a wide range of materials. Layered deposition in any AM process leads to generation of uneven wavy surface on the additively manufactured product. This becomes more problematic for its functional surfaces. Other factors that contribute to surface roughness include accumulations of semi molten powder particles adhered to the surfaces, solidification lines developed due to melt pools, molten pool overflow over the layer beneath, and formation of periodic menisci. This necessitates use of many post-manufacturing processes such as polishing, shot peening, sand blasting, heat treatment, and finishing thus increasing production time and cost. Minimization of dimensional inaccuracy and surface roughness developed during an AM process is always a challenge. Researchers have been working on this aspect and various techniques such as correlating process parameter with multi-physics of deposition geometry, statistical modelling, curve fitting techniques, etc. have been developed and/or used to predict and control dimensional inaccuracy and surface roughness. Liu and Li (2004) developed analytical model relating parameters of laser-based DED process namely laser power, travel speed, and laser beam orientation with the deposition track geometry. They used developed theoretical model to control real-time input parameters based on dimension of the previously deposited layer. Their study yielded a defect-free deposition with an improved surface quality and accuracy. Jhavar et al. (2014) found elliptical arc most suitable to describe geometry of single-track deposition by µ PTAMAM process and developed theoretical model for width and height of single-track deposition and overlap distance between two the adjacent tracks. They also reported that surface waviness can be minimized if area of the valley between the two adjacent deposition tracks is equal to area of overlap between them. Nikam et al. (2016) developed thermal model in terms of micro-plasma power, volumetric material deposition rate, worktable travel speed, and thermal properties of the deposition and substrate materials to predict width and height of single-track deposition by μ-PTAAM process. Their study also considered various forces acting during melt pool solidification. Xiong et al. (2018) experimentally examined impact of parameter of GMA-based DED process namely thermal gradient, wire feed rate, travel speed on the surface roughness. They found that surface finish can be enhanced by reducing inter-layer temperature, by increasing wire feed rate, and maintaining constant ratio of wire feed rate and travel speed. Hu et al. (2019) considered parabolic profile for single track and circular or parabolic profile for the adjacent tracks at different overlapping conditions to develop model for optimum overlapping distance for multi-track deposition by GMA based DED process. Their study found that optimum overlapping distance varied with the deposition geometry and found it to lie from 0.63 to 0.77 for different deposition geometry. They reported that use of optimal overlapping gives smooth surface on the top layer. Xia et al. (2021) used surface roughness data, gathered using laser sensor for wire arc AM process, to train different machine learning algorithms namely adaptive neuro-fuzzy inference system (ANFIS), extreme learning machine (ELM), and support vector regression (SVR) for prediction of surface roughness and found that ANFIS is better than other models