Abstract: For a die-cast front housing of a new energy vehicle, numerical simulation was applied to predict locations of potential defects such as porosity, shrinkage cavities, and cold shuts by comparing flow, filling, and solidification states of molten aluminum under different gating systems. Targeted optimizations were then made in design. Simulation results showed that an unreasonable gating system design led to problems such as unbalanced filling and cold material coiling at the end. Through iterative optimization of gating system, stable and orderly filling was achieved, significantly reducing risk of porosity and cold shuts at the end of flywheel cavity and water channel. Extrusion pin, movable block, cooling system designed based on simulation effectively controlled shrinkage cavities in thick-walled areas. Trial production verification showed that internal quality of casting was good, with no exposed porosity on key sealing surfaces and bearing holes after machining. Leakage tests met stringent standards, confirming reliability of process.
Aluminum alloys, due to their low density, have strength properties similar to gray cast iron but higher toughness, and excellent casting properties, making them suitable for forming complex thin-walled automotive parts. Meanwhile, aluminum alloys can be significantly strengthened with alloying elements to improve their strength. Furthermore, their lightweight and excellent heat dissipation properties make them well-suited for components such as gearbox housings, transmission housings, and motor housings operating in harsh environments. Most importantly, expanding application of aluminum alloys can significantly reduce vehicle weight, a crucial need in fiercely competitive automotive industry. Therefore, with development of new energy vehicles, application of aluminum alloys in automotive industry is becoming increasingly widespread.
Front housing, as one of the most important components in transmission of new energy hybrid vehicles, presents challenges in die-casting technology development due to its large size, complex shape, difficult molding, special working environment, high mechanical strength requirements. Numerical simulation of die-casting filling and solidification can effectively predict various defects, their size, and location, potentially occurring in casting during die-casting process design stage. This optimizes casting process design, reduces development difficulty, ensures casting quality, shortens trial production cycle, and lowers production costs. This study, focusing on complex structure, extremely high sealing requirements, and multiple thick-walled hot spots of front housing, uses numerical simulation to address key challenges of balanced filling of multiple gating systems and coordinated control of local hot spots.

Structure of front shell part of new energy vehicle is shown in Figure 1. Its outline dimensions are 609mm * 326mm * 581mm, its weight is 16.215kg, its base wall thickness is 4.2mm, and its projected area is 218398mm². Casting material is aluminum alloy, grade ADC12. This casting has several thick-walled hot spots, which are prone to shrinkage cavities. Product surface must be free of burrs, flash, and die-casting defects. All dimensions must conform to drawings and assembly requirements. After machining, one shrinkage cavity of Φ0.5mm-1.0mm is allowed on sealing surface, and two shrinkage cavities smaller than Φ0.5mm are allowed. No exposed shrinkage cavities are allowed in bearing holes after machining. Furthermore, product requires friction welding to form water channels, requiring high leakage control. Product leakage requirements: Oil channel 400kPa, leakage ≤24cc/min; Water channel 300kPa, leakage ≤1cc/min; Flywheel cavity/gear cavity/differential cavity 100kPa, leakage ≤16cc/min.

 

Figure 1 Front housing parts

Based on analysis of product structure and characteristics of die casting process, approximate gating positions and parting schemes are determined as shown in Figures 2 and 3, respectively.

 

According to parting scheme in Figure 3, there are two locations where pre-casting is not possible due to inability to demold: one is water outlet and adjacent threaded hole in water channel cavity, which are perpendicular to direction of movement of lower slide block, as shown in Figure 4(a); the other is bearing hole in differential cavity, which is perpendicular to direction of movement of side slide block, as shown in Figure 5(a).

 

Figure 4 Drain outlet structure and optimization scheme
Because pre-casting is not possible, these two locations have thicker walls. The thicker wall, the longer solidification time. As solidification progresses, feeding channels in these thicker areas are cut off because surrounding areas have already solidified. During subsequent solidification, shrinkage caused by phase transformation cannot be compensated, resulting in shrinkage cavities. These shrinkage cavities pose a significant risk of exposed pores and leakage through adjacent threaded holes. Therefore, improvements are needed in mold structure and die-casting process. An extrusion pin is added to side of water outlet, as shown in Figure 4(b), and differential cavity is designed as a movable block structure, as shown in Figure 5(b).

 

Figure 5 Differential cavity structure and optimization scheme
Extrusion pin essentially acts as a "feeding station" in thicker wall locations. During solidification, a hydraulic cylinder forces metal material from "feeding station" in, achieving feeding and eliminating shrinkage cavities. Its working principle is shown in Figure 6.

 

Figure 6 Working principle of extrusion pin
Similar to inserts, movable block is placed in mold before mold closing, as shown in Figure 7. After injection filling, it is encased in molten metal and solidifies, becoming an integral part of product. Difference is that movable block needs to be removed from product for reuse, as shown in Figure 8, while insert remains in product.

