As a researcher deeply involved in the advancement of casting technology, I have always been fascinated by the intricate challenges of designing effective riser systems for steel castings. The casting industry is a cornerstone of manufacturing, playing a pivotal role in economic development. The final quality of a casting is heavily dependent on the rationality of its process design, with riser design being one of the most critical aspects. For steel castings, which often have complex geometries and stringent quality requirements, the design of risers has been a persistent focus for engineers and technologists. The integration of computer-aided design (CAD) with riser design methodologies represents a significant evolutionary step, promising enhanced accuracy and efficiency. In this context, our team embarked on developing a comprehensive riser CAD system specifically for steel castings, built directly upon the source code of the domestic 3D graphics software SINOVATION. This endeavor aims to move beyond reliance on foreign commercial software and leverage indigenous technological capabilities to address the unique needs of steel casting production.
The core of our system revolves around three interconnected modules: a rapid hot spot analysis module using the point modulus concept, a local modulus calculation module for determining riser dimensions, and a 3D riser geometry generation module backed by a standardized database. Traditional methods for riser placement and sizing often rely on empirical rules or simplified calculations that may not fully capture the thermal dynamics of complex steel castings. Our approach seeks to embed scientific principles directly into the CAD environment, enabling a more analytical and automated design process. The significance of this work lies not only in the technical innovations but also in its contribution to fostering self-reliance in foundational industrial software for critical sectors like steel castings production.
The foundation of our riser CAD system for steel castings is the concept of point modulus, a powerful tool for rapid thermal analysis. The modulus, in casting science, represents the solidification time of a casting section. Extending this to every discrete point within a steel casting allows us to construct a complete solidification time field, revealing the sequence of solidification and identifying critical hot spots—regions most prone to shrinkage defects—that necessitate riser placement. For complex steel castings, this point-wise analysis is far superior to traditional geometric modulus calculations based on simplified sections. The point modulus, \( M_p \), at any location can be interpreted as an equivalent comprehensive heat dissipation distance. Its calculation is rooted in Fourier’s law of heat conduction. The heat conduction rate \( q_i \) in a direction \( i \) is given by:
$$ q_i = -\lambda A \frac{\Delta T}{\Delta x} $$
where \( \lambda \) is the thermal conductivity (W/m·K), \( A \) is the heat transfer surface area (m²), and \( \frac{\Delta T}{\Delta x} \) is the temperature gradient (K/m). The point modulus in direction \( i \), \( M_{pi} \), is inversely proportional to this conduction rate. We define it as:
$$ M_{pi} = – \frac{k_{\lambda}}{q_i} = \frac{d_i}{k_T} $$
Here, \( k_{\lambda} \) is a thermal conductivity coefficient, \( k_T \) is a heat storage capacity coefficient (the product of specific heat and density), and \( d_i \) is the distance from the center point of a computational grid cell to the mold wall in that direction. This formulation elegantly handles different boundary conditions. For instance, when a steel casting contacts a chill (like in direction 1 in our model), the effective thermal conductivity must account for the composite material path. If the heat transfer path consists of two materials with conductivities \( \lambda_1 \) and \( \lambda_2 \) over distances \( a \) and \( b \) respectively, the equivalent \( k_{\lambda} \) is calculated as:
$$ k_{\lambda} = \frac{a + b}{\frac{a}{\lambda_1} + \frac{b}{\lambda_2}} $$
In our implementation for steel castings, the casting and mold are discretized into numerous cubic grid cells. For each cell’s center point \( O \), the net heat conduction rate is the sum of differences between its outgoing rate and the incoming rate from adjacent cells in all 26 possible directions (considering face, edge, and vertex neighbors). The total point modulus \( M_p \) at point \( O \) is then the reciprocal of this sum, weighted by directional distance factors \( k_i \), which account for the varying distances to different neighbor types. The final comprehensive formula for the point modulus at any location in a steel casting is:
$$ M_p = \frac{1}{\sum_{i=1}^{26} \left( \frac{1}{M_{pi}(O)} – \frac{1}{M_{pi}(X)} \right) k_i } = \frac{1}{\sum_{i=1}^{26} \frac{k_T}{k_{\lambda}} \left( \frac{1}{d_i(O)} – \frac{1}{d_i(X)} \right) k_i } $$
where \( X \) is the adjacent cell center in the positive \( i \)-direction. By calculating \( M_p \) for every grid cell, we generate a solidification time field for the entire steel casting. Regions with higher \( M_p \) values solidify slower and are identified as hot spots. This process is computationally efficient; for a mesh of 10 million cells, the analysis completes in approximately 40 seconds, providing a rapid and accurate map of thermal centers. This capability is indispensable for initial riser positioning in complex steel castings, moving the designer from guesswork to data-driven decision making.
Once the hot spots in a steel casting are identified, the next step is to calculate the local modulus of the region requiring feeding. This local modulus is the primary input for dimensioning the riser using the modulus method. The challenge lies in accurately determining the volume \( V \) and the effective cooling surface area \( A \) of the isolated section of the steel casting that the riser must feed. The key is to correctly exclude non-cooling surfaces—areas where the section is in contact with another part of the casting itself rather than the mold. Our module in SINOVATION automates this. The designer can select the region of interest (the “local casting”) using a bounding geometry—a box, sphere, cylinder, or a custom solid. Let the total surface area of the entire steel casting be \( A_a \), the surface area of the local section be \( A_d \), and the surface area of the remaining casting be \( A_c \). The non-cooling interface area \( A_i \) between the local section and the rest of the casting is given by:
$$ A_i = \frac{A_c + A_d – A_a}{2} $$
Therefore, the effective cooling surface area \( A \) of the local section of the steel casting is:
$$ A = A_d – A_i = A_d – \frac{A_c + A_d – A_a}{2} = \frac{A_a – A_c + A_d}{2} $$
This calculation is performed automatically through programmed calls to SINOVATION’s geometric kernel functions for solid Boolean operations (like intersection), volume computation, and surface area calculation. Once \( V \) and \( A \) are obtained, the local modulus \( M \) for that section of the steel casting is simply:
$$ M = \frac{V}{A} $$
This value, along with the calculated volume \( V \), is stored and used as the basis for riser design. This automation eliminates the tedious and error-prone manual calculations traditionally associated with complex steel casting geometries.
