AISI 301 - SN Curves

[antoineL]

Hi everyone,

I am working on fatigue analysis with a AISI 301 material (austenitic stainless steel). I'm looking for a S-N curve which describe the behavior of this steel when it undergoes a reversal tensile stress (R=-1).
I've already looked in MIL-HDBK-5, unsuccessfully.
Is there another way to approach these data (theoritical plot, similar steel...)

Thanks in advance.

Best Regards,

Antoine L

 

[rb1957]

whilst i haven't ny info, you'll need to specify the hardness, this has a major impact on fatigue life

 

[TVP]

Here is some information from Outokumpu:

http://www.outokumpu.com/files/Group/HR/Documents/Fatigueproperties.pdf

Figure 3 shows the change in fatigue strength with increasing hardness (microstructure change due to strain hardening and martensitic transformation).

 

[antoineL]

Rb1957 -> I don't know the hardness, the only data i have are:

-Fty: 135 ksi (930 MPa)
-Ftu: 168 ksi (1158 Mpa)
-thickness: 0.5 mm


TVP -> Thanks for this article, i think i saw it but the curves didn't inspire me, I was only looking for typical S-N curve (Cycle/Stress).

 

[TVP]

Sorry I wasn't able to provide more inspirational data. If you can't use the data from the Outokumpu document to estimate the problem you need solved, then you probably need to review some background information on fatigue. The fatigue calculator website has some good fundamentals:

http://fatiguecalculator.com/

FYI, you are not really going to find free S-N curves for Type 301 stainless steel (spring temper) just lying around the ether. This sort of testing is incredibly expensive, costing ~ $500 or more per data point, which is why companies that pay for it don't subsidize the rest of the world and make it freely available. Good luck with your project.

 

[israelkk]

antoineL

Knowing:

-Fty: 135 ksi (930 MPa)-Ftu: 168 ksi (1158 Mpa) and knowing that R=-1 you can calculate the number of cycles given you are familiar with basic fatigue theory. If you don't, you should look for a qualified engineer or use any college grade book for machine design such as from Shigley, Faires, Norton, etc.

 

[antoineL]

Thanks all,

I've already plotted a theoretical SN curve with basic formulas such as Se=0.5*SU; Sf=1.62*Su*Nf^-0.085; etc.
But the fact is that my results (Life prediction in hours, Number of cycle...) hugely depends on the shape of the SN curve and modifying factors. Now my goal is to assess that this result is "accurate" enough. Of course the best way to do that would be to run some test but as TVP said, it's really expensive. So I want to compare a maximum of S-N curve in order to confirm my first result.

An other question: If fatigue tests have been conducted in rotational bending, is it acceptable to say (or conservative) that Se(axial)=0.7*Se(bending) & Te(torsion)=0.577*Se(bending).

Again, thanks

 

[israelkk]

The S/N curve is only the beginning. You have to take into account the factor numbers due to surface condition, roughness, size, stress concentration, etc,. In fatigue analysis a common factor of safety is 4 to 10. Finally, an actual cycle test may be needed to prove your design.

 

[rb1957]

can you do a fatigue test to verify the design ?

135ksi steel is just over 1/4 hard (that'd be 125ksi).

this should be pretty tolerant for fatigue, a fatigue cycle of 1/4 to 1/2 Fty should give you a reasonable life; depending on Kt. if you want infinite life, maybe 0.10Fty.