Linking foundry operations with computer-aided engineering.

Integration of foundry operations is being accomplished with the use of computer-aided engineering at CMI International Inc.

Dramatic improvements in quality and productivity in metalcasting are possible using existing computer technology. An example is Computer-Aided Engineering (CAE). It can raise casting quality, reduce casting weight and costs and virtually eliminate prototyping. Products are made market-ready faster.

The integration of CAE sets the groundwork for future continuous improvement in casting design, validation and product performance. Some specific CAE analysis areas addressed are:

* gating, risering and solidification;

* fluid flow and heat transfer;

* mechanical and thermal stress analysis;

* dimensional analysis;

* computer-aided design (CAD) and tooling manufacture.

Engineering Demands

Continuous change has characterized the status of U.S. foundries during the last 15 years, resulting in the need to improve efficiency and reduce casting weights by designing thinner walls and lighter sections. Substituting materials (aluminum, magnesium, composites) and utilizing new casting processes that optimize castability have helped implement these changes.

Other changes being investigated are ways to reduce weight and improve casting quality and performance. Castings now must withstand higher static and dynamic stresses, higher temperatures and more corrosive environments. These changes are the result of U.S. carmakers' efforts to build lighter vehicles to improve fuel economy and reduce emissions standards.

The use of CAE, CAD, coordinate measurement machines and new nondestructive testing techniques have focused attention on over-designed, out-of-tolerance, faulty castings.

Only when its customers redefined quality standards and started to look at alternative manufacturing methods, did the U.S. foundry industry react. As excessive product and tooling costs and a lengthy time-to-market, forced customers to other manufacturing processes, foundries began investigating new technologies.

It is evident that any new foundry technology must be judged on its abilities to:

* optimize design and weight;

* improve casting quality and performance;

* minimize implementation and validation time;

* reduce costs.

CAE offers a positive means of achieving all of these.

The Role of CAE

CAE in the U.S. has been broadly expanded from its origins as a tool for design and drafting. In the foundry industry, its application as a modeling, simulation and analysis tool has increased exponentially during the last five years. The exact level of utilization usually is dependent on the degree of sophistication and the needs of a user's customer base.

For simple parts, weight reduction and optimized product performance are not always the utmost concern. Where the customer is merely looking for functionality and low cost, advanced analysis capabilities may not be required and computerized drawing may suffice.

However, when performance, weight and cost become driving forces, the use of advanced technologies assumes greater importance. Where full customer support and high value products are needed, the foundry must be able to provide a diverse range of capabilities or its available market will be limited. Other services that may be supplied to customers by value-based metalcasters are:

* gating and risering optimization;

* design for manufacturability and assembly;

* statistical variation simulation;

* parametric analysis;

* fatigue analysis.

CAE also is used in the transition from product concept to casting design and to the manufacture and evaluation of foundry tooling. Programs have been developed that automatically configure the parting for a particular component to obtain optimum castability, maximizing parts per mold and optimizing gating and risering requirements. Metal shrinkage rates can be factored into the tool design and made nonuniform, if necessary, to ensure the universal dimensional accuracy of the part.

Development of machine tool cutter paths is computer-based and required for numerical control (NC) machining of patterns, coreboxes, permanent molds, dies and checking fixtures. After the tool build and the production of initial castings, reverse engineering techniques can be used to validate tooling and parts relative to the product design. Parts then can be tested and the results correlated to the original theoretical analyses using closed loop computer systems to continuously improve the accuracy of CAE models and predictions.

CAE Systems

CAE must be implemented as a system, with software and hardware compatible with those of the primary customers or inefficiencies will result. That means incompatible computer data must be redeveloped causing lost time and competitive advantage. The ability to manipulate data rapidly without loss of integrity is important and should be a major consideration in the selection of a system.

The use of computer networking has gained considerable support within the U.S. since networks address many of the shortcomings noted previously. A decision must be made, however, relative to the selection of a centralized mainframe computer system with branched workstations, or stand-alone, independent networked workstations.

Regardless of the hardware network selected, communications among software platforms are a necessity. U.S. and worldwide users have recognized this need and developed a number of specifications, such as IGES (initial graphics exchange specification), PDES (product data exchange using step), STEP (standard for exchange of product) and DMIS (dimensional measurement interface specification), to address the problem.

Translators have been developed based on these specifications. While these specifications are well-defined, problems persist. Software developer interpretations of written specifications continue to create shortcomings with one-to-one data transfers. Some specifications do not support data used in the CAE process. For example, solid models are required for thermal stress analysis but IGES does not include a provision for them. In such cases, it is still necessary to develop secondary data bases, adversely affecting analysis time and, ultimately, the time-to-market.

