Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Who have the fatigue analysis example? 3

Status
Not open for further replies.

thundersky

Automotive
Feb 16, 2005
5
0
0
JP
Hi

Now,I am trying to analyze the tractor frame by using ANSYS 8.0. And after I finished the static analysis ,I hope to analyze fatigue analysis of that model but I cannot find any example for it. Then,I beg everyone who have this example or know about it,please help me for it.

Yours Sincerely
 
Replies continue below

Recommended for you

You don't get fatigue damage from a static load case as fatigue depends on the stress variation through load cycles. First you need to know the load cycles and apply those to your model of the tractor. Then you need to read about fatigue and its evaluation and then refer to an appropriate design standard for fatigue assessment, such as BS 7608 or whatever is applicable to your country.

corus
 
You could also use a specific fatigue software like FE-Safe or FE-Fatigue that takes the results file from anays and uses traidtional fatigue theories to compute fatigue life and potential failures. I dont know if you have that software though.
 
thundersky

I'm exactly sure how you plan to do your analysis so I can be competly wrong now but this is one way.

Since you have a static load as a basis for your analysis I assume that this is some kind of max load, giving max stresses etc. Normally I would have another static load to get the min stresses. From these two you can get a stress span. Say, max stress 200 and min -100. This means stresspan 300.
Now if you have the number of cycles and a parameter describing the stressdistribution between different loadcycles you can do a design based om a code.
What I mean with parameter describing the stress distribution is this. Lets say that in ten cycles one cycle gives span 300, four cycles gives 200 and the remaining gives 100. Then the parameter would typically be below 1.0, if all cycles give 300 it would be 1.0. This is connected to Palmgren-Miners hypothesis of partial damage.

What you need from FEM-analysis for this codebased analysis is the stress distribution. If you want to do "real" fatigue analysis that's a bit more complicated.

Good Luck

Thomas
 
Hello Thundersky,

ANSYS does have the feature that it does the cycle count for a given loading events. You will need to specify the loading events and number of repetiotions associated with a particular loading event. ANSYS will take care of all the load cycles and will do a cycle count using Raiflow counting technique (note:- the accuracy of the result depends much on how you define the loading events, i mean the sequence) and will provide the fatigue damage as a ratio according to the Palmgren Miner Rule. Of course, you will require the allowable S-N curve for ur case.

The other way and I feel a better approach to evaluate fatigue life is by using the fracture mechanics tools in ANSYS that will provide you with the Intensity factors (K1, K2, K3) which can be use to evalulate the crack growth rate. But this more involved than the first method.

The ANSYS mannual will provide you with detail information.

hope it helps,

Deep
 
The best way would be a fatigue software, like FEMFAT or n-Code etc. These are very expensive, but as I know you can rent of them!
If you can understand German you can use the DIN15018 standard methodology. This is based on Simth diagrams and sx-sy-tx stresses. I had created a rutin for it, which collect the stress results in local coordinate system. You have to define the weld seam type and material than the program calculates with this theory!

Best regards,

Irwin
 
Hi,
What kind of stiffness analysis you are trying to carry out.Is it with respect to Cartesian or cylindrical coordinate system.
There are several methods for that but the best being Fourier Series method.
Let me know so that i could tell you how to proceed.

Cheers
Bye
Dinesh Kulkarni
Design Engineer,
Scandent Solutions,
Bangalore
India

 
Status
Not open for further replies.
Back
Top