output variable Mises

[abaqusian]

Hi,

I am doing a simulation in abaqus where pressure is applied over a rectangular solid model. I requested MISES output from one of the element where the pressure load is applied. My doubt is, " Is that possible to find the possible biggest model size above which the MISES stress is constant ? ".

Say for example, when area is (10*25) and the mises result shows 234.78
Then, i increase the area and check the mises stress (it changes accordingly). Is it possible that above certain value of the model area, the MISES stress remains same ? (i.e) however, you increase the area, MISES output remains same ?

note: Because, i checked that from manual, the MISES calculation involves sigma (stress) which depends on area.

 

[burningDiesel]

Well actually (if I understand your question right), it should always remain the same.
Stress is just force divided by area, and the same goes for pressure. E.g. if you put a pressure in one direction on an element, the stress in that direction will be the same as the pressure.
Mises stress however only takes into account deviatoric stresses. E.g. if you put a pressure on all directions on an element, mises stress will not be affected as yield is independent of hydrostatic pressure.

Anyway to answer your question, it depends on how you are loading your structure, and what boundary conditions you have at what locations.

 

[abaqusian]

@ sdebock: boundary conditions are such a way that X and Z coordinates are totally constrained. the is applied on the y-direction.

My doubt is : say for the model dimension (10*5*10) , i get MISES value for a particular element (say central element) as 240.
Now, I change the model dimension as (12*5*12) and i get different MISES value for the above same particular element as 242.

My question is, "Is it true that after certain increase in model dimensions, the MISES value would be constant. For example, at model dimension (20*5*20), i get MISES as 250 . and again for increase in model dimension (21*5*21), i get same MISES as 250.

Is such phenomenon possible , where MISES stress remains constant after a particular model size ?

 

[abaqusian]

*** the LOAD is applied on y-direction

 

[bgib]

If I understand properly what you say, the finite element analysis works by interpolating the stress values around the nodes. The bigger the element, the less accurate the results. For this reason, it is important to refine the mesh until reaching a point that the mesh is accurate enough to give the right stress value. It can be checked by using a wide mesh and refining it until finding the same stresses from one mesh to another. Then it's good for the simulation to use the wider mesh that gives an accurate result in order to reduce computational time.

I have a question of how to interpret von Mises results. Have you ever noticed that some elements with high stress (represented in red) are next to elements with low stress (blue one)? Do you know why does it happen? For me it makes no sense at all and I was wondering if it means that the mesh is not right or if it is just that the software is not able to give an appropriate result, but that the high stress value is right.

Thanks in advance.