Comparison of rigid-plastic and elasto-plastic finite element predictions of a tensile test of cylindrical specimens – Post 2 

 

 

In this continuing post, the results of elastoplastic finite element analysis of the tensile test will be presented.

 

The flow stress shown in Fig. 1, obtained by a material identification algorithm based on rigid-plastic finite element method, was also employed for elastoplastic finite element analysis of the tensile test. Modulus of elasticity and Poisson’s ratio are assumed 201GPa and 0,3, respectively.

 

 

Figure 1: Engineering stress-strain curve and its associated true stress-strain curve obtained by a tensile test of SCM435 cylindrical specimen

 

 

Predicted deformed shape of the tensile test specimen by the elastoplastic finite element method is shown in Fig. 2(b). It can be seen that it is very difficult to distinguish the predictions shown in Fig. 2(a)obtained by rigid-plastic finite element method from those shown in Fig. 2(b)obtained by elastoplastic finite element methods. Note that the material was assumed isotropic and to obey Huber-von Mises yield criterion, similarly to the rigid-plastic finite element method.

 

 

 

Figure 2: Predicted equivalent strain deformation history for the tensile test of SCM435

 

As shown in Fig. 3, tensile load and elongation obtained by the two approaches are nearly the same as a whole especially near the necking point. However, more in-depth investigation into Fig.4 reveals that elastoplastic finite element solutions reflect the input modulus of elasticity of 210GPa for the strain less than 0.001 while rigid-plastic finite element solutions cannot but causing plastic deformation at the starting point due to its theoretical limitation. It can be seen from Fig. 4 that the slope of tensile test predictions over the strain ranging from 0.001-0.002 is about 185GPa, which is slightly different from the input. Note it is due to a mere step size effect.

 

Figure 3: Comparison of tensile load-elongation curve between rigid-plastic and elastoplastic finite element methods

 

In addition, non-negligible difference between rigid-plastic and elastoplastic finite element methods can be observed especially in the necking region just before fracture point, that is, the difference in predicted tensile load increases as the elongation approaches to the fracture point. As shown in Fig. 5, the difference in predicted diameter at the necking point between the two approaches reaches about 6%.

 

It is very important to compare and evaluate the predictions obtained by the two approaches for understanding the related theory and the related application-oriented technologies. Thus, radius of point A as indicated in Fig. 5 was traced, revealing that it increased by 0.0010mm in the case of elastoplastic finite element method while it decreased by 0.0022mm in the case of rigid-plastic finite element method. The former is owing to elastic recovery caused from the decrease in tensile load while the former is due to numerical smoothing caused from the assumption of rigid-plastic behavior of material [12]. However, the difference is negligible compared with the radius difference of predicted tensile test specimen, that is, 0.1mm. As shown in Fig. 6, the radius difference at the necking point described in Fig. 5 is due to the radius difference over the elastic region involving point A. That is, in the case of elastoplastic finite element method, the elastic recovery in elastic region leads to relatively much shrinkage at the necked region since the input elongations in the methods were the same.

 

 

Figure 4: Detailed comparison of engineering stress-strain curve between rigid-plastic and elastoplastic finite element methods

 

 

Figure 5: Comparison of deformed shape at the final elongation

 

Figure 6: Comparison of radius at the point A in Fig. 5 between rigid-plastic and elasto-plastic predictions

 

It is thus concluded that this radius difference shown in Fig. 5 is due to radius increase caused from elastic recovery in elastoplastic finite element as shown in Fig.6.It can be seen from Fig. 4 and Fig. 5 that it is not easy to specify a necking point because the peak tensile force maintained for a certain time interval in which the elongation increased steadily without change of tensile force. Of course, this phenomenon is due to non-homogeneous deformation until the necking point [13], that is, the major deforming region moves from place to place due to numerical non-homogeneity and strain hardening effect even though the necking phenomenon does not occur. This discussion reveals that the necking point is the very end point of necking starting interval in which tensile force is stationary, i.e., the peak strain remains while the deformation region moves to the region with smaller strain.