With the hot spot location and the local modulus of the steel casting section known, we proceed to riser design. Our system is built upon the well-established modulus method, which is scientifically sound and widely applicable to steel castings of various complexities. The fundamental principle is that the riser’s modulus \( M_R \) must be greater than the casting’s modulus \( M \) to ensure directional solidification towards the riser. An expansion factor \( f \) is applied:
$$ M_R = f \cdot M $$
where \( f \) typically ranges from 1.1 to 1.3 for conventional risers in steel castings, and can be 0.7 to 0.9 if insulated riser sleeves are used. Foundries can select the factor based on their specific practices. To support this method, we established a comprehensive database of standard riser parameters for steel castings within an Access database. This database includes various types: standard top risers, side risers, blind risers, elliptical risers, etc. Each entry contains critical dimensions, the riser’s own modulus \( M_R \), its volume \( V_R \), and its feeding capacity (the maximum volume of steel casting it can feed for a given solidification shrinkage percentage \( \varepsilon \)). A subset of data for standard side risers is summarized in the table below, illustrating the structured information available to the CAD system.
| Riser Modulus \( M_R \) (cm) | Dimension d=a (mm) | Dimension b (mm) | Height h (mm) | Riser Volume \( V_R \) (dm³) | Feeding Capacity \( V_C \) (dm³) for ε=5% | Feeding Capacity \( V_C \) (dm³) for ε=7% |
|---|---|---|---|---|---|---|
| 1.89 | 100 | 53 | 150 | 1.04 | 2.2 | 1.4 |
| 2.08 | 110 | 59 | 165 | 1.38 | 2.9 | 1.4 |
| 2.26 | 120 | 64 | 180 | 1.79 | 3.8 | 2.4 |
| … | … | … | … | … | … | … |
| 11.71 | 620 | 329 | 930 | 248 | 523 | 329 |
The riser CAD system’s workflow is a seamless integration of these components. The process begins with the 3D model of the steel casting imported into SINOVATION. The rapid hot spot analysis module is executed to visualize areas of high point modulus. The designer then selects these hot spot regions, and the system automatically calculates their local modulus and volume using the bounding geometry method. Following the modulus method, the system queries the database: it first finds all risers whose modulus \( M_R \) satisfies \( M_R \geq f \cdot M \). From this filtered set, it selects the riser with the smallest \( M_R \) (for economic efficiency). It then verifies the feeding capacity; the riser’s available feed metal volume must be greater than the required feed volume for the steel casting section, which is \( V_{required} = \varepsilon \cdot V_{casting\_section} \). If the initially selected riser does not meet the feeding requirement, the system iteratively selects the next larger one until both modulus and feeding criteria are satisfied for the steel casting. Finally, the 3D riser generation module is invoked. Using the definitive riser type and dimensions (e.g., diameter, height), it calls SINOVATION’s native solid modeling functions to create the 3D solid model of the riser and positions it correctly on the steel casting. The entire flowchart is implemented within the SINOVATION environment, creating a cohesive and interactive design tool.
To validate the system, we applied it to a real-world component: a heavy-duty truck rear axle made of steel casting. This part features complex geometry with thick sections and intersecting ribs, making it a classic challenge for riser design. The point modulus analysis clearly revealed pronounced hot spots at the two supporting ends and on one side wall of the central section. Using the local modulus calculation, we determined the modulus and volume for these regions. The CAD system then automatically selected appropriate standard side risers from the database. Importantly, the system allows the designer to define a “feeding zone” larger than the immediate hot spot to ensure a proper temperature gradient for directional solidification. Therefore, the risers placed at the ends were designed to feed not just the hot spot itself but also adjacent sections of the steel casting. The final design, generated in minutes, included multiple risers with calculated dimensions and positions. The interface of the system during this operation provides intuitive selection tools and displays calculated parameters, making the process efficient. This case demonstrated a significant improvement in both accuracy, by using quantitative thermal analysis, and efficiency, by automating database lookup and 3D modeling tasks that would otherwise take hours for steel castings of such complexity.
In conclusion, the development of this riser CAD system within the SINOVATION platform represents a meaningful stride towards intelligent process design for steel castings. By working at the source code level, we achieved deep integration, allowing for the creation of specialized modules that leverage the software’s core geometric and computational capabilities. The system successfully combines advanced thermal analysis (via point modulus), automated local modulus calculation, a comprehensive standard riser database, and automated 3D geometry generation into a unified workflow. This not only enhances the scientific basis for riser design in steel castings but also dramatically reduces the time and skill threshold required for producing effective casting工艺. The ability to rapidly identify hot spots, accurately compute feeding requirements, and instantly generate compliant riser geometry empowers foundry engineers to focus on optimization and problem-solving rather than repetitive calculation and drafting. As the casting industry continues to evolve towards greater digitalization and precision, such indigenous software solutions tailored for critical components like steel castings will be instrumental in improving quality, yield, and competitiveness. Future work may involve integrating this CAD system directly with solidification simulation (CAE) for iterative validation and optimization, further closing the loop in the digital process chain for manufacturing high-integrity steel castings.