This paper presents a comprehensive study on the defect analysis and process optimization of an isolating switch shell produced by low-pressure die-casting. The shell is a large thin-walled aluminum alloy pressure vessel used in gas-insulated switchgear (GIS). The research focuses on three main aspects: development of a parameter management system for low-pressure casting, CAD/CAE-based gating system design and numerical simulation, and actual production validation with defect analysis and mitigation. Several sand foundry defects such as gas porosity, shrinkage, cold shuts, slag inclusions, and cracks were systematically investigated. The optimized process parameters include a pouring temperature of about 700°C, mold temperature of 200–300°C, lift pressure of 0.04 MPa, filling pressure of 0.055 MPa, boost pressure of 0.08 MPa, and holding time of 120 s. The final mechanical properties reached σb≥250 MPa, δ≥1.4, and HB≥90. The gas-tightness test passed at a rate exceeding 98%. This study demonstrates that the combination of a systematic process management system, numerical simulation, and defect analysis can effectively reduce sand foundry defects and improve product quality.
1. Introduction
With the rapid development of power transmission equipment, gas-insulated switchgear (GIS) has become widely adopted due to its compact size, high reliability, and immunity to environmental conditions. The aluminum alloy shell is a key component of GIS, and its manufacturing quality directly affects the performance of the entire switchgear. Traditional sand casting methods often result in low yield and inconsistent mechanical properties. Low-pressure die-casting has emerged as a superior alternative because it provides controlled filling, pressure-assisted solidification, and high material utilization. However, the occurrence of sand foundry defects remains a major bottleneck for mass production.
The objective of this work is to analyze the root causes of sand foundry defects in a large thin-walled aluminum shell and to propose effective solutions through process standardization, numerical simulation, and production validation. The shell is made of ZL114A aluminum alloy, which is a common Al-Si-Mg alloy with excellent castability and mechanical properties after T6 heat treatment. The shell has a complex geometry with four flanges and multiple thick bosses, making it prone to hot spots and shrinkage porosity.
To address these challenges, this study first developed a low-pressure casting process parameter management system using Visual Basic 6.0 and Access database. The system standardizes the design of gating systems, mold dimensions, and process parameters, thereby reducing reliance on empirical judgment. Secondly, CAD/CAE tools were employed to perform stress analysis of the shell under uniform internal pressure, identifying weak areas that are susceptible to failure. Two gating system designs were proposed: an “arborization” (tree-like) gating system and a circular crevice (slot) gating system. Numerical simulation was used to compare the filling and solidification behaviors and to predict defect locations. The better design was selected and validated in actual production.
2. Experimental Materials and Methods
2.1 Materials
The alloy used in this study is ZL114A, whose chemical composition is given in Table 1. The mechanical properties and thermophysical parameters are listed in Tables 2 and 3, respectively.
Table 1. Chemical composition of ZL114A (wt.%)
| Si | Be | Mg | Zn | Mn | Ti | Al |
|---|---|---|---|---|---|---|
| 6.5–7.5 | 0.04–0.07 | 0.45–0.60 | ≤0.10 | ≤0.10 | 0.10–0.20 | Balance |
Table 2. Mechanical properties of ZL114A (T6 condition)
| Casting method | Alloy state | Tensile strength σb (MPa) | Elongation δ (%) | Hardness HB |
|---|---|---|---|---|
| Metal mold, modified | T6 | 310 | 3 | 90 |
Table 3. Thermophysical parameters of ZL114A
| Density (kg/m³) | Specific heat (kJ/(kg·K)) | Latent heat (kJ/kg) | Thermal conductivity (W/(m·K)) | Solidus (K) | Liquidus (K) |
|---|---|---|---|---|---|
| 2.7×10³ | 0.964 | 29.3–1294 | 92.8–184 | 829 | 889 |
2.2 Auxiliary Materials
Argon gas was used for degassing the molten metal. A universal flux for Al-Si alloys, which acts as a covering, refining, and modifying agent, was used. Low-pressure die-casting oil-based release agents were used for the metal mold, and alcohol-based coatings were applied to the resin sand core to reduce gas evolution and improve surface finish.
2.3 Experimental Procedure
The overall experimental flow is illustrated in Figure 1. The procedure included: development of a process parameter management system, CAD/CAE-based gating system design, numerical simulation of filling and solidification, optimization of process parameters, and production validation with defect analysis. After casting, T6 heat treatment was performed, followed by mechanical testing, microstructure observation, and leak/pressure tests.
Figure (inserted reference): Typical sand foundry defects observed during casting production.
