Available S-N curves

[carlsobr]

Does anyone have a good source for finding different Stress and cycles curves (SN curves) for different materials. I'm mostly concerned with ferrous materials. I'm trying to relate fatique and number of cycles with stress in a purely analytical way without doing any physical testing.<br>
<br>
Thanks

 

[Irwin]

You can buy lots of books. Try the

http://www.sae.org

website's bookstore.<br>
<br>
Irwin

 

the problem you will have is not getting hold of the SN data but calculating the stress at the fatigue failure location in your part!! (unless you can be sure that the guy who generated the SN curve used the same method for calculating the stress as you and used similar geometry.)

- 99.9% of the time fatigue failures occur at stress concentrations (i.e. welds, sharp corners etc) - and the question to answer is how sure can you be of the stress you get from either FE or measurement with strain gauges - it's a million dollar question I'm afraid.

Most people use an average stress near to the region of the stress concentration and then multiply this by a stress concentration factor (a standard value for a standard geometry you can simply look up in a book (Peterson for instance)). This sounds great except that your geometry must be the same as the geometry in the book for it to work.

This is why there are endless research programs trying to get more accurate estimates of fatigue life for various bits of metal in different configurations - and why no one is able to come up with an analytical method that works well for every situation.

 

[Irwin]

Try to use FEMFAT! This can solve your problem! This program uses relative stress gradient for the SN curve modification. The relativ stress gradient is somewhere somehow similar with notch factor, but more general!

Best regards,

Ervin