In the production of large and complex steel castings, ensuring soundness and minimizing defects like shrinkage porosity and cavities is paramount. A critical challenge lies in the design of the feeding system, particularly the risers, which can account for 30% to 50% of the total liquid metal poured. The subsequent cutting and recycling of these massive risers represent a significant consumption of manpower and resources, directly impacting production costs. My research focuses on enhancing the efficiency of the feeding process in steel casting. The principle is straightforward: improve the feeding capability of a riser either by increasing its feeding pressure or by prolonging its solidification time. Among various methods, the use of exothermic insulating riser sleeves and toppings has proven to be one of the most effective techniques. This article details my first-hand application of this technology, supported by numerical simulation and validated through practical production, demonstrating its potential to reduce costs and improve the quality of steel castings.
Exothermic insulating risers represent a synergistic combination of insulating and exothermic materials. When molten metal rises into the riser, the exothermic materials react violently upon contact, generating substantial additional heat. Once this reaction subsides, the insulating materials take over, dramatically reducing heat loss to the mold. This dual-action mechanism significantly extends the solidification time of the metal within the riser. This sustained liquid pool ensures that thick sections and thermal centers within the steel casting receive timely feeding, effectively preventing the formation of shrinkage porosity and cavities, thereby guaranteeing the denseness of the final casting. The feeding efficiency of such risers can reach 30% to 60%, far surpassing that of conventional sand risers.
To analyze and optimize this process, I employed the ProCAST numerical simulation software, a powerful tool widely used in the foundry industry for its robust meshing capabilities, extensive material database, and advanced fluid flow and thermal criteria. ProCAST can accurately predict defects such as mistruns, cold shuts, gas entrapment, and, most importantly for this study, the location and morphology of shrinkage porosity and macro-shrinkage cavities.
Product Analysis and Initial Process Design
The subject of this study was a large beam casting, a crucial component for an open rolling mill. The 3D model is shown in the referenced figure, with overall dimensions of 5582 mm x 447 mm x 356 mm. The material specification was ZG230-450 (a common cast carbon steel), with a rough casting weight of 6300 kg. The casting modulus was calculated to be 9.1 cm. All surfaces were to be machined, and the casting specification demanded freedom from cracks, slag inclusions, shrinkage, and gas holes—defects that could compromise mechanical performance.
Given the elongated, bar-like shape of this steel casting, achieving an effective feeding distance was a primary concern. The initial casting process was designed with two top risers to feed the casting. To promote directional solidification towards these risers, three chills were placed: two at the natural end zones of the casting and one at the center to create an artificial end zone, effectively dividing the casting into two separately fed sections. The molding was done using pit molding with assembled cores. The pouring temperature was set between 1550°C and 1560°C. A bottom-gating system was designed to ensure a calm and complete fill, with a controlled pouring rate of 150 kg/s. After pouring, conventional exothermic topping compound was used for the original sand risers.
Numerical Simulation Methodology
Geometric Modeling and Process Comparison
Two distinct 3D simulation models were created to compare the performance of conventional sand risers against the new exothermic insulating risers. The fundamental design rule for a conventional sand riser states that its modulus should be at least 1.3 times the modulus of the casting section it is intended to feed. For exothermic risers, this factor can be reduced to between 0.8 and 0.92. For this initial trial, a conservative factor of 1.1 was chosen to ensure product quality. Therefore, the target modulus for the exothermic riser was set at 10 cm (9.1 cm x 1.1). The riser dimensions were designed accordingly. The conventional process used standard riser sleeves and topping, while the new process utilized pre-formed exothermic insulating boards (FT400 type) and a matching exothermic insulating topping compound, which can be flexibly assembled around the riser pattern.
