Numerical simulation of manufacturing processes has become in the last years an important tool to improve these processes reducing lead times and try out, and providing products free of defects and with controlled mechanical properties. Finite Element Method (FEM) is one of the most important methods to simulate metal forming. In tube drawing with fixed plug both the outer diameter and the inner diameter of the tube are properly defined if correct process conditions are chosen for the die angle, drawing speed, lubrication and area reduction per pass. These conditions have great influence on drawing loads and residual stresses present in the product. In this work, the cold drawing of tubes with fixed plug was simulated by FEM with the commercial software MSC.Superform to find the best geometry of die and plug to reduce the drawing force. The numerical analysis supplied results for the reactions of the die and plug and the stresses in the tube, the drawing force and the final dimensions of the product. Those results are compared with results obtained from analytic models, and used tooling design. Experimental tests with a laboratory drawing bench were carried out with three different lubricants and two different lubrication conditions

No abstract

No abstract

No abstract

The study of natural convection on flat plates is of great interest in the areas of Engineering, both for the simplicity of geometry and the wide variety of applications. In the study and definition of a numerical model, an ideal mesh configuration is the one that best represents physically, with minimal numerical influence and with the lowest computational cost, the problem addressed. The influence of two mesh configurations (non-uniform staggered and entirely uniform), at different refinement levels, was studied to evaluate natural convection heat transfer rates in flat plates of AR = 5; in isothermal conditions, in turbulent regime with the κ−ω SST RANS model and using free and open-source software OpenFOAM®. The physical-numerical methodology applied, and the numerical results obtained were validated from experimental results in the literature. The non-uniform staggered mesh configuration proved to be more adequate in precision, and computational cost to the problem situation studied. The entirely uniform mesh proved to be infeasible due to the high number of elements and computational cost demanded.