The lost foam casting process, recognized as an advanced and environmentally friendly manufacturing technology for the 21st century, offers significant advantages including high dimensional accuracy, excellent surface finish of castings, simplified sand removal, and high material yield. This process involves creating a foam pattern identical to the desired part, assembling it with a gating system, embedding it in unbonded dry sand within a flask under vibration, and subsequently pouring molten metal under a vacuum. The foam pattern vaporizes upon contact with the metal, which then fills the cavity and solidifies. This technique is widely applicable to ferrous and non-ferrous alloys such as cast iron, steel, aluminum, and magnesium, and is extensively used in automotive, mining, and heavy machinery sectors. In the development phase of new castings, employing numerical simulation technology to visualize the filling and solidification stages is crucial. It allows for the prediction of potential defects like shrinkage porosity, gas entrapment, and misruns, enabling the evaluation and optimization of casting process designs before physical trials. Compared to traditional trial-and-error methods, this simulation-driven approach substantially reduces development lead time, lowers production costs, and enhances overall efficiency.
This article details a comprehensive study conducted from a first-person research perspective on the process design for a critical cast steel component—a convex ring used in mining equipment. The primary challenge lay in its complex geometry and varying section thickness. Through meticulous lost foam casting process analysis and the application of numerical simulation, two distinct process schemes were designed, simulated, and compared. The optimized scheme was successfully validated through actual production, yielding high-quality castings that met all service requirements.
Component Analysis and Casting Challenges
The subject component is a large convex ring with an outer diameter of 435 mm and a height of 114 mm. Its nominal wall thickness is 25 mm, but it features a highly asymmetric profile. One face has a complex involute curve surface which is designated as a critical functional area; it must be free from any casting defects such as shrinkage cavities, porosity, cold shuts, or inclusions. The opposite face is flat. The geometry results in a significant height differential, with one side measuring only 21 mm and the opposing side the full 114 mm. This asymmetry and the requirement for impeccable quality on the contoured surface present substantial difficulties in controlling the solidification and feeding during the lost foam casting process. The material specification is ZG50CrMnSiA (a cast steel), with a casting weight of approximately 28 kg. The required chemical composition is detailed in Table 1.
| Element | Weight Percentage (wB / %) |
|---|---|
| C | 0.45 – 0.55 |
| Mn | 0.70 – 1.00 |
| Si | 0.40 – 0.60 |
| Cr | 0.80 – 1.20 |
| S | ≤ 0.020 |
| P | ≤ 0.030 |
The core of the lost foam casting process design revolves around managing the solidification sequence to ensure soundness in the critical sections. Given the geometry, a vertical orientation of the pattern was selected to position the critical involute surface on the side, which is generally easier to control than a top-facing surface. Two primary process schemes were conceived based on the orientation of the thick and thin sections relative to gravity and the gating system.
Lost Foam Casting Process Scheme Design and Numerical Simulation Setup
Proposed Process Schemes
Two distinct lost foam casting process layouts were designed:
- Scheme 1 (Thin-Section-Up): The pattern is oriented with the narrow (21 mm) section at the top and the thick (114 mm) section at the bottom. The gating system employs a step-gate design where the pouring cup connects to the top thin section, and the sprue base connects laterally to the thick section. The rationale relies on using the metal head from the sprue to feed the thick section from below via gravity.
- Scheme 2 (Thick-Section-Up): The pattern is inverted, with the thick section at the top and the thin section at the bottom. The step-gate design is modified so the pouring cup connects to the top thick section, and the sprue base connects laterally to the bottom thin section. This scheme intends to use the top-mounted pouring cup as a direct feeder (riser) for the thick section, which is the last to solidify.
Both schemes utilize a step-gating system to promote stratified filling, minimize turbulence, and facilitate the flotation of slag and gas bubbles into the pouring cup. The absence of a horizontal runner simplifies pattern assembly, a key consideration in the lost foam casting process.
Numerical Simulation Parameters and Governing Physics
The simulation of the lost foam casting process involves solving coupled equations for fluid flow, heat transfer, and foam degradation. The filling process is governed by the Navier-Stokes equations with source terms for momentum and energy sink due to foam decomposition. The energy equation is critical for modeling solidification:
$$
\rho c_p \frac{\partial T}{\partial t} + \rho c_p \mathbf{u} \cdot \nabla T = \nabla \cdot (k \nabla T) + S
$$
Where \( \rho \) is density, \( c_p \) is specific heat, \( T \) is temperature, \( t \) is time, \( \mathbf{u} \) is the velocity vector, \( k \) is thermal conductivity, and \( S \) is a source term accounting for latent heat release during phase change:
$$
S = -\rho L \frac{\partial f_s}{\partial t}
$$
Here, \( L \) is the latent heat of fusion and \( f_s \) is the solid fraction. The foam decomposition is modeled as a temperature-dependent process, creating a gap at the metal-foam interface that affects heat transfer. The key thermophysical parameters for the casting material, foam pattern (EPS), and molding sand used in the simulation are consolidated in Table 2.
| Material / Parameter | Casting (ZG50CrMnSiA) | Foam Pattern (EPS) | Molding Sand |
|---|---|---|---|
| Density (g/cm³) | 7.6 | 2.5 | 1.55 |
| Specific Heat (J/g·K) | 0.828 | 3.7 | 1.088 |
| Thermal Conductivity (W/m·K) | 22.015 | 0.15 | 0.804 |
| Latent Heat (J/g) | 251.2 | 100 | – |
| Liquidus Temperature (°C) | 1480* | 350 | – |
| Solidus Temperature (°C) | 1309* | 330 | – |
*Calculated via empirical formulae.
