Abstract: To address issues such as porosity and shrinkage cavities during high-pressure die casting of aluminum alloy differentials, high-pressure die casting process of differential housing was numerically simulated using finite element software ProCAST. Flow field, temperature field, and velocity field were solved to predict die casting defects, and die casting process was optimized through orthogonal experiments. Results show that air entrapment occurs in the bottom cylinder and side support areas during filling; a large temperature gradient in molten metal during solidification leads to uneven solidification and premature solidification of gating system, resulting in shrinkage cavities upon complete solidification. Considering combined effects of die casting parameters on solidification time, air entrapment, and shrinkage cavity volume, optimal die casting parameters were determined to be a pouring temperature of 650 ℃, a mold temperature of 220 ℃, and an injection speed of 5 m/s. Die casting experiments verified reliability of these parameters.
Differential housing is a large and complex housing component, characterized by its large size, uneven wall thickness, and complex overall structure. It is primarily used in power transmission of four-wheel-drive SUVs. Vehicles operate with high power output and large starting torque, requiring differential housings to withstand harsh service environments such as heavy loads, high impacts, and high stresses. Traditional differential housings, typically made of ductile iron or gravity casting, suffer from problems such as high overall weight, low dimensional accuracy, long production cycles, and large gating gates.
Currently, differential housings have undergone a lightweight structural design using ribs instead of solid components, effectively reducing their weight while maintaining structural strength. High-strength aluminum alloys have been used instead of traditional ductile iron for related components, this dual weight reduction through materials and structure satisfies requirements for lightweight differential housing design. However, this lightweight design presents challenges in coordinated control of differential housing shape and properties. Therefore, seeking advanced processes for efficient and high-quality differential housing production has become a key research focus.
This paper proposes a high-pressure die-casting process for aluminum alloy differential housings, using lightweight A380 aluminum alloy and advanced manufacturing technology. However, differential housing has a complex structure, and during die casting process, eddy currents and uneven metal feeding can occur, leading to defects such as porosity and shrinkage cavities inside casting. Therefore, this paper establishes a finite element model of high-pressure die-cast differential housing, numerically simulates die-casting process, uses solution of flow field, temperature field, and velocity field to predict casting defects. Through orthogonal experiments and statistical data analysis, effects of pouring temperature, mold temperature, and injection speed on filling quality of casting are studied. Reliability of optimal process parameters is verified through die-casting experiments, providing guidance for lightweight production of differential housings.

Designed three-dimensional model of differential housing (Figure 1a) is imported into ProCAST simulation software as an igs format file. Differential is meshed in mesh module of ProCAST. Considering different wall thicknesses of casting, gating system, and overflow system, different mesh sizes are used for casting, gating system, overflow system, and mold. Casting mesh size was set to 1.5 mm, gating and overflow systems mesh size to 3 mm, and mold mesh size to 10 mm. Resulting mesh size was approximately 17.5 million elements, and finite element model is shown in Figure 1b.
Table 1 Chemical component of A380 Al alloy wB /%

 

Fig. 1 FEM of high pressure die casting
Differential housing material was selected as A380 aluminum alloy, with chemical composition shown in Table 1. Thermal properties of A380 aluminum alloy were provided by software, with a solidus temperature of 508 ℃ and a liquidus temperature of 589 ℃. Mold material was selected as H13 alloy steel. In die-casting numerical simulation, die-casting mold only served as a heat exchanger and did not participate in flow field calculation. Various heat transfer phenomena are involved in die-casting process, mainly including heat conduction between casting and mold, between molds, heat flow between air and outer surface of mold. Heat transfer coefficient between molds was set to 1000 W/(m²·K), heat transfer coefficient between mold and air was set to 10 W/(m²·K). Heat transfer coefficient between casting and mold was expressed as a curve (Figure 2). High Pressure Die Casting Initial Process Parameters: Pouring temperature 660 ℃, mold temperature 220 ℃, injection speed 4 m/s.

