Development of a CAD System for Sand Casting Foundry of Shaft Parts Based on Pro/Engineer

This thesis presents a comprehensive investigation into the development of a Computer-Aided Design (CAD) system specifically tailored for sand casting foundry processes of shaft-type components, built upon the Pro/Engineer platform. The research addresses the critical gap between generic CAD tools and the specialized needs of the casting industry, particularly focusing on the automation of casting process design. By leveraging the powerful secondary development capabilities of Pro/Engineer through its Pro/Toolkit application programming interface, this work establishes a robust framework that integrates three-dimensional solid modeling with casting-specific design methodologies. The system encompasses essential modules including part classification, casting type selection, process parameter determination, gating system design, and riser system calculation. A significant contribution of this research is the development of a comprehensive casting process database using Data Access Objects technology, which stores empirical data and standard parameters derived from casting handbooks and industrial practice. The practical application of this system is demonstrated through a case study involving the design of a ZG270-500 crankshaft casting, showcasing the system’s capability to streamline the entire casting process design workflow. The findings indicate that the developed system significantly reduces design time, minimizes defects, and enhances the overall efficiency and quality of sand casting foundry operations.

Introduction to Casting Technology Development

Metal casting represents one of the most ancient manufacturing techniques known to humanity, yet it remains fundamentally important in modern industry. The process involves transforming materials through a solid-liquid-solid cycle, which presents unique challenges for process design due to the complex physical and chemical transformations involved. Traditional casting process design has historically relied heavily on empirical knowledge and accumulated experience, often resulting in extended design cycles, numerous modification iterations, and a lack of scientific rigor. The advent of computer technology, however, has opened new avenues for addressing these long-standing challenges that have impeded the advancement of casting production.

The integration of computer technology into the foundry industry has become an inexorable trend, driven by the need to enhance product quality, reduce material consumption, improve production efficiency, and meet increasingly stringent environmental standards. As a major casting-producing nation, China has witnessed remarkable growth in its foundry sector, with annual production exceeding 17 million tons in 2002, ranking first globally for three consecutive years. However, the industry faces persistent challenges including high defect rates, excessive material waste, and significant quality inconsistencies. The escalating costs of raw materials, including scrap steel, pig iron, and coke, have further exacerbated pressures on casting enterprises, compelling them to seek innovative solutions.

Computer-aided engineering (CAE) has emerged as a transformative force in modern casting technology. Casting CAE encompasses three principal modules: pre-processing, central processing, and post-processing. The pre-processing module handles three-dimensional solid modeling and mesh generation, while the central processing module performs numerical simulation of mold filling and solidification processes, predicting shrinkage porosity, shrinkage cavities, and thermal stress distributions. The post-processing module visualizes these results, enabling engineers to identify potential defects and optimize process parameters iteratively until sound castings are achieved. Within this framework, casting process CAD serves as a critical component of the pre-processing phase and represents the ultimate objective of casting CAE implementation.

The global research landscape in casting technology has shifted toward several key directions, including precise forming technologies, advanced simulation techniques, and intelligent process optimization. In particular, computer simulation and virtual manufacturing have emerged as indispensable tools for replacing traditional trial-and-error methodologies. The United States’ Next Generation Manufacturing initiative identified advanced manufacturing processes and pervasive modeling/simulation as critical foundational technologies, underscoring the strategic importance of computational approaches in modern manufacturing.

Selection of Development Platform and Programming Language

The selection of an appropriate CAD platform and programming language constitutes a fundamental decision that significantly influences the success of secondary development efforts. Unlike conventional software development that builds systems from the ground up, secondary development extends and enhances existing software capabilities. Consequently, the performance and functionality of the developed system are substantially constrained by the underlying support software’s capabilities and openness. This necessitates a thorough evaluation of available options before committing to a specific development path.

Commercial CAD/CAM software systems have evolved considerably over recent decades, and several prominent products dominate the market. Table 1 provides a comprehensive comparison of major commercial CAD systems considered for this research.

