Guys,
How do you evaluate the stress result of FEA? We usually need to compare the existing/new design and see how much advantage we get from new design. But I have no idea how to general a report to my boss how much better I have done on stress.
In displacement, we can easily give a result like % advantage = 1 - max displacement of new design / max displacement of existing design.
But I can't do this to stress. Any idea about this?
Thanks,
There is no simple answer to that... it all depends on what you are trying to achieve.
Is there a stress concentration? Are you attempting to reduce weight? (make your design more efficient), is fatigue a factor? ....
Also, only use stress as a guide if you are working with linear materials. For plastics it is better to look at strain.
thanks for you answer, Tim.
We are a company who works in medical device industrial. So we need to take documentation for all of our design. In other word, if we have some design modification, we need to document how we did the change, and how it works better than before. For myself, I can tell better or worse by comparing the different stress fatigue. But I just want to find out a way how to document it to a simple comparison how much % advantage we get from new design by a set numbers from stress result. Because the FDA guy won't be an engineer, numbers will make more sense to them.
how much "safety" is what I concern.
But unfortunally, I have no idea how to do it.
anyway, thanks again,
Look at Factor of Safety (FOS). It is an easy way to give a relative comparison, although it does have some limitations.
I suggest investigating if it will help you show your results in a way you want.
Lachlan
Thanks Lochlan
I thank FOS is part of the result I can show to FDA guys. Howerer, FOS is a different expression of stress. It still has same problem to show the % advantage.
Hi jack,
I think if you want to compare the stresses I would suggest you to compute sumation of squares of the stress in all three orthogonal directions i.e x,y z all the normal/principle stresses which should be equal to resultant stresses .
use this formula
(S1-S2)^2 + (S2-S3)^2 + (S3-S1)^2 = 2 Se^2
where
S1 ---- Stresses in X -direction -----these are calculated by boundary conditions applied you can get each of these from solidworks
S2------ Stresses in Y -direction
S3------- Stresses in Z-direction
& Se is Resultant/Equivalent stresses across the model
Hope this will help you & let me know if it works or not
Thanks Susheel,
I'm still trying to find the S1 S2...from soildworks. Can you please give me some hint where I can find it or can compute it?
Hi Jack,
Good to here from you,for S1,S2 S3 these are stresses in respective X,Y,Z directions once you complete analysis in solidworks you will find stress,displacement and strain in the tree coloum of results kindly right click on the stress and say edit defination a window in left area will display click the first VB tab you will find SX,SY,SZ and many other ways you want to derive stresses in respective directions.
Hope this will solve your issues,feel free to ask if have any difficulties.
Good Luck
Hi jack,
I think if you want to compare the stresses I would suggest you to compute sumation of squares of the stress in all three orthogonal directions i.e x,y z all the normal/principle stresses which should be equal to resultant stresses .
use this formula
(S1-S2)^2 + (S2-S3)^2 + (S3-S1)^2 = 2 Se^2
where
S1 ---- Stresses in X -direction -----these are calculated by boundary conditions applied you can get each of these from solidworks
S2------ Stresses in Y -direction
S3------- Stresses in Z-direction
& Se is Resultant/Equivalent stresses across the model
Hope this will help you & let me know if it works or not