 

Fig. 2 Heat transfer coefficient of A380-H13

Casting filling process is shown in Figure 3. As can be seen from figure, when filling rate is 30%, molten metal in central gating begins to fill cavity through ingate, while molten metal in side gatings is far from punch and has not yet reached ingate. When filling rate is 50%, because molten metal in central gating fills faster, it flows back into right cylindrical area of cavity after reaching far gating and interacting with molten metal in right gating, potentially causing more defects in right cylindrical area. When filling rate is 70%, molten metal in central gating completes filling of middle of cavity, while molten metal in side gatings fills side supports. When filling rate reaches 80% to 90%, filling process at upper right support becomes more chaotic. This is last point where molten metal reaches mold cavity, requiring a longer filling distance. At this stage, molten metal loses a significant amount of heat, reducing its fluidity. Simultaneously, due to unfilled areas in thicker parts of casting wall, molten metal flows back after reaching end of cavity, causing air entrapment in thicker areas. This backflow and air entrapment can lead to porosity defects in later stages of casting solidification.

 

Fig. 3 Filling process of liquid metal
Filling time and temperature distribution of molten metal are shown in Figure 4. Based on time distribution of filling process in Figure 4a, differential housing filling process can be divided into four parts sequentially: blue, green, yellow, and red.

 

Fig. 4 Distributions of time and temperature during mold filling
Blue area mainly includes sprue and area near sprue of casting; green area mainly includes bottom of two side supports; yellow area mainly includes platform of casting and upper right support, which is farther from ingate; red area is mostly located in overflow groove of casting, and a small part is located in the area where molten metal flows back at wall thickness after reaching end of cavity. Overall, filling sequence of differential housing mainly follows rule of from near to far, but upper right support of casting is filled later, resulting in poor filling quality at this point. Figure 4b shows temperature distribution during casting filling process. During filling process, molten metal in cavity remains at a high temperature, with the lowest temperature area in cavity being bottom cylinder of casting and two side supports, at approximately 625 ℃. At the end of filling, temperature of molten metal in cavity is above A380 liquidus line (598 ℃), and no premature solidification occurs. Throughout filling process, temperature of molten metal in mold cavity is relatively high. Temperature of molten metal at far end of gating system is slightly lower than that near gating system, but temperature gradient is small, the overall distribution is reasonable, and molten metal has good fluidity.

Fig. 5 shows temperature distribution and solid fraction changes during casting solidification process. As can be seen from Fig. 5a, when solidification reaches 60%, temperature decreases rapidly in thin-walled areas of casting, indicating a coexistence of solid and liquid phases. In contrast, temperature decreases slowly in thick-walled areas, which remain in liquid phase. When solidification reaches 80%, temperature of molten metal in most areas of mold cavity has reached below solidus line. However, temperature level of molten metal in thick-walled areas remains high, maintaining a large temperature gradient (90 ℃) with surrounding areas. This can easily lead to thermal stress concentration, resulting in hot cracks and shrinkage cavities in thick-walled areas of casting.

 

Fig. 5 Temperature distribution and solid fraction during casting solidification
Solid fraction variation during casting solidification process is shown in Figure 5b. Solid fraction distribution follows a pattern of high solid fraction in thin-walled areas and low solid fraction in thick-walled areas. Solid fraction at upper right support and in thick-walled areas is lower than that in surrounding areas. When solid fraction in surrounding areas reaches 0.8, solid fraction in thick-walled areas is only 0.5, even lower than that in ingate. According to metal solidification theory, when solid fraction reaches a certain level, dendrite growth forms a closed framework, cutting off pressure transmission and liquid phase feeding channels. This uneven solidification leads to formation of isolated liquid phase regions during solidification, ultimately resulting in shrinkage porosity and shrinkage defects in thick-walled areas of casting.
Casting solidification time is shown in Figure 6. As can be seen from figure, solidification sequence follows a pattern from far to near. Thin-walled areas and far gate solidify first, followed by overflow groove and supports on both sides of casting, and thick-walled areas solidify last. This is because thin-walled areas, due to their lower temperature and sufficient heat exchange with mold, solidify preferentially, while thick-walled areas, with their greater thickness and slower heat transfer, solidify last.