Table 1: Comparison of Major Commercial CAD Systems

Several criteria guide the selection of a suitable CAD system for secondary development. System functionality and configuration capabilities must align with the intended application domain. The price-performance ratio represents a crucial economic consideration, particularly for small and medium-sized enterprises. Hardware compatibility ensures smooth operation across different computing environments. The availability and sophistication of secondary development tools determine the feasibility of creating customized functionality. System openness facilitates integration with other software and databases. Finally, vendor reliability, support services, and upgrade paths influence long-term sustainability.

After comprehensive analysis, Pro/Engineer was selected as the development platform for this research. This decision was influenced by several factors. Pro/Engineer’s parametric design technology provides an ideal foundation for CAD system development, enabling flexible and efficient design modifications. The software’s integrated finite element analysis capabilities align well with the requirements of casting CAE applications. The research team’s prior experience with Pro/Engineer and ProCAST integration provided valuable foundational knowledge. Furthermore, Pro/Engineer 2001 offers comprehensive secondary development interfaces, making it highly suitable for creating customized casting process design tools.

For the programming language, Visual C++ 6.0 was chosen based on its exceptional development environment, comprehensive MFC class library support, and seamless integration capabilities. The Visual C++ 6.0 Developer Studio provides an integrated suite of tools including text editing, resource management, project building, optimization compilation, incremental linking, source code browsing, and debugging facilities. The MFC library offers robust support for Windows application development, including sophisticated graphical user interface components and database access functionalities. The Data Access Objects (DAO) technology integrated within Visual C++ 6.0 enables efficient database application development, which proved essential for creating the casting process parameter database.

Pro/Engineer Secondary Development Technology

Pro/Engineer, developed by Parametric Technology Corporation (PTC), has achieved widespread adoption in manufacturing industries worldwide. Its feature-based parametric design philosophy and complete associativity distinguish it as a leading solution for three-dimensional product development. The software was introduced in 1986 and has since evolved to incorporate advanced theories and technologies in CAD, establishing a unified data structure across all functional modules.

Several secondary development approaches exist for Pro/Engineer, each with distinct characteristics and applicability. The Pro/PROGRAM module provides a BASIC-like programming environment for controlling model parameters and features. Pro/Toolkit, developed in the C language, offers comprehensive application programming interfaces for deep integration with the Pro/Engineer database. J-Link enables Java-based customization and extension. Automation Gateway facilitates Visual Basic-based development. Additionally, Family Tables and User-Defined Features provide table-driven and feature-based customization approaches, respectively.

Table 2: Pro/Engineer Secondary Development Methods

Pro/Toolkit emerged as the preferred choice for this research due to its comprehensive interface capabilities and alignment with our C/C++ programming environment. Pro/Toolkit provides functions for menu customization, information display, database access, feature manipulation, geometry retrieval, assembly management, and section generation. The architecture employs an object-oriented approach where functions operate on specific object types, with naming conventions following the pattern: “Pro” prefix + object name + action name.

Two primary development modes exist for Pro/Toolkit applications: synchronous and asynchronous modes. The synchronous mode, controlled entirely by the main Pro/Engineer process, offers higher communication efficiency and is generally preferred. Within synchronous mode, two implementation strategies are possible: dynamic link library (DLL) mode and multi-process mode. DLL mode generates an object file that loads with Pro/Engineer, providing rapid execution. Multi-process mode produces an independent executable that runs as a sub-process of Pro/Engineer, which is particularly advantageous during development and debugging. For this research, the multi-process mode was employed during development to avoid restarting Pro/Engineer for each debug iteration, while DLL mode was intended for final deployment.

Framework Development for Casting Process CAD System

The overall architecture of the sand casting foundry CAD system was designed to follow the conventional casting process design workflow. The system comprises two primary components: three-dimensional solid modeling and casting process computer-aided design. The solid modeling component operates within the standard Pro/Engineer environment, while the casting process CAD component represents the core innovation of this research.