 

Figure 7. Placement of loose block on fixed mold

 

Figure 8. Pressing out a movable block
A movable block is generally used in areas where product cannot be demolded. In this product, direction of differential bearing hole is inconsistent with direction of slider's movement, making pre-casting impossible. Using a movable block reduces casting wall thickness and subsequent machining allowance, as shown in Figure 9. Simultaneously, because movable block temperature is approximately 100℃-120℃, far lower than molten metal temperature, it acts as a "chill," accelerating solidification speed of bearing hole, increasing density of bearing hole surface layer, and ensuring internal and external quality of this area.

 

Figure 9 Influence of workpieces on product casting wall thickness and machining allowance
Based on product characteristics and above analysis, gate location and gate type were roughly determined. Two different gating schemes were initially pre-designed, as shown in Figure 10. Table 1 compares specific parameters of two gating schemes for front shell.
Table 1. Front Shell Casting Scheme

 

Figure 10 Two pre-casting schemes

MAGMA simulation software was used to perform filling simulation analysis on product. H13 mold steel was selected as mold material, ADC12 as die-casting material, and molten aluminum temperature was set to 670℃, while mold temperature was set to 180℃. Injection process parameters are shown in Table 2.
Table 2 Simulation Parameter Settings

Figure 11 shows molten metal tracking results during filling process of Scheme 1 (molten metal tracking mainly reflects final filling position of molten metal passing through each ingate). It can be seen that filling speed of middle gate is slow, while side gates are fast. This causes molten aluminum and gas in the middle gate to be trapped in the middle of product, preventing cold material and gas at the beginning of filling from escaping. Therefore, there is a greater risk of cold shuts and porosity at the end of flywheel cavity filling, as shown in Figure 11(b). In addition, filling time in differential cavity and water channel is long, resulting in excessively low aluminum temperature and formation of cold material. This cold material cannot be discharged, leading to failure to fuse and formation of cold shuts, resulting in a greater risk of cold shuts and porosity, as shown in Figures 11(e) and 11(f).

 

Figure 11: Liquid molten material tracking results during filling process of Scheme 1 (different colors represent material flow from different gates)
Figure 12 shows gas pressure, air contact time (reflecting severity of cold material), and filling time after filling of Scheme 1 is completed. It can be seen that areas with high gas pressure and severe cold material are mainly concentrated at the end of flywheel cavity filling and friction welding positions of differential cavity and water channel, consistent with molten metal tracking analysis results.

 

Figure 12 Simulation results of filling process of Scheme 1
Based on above analysis, main problems with Scheme 1 are that side runners fill too quickly and differential cavity region fills too late. Therefore, side runners were eliminated to slow down filling of aluminum liquid on both sides. Simultaneously, a runner was added to differential cavity to optimize filling sequence and temperature field in differential cavity region, resulting in Scheme 2's gating system. Molten aluminum tracking results during filling process are shown in Figure 13. It can be seen that aluminum liquid passing through several runners maintains a relatively parallel filling, which is a significant improvement compared to Scheme 1. However, aluminum liquid on the left side still fills slightly faster, as shown in Figure 13(b), requiring further optimization.

 

Figure 13: Liquid molten material tracking results during filling process of Scheme 2 (different colors represent material flow from different gates)
Figure 14 shows gas pressure, air contact time, and filling time results after Scheme 2's filling is completed. As can be seen, there is a significant improvement compared to Scheme 1: peak air pressure at the end of flywheel cavity filling position decreases from approximately 5300 mbar in Scheme 1 to approximately 3000 mbar in Scheme 2; peak air contact time decreases from approximately 590 ms in Scheme 1 to approximately 300 ms in Scheme 2. Peak air pressure at water channel orifice decreases from approximately 1900 mbar in Scheme 1 to approximately 1700 mbar in Scheme 2. Although peak air contact time changes little between two schemes, Scheme 2 reduces filling time by approximately 50% compared to Scheme 1, from approximately 100 ms in Scheme 1 to approximately 80 ms in Scheme 2. However, based on the air contact time results, although range of cold material at water channel outlet is reduced, cold material is still relatively close to water channel outlet, and risk of cold shut-off due to air bubbles at outlet remains relatively high, requiring further improvement.

 

Figure 14 Simulation results during filling process of Scheme 2
Based on analysis of two schemes above, scheme was further optimized: flow rate of third gating channel on the left was reduced to slow down filling of aluminum liquid on the left; a 4mm bridging was added to middle through-hole to ensure smooth filling of aluminum liquid in the middle; an additional gating channel was added to differential chamber to improve filling of cold material near water outlet. Simultaneously, an overflow system was arranged at the end of filling process and at convergence point of multiple gating channels. After multiple rounds of iterative optimization and design simulation, Scheme 3 was designed. Material tracking simulation during filling process is shown in Figure 15. Results show stable filling and significant optimization effect.