 

Fig. 6 Solidification time of casting

Distribution of air entrapment inside casting is shown in Figure 7a. As can be seen from figure, a significant amount of gas exists in overflow channel of casting, effectively removing gas and demonstrating rationality of overflow system design. Average air entrapment rate inside casting is approximately 0.0003 g/cm³, but air entrapment rate in upper right support and left support areas is approximately 0.0006 g/cm³. This is because air entrapment and backflow occurred in this area, leading to poor filling in this region.
Distribution of shrinkage cavities inside casting is shown in Figure 7b. As can be seen from figure, defect distribution is consistent with analysis of casting solidification process. Most of shrinkage cavities inside casting are concentrated in the thickest wall areas, with a smaller number located in upper right support and bottom cylinder. Shrinkage cavity volume of casting was measured using VE software, and defect volume was approximately 3.146 cm³. Since upper right support and bottom cylinder of casting are areas with high quality requirements, shrinkage cavity defects are not permitted. Therefore, it is necessary to optimize die-casting process to eliminate shrinkage cavity defects.

 

Fig. 7 Prediction results of defect

This study selected pouring temperature, mold temperature, and injection speed as main influencing factors in orthogonal experiment. A factor-level table for orthogonal experiment was designed, as shown in Table 2.
Table 2 Orthogonal test factor level table

Orthogonal experimental design was a three-factor, three-level experiment. According to rules for using orthogonal tables, L9(33) orthogonal table was selected, and a total of 9 orthogonal experiments were conducted. Numerical simulation analysis was performed on each experimental scheme in turn. Solidification time, shrinkage cavity volume, and air entrapment volume in simulation results were used as experimental results. Experimental results for each group are shown in Table 3.
Table 3 Orthogonal test plan

To determine order of influence and variation law of three factors on solidification time, range analysis was performed on simulation results. Table 4 shows range analysis table of solidification time, and Figure 8 shows main effect diagram of die casting process parameters on mean solidification time. According to Table 4, degree of influence of die casting process parameters on solidification time is: pouring temperature (A) > mold temperature (B) > injection speed (C). For solidification time, pouring temperature and mold temperature have a significant impact, while injection speed has a limited effect. This is because excessively high temperature parameters lead to a longer time for molten metal to reach solidus, while injection speed has a relatively small impact on temperature change of molten metal. Therefore, within a reasonable range of process parameters, a lower temperature level can be selected to improve die casting production efficiency. As shown in Figure 8, when pouring temperature is selected as A1, mold temperature as B1, and injection speed as C3, solidification time of casting is minimized. Therefore, when solidification time is used as single criterion for evaluating casting quality, optimal die casting process parameters are A1B1C3.
Table 4 Solidification time range analysis table

 

Fig. 8 Main effect diagram of die-casting process parameters on mean value of solidification time
To determine order of importance and variation of three factors' influence on amount of air trapped inside casting, a range analysis was performed on simulation results. Table 5 shows range analysis table for air trapped volume, and Figure 9 shows main effect diagram of die casting process parameters on mean air trapped volume. According to Table 5, degree of influence of die casting process parameters on amount of air trapped is: pouring temperature (A) > injection speed (C) > mold temperature (B). For air entrapment, pouring temperature and injection speed have a significant impact, while mold temperature has a limited effect. This is related to fact that a reasonable pouring temperature and injection speed can provide good molten metal flowability. Within a certain range, the higher pouring temperature and the faster injection speed, the better flowability. As shown in Figure 9, when pouring temperature is A3, mold temperature is B3, and injection speed is C3, air entrapment in casting is minimized. Therefore, when air entrapment is used as single evaluation criterion for casting quality, optimal die-casting process parameters are A3B3C3.
Table 5 Air entrainment range analysis table

 

Fig. 9 Main effect diagram of die-casting process parameters on mean value of air entrainment
To determine order of importance and variation of three factors' influence on internal shrinkage cavity volume of casting, range analysis was performed on simulation results. Table 6 shows range analysis table for shrinkage cavity volume. Figure 10 shows main effect diagram of die-casting process parameters on mean shrinkage cavity volume. According to Table 6, degree of influence of die-casting process parameters on shrinkage cavity volume is: mold temperature (B) > pouring temperature (A) > injection speed (C). Pouring temperature and mold temperature have a significant impact on shrinkage rate, while injection speed has a limited effect. This is because shrinkage defects inside casting are caused by insufficient feeding of molten metal during solidification. Optimal temperature parameters can improve solidification process, while injection speed has a relatively small effect on improving it. As shown in Figure 10, when pouring temperature is A2, mold temperature is B3, and injection speed is C2, shrinkage volume of casting is minimized. Therefore, when using shrinkage volume as sole criterion for casting quality, optimal die-casting process parameters are A2B3C2.

