Fatigue Life calculation on asymmetric cycle

tonystark
tonystark New Altair Community Member
edited October 2020 in Community Q&A

Hello,

trying to calculate how optistruct gives results about fatigue. 
I've tried with an easy loading before, symmetric cycle with R=-1 and TABFAT 1.0 -1.0 and it gave me the correct sigma_range. I used 415 as UTS on Hypermesh to generate the wohler curve, giving me SR1=1769,15 b=-0.125. Using the formula Nf = (2*SGVON / 1769) ^ (1/-0,125) I found the cycle number obtained by Optistruct. Where SGVON is the value of the sigma obtained on Hyperview.

 

Ok then I've tried an asymmetric cycle with TABFAT 1.0 -.5. I've read the stress and using Goodman formula I should have 1.6M cycles...giving a wrong result. Instead Optistruct tells me that the element survives 18224 cycles.

 

I read for example sigma max = 286MPa. Sigma Min 143 (it's the half of the max..)

Sigma alternate=71

Sigma mean=214,5

sigma ultimate = (415MPa)

Goodman = sigma alternate / (1-sigma mean / sigma ultimate) = 148MPa

 

Sigma (range) = 2* Sigma (alternate) = 296MPa -> giving 1.6 million cycles??

 

How do I calculate this value in case of asymmetric cycle? (-1<R<0) ? What am I missing?

Thanks in advance,

Antonio

 

 

Tagged:

Answers

  • Rahul Rajan_21763
    Rahul Rajan_21763 New Altair Community Member
    edited June 2019

    Can you share the .fem and .out file ?

  • tonystark
    tonystark New Altair Community Member
    edited June 2019

    not needed anymore ..thanks :). It was just a simple supported plate. I've found my calculation error...I was using the sigma amplitude in Goodman formula..not Sigma alternate as I wrote here. Doing that I get the same number of cycles that Optistruct gives. Now the problem is that I need to do a dynamic fatigue calculation. I've received an acceleration time history in an RSP file and a # of cycles that the profile must be repeated. Can I use only Optistruct 2017.2 to solve this problem with latest Hypermesh 2017.3 ?

    Shall I do a transient analysis before and then a fatigue subcase calling the transient analysis?

    Thanks

Categories