Material Properties and Boundary Conditions
The accuracy of a casting simulation heavily depends on the thermophysical properties of the materials involved. For the steel casting material ZG230-450, which was not in the standard ProCAST library, a custom material was created. Key properties like thermal conductivity and density as a function of temperature were calculated by the software based on the alloy’s chemical composition, as shown in the formulas below where $T$ is temperature in °C.
$$ k(T) = 40.0 + 0.02 \times T \quad \text{(for solid)} $$
$$ k(T) = 30.0 \quad \text{(for liquid)} $$
$$ \rho(T) = 7850 – 0.5 \times (T-1500) \quad \text{(approximation)} $$
The critical parameters for the exothermic insulating material, including its thermal conductivity, specific heat, density, and the kinetics of its exothermic reaction (heat release rate and duration), were provided by the manufacturer and input into a custom material database in ProCAST. The boundary conditions and initial parameters are summarized in the table below.
| Boundary Condition | Cast-Mold | Cast-Insulator | Mold-Chill | Cast-Chill |
|---|---|---|---|---|
| Heat Transfer Coefficient (W/m²K) | 700 | 200 | 500 | 2000 |
| Initial Condition | Material | Initial Temperature (°C) |
|---|---|---|
| Casting | ZG230-450 | 1560 |
| Sand Mold | Furan Resin Sand | 20 |
| Exothermic Riser Lining | FT400 Insulating Board | 20 |
| Chills | Low Carbon Steel | 20 |
Meshing and Computational Setup
An accurate mesh is crucial for simulation fidelity. In the ProCAST Mesh module, both models were discretized into tetrahedral volume elements. To balance computational accuracy and efficiency, a local mesh refinement technique was applied to areas of interest like the risers, casting hot spots, and gating system. The conventional riser model resulted in approximately 2.32 million volume elements, while the exothermic riser model had about 2.67 million elements. The filling simulation was set up with the defined pour rate and temperature, and the solidification simulation accounted for the exothermic reaction in the new riser design.
Simulation Results and Analysis
Filling and Solidification Sequences
The filling simulation confirmed the gating design was sound. Metal flowed smoothly from the pouring basin down the sprue, through the runner, and into the mold cavity via the ingates with minimal turbulence. The cavity filled completely in approximately 55 seconds without any predicted mistruns or cold shuts.
The solidification analysis revealed the most significant differences between the two processes. The temperature field evolution clearly showed directional solidification in both cases: starting from the chills and progressing towards the risers, with the central chill effectively splitting the thermal field. However, the timing was drastically different. The simulation predicted a total solidification time of about 10.2 hours for the casting with conventional sand risers. In contrast, the casting employing exothermic insulating risers required approximately 14.3 hours to solidify completely. This 40% increase in solidification time is directly attributable to the insulating and heating effect of the special riser lining. Cross-sectional views of the final temperature fields distinctly showed the high-temperature “hot spot” within the exothermic riser was sustained higher up in the riser compared to the conventional one, indicating a more efficient thermal profile for feeding.
Prediction of Shrinkage Defects
The primary metric for evaluating riser performance is its ability to concentrate shrinkage porosity within itself and away from the casting. ProCAST’s shrinkage prediction module provided clear visualizations of the expected macro-shrinkage cavity (primary pipe) and micro-porosity regions.
For the conventional sand riser, the predicted shrinkage cavity had the classic “V” shape, penetrating deeply into the riser. The calculated “safe feeding distance” – the height from the top of the casting to the lowest point of the shrinkage cavity – was about 73 mm. Micro-porosity was also predicted slightly closer to the casting-riser junction.
The results for the exothermic insulating riser were markedly better. The predicted macro-shrinkage cavity transformed into a shallower, wider “U” shape. More importantly, the entire cavity was pushed upwards. The safe feeding distance increased dramatically to approximately 240 mm. This upward shift of the shrinkage cavity and the reduction in secondary micro-porosity mean the casting body is far less likely to contain any shrinkage-related defects. The underlying mechanism is the reduced radial temperature gradient and heat loss from the riser walls, maintaining metal fluidity longer and allowing for more efficient mass feeding to compensate for solidification shrinkage throughout the entire process.
The relationship between riser efficiency and solidification time can be conceptually described by Chvorinov’s rule, where solidification time $t$ is proportional to the square of the volume-to-surface area ratio (modulus, $M$), modified by the insulating effect:
$$ t = k \cdot M^n $$
where $n$ is typically 2 for sand molds, but the constant $k$ is much larger for an exothermic insulating sleeve, effectively increasing $t$ for a given riser size.
Engineering Validation and Production Trial
Based on the promising simulation results, the new process with exothermic insulating risers was implemented in the actual production of the beam steel casting. The mold was prepared in a pit using the assembled core method, following the simulated layout including chills and the new riser assemblies.
