A model of tool wear in electrical discharge machining process based on electromagnetic theory

2017 
In electrical discharge machining (EDM) process, tool wear is an inevitable phenomenon that adversely affects the geometrical accuracy of machined features. A theoretical model accounting for tool wear during EDM process is hence the basis study for high precision machining. However, in most modeling studies on tool wear and electrode shape, the sparking process is only factorized by the geometric configuration, i.e. the distance between electrodes. The real sparking process related to the fundamental physics is not addressed in these geometric models, which can produce large discrepancies with the experimental results. In this paper, a model of tool wear in EDM is proposed, which accounts for the electric field inside the dielectric fluid using electromagnetic (EM) theory. The spark is proposed to occur at the position where the local electric intensity reaches maximum and exceeds the breakdown strength of the dielectric fluid. This model is shown to provide the physical insight of the real EDM situation, and to give a more accurate prediction of tool wear compared with traditional geometric property based modeling. With these merits, this proposed model can be applied to predict tool wear in various machining processes. To evaluate this model, simulations of EDM die sinking and ED milling are carried out. The results by this electric field model were compared with both geometric model and experiments. By analyzing the profiles of the tool end, the differences in mechanism between the electric field and geometric model are identified. In addition, this electric field model is also applied to simulate the conic tool forming process in the fix-length compensation with micro-milling, which cannot be thoroughly addressed by the geometric model. The model presented in this paper is capable of capturing the key features of the tool wear in a variety of machining processes.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    20
    References
    10
    Citations
    NaN
    KQI
    []