Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

FEA fatigue 3

Status
Not open for further replies.

udar

Structural
May 25, 2006
8
Hi,

Which program performs fatigue analyses by FEM, both elastic and elasto-plastic domain? I am interested in thermo-mechanical fatigue.
Thank you!
Udar
 
Replies continue below

Recommended for you

To my knowledge (I may be putting my foot in my mouth here) there are no FEA programs which do fatigue calculations via the FEM. Fatigue calculations are typically performed by taking user-specified criteria (mean and alternating stress life curves, etc) and using this data a calculation is performed on the areas of interest from an analysis. I think if you search this forum a little bit you'll find some more postings similar to your question. The FEM is useful for predicting failures such as fracture, delamination, and catastrophic events (such as a bladeoff or bird strike within a turbine engine).
 
There are several fatigue softwares that can directly take FEM stress results for use in a spectrum of cases and perform fatigue damage or life calculations on every node, and then pass the results back to your post processor for viewing. I can suggest that you take a serious look at which can do either stress-life or strain-life based fatigue analysis and out performs most if not all of its more expensive competitors.
 
Sorry I should have been more clear in my posting. Almost every major FE software company has some sort of fatigue capability available. Fatigue is not calculated via the FEM however.
 
FEM is based or directly derived from theory of structures (or whatever phenomena you are modelling) using well established formulae. Fatigue is empirically based, that is there is no theory to follow, just a bunch of rather strange (IMHO) methodologies that appear to fit remarkably well with test data, e.g. rainflow analysis, Miner's rule, Goodman equivalent stress and so on, which really has nothing to do with FEM, since the source of stress history is quite commonly taken from hand calculations or strain gauge readings. For this reason you will only find fatigue software in the form of add on modules when using FEM.
 
Hi,
adding to JohnHors:
- Fatigue add-ons of commercial FEA programs performs automatically post-processings you can perform yourself "by hand" (well, at least with Excel...)
- without having a Fatigue add-on, you can proceed like this:

1- a) run a solution for every loadcase which composes your load history, or b) run a transient analysis in which you save the result data for every time step
2- for every loadcase, have your program list the nodal or element stress values you are interested in
3- import the data in Excel (if you have more than 65000 nodes, split the results files)
4- set up in Excel the Voeller or Design Fatigue curve, the Haigh correction, etc, as needed
5- calculate the cumulative damage node by node, and then find the MAX value.
You're done ;-) !!

Regards
 
Thanks to all for advises!
What dou you think about fe-safe software? (which works with major FEA programs) see:Does anybody work with it?
tanks in advance
udar
 
Fe-safe is worth considering but it's best to go on one of their training courses, run by John Draper who started the business, before clicking on buttons to produce pretty pictures. CosmosWorks Advanced Professional also produces a life prediction based upon SN data (from user input or library data) and a load history, though I doubt if it's advanced as FE-Safe for looking at biaxial stress, for instance.


corus
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor