1. Introduction and Background
The foundry industry serves as a fundamental pillar of the manufacturing sector, providing essential semi-finished components for virtually every industrial domain. As global markets become increasingly integrated and competitive, traditional approaches to casting process design that rely heavily on empirical knowledge and trial-and-error methods are no longer sufficient to meet the demanding requirements of cost reduction, quality improvement, and rapid product development cycles. The need to minimize casting defect incidence rates while simultaneously improving yield and reducing development time has driven the adoption of advanced computational tools.
Computer simulation technology, specifically the numerical simulation of mold filling and solidification processes, offers a transformative vehicle to address these challenges. By creating a virtual representation of the entire casting process, engineers can visualize the flow of molten metal, track temperature evolution, predict the formation of casting defects such as shrinkage porosity and gas entrapment, and optimize riser and gating system designs before any physical trial is conducted. This approach fundamentally shifts the paradigm from a reactive, production-based method to a proactive, predictive engineering discipline.
This research work focuses on the application of a comprehensive CAE platform, primarily the MAGMASOFT software, integrated with SolidWorks for three-dimensional modeling, to analyze and optimize the casting processes for a range of complex valve bodies. These components, including a large coking tower bottom valve, a valve cover, and a high-temperature plate gate valve, present unique challenges due to their intricate geometries, thick sections, and stringent quality requirements. The investigation encompasses the complete process from initial three-dimensional modeling and mesh generation to the simulation of mold filling, solidification, and the prediction of macro and micro shrinkage porosities. The ultimate goal was to establish a robust, efficient engineering methodology that enables the virtual development and validation of casting processes, effectively minimizing casting defect risks in an industrial setting.
2. Foundry Process Design Fundamentals
Before delving into the simulation techniques, it is essential to establish the theoretical framework governing traditional casting design. The primary objective is to engineer the solidification sequence such that all volumetric contractions are compensated by liquid metal feeding, thereby preventing the formation of casting defects.
2.1 Riser Design Principles for Steel Castings
Riser design is critical for ensuring that the molten metal can flow into the solidifying casting to compensate for volume shrinkage. The fundamental prerequisites for an effective riser are threefold:
1. Existence of a continuous feed path from the riser to the region being fed.
2. The solidification time of the riser must be greater than or equal to the solidification time of the casting section it is meant to feed.
3. Sufficient liquid metal volume must be available in the riser to supply the required feed metal.
The modulus concept, defined as the ratio of volume to cooling surface area, is a widely used tool for riser design. For a casting to be sound, the following conditions must be met:
$$ M_{riser} \ge M_{neck} \ge 1.2 M_{casting} $$
$$ M_{riser} \ge 1.2 M_{casting} $$
The integral for riser sizing is often based on the principle of equivalent modulus.
The feeder’s capacity to compensate for shrinkage is evaluated by comparing the total volumetric contraction of the casting with the available liquid metal in the riser. This is summarized by the equation:
$$ V_{shrinkage} = \varepsilon_{s} \cdot V_{s} \cdot (1 – \eta) $$
Where \( V_{shrinkage} \) is the volumetric deficit, \( \varepsilon_{s} \) is the solidification shrinkage factor, and \( V_{s} \) is the casting volume being fed. The factor \( \eta \) represents the feeder efficiency, which is a function of the riser type and is critical for estimating the necessary riser volume.
| Riser Type | Cylindrical & Side Riser | Spherical Riser | Insulated Riser | Exothermic Riser |
|---|---|---|---|---|
| Efficiency (\(\eta\)) | 0.14 | 0.15 | 0.20-0.30 | 0.30-0.50 |
2.2 Gating System Design
The gating system controls the flow of molten metal into the mold cavity. Its design is crucial to prevent turbulence, gas aspiration, and slag inclusion, which are primary sources of casting defect. The types of gating systems are commonly classified by the location where metal enters the casting:
- Top Gating: Promotes favorable temperature gradients for directional solidification, aiding riser feeding. However, it can lead to turbulence and splashing.
- Bottom Gating: Provides smooth, quiescent filling, reducing oxidation and mold erosion. However, it can result in an unfavorable temperature gradient with the top of the casting being cooler.
- Intermediate Gating: A balance between top and bottom gating, often used for medium-sized castings.
- Step Gating: Metal enters at multiple levels, allowing for smooth filling while promoting an upward temperature gradient.
The critical cross-sectional area of the gating system can be calculated using the fluid dynamics-based Bernoulli equation, which relates the flow rate to the static pressure head and flow resistance:
$$ A_{min} = \frac{m}{\rho \cdot \mu \cdot t_s \cdot \sqrt{2 \cdot g \cdot H_p}} $$
Where:
- \( A_{min} \) = minimum choke area
- \( m \) = mass of liquid metal flowing
- \( \rho \) = density of the liquid metal
- \( \mu \) = discharge coefficient (flow losses)
- \( t_s \) = pouring time
- \( g \) = gravitational acceleration
- \( H_p \) = effective static head pressure
2.3 Chills Design and Application
Chills are employed to accelerate the local solidification rate of thick sections, thereby promoting directional solidification towards a riser or eliminating hot spots that lead to macro-shrinkage. The heat absorbed by a chill increases the effective cooling surface area of the casting, reducing its local geometric modulus. The necessary chill weight can be estimated based on a heat balance between the excess heat in the casting hot spot and the heat absorbed by the chill:
$$ W_{chill} = \frac{(V_c – V_n) \cdot \rho_s \cdot [L + C_s \cdot \Delta T_s]}{C_c \cdot \Delta T_c} $$
Where \( V_c \) and \( V_n \) are the volumes of the hot spot and adjacent sections, \( \rho_s \) is the density of the steel, \( L \) is the latent heat, and \( C_s \) and \( C_c \) are the specific heats of the steel and chill material, respectively. The end effect on the freezing range can also be characterized by an increase in the apparent surface area, described by the factor \( \theta \). Chills are instrumental in controlling the solidification morphology and are a strategic tool in reducing reliance on multiple risers.
3. Casting Process Simulation: Mathematical Models and Methodology
To fundamentally understand and predict the emergence of casting defects, we must mathematically represent the physical phenomena occurring during mold filling and solidification. The governing equations for these processes, as solved by the simulation software, are as follows.
3.1 Energy Transport
The transient temperature field is governed by the non-linear heat conduction equation with a source term for latent heat evolution:
$$ \frac{\partial (\rho H)}{\partial t} = \nabla \cdot (k \nabla T) + S $$
Where \(\rho\) is density, \(H\) is enthalpy, \(T\) is temperature, \(k\) is thermal conductivity, and \(S\) represents the latent heat source term. This formulation is a non-linear transient equation. When considering fluid flow, the model can be extended to include convective heat transfer, which describes the coupled fluid flow and heat transfer during the filling process:
$$ \frac{\partial (\rho H)}{\partial t} + \nabla \cdot (\rho \mathbf{u} H) = \nabla \cdot (k \nabla T) $$
3.2 Fluid Flow and Momentum Transport
The flow of molten metal is modeled as an incompressible Newtonian fluid. The governing equations, known as the Navier-Stokes equations, are solved in their transient, non-linear form. These are essential for predicting flow patterns and identifying areas of high turbulence, which can lead to gas entrapment and mold erosion – key catalysts for casting defect formation.
$$ \rho \left( \frac{\partial \mathbf{u}}{\partial t} + (\mathbf{u} \cdot \nabla) \mathbf{u} \right) = -\nabla p + \mu \nabla^2 \mathbf{u} + \rho \mathbf{g} $$
In this equation, \( \mathbf{u} \) is the velocity vector, \( p \) is pressure, \( \mu \) is dynamic viscosity, and \( \mathbf{g} \) represents gravitational acceleration. The simulation solves for the transient state of these equations.
3.3 Mass Conservation (Continuity)
The conservation of mass for incompressible flow demands that the divergence of the velocity field be zero:
$$ \nabla \cdot \mathbf{u} = 0 $$
This equation ensures consistency between detailed flow fields and is applied across the modeling grid. The Mass Transport equation is a measure of how a particular species or material, such as a tracer particle, moves through the fluid domain.
4. CAE Simulation Platform and Workflow
The integration of a robust simulation software into the product development workflow represents a significant upgrade over traditional, trial-and-error-based foundry practices. The goal of this approach was to accurately predict casting defects prior to physical production, allowing for the virtual correction of process parameters.
4.1 Integrated Simulation Workflow
Utilizing SolidWorks as the primary 3D CAD tool, a solid model of each component was created. The model was then exported in a neutral file format for meshing and simulation. The main process flow involves:
- 3D Modeling: The casting, including all cores, gating systems, and risers, is fully designed in SolidWorks.
- Data Transfer: The solid model of each component is converted into a compatible file format for import into the simulation software.
