Acceptance Criteria 2

[DDP]

What do you guys use for the bottom line in your analysis?
Do you just check Von Mises or other paramters also?

 

[mloew]

DDP,

The short answer is "it depends." There are many factors in determining what failure criteria to consider. Materials/manufacturing process, failure mode, etc. Can you elaborate on what you are trying to solve?

Best regards,

Matthew Ian Loew

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora

 

[corus]

To check a design then it is wise to check that against the appropriate design standard. These will likely limit stress intensity to some factor of yield, limit compressive stresses to prevent buckling, stress ranges to prevent fatigue, and failure against creep rupture, and/or cumulative creep-fatigue damage.

 

[DickButkus]

Be sure that you check your values from your postprocessor with the integration point values. If there is a big discrepancy, your mesh is not refined enough. Remember, that the only real values in your analysis are integration point values. Everything you see in the post processor is an extrapolation from the integration points (unless you specifically tell the code you want results at the nodes - in wich case the analysis code extrapolates for you).

 

[jxc]

MLeow statement is the best. You began your analysis with a purpose to prove or disprove(?) a phenomenon particular to you or your company/group. What you chose to use to say that analysis was successful depends on if the analysis answered your initial question. This may mean using VonMises results, principle stress or strain results (my favorite) or displacements that correspond to physical test results. What I am attempting to say, what I check is dependent on what type of analysis I am performing.

Hope my ramble helps.
jxc

 

[neetha]

hi,
what is integration point value? how can i get this value?
is that true that the von mises stress in a particular element will continue to increase if i increase the loading? if yes, why sometimes in my post processing result the value decreases? is that because the value shown is an extrapolation and not the real value?

 

[DickButkus]

what's an integration point? Within a finite element the stresses are only calculated at a couple descrete points. You may have noticed that you have a choice between fully integrated and reduced integration elements. For example, a typical 4 node plain strain quad element has 4 integration points within it. A reduced integration 4 node quad only has a single integration point within it.

How do I get my integration point values? You should be able to obtain the actual integration point values from within your post processor. In ABAQUS, I use the "probe" function to get this. The probe is accessable from the icon of the i with the circle around it.

From other software packages, I am not sure how to get the integration point values, but, if you have the values printed to a text output file, those values should be integration point values. In fact, it will probably have the integration points numbered and a table of values below it.

Is it possible that the Von mises stress in a particular element will decrease if I continue to increase the loading? Absolutely! There are many cases where this is possible, such as a non-linear material definition. After a certain point your material may be loosing stiffness as it begins to fail. Another case would be if your part comes in contact with something, begins to buckle, or simply moves enough that your load path changes. Linear codes will not pick this up, but non-linear codes will.

Best regards and Merry Christmas!