Another factor to be considered with the use of a CAE system is the technical level at which to start. While customers may be driving the foundry to use advanced approaches for analysis and design, it is advisable to start slowly. For CAD, it is recommended that the foundry begin with 2-D drawing and design and progress into 3-D wireframes and, finally, surface models. With this type of design data, NC machining cutting paths can be developed for use with computer-aided manufacturing of foundry tooling.

Simplified programs for creating solid models should be used, as well as software that has automeshing capabilities for finite element analysis (FEA). As more complex analyses are required, more sophisticated methods can be implemented.

Application of CAE

The application of CAE simplifies the analysis, design, tooling construction and casting of an optimized product design. For example, a connecting rod had long been produced using conventional forging techniques. The customer, a manufacturer of automobile engines, wanted to significantly reduce rod weight and initiated a design competition among suppliers of forgings, castings and powder metal products. The contest was to determine who could supply the lightest part able to meet all performance requirements at the lowest cost in the shortest time. CAE was the key.

Modeling

Based on design criteria supplied by the customer (tensile and compressive loading during engine operation, specified safety factors, etc.), an initial 2-D concept design was developed followed by a 3-D wireframe model. This data base was transferred to two other software platforms, one for solidification analysis and the other for FEA. The latter was conducted to better define the loading on the part and determine if design or material refinements were necessary.

Initially, the wireframe part model is converted into a solid computer model. From this, the various mass properties of the part, such as weight and center of gravity, are identified. If the design is acceptable, the model can be meshed for finite element modeling. Depending upon the particular program, automeshing capability may be available to generate a broad mesh, which can then be refined for a particular design.

After meshing, loads are applied, forming a free body diagram. Stress analysis follows using various failure theories related to given safety factors. Color or shade-coded analyses of stress and/or deformation levels can be provided. Some systems allow easy design and material refinements and gives the engineer a "what if..." analysis using different design criteria.

Solidification Modeling

Concurrent solidification modeling, using the original data base is possible at a different workstation. Depending on the complexity of the part, either 2-D finite difference or 3-D finite element programs are used. The program also allows the mold material to be altered and the heating and cooling of the mold to be investigated. The 3-D analysis is useful for more complex designs and zones in which problems might be expected.

Once the stress and solidification evaluations are complete, and the part is optimized, a revised concept is developed. Often a series of design iterations is necessary and compromise between design, manufacturing and assembly concerns is required to complete the part design.

By providing rapid input and making even subtle refinements at this early stage of design, the use of CAE reduces prototyping and time to produce a sound part meeting the specified performance criteria.

Having completed the connecting rod design, surface models are created. Often these are shaded using the CAD system to give a lifelike appearance to the part that offers the casting engineer one last opportunity to "see" the casting, and review the parting and other features without the cost and lost time necessary to physically build a model.

Tooling

After accepting the part, CAD of the pattern tooling and coreboxes can begin. This confirms all partings and optimizes the mold layout. Once completed, a tool cutting path model is developed for direct computer numerical control machining (CNC) of the tooling. Use of CNC machining centers with precision cutter paths allow rapid, accurate and repeatable tooling construction and complete exploitation of the capabilities of CAE and design.

To ensure that tooling and product are produced to the design intent, detailed dimensional and statistical analyses are conducted on a computer controlled coordinate measurement system. This equipment is linked via a closed loop network to both the CAD and CNC machining centers to provide direct data comparison.

The coordinate measurement system also has the potential for design refinement since it is capable of providing reverse engineering. Design changes made manually can be digitized on the coordinate measurement machine and the new data then transferred back to the CAD, resulting in an updated data base. Cutter paths then can be regenerated and tooling revisions easily made. Now, molds can be produced duplicating the revised CAD model and customer requirements.

CAE Analysis

The use of CAE described above outlines the simplest application of these systems and technologies to the manufacture of high volume production casting. More sophisticated software programs further enhance the beneficial aspects of CAE to improve product quality and manufacturability.

Mechanical Stress--The prediction and analysis of mechanical stress is not only important in simulating static conditions, but also in evaluating dynamic situations where there might be impact or similar loading. Figure 4 demonstrates the use of FEA to evaluate the stress developed on a ductile iron lower control arm casting. Analysis also can show the suspension system stresses of the arm as it hits a curb or a pothole.

Finite element stress analysis also can be used to evaluate failures. For example, a failed ductile iron axle had no metallurgical or structural basis for failure at the fracture location. A stress analysis determined that the part was underdesigned by the customer for the material utilized.