 

Tensile deformation basically causes expansion of material and thus the total volume should increase in the case of elastoplastic finite element method. In fact, the volume increase rates were 0.001% and 0.01% in rigid-plastic and elastoplastic finite element methods, respectively. The former is negligible and meaningless because of the assumption of incompressibility in the theory of rigid-plasticity. Of course, its small value guarantees the reliability and validity of numerical schemes in the rigid-plastic finite element method. Note that the volume increase of elastoplastic finite element predictions are indispensable because mean stresses at any point are positive. 

 

In the series of these 2 posts, experiments of a tensile test of cylindrical specimen ofSCM435 which was spheroidized to be forged were compared with predictions obtained by both rigid-plastic and elastoplastic finite element methods to reveal the similarity and difference of the two finite element methods.

It was seen that two different finite element methods predicted nearly the same solutions even though the tensile test is relatively sensitive to the theories on which numerical approaches are based.

 

The detailed investigation into the tensile load and deformed shape revealed more or less difference between the two finite element methods, which caused from the related theories. However, the difference can be acceptable, considering that numerical simulation of a tensile test is very sensitive to numerical schemes and conditions including finite element mesh, time step, minimum allowable effective strain rate in the rigid-plastic finite element method and the like.

 

Tensile test simulation by the rigid-plastic finite element method is vulnerable because the effective stress in the elastic region just after the necking point is quite high, which causes artificial numerical deformation in the elastic region. However, since the rigid-plastic finite element method is more pragmatic in most cases, the knowledge drawn from the comparison between rigid-plastic and elastoplastic finite element method are very meaningful for the process design engineers to get more valuable information from the predictions by the rigid-plastic finite element method.

 

References

 

[1] Bridgman P.W.:Studies in large flow and fracture, McGraw-Hill, New-York, 1952.

[2] ZhangK.S.:Fracture prediction and necking analysis,Eng.Fract.Mech.52 (1995) 575-582.

[3] Komori K.:Simulation of tensiletest by node separation method,J.Mater.Process.Technol.125-126 (2002) 608-612.

[4] MironeG.:A new model for the elastoplastic characterization and the stress-strain determination on the necking section of a tensile specimen,Int. J.Solids.Struct.41 (2004)3545-3564.

[5] Springmann M., KunaM.: Identification of material parameters of the Gurson-Tvergaard-Needleman model by combined experimental and numerical techniques,Computational Materials Science 32 (2005) 544-552.

[6] M. S. Joun, J. G. Eom, M. C. Lee:A new method for acquiring true stress-strain curves over a large range of strains using a tensile test and finite element method, Mechanics of Material40(2008) 586-593.

[7] J. G. Eom, Y. H. Son, S. W. Jeong, S. T. Ahn, S. M. Jang, D. J. Yoon, M. S. Joun:Effect of strain hardening capability on plastic deformation behaviors of material during metal forming, Material and Design54(2014)1010-1018.

[8] C. H. Lee, S. Kobayashi: New Solutions to Rigid-Plastic Deformation Problems Using a Matrix Method, J. Eng. Ind. 95 (1973) 865-873.

[9] M. C. Lee, S. H. Chung, S. M. Jang ans M. S. Joun: Three-dimensional simulation of forging using hexahedral and tetrahedral elements, Finite. Elem. Anal. Des., 45 (2009) 745–754.

[10] W. J. Chung, M. S. Joun:Elastoplastic module of AFDEX 2D, Proc. of 2013 KSTP AutumnConference(2013)57-60.

[11] J. R. Hughes:The Finite Element Method:Linear Static and Dynamic Finite Element Analysis,Prentice-Hall, New Jersey, 1987.

[12] M. S. Joun: Case study on major defects in cold forging by finite element method (1) –Accuracy of forging simulation, KSTP Autumn conference(2011) 223-227.

[13] M. S. Joun, I. S. Choi, J. G. Eom, M. C. Lee:Finite element analysis of tensile testing with emphasis on necking,Computational Materials Science41(2007) 63-69.