2.4 Heat Treatment
The T6 heat treatment involved solution treatment at 535°C ± 5°C for 8 hours, quenching in water at 60–80°C, natural aging for 12 hours at room temperature, then artificial aging at 155°C ± 5°C for 6 hours followed by air cooling.
2.5 Testing Methods
Tensile tests were conducted on a WDW-100 electronic universal testing machine. Hardness was measured using a HB-1875 Brinell hardness tester with a 2.5 mm ball indenter and 62.5 kgf load. Microstructural observations were performed using an XJL-02 optical microscope. Nondestructive testing was carried out by X-ray radiography. Helium leak detection and hydrostatic pressure tests were used to evaluate gas-tightness and pressure resistance.
3. Development of Low-Pressure Casting Process Parameter Management System
3.1 Design Philosophy
Traditional low-pressure casting process design relies heavily on the experience of engineers, which often leads to inconsistent quality and long development cycles. To overcome this, a parameter management system was developed on the Visual Basic 6.0 platform with an Access database. The system integrates knowledge from experienced engineers and standardizes the design process for GIS shell castings. It covers alloy composition calculation, gating system design, mold design, process parameter selection, heat treatment schedules, and defect analysis.
3.2 System Architecture
The system consists of eight main modules: file management, alloy mixing, gating system design, mold design, low-pressure casting process parameters, lift tube connection, heat treatment, and defect analysis. Each module includes sub-modules that guide the user through the design steps and calculate key parameters based on input data.
3.3 Key Functional Modules
File module: This module provides access to basic knowledge about low-pressure casting, saves process cases, and allows querying and printing of process cards. The database can be searched, edited, and expanded.
Alloy mixing module: Based on the desired alloy grade and the capacity of the crucible, the module calculates the required weights of pure metals and master alloys. It also accounts for the addition of foundry returns and auxiliary materials such as fluxes and degassing agents.
Gating system design module: This module includes sub-modules for determining the filling time, the cross-sectional area of the lift tube, sprue, runner, and ingates. The fundamental relationship follows the principle of sequential solidification, where the cross-sectional areas are designed to ensure proper feeding. The recommended area relationship is given by:
$$ \sum F_{\text{outlet}} \ge \sum F_{\text{runner}} \ge \sum F_{\text{ingate}} \quad \text{(open system)} $$
For easily oxidized alloys, an open gating system is preferred, and each cross-section should be large enough to allow the metal to flow without excessive turbulence.
Mold design module: This module helps determine shrinkage allowance, machining allowance, draft angle, minimum wall thickness, mold type, parting surface, cavity dimensions, and mold wall thickness. The mold wall thickness is typically twice the casting wall thickness to provide sufficient chilling effect and heat capacity.
Process parameters module: The module calculates important parameters such as lift pressure, filling pressure, boost pressure, holding time, and pouring temperature. The theoretical basis is the balance between gas pressure and static head, expressed as:
$$ p = \rho g h $$
where p is the applied pressure, ρ is the alloy density, g is gravitational acceleration, and h is the metal rise height. The module also generates a pressure-time curve as shown in Figure 2.
Table 4. Optimized low-pressure casting parameters
| Parameter | Value |
|---|---|
| Pouring temperature | 700 °C |
| Mold temperature | 200–300 °C |
| Lift pressure | 0.04 MPa |
| Lift time | 15 s |
| Filling pressure | 0.055 MPa |
| Filling time | 30 s |
| Boost pressure | 0.08 MPa |
| Boost time | 15 s |
| Holding time | 120 s |
4. CAD/CAE-Based Low-Pressure Casting Process Design for the Isolating Switch Shell
4.1 Casting Process Analysis
The isolating switch shell is a large thin-walled pressure vessel with an average wall thickness of 12 mm and a height of 832 mm. The casting mass is approximately 61.7 kg. The structure includes four cylindrical flanges with diameters of 415 mm and several thick bosses that create local hot spots. The shell must withstand an internal pressure of 0.5 MPa under service conditions, and the design pressure for testing is 3.35 MPa. The required mechanical properties are σb≥250 MPa, δ≥1.4, and HB≥90.
4.2 Gating System Designs
Two gating system configurations were proposed:
Design 1: Arborization (Tree-like) Gating System – This system uses a single lift tube connected to a central sprue located inside the sand core. Multiple ingates are designed to feed the hot spots and flanges directly. The advantages include excellent feeding capacity under applied pressure and the absence of additional cooling devices. However, because the sprue is embedded in the resin sand core, the coating of the sprue walls is difficult, which may lead to gas evolution and the formation of pinholes. To prevent this, ceramic tubes can be used as the sprue material.
