In situ image processing for process parameter-build quality dependency of plasma transferred arc additive manufacturing

• Published in: International journal of advanced manufacturing technology Vol. 119; no. 11-12; pp. 7557 - 7577
• Main Authors: Gajbhiye, Raviraj V., Rojas, Jose Guadalupe Mercado, Waghmare, Prashant R., Qureshi, Ahmed Jawad
• Format: Journal Article
• Language: English
• Published: London Springer London 01.04.2022; Springer Nature B.V
• Subjects: Additive manufacturing; CAE) and Design; Component reliability; Computer-Aided Engineering (CAD; Deposition; Engineering; Feed rate; Flow stability; Geometry; High speed cameras; Image analysis; Image processing; Image quality; Industrial and Production Engineering; Manufacturing; Mechanical Engineering; Media Management; Metal powders; Multilayers; Near net shaping; Original Article; Plasma; Plasma currents; Plasma jets; Plasma transferred arc process; Process parameters; Stability analysis; Substrates; Surface properties; Tungsten carbide; Variance analysis; Analysis of variance (ANOVA); Image processing technique; Additive manufacturing; PTA-AM; High-speed imaging; Plasma transferred arc (PTA)
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-021-08643-6

Plasma transferred arc additive manufacturing (PTA-AM) is a high volume and high speed directed energy deposition (DED) technique that uses metal powder as a feedstock material to build up 3-dimensional (3D) components, layer by layer. In the PTA-AM process, the quality of the build is a function of several parameters, including but not limited to the powder feed rate, the torch travel speed, stand-off distance (SOD), and plasma arc current. Unlike the laser DED systems, the melt pool and laser visualization are not possible through inline imaging systems. These processes have difficulty sustaining uniform evenly distributed powder flow between the nozzle and the substrate. For the constant operating parameters, the process requires in-depth attention to inspect the quality and geometrical features of bead deposition, which may alter due to continuous mechanical connection with the deposition process. Surprisingly, no work is reported in plasma-based DED systems to determine the process parameter influence on build quality using image processing technique. Therefore, in this work, high-speed imaging is employed to analyze and quantify the process parameters of the PTA-AM and their effects on the bead quality and geometry. The experiments of single-track multi-layer bead deposition of Tungsten Carbide (WC)-reinforced NiCrBSi composites are carried out and recorded with a high-speed camera. Based on captured 2D image data, the image processing techniques are developed using Matlab and Image-Pro premier software. The proposed techniques optimize and validate the SOD to print the stable bead and maintain the higher powder catchment efficiency at the deposition zone. It is also used to assess the process stability (uniform powder flow) and the impact of powder flow variation on the bead height. The powder stream distribution, the statistical dispersion of powder particle count, and their effect on the build quality are described based on powder flow images, using an image processing approach. The significance of relationships between bead height variation and powder particle counts is analyzed using statistical methodologies such as analysis of variance (ANOVA) and two-tailed t -test. The effect of PTA-AM process parameters on bead geometry reveals that the plasma arc current and powder particle count significantly affect the bead geometry. This work validates the benefit of using 7-mm SOD for improving part quality. Using image analysis, the uniformity of powder flow is evaluated. In addition, the finding shows that the increment in plasma current by 40% increases the mean bead width by 85% and decreases its height by 24%. By determining the effect of processing parameters on the bead quality and geometry, this PTA-AM process can build near-net-shape parts and increase the reliability and reproducibility of the process. Thus, the proposed in situ image processing methods can be adopted in the PTA-AM process to analyze manufactured parts’ quality and characterize the process parameters responsible for the deviation in surface quality and dimensional geometry.