Table 3: System Architecture Components

The menu structure of the casting process CAD system was designed based on the casting process design flow and casting engineering database requirements. The system navigates through three main modules: part type, casting type, and casting part module. The part type module enables selection and management of part categories, including shaft-type, wheel-type, and plate-type components, providing functionality for adding, querying, retrieving, and modifying parts. The casting type module configures the casting method, such as sand casting, metal mold casting, or investment casting, with each type associated with corresponding databases. The casting part module varies according to part category and encompasses several essential functions.

Material definition enables specification of alloy type and associated physical and mechanical properties, leveraging Pro/Engineer’s native material definition capabilities. Parting surface determination interfaces with Pro/Engineer’s datum plane functions to establish mold parting lines, which are critical for determining pouring position and casting orientation. Casting process parameter setup addresses machining allowance, shrinkage rate, draft angle, and casting fillet radius by querying the parameter database and applying modifications directly to the model. The riser system module guides users through heat zone identification, modulus calculation, riser type selection, riser size determination, riser count calculation, riser installation, and feed distance validation. The gating system module encompasses pouring type classification, gate count determination, runner design, and sprue calculations, ensuring proper metal flow characteristics.

Table 4: Casting Process Parameter Modules

Casting Process Database Development

The development of a comprehensive casting process database represents a significant achievement of this research. The database serves as a centralized repository for the extensive empirical data and standardized parameters required during casting process design. Using Microsoft Access 2000 as the database management system and Visual C++ 6.0 with DAO technology for the front-end application, the database provides efficient management and maintenance of casting process parameters.

DAO (Data Access Objects) provides a robust framework for database access within Microsoft Windows applications. The MFC library fully encapsulates DAO objects, offering classes such as CDaoDatabase, CDaoRecordset, CDaoTableDef, and CDaoFieldInfo that simplify database operations. The database design process involved creating tables for various casting parameters based on data from established casting handbooks and industrial experience.

Table 5: Database Structure for Sand Casting Parameters

An important technical consideration involves ensuring backward compatibility of the Access database file. Since the development environment may require earlier Access versions, the database file must be converted using the database utility functions in Access, specifically the “Convert Database to Earlier Access Database Version” option under the Tools menu.

The interface between Pro/Engineer and MFC applications was implemented through a carefully designed communication architecture. A primary executable (A.exe) hosts the Pro/Toolkit application, while a dynamic link library (B.dll) contains the MFC dialog classes and database functionality. The Pro/Toolkit program within A.exe calls exported functions from B.dll, enabling seamless integration of the casting process database with the Pro/Engineer environment. This approach leverages the DLL communication mechanism to provide rapid function calls and efficient data exchange.

Table 6: Interface Implementation Architecture

The database interface utilizes spawn mode loading for the Pro/Toolkit application within Pro/Engineer, followed by calls to the MFC application through the DLL. This architecture was validated through successful operation of the system, demonstrating the feasibility of combining spawn mode, DLL technology, and MFC dialog invocation within the Pro/Engineer environment.

Implementation of Functional Modules

Building upon the fundamental framework and database infrastructure, individual functional modules were implemented to support the complete casting process design workflow. The part type module, casting type module, and casting part module work in concert to guide users through the design process.

The casting part module represents the heart of the system, providing step-by-step guidance for casting process design. The material definition function allows users to select materials and input corresponding physical and chemical properties. The parting surface module connects with Pro/Engineer’s datum plane creation tools, enabling visual and intuitive part orientation. The casting process parameters module provides access to databases containing recommended values for machining allowances, shrinkage rates, draft angles, and casting fillets.

In the riser system module, the modulus method was implemented as the primary calculation approach. The modulus of a casting section is defined as the ratio of volume to cooling surface area, which governs solidification time. For complex castings, the component is subdivided into simpler geometric elements, and each element’s modulus is calculated individually. The module provides designers with algorithms for basic geometric shapes, enabling accurate modulus determination for various casting geometries.