 

Fig. 10 Main effect diagram of die-casting process parameters on mean value of shrinkage pore volume.
Table 6 Total shrinkage volume range analysis table

Based on above analysis, when solidification time is used as evaluation index, optimal process parameters are: casting temperature 630 ℃, mold temperature 180 ℃, and injection speed 5 m/s; when air entrainment is used as evaluation index, optimal process parameters are: casting temperature 670 ℃, mold temperature 220 ℃, and injection speed 5 m/s; when shrinkage pore volume is used as evaluation index, optimal process parameters are: casting temperature 650 ℃, mold temperature 220 ℃, and injection speed 4 m/s. According to selection principle of actual production process, it is necessary to improve production efficiency while ensuring product quality. Therefore, order of importance of quality indicators in orthogonal experiment is: shrinkage pore volume, air entrainment, and solidification time. After comprehensive analysis, optimal die-casting process parameters for differential housing are finally determined to be: casting temperature 650 ℃, mold temperature 220 ℃, and injection speed 5 m/s.

High-pressure die-casting of differentials was conducted using a Bühler 1300 t die-casting machine. Optimal die-casting process parameters were selected: pouring temperature 650 ℃, mold temperature 220 ℃, and injection speed 5 m/s. Two high-pressure die-cast differential parts were obtained, as shown in Figure 11. After trimming to remove overflow grooves, venting grooves, and flow channels, net weight of parts was 4.4 kg. Differential housing surface was smooth, with clear contours, free from defects such as cracks, flash, and cold shuts, indicating good quality. X-ray flaw detection was performed on upper and lower brackets and bottom cylinder of differential, which have high quality requirements. X-ray images showed no obvious porosity or shrinkage defects inside castings, verifying that optimized die-casting process met production requirements.

 

Fig. 11 Differential housing die-casting physical map and X-ray results
Mechanical property tests were performed on upper and lower brackets and bottom cylinder of differential housing casting, as shown in Table 7. Table shows that mechanical properties differ at different locations, but generally follow pattern that samples near gating system have better mechanical properties than those far from gating system, and samples with thinner walls have better mechanical properties than those with thicker walls. During casting process, as extrusion pressure gradually increases, primary grain size decreases, significantly improving mechanical properties of casting. As flow field analysis above shows, upper support and bottom cylinder of sample are located near gating system, where pressure is higher and mechanical properties are better. Conversely, lower support is located far from gating system, where pressure is lost during transmission, resulting in lower mechanical properties. However, the overall mechanical properties of differential housing casting are good and meet production requirements.
Table 7 Mechanical properties of differential housing castings

(1) Based on finite element model of high-pressure die-cast differential housing casting established in ProCast software, high-pressure die-casting filling and solidification process was simulated. By solving flow field, temperature field, and velocity field, location and cause of internal defects in casting were predicted.
(2) During differential housing filling process, air entrapment occurred in the bottom cylinder and side support areas, especially in upper right support and left support areas, where air entrapment volume exceeded average by 0.0003 g/cm³. During solidification, maximum internal temperature gradient of casting was 90 ℃, leading to uneven solidification of molten metal and resulting in multiple isolated liquid phase zones in localized areas. Therefore, shrinkage cavities appeared in areas with thicker casting walls and at upper right support and bottom cylinder, with a defect volume of approximately 3.146 cm³. Optimization of die-casting process is needed to eliminate air entrapment and shrinkage cavities.
(3) Combining finite element orthogonal experiments and high-pressure die-casting experiments, optimal die-casting process is: pouring temperature 650 ℃, mold temperature 220 ℃, and injection speed 5 m/s.