- Mesh Generation: The model is discretized into a finite difference mesh, capturing all geometric details. The mesh density is optimized to balance accuracy with computational time.
- Parameter Definition: This is a critical step, involving the assignment of materials (e.g., cast steel, sand), thermal properties (such as conductivity and specific heat), and process parameters (e.g., pouring temperature, filling time).
- Numerical Simulation: The software solves the governing equations to simulate the filling and solidification process, calculating temperature, velocity, and pressure fields at each time step.
- Result Analysis: The post-processor allows for the visualization of the filling sequence, temperature gradients, and the prediction of solidification shrinkage using various criteria.
- Process Optimization: Based on the simulated results, engineers can identify the risk of casting defects (such as shrinkage and porosity) and modify the process design—for instance, by adding chills or repositioning risers—to mitigate those risks.
5. Case Study I: Coking Tower Bottom Valve Body
5.1 Component and Initial Simulation
The first case study addresses a large coking tower bottom valve body weighing nearly nine tons. Complex geometry with significant mass concentration creates a high risk for casting defects. The standard foundry practice involved designing a gating and riser system. In the initial design, the gating system was a bottom-gate setup, and the riser placement was based on the conventional modulus calculations. The initial simulation was run to validate the proposed casting process.
| Property | Cast Steel (GS-20Mn5V) | Molding Sand (Sodium Silicate) |
|---|---|---|
| Pouring Temperature (°C) | 1550 | – |
| Liquidus Temperature (°C) | 1490 | – |
| Solidus Temperature (°C) | 1350 | – |
| Initial Temperature (°C) | – | 25 |
5.2 Analysis of Initial Design and Identification of Casting Defects
The initial simulation results, without the addition of risers, indicated that the last regions to solidify were located within the flanges. The ANSYS criterion, a dimensionless parameter evaluating temperature gradient vs. cooling rate, predicted a high likelihood of shrinkage porosity in these isolated hot spots. The use of tracer particles showed that the metal flow from the four bottom gates was prematurely concentrated within the end flanges, creating a high-temperature zone that would not be adequately fed.
The simulation was repeated with the initial riser and gating system design proposed by the foundry. The results showed that while the flanges remained the last to solidify, the risers were not positioned correctly and were unable to establish the proper feeding path. This condition is a classic trigger for casting defects. The “negative temperature gradient” was identified, meaning the cavity’s feeder system was solidifying earlier than the casting areas they were intended to feed. The simulation confirmed the need for a more rigorous approach to riser sizing and placement.
5.3 Improved Casting Process Design
The simulation was instrumental in guiding the following changes to the process:
- Modified Gating System: The design was changed to a slow bottom-gating system, with ingates distributed more evenly to minimize localized heating and promote smoother filling.
- Optimized Riser Placement: Based on the heat-solidification simulation, risers were placed directly above the end flanges to provide the required feeding path. A blind riser was also installed at the top to address the flange area.
- Utilization of Chills: Chills were added to the bottom flanges to accelerate cooling and consolidate directional solidification from the bottom upwards, thereby drawing the feeding demand toward the risers.
The subsequent simulation of the modified design showed a substantial improvement in solidification behavior. The simulation results, characterized by the solid fraction and temperature gradients, revealed that the final solidification zones had moved into the risers, confirming their effective feed potential. The casting defect indicators showed a marked reduction in shrinkage porosity in the critical flange areas. This virtual optimization demonstrated the power of simulation in achieving a sound casting design prior to physical production.
6. Case Study II: Coking Tower Bottom Valve Cover
6.1 The Product and Its Challenges
This case study focuses on the valve cover used in the coking tower bottom valve. It is a complex, thick-walled steel casting (roughly 1200 kg) and had a high rejection rate due to severe shrinkage porosity in the root of the cross-rib structure and leakage at the large flange after machining. The original casting process was a horizontally-oriented casting with four risers to feed the thick sections, but no chills were used. A ceramic tube bottom gating system was employed.
6.2 Simulation as a Diagnostic Tool
Computer simulation of the original process (validated by the strong correlation with the observed defects) provided a clear visualization of the underlying cause. The simulation showed that the cross-rib intersections and the mid-to-lower sections of the large flange formed isolated hot spots. After the surrounding thin sections had solidified, these hot spots were no longer able to be fed by the risers, resulting in internal shrinkage porosity. The criterion shown in the simulation identified these regions, which become the primary source of casting defect.