In a similar manner, these tools can be used to evaluate casting failures in subsequent processing operations where stresses not associated with the casting process or inferior materials may develop.

Heat Transfer Analysis--This is important to the foundryman in two areas. It can be used to evaluate processing conditions to ensure that, for example, the correct mold materials are being used and desired cooling rates are being achieved. Figure 5 shows an analysis of cooling in an engine block casting that was produced using different molding media and chills to uniformly speed the cooling rate and develop specific properties to minimize residual stresses.

Heat transfer within the product itself also is an important consideration because it may affect various casting parameters like wall thickness and material selection for specific products, i.e., engine blocks and heads requiring water cooling.

Heat transfer in components such as exhaust manifolds is important because of increased emphasis on catalytic converter light-off for emissions control. While wall thickness reductions and alternate casting materials are being considered, the use of insulating coatings also is gaining interest and relies on heat transfer measurement for performance.

Fluid Flow--Fluid flow can be computed from both a process and product perspective. Metal flow in gating systems and within a mold can be analyzed readily. Also, air and water flow within a particular component, such as the effect of water flow around the cylinder bores in an engine block and can be estimated using CAE techniques.

Thermal stress analysis--Thermal stress analysis represents one of the most complex applications of CAE available to the casting designer. It combines the technologies of fluid flow, heat transfer and mechanical analysis to predict the development of thermal stresses within a part. Used to evaluate a number of different engine castings, it has been particularly useful in optimizing the design of exhaust manifolds.

The thermal stresses developed in exhaust manifolds are based on operating temperature and temperature variation within the part. Knowing an inlet gas temperature and flow rate, gas flow within a manifold can be modeled. From this, the temperature attained in different sections of the manifold can be predicted. It is then possible to use the temperature data to establish the thermal stresses developed in the part because of temperature differences in adjoining sections due to geometric variations.

The foundryman's objectives at this point are to use this data to ensure that the design and material selected will meet performance, manufacturability and quality standards. In this way, he assures his customer's requirements and his foundry's profit goals are met.

In the example illustrated in Fig. 6, the customer wanted the manifold to weigh less. Moreover, the heavy sections were prone to shrinkage defects and the foundryman also could benefit from weight reduction in selected areas as long as performance, from a thermal durability standpoint, was not compromised. As a result, the design of the part was reevaluated and a series of thermal stress analyses was conducted to ensure that similar or better levels of performance were attained.

Without compromising thermal durability, the CAE redesigned exhaust manifold weighed 37% less and was more easily cast. Conducting these analyses also eliminated the costly, recurring prototype expenses and shortened the time-to-market.

Solidification--Using sophisticated gating and risering analysis, the foundryman can optimize the feeding system and maximize the number of good parts per mold. From the aspect of solidification, the ability to predict and control structures is becoming increasingly important to customers.

For example, specifications for microstructures have been formulated for the high stress levels developed in the combustion chambers of aluminum cylinder heads. To achieve these refined structures in the areas designated, very high solidification rates must be reached. Solidification analysis thus becomes useful in developing casting designs and molding parameters that meet these requirements.

Variation Simulation--This also is a valuable tool for the foundryman in assessing manufacturability. Using CAE software with the ability to predict causes and the amount of variability, it is possible to estimate tolerance stack-ups and dimensional capability, as well as minimum wall thicknesses for a particular molding/core process.

This software, uses the CAD of the casting and selected process capability information, has the ability to isolate sources of variability and then rank their relative importance using pareto analysis to direct corrective actions. Evaluations can vary from simple to complex. The objective is to minimize the larger sources of variability to improve the overall robustness of the production process.

Among the advantages of utilizing computer technologies, is the reduction in time and effort needed to attain high levels of quality even with the introduction of new products. It involves the metalcaster in a leadership position in the design of a part that assures manufacturability, productivity and profit.

By having a working knowledge of the product and providing full service capabilities, the metalcaster is in a better position to assist his customer thereby differentiating himself from other competitors who merely manufacture castings.

The foundryman of the future must be more than just a source of cast metal components. The technologies required of today's full service metalcaster demand not only an investment in technology, but increased emphasis on a more educated and trained human resource. The competition for this more advanced labor force will come not only from within the metalcasting industry, but also from the entire manufacturing community.

In order to take advantage of computer-based technologies, the metalcasting industry must continue to invest in both research and education. There are significant advantages and benefits for those who have the vision to learn from the past, take advantage of the present and invest in the future.

Editor's Note: This article is an edited version of the official U.S. exchange paper which was presented at the 58th World Foundry Congress in Cracow, Poland on September 15-19 1991.