The mathematical foundation for riser design involves the relationship:

$$M_{riser} = f \cdot M_{c} $$

where $M_{riser}$ is the riser modulus, $M_{c}$ is the casting modulus at the location requiring feeding, and $f$ is a safety factor. For top risers, $f$ equals 1.2, while for side risers, $f$ equals 1.1. Once the required riser modulus is established, the system provides access to a standard riser database containing geometric data for various riser types, enabling designers to select appropriate riser dimensions that satisfy the modulus requirement.

Table 7: Standard Riser Selection Criteria

| Riser Type | Modulus Range (cm) | Max Feeding Weight (kg) | Application |
|————|——————-|————————|————-|
| Top Riser | 8-15 | 2,000-8,000 | Large castings |
| Side Riser | 6-12 | 1,500-6,000 | Medium castings |
| Blind Riser | 5-10 | 1,000-4,000 | Compact castings |

The gating system module calculates runner cross-sections based on the choke area method. The fundamental formulas govern the relationship between pouring time, filling rate, and gating system dimensions. The pouring time calculation follows the equation:

$$ t = \frac{c}{\sqrt{G}} $$

where $t$ is the pouring time in seconds, $c$ is an empirical coefficient related to casting thickness, and $G$ is the total casting weight in kilograms. The choke area is then determined from:

$$ F_{chok} = \frac{G}{t \cdot \mu \cdot v_{fill} \cdot \gamma} $$

where $\mu$ is the flow coefficient, $v_{fill}$ is the filling velocity, and $\gamma$ is the liquid density. The system provides databases containing recommended runner ratios and section dimensions, with typical gating ratios for steel castings being:

$$ F_{sprue} : F_{runner} : F_{gate} = 1 : 1.2 : 1.4 $$

Case Study: Crankshaft Casting Process Design

To validate the developed system, a comprehensive case study was conducted on a crankshaft component processed by a company in Lanzhou. The crankshaft had a maximum diameter of 1270 mm and a total length of approximately 1720 mm, featuring three substantial throws with significant thermal nodes formed at the intersections of throws and main journals.

The production conditions specified small-batch manufacturing using ZG270-500 material. Mechanical performance requirements included yield strength of at least 585 MPa, tensile strength exceeding 725 MPa, elongation minimum of 17%, and reduction of area minimum of 35%. The factory mandated normalizing plus tempering treatment to achieve hardness ranging from 217 to 248 HB. Surface quality requirements prohibited sand adhesion, oxide scale, and harmful surface defects.

The casting process design commenced with three-dimensional solid modeling using the developed system. The part library was consulted to identify similar crankshaft configurations, and since the specific geometry was not readily available, manual modeling was performed using Pro/Engineer’s comprehensive solid modeling tools. The created model accurately represented the crankshaft geometry including all complex features and transitions.

Material definition was established early in the design process. The unit system was configured to mm-tonne-second notation, and steel was selected as the material. Material properties including density and relevant physical constants were input to enable subsequent calculations. The parting surface was determined using Pro/Engineer’s datum plane insertion capability, selecting the plane that best suited the casting orientation and mold configuration.

The casting process parameters module was employed to establish machining allowances, shrinkage rates, draft angles, and casting fillets. For steel casting with ZG270-500, the shrinkage rate was determined by querying the material-specific database. The machining allowances were applied based on part classification and required surface finish specifications.

Gating system design for the crankshaft was performed using the steel flow velocity calculation method. This method determines the appropriate rising velocity of molten steel in the mold cavity and utilizes empirical databases to establish recommended velocity ranges. The system calculates pouring time based on steel weight, ladle capacity, and number of pouring orifices using the relationship:

$$ t = \frac{G}{N \cdot n \cdot V} $$

where $G$ is the total steel weight (kg), $N$ is the number of ladles, $n$ is the number of pouring orifices per ladle, and $V$ is the steel pouring velocity (kg/s). The steel rising velocity is subsequently verified using:

$$ v_{rise} = \frac{C}{t} $$

where $C$ is the casting height in the pouring position (mm) and $t$ is the pouring time (seconds). If the calculated velocity falls below the recommended database value, the system prompts adjustment of ladle capacity or orifice diameter.

