Fig. 2  Equivalent stress distribution on the Piston rod.
The problem can be divided in subtasks:
 Traditional structural analysis of the designed piston rod to verify which are the critical zones where there are the maximum compressive and tensile stress and to evaluate the stress level compared to the ultimate tensile strength of the material;
 Optimization of the production process (take advantage of Automatic Optimization technology) for defining the dies and process parameters suitable for the production of Piston Rod maximizing its quality;
 Advanced structural analysis by exploiting the mechanical properties of the cast from the process analysis to make the final structural verification and validating the production process.
The article aims to show the innovative techniques for the design using the integration of virtual simulation tools to optimize the quality of the finished product, indicating the basic.
Introduction
The applications of light alloys regards engine and also frame parts with a potential weight saving of 4050 percent compared to traditional materials used in industry, such as steel and cast iron.
It’s therefore important to design automotive components, and in general all the components, in combination with the processes in order to minimize costs, scraps and maximize the mechanical performance.
In such scenario, the design activities of the “processproduct” is determinant to ensure the maximum reduction of waste and the considerably shorttime to finished product as well as to guarantee the casting characteristics required.
Typically the High Pressure Die Casting is the most widely used process for that king particulars. Similarly to other casting processes, also the High Pressure Die Casting process can introduce defects that can reduce the mechanical qualities of the product determining the scrap. The very high speed of injected metal that govern the process of filling influences the flow dynamics, generating turbulence effects that can lead to the formation of air entrapment or missed fills.
Although HPDC is a highly automated process, the process denotes the multiplicity and variety of parameters with hard identification of the causes that generate a micro or a macro defect in view of corrective actions. For mass production, it is essential to search the optimal solution in terms of quality and production efficiency: the reduction of a few seconds of the cycle time has a positive effect on the number of units produced daily; the alloy mass for the single injection can drastically reduce the cost of raw materials, energy and recycling; the reduction of scrap percentage may result in attractive economic benefits, and finally the extension of the dies life time reduces the incidence on the cost of the single cast. The multiobjective optimization techniques, such as those applied in this case, support the determination of optimized and robust solution that has to satisfy all these objectives often conflicting with each other. The high knowledge of the defects and the appropriate interpretation of the virtual results is a prerequisite to the use the numerical simulation and optimization algorithms. The document published on the defects classification by AIM provides an important contribution to the knowledge assessment and it is worth considering as advanced study of scrap production causes.
To obtain the best quality of casting, the Design Chain approach, for metal casting processes and mechanical properties behavior is increasingly used in consideration of the solutions and the process optimization provided.
Preliminary structural analysis
The first phase regards the elastoplastic structural analysis of the Piston Rod to identify any critical issues for the reference geometry and the nominal and homogeneous mechanical properties of the selected Al alloy. The aim is to provide information regarding the critical application of the rod and the useful information to the designer of the production process that will define the process ensuring the highest possible quality of the rod and in particular in the areas identified with high stresses.
The work regards the virtual structural analysis of the rod coupled with foot cap to which is applied the constraints, loads and boundary conditions (Fig. 1).
The conclusion of the verification allows to identify a maximum stress concentration in the head and foot part (Fig 2), providing useful information for the design of the production process. In fact it will be necessary to pay the greatest attention in the two areas indicated in order to avoid introducing defects typical of the production process, responsible for possible reductions of mechanical characteristics of the cast.
Die casting process optimization
The high numbers of variables which define the foundry process of High Pressure Die Casting (HPDC) definitely make it one of the most difficult problems to optimise since, generally speaking, there is no single solution which will satisfy the stated aims in the best possible way. In order to select the best solution, it is necessary to evaluate and compare a large number of potentially possible solutions. The strategy for searching for the optimal solution characterises the optimisation technique.
In the case of foundry, it is not possible to analyse the entire range of solutions with a high number of variables and goals by means of a single  objective tool. Despite dividing the problem into subproblems of a lesser entity (filling, solidification, thermal aspects, residual stress, etc.), the goals are frequently in contrast or interlinked with each other and must therefore be pursued separately without attributing individual degrees of incidence a priori.
The multiobjectives optimisation algorithm (e.g. MOGA) must look for a robust solution which is accurate and respects the design constraints affecting the problem, and establishes the scope of the solutions which can be achieved.
Fig. 3  Examples of geometries provided by a combination of parametric geometry variables
of the casting system
The main objective of the optimization phase is to design, with a fully automatic mode, the best configuration of the process. The aim is pursued using MAGMAfrontier that change the relevant geometric and process parameters to simulate a large number of variants.
The optimization phase is based on an initial population of configurations to simulate (the DOE, Design of Experiment) selected through the use of the algorithm “Reduced Factorial” (this algorithm grounds on two distinct levels of “full factorial” algorithm to cover the extreme of the intervals considered) in combination with the algorithm “Sobol” (this algorithm creates design “quasi random” ensuring that all the factorial design space is covered more uniform as possible). With the combination of these algorithms, it is possible to completely cover the space vector of input variables guaranteeing the representation of the complete design space. Subsequent generations are created by the, above mentioned, genetic algorithm called MOGA (Multi Objective Genetic Algorithm), which allows the user to define new additional designs based on elitism and mutation attributes. Despite the complexity of the topic, the user define only the number of configurations of the first population and the number of successive generations as a function of the available time.
In the case of the die casting process, the design of optimal gating geometry and the corresponding flux curve, to control the plunger movement during the filling phase, are the goals of the project.
Fig. 4  Optimization Analysis: a) scatter charts, b) Airpressure results in a section
Some parametric variables (fig. 3) can define the shape of the gating which connects the shot chamber to the ingates and the ratio between the sections. The main objective is the minimization of the presence of air in the gating system for a reestablished ideal injection curve keeping the melt front compact. The output variables, on the basis of which the goals are formulated, are the criterion which represents the pressure of the air entrapped in the gating system (objective 1) and in the cast (objective 2), both to be minimized.
The optimization results, 2840 computed configurations, identify a Pareto frontier representing the best set of solutions that minimize the two objectives. The comparison between the different configurations in a twodimensional diagram allows to identify the best design (Fig. 4), in terms of principal quality indicators appreciating the improvement in the casting.
The used criterion (Airpressure) is typically used to predict the maximum concentration of air bubbles and their internal pressure. It is also important to reduce the air entrapment in the gating system that can occur during the generation of particularly extreme geometry of gating.
Although the cold shuts is typically another objective of the project, in the case of small Rod the absence of the cooled melt during the cavity can be checked diligently only in the final optimal solution (fig. 5a) as well as for the velocity distribution correlated to turbulence and soldering (fig. 5b).
The steady state simulation allows to verify the solidification behavior ensuring the highest quality of the cast in terms of shrinkage and the durability of the dies (fig. 5c).
Fig. 5  Visualization of the detailed results of the optimum design: a) temperature filling, b) velocity filling, c) solidification time distribution
Mechanical properties prediction stress analysis and
advanced structural analysis
The local mechanical properties can be correlated to defects mapping of the casting using the appropriated criteria in a polynomial function [f1].
The implementation of the prediction model of the local Ultimate Tensile Stress (UTS) and the relative Elongation to fracture is made possible by exploiting the curve ‘StressStrain’ of the material used for the Piston Rod.
The used formulation considers to calculate UTS as a function of the maximum nominal value provided by the curve (Fig. 6) decreased by a specific amount depending on the presence of defects due to the filling and solidification phases.
Fig. 6  StressStrain curve of the EN AB 46100 material used for the Piston Rod
The used formulation is represented by the following function:
UTSfinal = UTSnominal – (a*Airpressure)A  (b*FlowLength)B 
(c*Aircontact)C (d*Coolrate)D [f1]
where:
 Airpressue: is the criterion that allows to identify the presence of air bubbles inside the cast as a function of an increase in air pressure compared to the value defined by the production conditions;
 Flowlength: is the criteria to identify the distance traveled by the alloy during the filling. This result contains the information concerning the risk of an excessive temperature reduction;
 Aircontact: is the criteria to identify which areas are critical for the oxides formation. This criterion is based on time of alloy contact with the air until the end of the filling.
 Coolrate: is the criteria to identify which areas are critical for any delay during solidification. This criterion is based on the cooling rate of each zone during the solidification phase.
 a, b, c, d, A, B, C, D are constants
The figure 7 and figure 8 illustrates the final local UTS and the corresponding Elongation based on the Al alloy stressstrain curve. The identified values show a lowering of the tensile strength without interfering with the values relating to the Yield Stress.
The end of the process analysis consist of the calculation of residual stresses generated at the end of the production process on the Piston Rod (Fig. 9). The measured values are particularly low, but with the mechanical properties are a good starting point for the final Structural Analysis
The final interpolation and export of the local mechanical and residual stress results on the FEM mesh is the appreciated linking step from the process to the product to address the mechanical response starting from realworld “daughter” of the manufacturing process (Fig. 10).
The final mechanical simulation (Fig. 11), carried out with the innovative process described above, evaluates the realistic performance of the Piston Rod subjected to the operating loads considering the properties deriving from the production process. The processproduct approach is based on more accurate method providing the possibility of design the component with the correct admissible stresses and strain reducing the security factor and of course the weight of the body. The final analysis show how the areas previously tested critical, the foot and head of the piston rod, does not reach the values for breaking.
Fig. 7  Mechanical Properties prediction in terms of local UTS

