Is a "traditional" elastic 2D element.

KEYOPT(3)

0 - Plane stress

1 - Axisymmetric

2 - Plane strain (Z strain = 0.0)

3 - Plane stress with thickness input

http://www.ansys.stuba.sk/html/elem_55/chapter4/ES4-42.htm

Hi Nicolas,

I know these options. I guess I didn't clearly mention what I wanted to know. Let me try to explain.

PLANE42 is a Bi-Linear Lagrange element (4 nodes) and uses 2X2 integration for quad elements. I know these details well. But I wanted to know which formulation was used to derive the element stiffness matrix. For ex. if you analyze a plane strain problem with nearly incompressible Elasto-Plastic material and with small strain assumption, it exactly predicts 'slip-zones', if you use enough mesh density in the plastic zone. I simulated such a problem and got some result(it matches well with theory).

But when I ran the same problem using a code developed here in the university, using 4noded Lagrange element(with pure displacement formulation) and I did not get the same result. This is because, simple displacement based elements, that too linear elements like 4-noded one are very poor in predicting the behavior of incompressible materials. There must be some enhanced formulation(Bbar or mixed or EAS etc) associated with it in calculating stiffness matrix and residuals. I dont see any such information in ANSYS help.

So, by element formulation, I mean whether it is purely displacement based element or some kind of improved element. I suspect that its not purely displacement based. I want to know if its true and if true what exactly is that formulation used with PLANE42 element in ANSYS.

Hello Kadapa,

Did you check Ansys theory manual? For plasticity, the material behaves in an incompressible manner, so that the bulk modulus of the material goes to infinity, resulting in ill-conditioned stiffness matrices, and volumetric locking occurs with fully integrated, displacement based finite elements.

As bulk modulus approches to infinity and volumetric strain to zero, the hydrostatic part of the stress should be a finite value (simply hydrostatic stress= bulk modulus.volumetric strain). however, because the strains are calculated by derivation of displacement, volumetric strains calculated are not accurate, and any small error in strain will result in very huge error in hydrostatic stress which in turn will result in mesh locking.

in some elements in Ansys, to my knowledge, hydrostatic stress is a seperate degree of freedom using Lagrange multipliers. This is called mixed U P formulation. I think you should start from here; just make some google on mixed U P formulation in Ansys for incompressible materials.

I wish this helps.

Hi Fatih,

Thanks a lot for the information.

Actually I know the theory behind dealing with incompressible materials, advanced element formulations like different mixed u-p formulations, Bbar, Fbar methods.

I did not find any information in ANSYS help mentioning the formulation of PLANE42. It says PLANE42 is a legacy element. I know that there are advanced elements like PLANE182 which explicitly support mixed u-p formulation and used them in the past.

My problem is that when I simulated a problem in ANSYS using PLANE42 elements with nearly incompressible Elasto-Plastic material (nu = 0.48) I could see slip lines clearly. But when I simulated the same problem(with same mesh) using the code developed here in the university with simple 4noded elements, I did not get the same result. We obviously needed to go for enhanced formulation because of locking, to get the correct results. I cant suspect the code here because it was developed by the pioneers of FEM. And I don't suspect ANSYS either. ANSYS might have some means of dealing with locking. That is the missing information I want to know.

Hope someone knows about it.

What about asking ANSYS people?

@Fatih: Yeah ANSYS help manual does not provide much information. I may try in ABAQUS, but for that I need to learn and I think it would be a waste of time.

@Nilanjan Das: I think you have not completely read my previous comments. I already know ANSYS has such capability and said it in my last comment. The question is what is that ANSYS use ?

Hello,

What is the replacement of Quad 4 node 42 in ANSYS 19?

Hi Lee Kai Yan,

In ANSYS 19, you can use this element type: SOLID -> Quad 4 node 182. It can be used as either a plane element.

What is the URL for ANSYS PLANE42 element specifications and theory?

Hi Masoud Sanayei , the URL for ANSYS PLANE42 element specifications and theory is:

https://www.mm.bme.hu/~gyebro/files/ans_help_v182/ans_elem/Hlp_E_PLANE182.html