Fatigue Analysis with MecWay

Olalleitor

Hello from Spain.

I'm taking my first steps with MecWay.

What is the best way to carry out an fatigue analysis using MecWay?
Constant Cicle. Simetric Traction and Compression. GRP Material.

Best regards.

Olalleitor

Nobody can help me?

prop_design

hi,

i do fatigue analysis with mecway. however, i didn't respond to your post because you are doing things i haven't done. i have several videos on youtube showing how to use mecway. they include fatigue analysis. but again, probably not gonna help you out.

basically, you can calculate stress, strain, etc with mecay. victor added custom formulas not long ago. you can use the custom formulas to do many things. so you can set those up to do the fatigue analysis the way you want. you also have to have your own material properties. the old phrase of garbage in garbage out is very appropriate to fea and fatigue analysis. without the 'real' material properties, you will just be wasting your time anyways.

victor and others may be able to help more.

anthony

Victor

I can't help any more than prop_design, sorry. Mecway doesn't have any explicit tools for fatigue so it would be a matter of using the existing results like stress in your fatigue model if that's possible.

mmartin

I am also interested in fatigue analisys.I think that it has to be solved comparing stresses between events . For doing this you will need python script or the MW spreed sheet(tables)

JohnM

We use Mecway to perform basic fatigue analysis. The slide and notes attached shows the basic idea for high-cycle fatigue for typical engineering materials. I don't expect this to be the "lock" on how to do this, but it is a good start.

Our current method drags the data out from FRD files using stone-age FORTRAN.
If the quarantine has made you an ambitious to flex your Python skills, I am happy to discuss the details for upgrading this method. My only request is that the resulting PY code be shared with this community!

Given two stress results, S1 and S2,

To calculate the alternating stress from the multiaxial stress result, starting with "node1", first calculate the delta stress for each component:

delta_sx =(sx1-sx2)/2
delta_sy =(sy1-sy2)/2
delta_sz =(sz1-sz2)/2
delta_sxy=(sxy1-sxy2)/2
delta_syz=(syz1-syz2)/2
delta_sxz=(sxz1-sxz2)/2

then calculate the vonMises equivalent stress from these components.

This now represents Sa (alternating stress) for node1 (now do it for node2-N)
This method is outlined in the ASME Boiler&Pressure Vessel Code, Section VIII



Repeat the above process but calculate the mean stress between S1 and S2, for example

mean_sx =(sx1+sx2)/2
...
...

In this case, there are several options for determining the mean stress state.
You can:

1- Calculate von Mises (but has not sign, so can very conservative)
2- Calculate vonMises, but assign the result the sign of the largest principal stress (can sometimes show strange results)
3- Calculate the max principal stress, use this as mean stress state (popular in "practical" powergen applications)

There are (many) more variations, but I recommend (3) for its simplicity.



Now that you have Sa and Sm, you simply apply the formula in the graphic:

SF= 1/(Sa/Se+Sm/Su)

where
Se is Endurance Strength
Su is Ultimate Strength


(if Sm is negative it is set to zero)


If you have many cases, you simple repeat the calculation:

S1 to S2
S1 to S3
...
S1 to SN


S2 to S3
S2 to S4
...
S2 to SN


As you calculate safety factor, you remember which two stress states represents the worst safety factor. Keep that node, Sa, Sm, FS.

To view the results, you can reformat and FRD file, or run a dummy thermal run where you set the temperatures to the Sa values, then the Sm, then FS values, so you can view the minimum safety factors and see the relevant Sa and Sm.

disla

Very good post JohnM.
Covid's fatigue challenge.

Olalleitor

Thanks!