Fig. 8  Mechanical Properties prediction in terms of local Elongation

Fig. 9  Residual stresses at the end of the die casting process

Fig. 10  Example of Residual stress transfer from MAGMA (a) to ANSYS (b)

Fig. 11  Final Structural Analysis: Comparison between a) UTS final distribution and b) Von Mises stresses

Conclusions
The present work describes an integrated approach between complete process simulation and structural analysis for the Piston Rod component, in order to verify virtually, as realistically as possible, the life of a component from the stage of production until the exercise test.
The described methodology is addressed to the High Pressure Die Casting, as shown in its practical application on the component studied and it is aimed at the product design, HPDC foundry and industrial endusers.
The innovative aspect of the adopted integrated method does not lie in the use of computer codes specialized on the process or on the mechanical response, considered as reference in the two areas of application, but it is found in the prediction of mechanical properties as a function of process and product quality as well as in the dialogue between different codes. It offers the ability to view the real mechanical behavior of the cast HPDC produced, and consequently the ability to validate and to correctly predict the response of the service casts.
The design chain method intend to meet the continuing needs and requirements of higher quality, more volume, timeliness, costeffective production method, especially for the automotive industry. The ability to integrate engineering and manufacturing processes, allowing a more accurate and reliable design of the components, with a view to increasing demand for service performance, can meet the modern requirements of economic efficiency (e.g. reduced energy consumption and recyclability), security (impact, analysis of noise and vibration) and ecosustainability.
Acknowledgement
The optimization work was carried out in close collaboration with the technicians of the A.B.O.R S.p.A., it is thanked in particular the availability of Luca Bellati.
The technology also has been developed using both experimental and computational tests, carried out in collaboration with the producers of the alloy, university laboratories, foundries, endusers in the automotive sector, in the context of the NADIA European Research Project.