How to get Von Mises stress for whole section and not just indiviual elements

[cfoley]

I am doing FEA analysis using Abaqus and I need to get the average/min/max Von Mises stress for a section rather than the Von Mises stress for each individual element.

I have tried using python scripts such as

http://www.eng-tips.com/viewthread.cfm?qid=184288

however I have had no success as of yet.

Has anybody any suggestions on how I can go about this ?

 

[rb1957]

i'm trying to think of a reason why i'd want to average von Mises over an entire section ?

the most i can get is if i had a relatively compact loadpath under a relatively consistent load/stress, but why then ?
(there shouldn't be much variation in the stresses that there'd be much difference from max to average)

another day in paradise, or is paradise one day closer ?

 

[cfoley]

we are doing an advanced brain model under dynamic impact and we wish to measure the maximum stress or strain in each part of the brain at each time increment as this is critical in the study of concussion. There are many parts in the brain all section off and we do not want to know the stress for every single element.

If you could help it would be appreciated.

 

[rb1957]

is von mises an appropriate stress for brain tissue ?

how will the different directions (of vM stress in different elments) affect the average ?

maybe you should extract the normal and shear stresses in a specific direction (along a co-ord sys) ?

another day in paradise, or is paradise one day closer ?

 

[cfoley]

I asked for stress because I thought I would get a responce quickly hopefully with a usefull python script, once we have that we can alter it to figure out whatever else is needed. However for certain parts of the brain the Head and Brain the von mises is important.

 

[dhengr]

Cfoley:
Von Mises must be rolling over in his grave and developing a real/live concussion just thinking about all of the crazy ways his theories have been used and bastardized since he thought, what he thought, he thought and developed. If he were alive today, I suspect he would never have published his work, for all the confusion and bastardization it has caused. You’ve got me stumped. But maybe, look for the maximums. Then consider areas or volumes in that immediate region, as a function of what you know about concussions and average only a finite number of stress levels in that immediate area or volume. I’m wondering, there must be an important difference btwn. a portion of the brain which sorta impacts the skull or another volume (mid-brain volume) which just accelerates and decelerates within a larger semi-fluid mass, which may be another area/volume which you wanted to study and do a finite average over, for a slightly different reason. I’m sure you’ve seen other, prior, papers studying this exact subject, or at least trying to indicate (infer) brain damage as a function of head impacts, I have. I’m wondering, an impact to the forehead must have very different affects on that immediate region than the affects on a finite volume in the back of the brain, which is at least somewhat cushioned. The closest I’ve come to what you want is to look for maximums in a region of interest and then extract principal stresses, max. shear stresses and strains, and the like, for structural/mechanical stress analysis purposes, a Theory of Elasticity and Strength of Materials problem, viewed through a FEA lense.

 

[rb1957]

@dh, i guess you didn't read he reply to me ... he's not so interested in vM as averaging some output parameter.

what is the yield stress of brain tissue ?

another day in paradise, or is paradise one day closer ?

 

[cfoley]

Thanks for your input, but the advice I am looking for is how to pull the results from abaqus, not on how to go about analyzing the brain.

I am looking for maximum principal strain, and have a brain model consisting of 11 different materials all with varying properties.

If anyone has a python script to pull this out of a part rather than individual elements it would be greatly appreciated.

 

[Ussuri]

I would suggest this is better asked in the Abaqus Forum.

However, I think your best bet is to contact Simulia directly. I think that would be your quickest way to getting an answer.

 

[geesamand]

I would look to see if Abaqus will offer you pressure loads over a defined area.

As has been stated, Von Mises stress varies throughout the volume and averaging it makes it into something that is no longer a Von Mises stress. Call it what you want but don't call it a VM stress.