The initial and boundary conditions were defined as follows: pouring temperature of 1640°C, mold vacuum of 0.06 MPa, ambient temperature of 20°C, and a pouring time of 10 seconds. The interfacial heat transfer coefficient (HTC) is a critical parameter. For the metal-sand interface, an HTC of 800 W/(m²·K) was used. For the foam-sand interface during filling, a lower HTC of 100 W/(m²·K) was applied to model the insulating gas gap.
Numerical Simulation Results and Comparative Analysis
Scheme 1: Thin-Section-Up Orientation
The simulation of the filling process for this lost foam casting process scheme showed a stable fill with minimal turbulence. Metal initially entered the thin top section, while the thick bottom section began filling later as the underlying foam decomposed. The end of filling was characterized by the complete filling of the casting cavity followed by the pouring cup. The total fill time was approximately 12.9 seconds.
The solidification analysis revealed the critical flaw in this design. As shown in the temperature field progression, the thin top section solidified rapidly. The thick bottom section, intended to be fed from the sprue, began solidifying from its outer walls. While the connection between the sprue base and the thick section remained liquid for a period, it eventually isolated into a localized hot spot. Crucially, the metal in the sprue above this connection solidified before the thick section, severing the feeding path from the main reservoir of liquid in the pouring cup. The final solidification shrinkage was consequently predicted to manifest as a macro-porosity in the sprue base and potentially as micro-shrinkage (shrinkage porosity) within the thick section of the ring itself, despite the latter not being explicitly flagged as a gross defect in the Niyama criterion output. The last area to solidify was the isolated pool at the sprue base, confirming an inadequate feeding mechanism for the thick section in this lost foam casting process layout.
Scheme 2: Thick-Section-Up Orientation
The filling process for this lost foam casting process scheme was also stable but followed a different sequence. Metal first filled the thick top section via the pouring cup and then flowed downward through the sprue to fill the thin bottom section. The cavity was filled in approximately 12.1 seconds, slightly faster than Scheme 1.
The solidification behavior was fundamentally superior. The thin bottom section solidified first, followed by progressive solidification upward through the ring walls. Most importantly, the thick top section of the ring and the pouring cup remained liquid the longest, forming a continuous thermal mass. The temperature field snapshots clearly demonstrated that the pouring cup acted as an effective riser, providing directional solidification towards itself. The thermal gradient was favorable, with the ring’s thick section feeding the thinner walls below, and the pouring cup feeding the ring’s thick section. The final predicted shrinkage was confined to the center of the pouring cup—the intended location for such defects. No significant shrinkage was predicted within the critical body of the convex ring casting itself. This confirmed the efficacy of this lost foam casting process scheme in establishing a sound thermal gradient and feeding path.
The comparative results can be summarized using a key solidification metric, the Local Solidification Time (LST), which is indicative of feeding demand. A simplified representation of the thermal condition for soundness can be derived from the thermal gradient (G) and solidification rate (R). While the full Niyama criterion ( \( G / \sqrt{R} \) ) was used in simulation, the comparison of LST is insightful.
For a simplified cylindrical feeding zone, the solidification time \( t_f \) can be related to the section modulus (Volume/Surface Area):
$$
t_f \propto \left( \frac{V}{A} \right)^n
$$
Where \( n \) is a constant (often ~2 for sand castings). In Scheme 1, the thick section (high V/A) lost its feeding source prematurely. In Scheme 2, the thick section was directly connected to the largest thermal mass (the pouring cup, with the highest V/A), ensuring:
$$
t_{f,\text{cup}} > t_{f,\text{thick section}} > t_{f,\text{thin section}}
$$
This hierarchy is essential for a defect-free lost foam casting process. The simulation quantitatively verified this hierarchy for Scheme 2 but not for Scheme 1.
Production Validation and Conclusion
Based on the decisive numerical simulation results, Scheme 2 (Thick-Section-Up) was selected as the optimized lost foam casting process for production. The EPS patterns were manufactured and assembled accordingly. The actual castings produced using this validated lost foam casting process were found to be fully sound upon visual and radiographic inspection. The critical involute surface was free from defects, and the dimensional accuracy met all specifications. The components successfully passed subsequent heat treatment and are performing satisfactorily in field service within the mining equipment.
In conclusion, this study successfully demonstrates the integral role of numerical simulation in the design and optimization of a complex lost foam casting process. For the asymmetric cast steel convex ring:
- The lost foam casting process scheme with the thick section oriented upwards and directly fed by a top-mounted pouring cup (Scheme 2) was numerically proven to establish a favorable solidification sequence. This scheme promoted effective directional solidification towards the feeder, thereby eliminating shrinkage defects in the critical casting body.
- The alternative scheme (Thin-Section-Up), while having a stable fill, failed to maintain a continuous feeding path to the thick section, leading to a high risk of internal shrinkage.
- The simulation-driven approach enabled a scientific comparison, eliminating the need for costly and time-consuming physical prototyping trials. The accuracy of the prediction was confirmed through successful production validation.
This workflow—combining astute lost foam casting process design principles with advanced numerical simulation—provides a robust and efficient methodology for developing high-integrity cast components, particularly those with challenging geometries prone to solidification-related defects. The successful outcome underscores the value of simulation as an indispensable tool in modern foundry engineering for optimizing the lost foam casting process.