Design 2: Circular Crevice (Slot) Gating System – This system employs two lift tubes and a stepped circumferential slot that feeds metal from the sides. The slot provides effective feeding for the flanges, and the top uses a blind riser for the smaller flange and for venting. The advantages are lower risk of resin sand gas defects and reduced filling time. However, the four thick bosses cannot be directly fed, requiring the addition of chillers, which complicates the process. Also, controlling two lift tubes simultaneously is more difficult, and uneven filling may occur.
4.3 Stress Analysis of the Shell
A finite element model of the shell was created from the 3D CAD model. The material properties used are listed in Table 5. The boundary conditions fixed both ends, and an internal pressure of 3.35 MPa was applied uniformly on the inner surfaces.
Table 5. Material properties used in FEA
| Property | Value |
|---|---|
| Density (g/cm³) | 2.7 |
| Elastic modulus (MPa) | 70000 |
| Poisson’s ratio | 0.33 |
The stress distribution analysis revealed that the maximum first principal stress occurs at the transition zones between the thin cylindrical wall and the thick flanges. These areas are therefore the weakest parts of the casting and are more prone to defects such as shrinkage porosity and cracks. Figure 3 shows the first principal stress contour. The simulation result matched the actual defect locations observed during hydrostatic testing, where minor leakage occurred due to internal shrinkage at the flange-to-wall transitions.
4.4 Numerical Simulation of Filling and Solidification
Both gating systems were modeled and meshed using tetrahedral elements. The filling and solidification processes were simulated with a commercial finite element code. The boundary conditions included the pressure-time curve, pouring temperature, mold temperature, and heat transfer coefficients between the casting and the mold.
Filling simulation: For Design 1, the metal flows steadily through the arborization runner and fills the cavity smoothly without turbulence. The entire filling takes about 30 s. For Design 2, the metal rises through the two lift tubes and the stepped slot, also exhibiting smooth filling with no visible splash. Both designs show acceptable filling behavior, but Design 1 provides more direct feeding to the hot spots.
Solidification simulation: The solid fraction evolution is shown in Figure 4. In Design 1, the thin walls solidify first, while the flanges and the internal runner remain liquid longer, allowing pressure feeding to be effective. In Design 2, the four thick bosses solidify last without any direct feeding, leading to high shrinkage porosity risk. Therefore, Design 1 was selected for production.
4.5 Defect Prediction
Shrinkage porosity prediction indicated that Design 1 produces almost no defects in the critical regions, while Design 2 shows significant porosity in the four thick bosses. The simulation also predicted mold temperature distribution, which showed the highest temperatures at the flange-wall junctions, confirming that these areas require enhanced cooling or controlled coating thickness to avoid local overheating.
4.6 Optimization of Process Parameters Using Orthogonal Experiments
An L9 orthogonal experiment was designed to optimize three key parameters: pouring temperature, mold temperature, and filling pressure. The factor levels are shown in Table 6. The simulation results were used to evaluate the quality of the castings. The optimal combination was determined as pouring temperature 700°C, mold temperature 260°C, and filling pressure 0.055 MPa. The final optimized parameters are summarized in Table 4.
Table 6. Orthogonal experiment factors and levels
| Level | Pouring temp. (°C) | Mold temp. (°C) | Filling pressure (MPa) |
|---|---|---|---|
| 1 | 680 | 220 | 0.050 |
| 2 | 700 | 260 | 0.055 |
| 3 | 720 | 300 | 0.060 |
5. Production Validation and Defect Analysis
5.1 Production Process
The manufacturing process includes alloy melting, refining, modification, core making, mold preheating, coating, pouring, solidification, shakeout, and heat treatment. The J4510 low-pressure casting machine was used, equipped with a CLP-8 pressure control system. The sand core was placed in the mold, and the mold was closed before pouring. The actual pouring process is shown in Figure 5.
5.2 Sand Foundry Defects Observed and Their Solutions
During the production trials, several sand foundry defects were encountered. These defects are common in low-pressure die-casting, especially when dealing with large thin-walled shells in sand molds or resin sand cores. The following defects were identified and analyzed:
5.2.1 Gas Porosity
Gas porosity was found on the flange surfaces and inside the casting. The main causes were inadequate venting, insufficient degassing of the melt, and gas evolution from the resin sand core. The defects appeared as “pebble-shaped” pores after machining, “rain-like” pinholes on the surface, and dispersed pinholes in the internal cavities. Solutions included: strict control of melting practice, argon degassing, optimizing core venting by placing exhaust ropes inside the core, reducing the binder content of the resin sand, and applying a proper refractory coating to the core surface. Additionally, the mold venting system was improved by installing exhaust plugs and grooves at the parting line.