Riser Cutting and Macro Examination
After shakeout, one of the two risers was removed via flame cutting for analysis. The macroscopic examination of the cut riser face revealed a shrinkage cavity concentrated in the upper portion of the riser. Its profile was a clear, shallow “U” shape, closely matching the simulation prediction. This visual confirmation was the first strong indicator of the simulation’s accuracy and the effectiveness of the exothermic material.
Non-Destructive Testing (NDT)
The cut riser section was then machined to create a smooth surface for non-destructive testing. The surface was inspected via Liquid Penetrant Testing (PT) according to GB/T 9443-2007 standards. Furthermore, Ultrasonic Testing (UT) was performed. Both NDT methods confirmed the absence of any shrinkage defects exceeding the acceptance criteria (Level 3, φ3 mm sensitivity) down to a distance of approximately 250 mm from the casting-riser junction. This measured “safe height” was consistent with, and even slightly better than, the 240 mm predicted by the simulation. Key measurements from the production riser are tabulated below.
| Measured Parameter | Value |
|---|---|
| Actual Riser Height after Pouring | 612 mm |
| Riser Height after Cutting | 585 mm |
| Measured Safe Feeding Height | ~250 mm |
| Shrinkage Cavity Depth | 165 mm |
| Estimated Cavity Volume | 0.02117 m³ |
Economic and Technical Analysis
The most direct benefit was the reduction in riser size and weight. The comparative data between the original and new riser designs is compelling.
| Riser Type | Dimensions (R x B x H) mm | Calculated Weight* | Modulus | Modulus Ratio (Riser/Casting) |
|---|---|---|---|---|
| Conventional Sand Riser | 280 x 230 x 650 | ~1780 kg | ~11.12 cm | 1.22 |
| Exothermic Insulating Riser (Design) | 250 x 200 x 550 | ~1190 kg | ~9.72 cm | 1.07 |
| Exothermic Insulating Riser (Actual) | – | ~1324 kg | ~10.08 cm | 1.11 |
*Density of steel taken as 7.3 t/m³ for calculation.
The use of exothermic insulating risers led to a reduction in single riser weight from 1780 kg to 1324 kg, a saving of 456 kg per riser. For the two risers on this steel casting, the total saving was 912 kg of liquid steel. Factoring in the cost of the exothermic materials versus the saved steel, the net cost reduction for this single casting was approximately 3000 CNY. For alloy steel castings, the savings would be substantially greater. Furthermore, the NDT-confirmed safe height of 250 mm suggests the riser design was conservative. The height could potentially be reduced by 100-150 mm for similar steel castings in the future, which would increase the yield (weight of sound casting / total poured weight) from the current ~65% to between 66.9% and 67.6%, offering further cost reductions. This technology has since been successfully applied to other complex steel castings like valve bodies and upper half turbine casings with equally positive results.
Conclusion
This integrated study, combining numerical simulation with physical production trials, successfully demonstrates the significant advantages of using exothermic insulating risers in the production of heavy-section steel castings. The ProCAST software proved to be an invaluable tool for accurately simulating the filling and solidification processes, predicting shrinkage defect morphology, and optimizing the riser design before any metal was poured.
The key findings from my work are:
- The application of exothermic insulating riser linings and toppings effectively prolongs the solidification time of the riser by over 40%, creating a more favorable thermal gradient for directional solidification.
- This extended feeding capability transforms the shrinkage cavity from a deep “V” shape to a shallow “U” shape and, most critically, pushes the cavity and associated micro-porosity further away from the casting body. The safe feeding distance increased from ~73 mm to over 240 mm, substantially reducing the risk of shrinkage defects in the final steel casting.
- For the studied steel casting with a modulus of 9.1 cm, designing the exothermic riser with a modulus ratio of 1.1 (riser modulus / casting modulus) was successful and conservative. The confirmed large safe height indicates that for future projects, this ratio could be pushed towards the lower end of the recommended range (0.9-1.0), enabling even smaller risers and higher yield.
- The direct economic benefit was a reduction in riser weight by 25.6%, saving nearly a ton of liquid steel per casting and reducing total cost. The improvement in guaranteed quality and reduction in potential rework/scrap provide additional, less quantifiable value.
In summary, the adoption of exothermic insulating riser technology, guided by numerical simulation, presents a highly effective strategy for enhancing the quality, reliability, and cost-efficiency of steel casting production, particularly for large and critical components where internal soundness is non-negotiable.