Table 8: Steel Pouring Velocity Database Values

| Casting Weight (kg) | Minimum Rise Velocity (mm/s) |
|——————–|——————————|
| < 5,000 | 25 |
| 5,001-15,000 | 20 |
| 15,001-35,000 | 15 |
| 35,001-65,000 | 10 |
| > 65,000 | 8 |

The riser system was designed using the modulus method. The large end of the crankshaft was identified as a critical feeding location, where thermal nodes could lead to shrinkage defects. Initial modulus calculations yielded a casting modulus of 10.02 cm. Using the riser safety factor of 1.2, the required riser modulus was calculated as 12.02 cm. Consultation of the standard riser database identified an appropriate riser with modulus 11.96 cm and maximum feeding weight of 4,340 kg, which exceeded the required feeding weight of 4,338 kg.

The gating and riser systems were strategically positioned on the crankshaft, with risers placed at the two substantial throws where defects were most likely to occur. The smaller throw, positioned at a lower elevation, benefited from self-feeding by the casting itself and did not require separate riser installation. The final casting assembly included both gating and riser systems, enabling complete process visualization before actual production.

The calculated process yield for the crankshaft casting reached approximately 60 percent, demonstrating the effectiveness of the system in optimizing foundry economics. This yield calculation considers the ratio of casting weight to total poured weight, including gating and risering systems.

Summary and Future Directions

The primary contributions of this research are the successful development of a casting process CAD system built upon Pro/Engineer leveraging Pro/Toolkit and the development of a comprehensive casting process database using DAO technology within the Pro/Engineer environment. The multi-process development mode proved highly effective in reducing debugging time by eliminating the requirement to restart Pro/Engineer for every application iteration. This practical discovery enhances developer productivity substantially and provides valuable guidance for future Pro/Engineer development projects.

Key technical achievements include the integrated use of spawn mode and DLL technology for MFC dialog invocation, which enables sophisticated user interface components directly within the Pro/Engineer environment. This approach successfully bridges the gap between the complex Pro/Toolkit function library and the more accessible Windows user interface paradigms. The development of the casting process database established a robust data foundation for process automation, incorporating empirically validated parameters from established standards and industrial practices.

The system was successfully applied to the crankshaft casting design, validating its suitability for engineering application in sand casting foundry production. The system effectively supports computer-aided design for shaft-type steel castings, producing optimized process designs that minimize defects and improve casting quality.

Future research directions focus on several areas of enhancement. The casting process CAD system should be expanded to integrate with Computer-Aided Process Planning (CAPP) and Computer-Aided Manufacturing (CAM) technologies, creating a more comprehensive digital manufacturing chain. The system should incorporate additional modules including chill design, core design, and tooling design capabilities. The current modular architecture should evolve toward universal applicability rather than being restricted to classified component types. Future implementations could decompose any component into combinations of fundamental shapes such as shafts, plates, and cylinders, then apply relevant design modules to each component before integrating all design outputs.

The database infrastructure requires further refinement, with consideration given to network-enabled database platforms that support collaborative design and distributed manufacturing environments. The continuing evolution toward platform-independent casting process CAD systems will further broaden the accessibility and applicability of these valuable design tools, potentially enabling seamless integration across different CAD environments.

In conclusion, the growing complexity of modern casting production demands robust computational tools that enhance efficiency, quality, and sustainability. The development of specialized CAD systems such as the one presented in this research represents an essential step toward modernizing sand casting foundry practices and addressing the technological challenges of contemporary manufacturing. The continued evolution of these tools will play an increasingly vital role in the sustainable advancement of the global foundry industry.