Minimizing the Casting Defects in High Pressure Die Casting Using Taguchi Analysis

Abstract

High-Pressure Die Casting (HPDC) is one of the major production processes of automotive industry, widely used to manufacture geometrically complex nonferrous castings. The mechanical strength and microstructure of HPDC-manufactured products changes with variation in several process parameters such as injection pressure, molten temperature, 1 st and 2 nd stage plunger velocity, cooling temperature etc.Since these process parameters directly affect casting quality, therefore, their optimum combination is needed to maximize productivity of the process and minimize casting defects such as porosity, pinholes, blowholes, etc. Hence, to tackle this problem, an approach is presented in this paper that minimizes the major casting defect, i.e., porosity, in the HPDC process by optimizing parameters through Design of Experiments (DOE) in combination with a Taguchi Analysis. The results obtained showed that cooling time, injection pressure, and 2 nd stage plunger velocity have a major influence on the response factor (density of the cast part). It was further concluded that, by using: a 178 bar injection pressure, 665 °C molten temperature, 5 seconds of cooling time, 210 °C mold temperature, 0.20 m•s −1 1 st stage plunger velocity, and 6.0 m•s −1 2 nd stage plunger velocity, the rejection rate of the selected part due to porosity was reduced by 61%.

Machine Learning Methods for diagnosing the causes of die-casting defects

“…High-pressure die casting is a process that requires the non-destructive testing of critical components using modern technology and advanced data analysis methods. Excellent examples are the Pareto analyses which have been conducted, in which critical areas were identified by performing a case study and prioritizing casting defects, for which a causal analysis was then performed (Bharambe et al, 2023;Govindarao et al, 2022) or an analysis using ANOVA and Taguchi analysis (Tariq et al, 2021). Analyses using artificial intelligence methods such as image analysis (Parlak & Emel, 2023) or artificial neural networks (De-Jian & Young-Peng, 2021), which are very relevant in modern digital reality, are also conducted.…”

Archives of Foundry Engineering

“…Regarding the basis of the Industry 4.0 concept, considering the level of advancement of foundry process automatization and their data collection capabilities, it can be noted that high-pressure die-casting foundries (HPDC) probably represent the highest level [7]. Additionally, die casting is one of the main manufacturing processes in the automotive industry [9], as it requires high precision in the production of geometrically complex non-ferrous castings such as aluminum [10], [11]. It is characterized by the highest quality part production with the highest dimensional accuracy and reduced cost per part [12].…”

“…Selected castings, such as safety-critical automotive components, are 100% quality controlled in the further stages of production, as in their case the consequences of failure are severe, and the additional cost associated with quality control is high but also justified [2]. However, this process is highly dependent on individual process parameters affecting casting quality, so it is necessary to obtain their optimal combinations to minimize the formation of casting defects [10]. It should also be taken into account that the die casting process is considered to be inflexible, with low adaptability to changes introduced in the technological properties of the product designs implemented to evolve and progress the process [15].…”