The finite element method (FEM) is usually used for process design in hot forging. With FEM simulations, the forming processes are initially mapped and designed virtually, with the objective of optimizing them as realistically as possible. Susequently, the forming tools are actually manufactured. The FEM simulation enables a considerable reduction in the costs assiciated with process design.
Thus far, friction has only been considered in a simplified form in these simulations. Previous numerical modeling approaches assumed constant friction conditions in terms of time and location, which meant, for example, that tears or the consumption of the lubricating film were neglected. During hot forging, accumulations and residues of the lubricant used can be detected on both components and tools. Such phenomena are caused by changes in the lubricant film during the forming process. In the forming of aluminum in particular, altered friction conditions can occur, which can lead to adhesions and forging defects. For process design using FEM simulations, such local changes could not previously be mapped and taken into account. Furthermore, the exact correlations between the contact variables of hot forging and the behavior of the lubricating film have not yet been researched.
For this reason, the Institute of Forming Technology and Machines (IFUM) at Leibniz Universität Hannover has developed a novel analogy test and a numerical friction model and implemented the new modeling approach in commercial FEM software.
Further development of the rod drawing test according to Pawelski
IFUM has developed an analogy test for determining the friction coefficient as a function of local contact variables based on Pawelski’s rod drawing test . As shown in Figure 1, a round rod sample with defined lubrication is pulled through a feed die and the tensile force is measured. A change in the friction conditions can be detected by a change in the force. This setup makes it possible to characterize the behaviour of a lubricant by varying the sliding distance, the contact pressure, the sliding speed and the temperature. In contrast to previous approaches to friction coefficient measurement, this method allows contact variables to be constant over time with a homogeneous distribution in the value range of hot forging.
Extended friction model as a function of the contact variables
The experimental force curve of the rod drawing test is shown in Figure 2 and can be divided into the three states “fully lubricated”, “unlubricated” and a transient transition state. The fully lubricated state represents the initial conditions, from which the transient transition region reflects the changes in the lubricating film up to the unlubricated state.
To characterize the individual areas, the rod drawing test is simulated using FEM simulations. Based on the numerical and experimental results, the “fully lubricated” and “unlubricated” states can be described by constant friction conditions. Time-resolved contact variables and friction coefficients are combined to describe the transition behavior and serve as the basis for modeling. A contact size dependent friction model is parameterized from the contact variables and the friction coefficients of the transition behaviour for different parameter configurations. Figure 3 shows the friction model with the parameters that were determined for the test material and lubricant.
Implementation and user interface
In order to make the developed friction model accessible for industrial application, IFUM implemented the newly developed friction model in the FORGE® FEM software from Transvalor and developed a user tool for user-defined adaptation of the friction model. The implementation of the friction model in the substructure of the software also enables regular users to apply the new model to their processes. A user tool was created to easily modify the parameters of the model for varying combinations of tool and component materials with different lubricants. As shown in Figure 4, the tool can be used to visualize the model for user-defined coefficients and generate the corresponding output file for the FEM software.
In the future, industrial users will be able to parameterize and use friction models tailored to their processes in order to improve the predictive accuracy of their simulations. A more precise simulation can in turn save costs in the context of process design.