5.2.2 Shrinkage Porosity and Cavity
Shrinkage defects occurred at thick sections and at the transition between thick and thin walls. X-ray inspection revealed internal shrinkage at the lower flange, and hydrostatic testing showed leakage at the wall-thickness transition regions. The root cause was inadequate feeding and non-ideal solidification sequence. The remedies were to increase the boost pressure, adjust the ingate locations to reduce local overheating, and modify the coating thickness to control the cooling rate. The eventual implementation of the arborization gating system with direct feeding to the hot spots greatly reduced such sand foundry defects.
5.2.3 Cold Shuts
Cold shuts were observed on the upper wall of the shell. These were caused by low pouring temperature, slow filling, and poor mold venting. The solutions included increasing the pouring temperature to about 700°C, improving venting, and adjusting the filling pressure to ensure complete fusion of the metal streams. The use of a higher mold temperature (200–300°C) also minimized cold shuts.
5.2.4 Slag Inclusions
Slag inclusions appeared as rough surfaces on the internal walls or as embedded particles after machining. The sources were dross from the melt, eroded sand from the core, and oxidized films. To eliminate slag inclusions, the melt was filtered using ceramic foam filters placed at the lift tube outlet, the sand core strength was increased by using higher-quality sand and resin, and the mold cavity was cleaned with compressed air before closing. Furthermore, the lift tube was regularly sandblasted to remove oxides.
5.2.5 Cracks
Both cold cracks and hot tears were observed. Cold cracks occurred at the parting line due to premature mold opening before the casting had completely solidified. Hot tears appeared at the fillet radii of the bosses as a result of thermal stresses during cooling. The solutions were to extend the holding time to ensure complete solidification, adjust the ingate positions to avoid thermal concentration, and use a hollow core structure that provides better collapsibility. Proper coating thickness was also applied to balance the cooling rate.
5.3 Microstructure and Mechanical Properties
Samples were taken from the actual casting and subjected to T6 heat treatment. The microstructure after T6 treatment showed a fine and uniform distribution of spheroidized eutectic silicon in the aluminum matrix, without dendrite segregation. The grain structure was compact and free from microporosity. The mechanical test results are given in Table 7. All values exceed the required minimums.
Table 7. Mechanical properties of the produced shell
| Sample | Tensile strength (MPa) | Elongation (%) | Brinell hardness (HB) |
|---|---|---|---|
| 1 | 309 | 2.5 | 98 |
| 2 | 311 | 2.6 | 99 |
| 3 | 308 | 2.4 | 96 |
5.4 Nondestructive Testing and Gas Tightness
X-ray radiography of the production castings showed no significant internal defects. Helium leak testing was performed on the machined shells; the qualified rate exceeded 98%. The hydrostatic pressure test was conducted at 1.0 MPa for observation and then increased to 3.35 MPa and held for 10 minutes. No leakage or rupture occurred, demonstrating that the castings meet the requirements for GIS applications.
6. Conclusion
This study successfully addressed the challenge of sand foundry defects in the low-pressure die-casting of large aluminum alloy GIS shells. The following conclusions can be drawn:
- A low-pressure casting process parameter management system was developed, which standardizes the design of gating systems, molds, and process parameters. The system improves efficiency and reduces reliance on empirical knowledge.
- Two gating system designs were evaluated. The arborization (tree-like) gating system proved superior to the circular crevice system in terms of feeding capacity and defect reduction. This was confirmed by numerical simulation and actual production.
- The optimized process parameters are: pouring temperature 700°C, mold temperature 200–300°C, lift pressure 0.04 MPa, lift time 15 s, filling pressure 0.055 MPa, filling time 30 s, boost pressure 0.08 MPa, boost time 15 s, and holding time 120 s.
- Defect analysis of production castings identified the main sand foundry defects as gas porosity, shrinkage, cold shuts, slag inclusions, and cracks. Solutions included improved degassing, better venting, optimized gating, controlled coating, and appropriate process parameter adjustments.
- The final castings exhibit dense microstructure and mechanical properties exceeding the requirements: σb≥250 MPa, δ≥1.4, and HB≥90. Nondestructive inspection and pressure tests confirmed that the shells are free of significant internal defects, and the gas-tightness test passed at a rate of 98%.
This work demonstrates that a systematic approach combining process management software, numerical simulation, and rigorous defect analysis can effectively minimize sand foundry defects and enable reliable production of complex aluminum alloy GIS shells.
（The End）