6.3 Process Modifications and Results
Guided by the simulation, a modification to the casting process was undertaken. The objective was to control the solidification sequence. The key improvements were:
- Strategic Chill Placement: Numerous chill plates were added—not only at the flanges but also at the intersection of the cross-ribs—to promote faster cooling in these specific zones.
- Flange Cooling: The chills at the large flange were designed to force the bottom section to solidify first, establishing the necessary directional solidification pattern towards the risers.
The simulation of this revised setup showed that the addition of chills effectively broke up the large hot spots. The cross-rib intersections solidified in a synchronized manner, and the large flange solidified in a bottom-to-top sequence. The simulated solid fraction and temperature fields demonstrated that this directed cooling successfully eliminated the macroscopic shrinkage porosities, changing the solidification pattern from “simultaneous” to “directional.”
7. Case Study III: Addressing Leakage in a High-Temperature Plate Gate Valve
7.1 Problem Definition
This case study involves a high-temperature plate gate valve used in the petrochemical industry. The component experienced persistent leakage in pressure testing, which was traced to porosity in the body casting. The original design had four large flanges cooled with top risers, a central blind riser, and a step-gating system. It was theorized that the step gating system caused premature cooling of the metal and a “negative temperature gradient,” reducing the effectiveness of the risers.
7.2 Numerical Diagnostics
The simulation of the initial process reproduced the problem. The results showed that the feeding paths from the risers to the flanges were prematurely blocked, creating isolated liquid pools within the flange body, which upon solidification, resulted in shrinkage. The simulation of the solidification sequence revealed the ineffective role of the central blind riser, which was being fed by the casting, rather than vice versa. The temperature gradient prior to complete solidification reflected a non-ideal state with higher temperatures in the casting body than in the risers—a direct feeding impediment.
7.3 Iterative Improvement towards a Defect-Free Casting
To solve this issue, the process was modified iteratively.
First Improvement:
In the first iteration, the original gating scheme was simplified to a bottom gating design In addition, the central blind riser was switched to an open riser to increase its cooling capacity and head pressure. The inclusion of chills was intended to improve the feeding of the flanges. Simulation of this initial improved process showed clear progress: the flange solidification progressed correctly from the lower regions upward, toward the riser. However, a unsolidified island remained in the region near the bottom of the central inner ring, suggesting another potential location for casting defects.
Second Improvement:
Building on the previous analysis, chills were strategically placed at the lower outer edge of the central internal ring to eliminate the remaining hot spot. This modification successfully eliminated the localized hot spot, allowing the central section to solidify in the correct sequence. The final simulation displayed a well-engineered solidification sequence where final solidification was confined to the risers only, and solidification shrinkage was minimized. This process design, validated by simulation, yielded a valve body that passed the water pressure test with no reported leakage.
8. Conclusions
This comprehensive research project successfully integrated 3D modeling and process simulation to analyze, optimize, and thereby minimize the incidence of casting defects in complex steel valve components. The key findings and outcomes of this work are:
1. Advanced Engineering Methodology: An efficient engineering workflow was established, facilitating the creation of accurate 3D solid models of complex valve bodies. This technique made challenging modeling tasks simple and reliable, ensuring a one-to-one correspondence between the virtual model and physical component.
2. Predictive Power of Simulation: The use of CAE tools proved successful in predicting the formation of casting defects such as shrinkage porosity and gas entrapment. This was confirmed through multiple case studies where simulation results were directly correlated with physical defect locations. This reinforces the dependability of computer simulation as a diagnostic tool.
3. Effective Process Optimization: By virtually testing alternative riser designs, gating systems, and chill placements, the research successfully engineered processes that eliminated the occurrence of casting defects. The result was a cast valve with a more homogeneous, dense structure that passed strict hydrostatic testing.
4. Practical and Economic Benefits: The substitution of physical experiments with validated simulations directly cut development lead time, reduced the cost of trial-and-error in scrap, and enhanced the overall productivity and competitiveness of the foundry. The research served as a successful validation of using CAE to guide robust production processes, establishing a framework for future process design work.
5. Strategic Implementation: The successful application of these simulation techniques demonstrates a shift from a trial-and-error approach to a science-based, predictive engineering discipline. This technological adoption is paramount to meeting the ever-increasing quality and cost-efficiency demands of the modern manufacturing landscape.
In summary, the integration of computational process simulation into the working methodology of the foundry has significantly improved the ability to identify and mitigate casting defect risks proactively. This results in a direct positive impact on cost, lead time, and quality, underscoring the indispensable role of CAE in the modern foundry industry.