 

Figure 15 Material tracking results during filling process of Scheme 3 (different colors represent material flow from different gates)
Figure 16 shows gas pressure, air contact time, and filling time after filling is completed in Scheme 3. It can be seen that compared to the first two schemes, optimization effect is significant: peak gas pressure of the entire product decreased from approximately 5300 mbar in Scheme 1 to approximately 70 mbar in Scheme 3; peak air contact time decreased from approximately 590 ms in Scheme 1 to approximately 200 ms in Scheme 3; and filling time at water outlet decreased from approximately 100 ms in Scheme 1 to approximately 50 ms.

 

Figure 16 Simulation results during filling process of Scheme 3
Figure 17 shows solidification simulation results for Scheme 3. It can be seen that the entire solidification process lasts 56.81 seconds. Final solidification locations are mainly runner and areas with thicker product walls near gate, as indicated by red circles in Figure 17. These locations have the highest risk of shrinkage cavities. Based on simulation results, during mold development, appropriately arranging cooling in areas with long solidification times and increasing cooling rate at these locations can ensure product's appearance and internal quality while shortening production cycle.

 

Figure 17 Simulation diagram of solidification process in mold direction of Scheme 3

 

Figure 18 Simulation diagram of moving mold direction during solidification process of Scheme 3
Based on solidification simulation analysis results in Figures 17 and 18, direct cooling is arranged at runner A, primarily to shorten mold cooling time, reduce production cycle, and minimize deformation caused by runner shrinkage. Ninety high-pressure cooling points are arranged around thick-walled areas such as B, C, D, E, and F. Mold cooling layout is shown in Figure 19.

 

Figure 19 Mold cooling water system layout
Cooling should be applied as close to mold surface as possible to better dissipate heat and improve cooling capacity. However, this requires a certain mold wall thickness to ensure mold strength and lifespan; typically, wall thickness to mold surface is 20mm-25mm. For areas with high cooling capacity requirements, it's necessary to sacrifice mold wall thickness to guarantee cooling performance. In this case, mold can be inlaid. As shown in areas B and C of Figure 20, an inlaid structure is used to shorten distance between cooling channels and cavity surface to 15mm, thereby enhancing heat exchange and improving cooling efficiency.

 

Figure 20 Inlay Distribution Diagram of Fixed Mold
Final product development follows Scheme Three. According to this scheme, the total projected area of the entire casting in mold opening direction is approximately 3650cm², casting pressure is designed to be 80MPa, and die-casting machine tonnage is designed to be a 3500T die-casting machine. Furthermore, the overall cavity volume (including pressure chamber) is large, requiring a large amount of gas to be expelled during filling. Therefore, mold design uses a "hydraulic vacuum valve + M-type venting block" method for overflow system.

Front shell production line used a domestic brand 3500T die casting machine. Aluminum material grade was ADC12, pouring temperature was 660℃, cavity vacuum degree was 50mbar-80mbar. Injection parameters are shown in Table 3. Compared with other products, die casting process for front shell differs mainly in that a loose block placement step was added after spraying and before mold closing, a loose block removal step was added during gate removal. Loose blocks are recycled. During use, it is important to control temperature of loose blocks (100℃-120℃) apply an appropriate amount of release agent to facilitate placement and removal of loose blocks.
Table 3. Front Housing Injection Parameters

 

The overall appearance quality of product is good, but there are localized areas with burn marks and aluminum adhesion, as shown in Figure 21. These mainly occur at sharp corners of some material reduction areas. These areas have thick bottom walls and are close to the gate, resulting in long solidification times, making them prone to burn marks and aluminum adhesion. Further optimization of mold cooling and adjustment of spraying process are needed.

 

Figure 22 X-ray inspection results of front shell
Figure 22 shows X-ray inspection results of front shell, indicating that internal quality meets CT inspection standards and no defects exceeding standard are found. After processing 200 pieces, a full inspection of appearance and leak testing was conducted. Figure 23 shows processing results of sealing surface of product. Figure 24 shows that there are no exposed pores on sealing surface and bearing hole surface of product, and leak test is qualified. Finally, product quality is determined to meet requirements.

 

Figure 23. Sealing surface machining result

 

Figure 24. Machining result of sealing surface of bearing hole

Numerical simulation can quickly and accurately help designers conduct casting feasibility analysis, thereby optimizing product structure design; comparison of simulation results of multiple gating schemes can quickly identify the most suitable gating system; by balancing multiple branches of gating system, problems of cold shut and gas entrapment at filling end are solved; solidification simulation analysis results can provide accurate theoretical basis for design of mold cooling system; combination process of "extrusion pin + live block + high-pressure point cooling" eliminates shrinkage defects in thick-walled hot spots. Trial production verification shows that process scheme based on numerical simulation iterative optimization achieves no defect exposure in key areas, a 100% leak test pass rate, and shortens development cycle.