Fatigue loading of pin in single direct shear

[redroab]

Hello All,

I am attempting to analyze a single pin (stainless) in fatigue that is loaded in single direct shear. The pin locates two plates with respect to each other. The plates are clamped together, so the only loading should be "true" direct shear. The cycles are fully reversed.

How does this analysis differ from a bending, torsion or axial fatigue loading? I've gathered that Sy should be corrected to about 0.577*Sy for shear, but otherwise I haven't found any different loading factors or approaches. Should the endurance limit be similarly modified? Are there any other important considerations that are unique to shear? I'm fairly surprised that I haven't been able to find any examples of shear fatigue loading in design examples.

I've poured over Shigley's, roymech and this forum ( this thread was *some* help), but to no avail.

Any input you could provide would be greatly appreciated,

Redroab

 

[rb1957]

consider mohr's circle ... pure shear is equivalent to a state of tension-compression (ie max principal stress = min principal stress = max shear stress.

you might neglect the compression (consider it as uni-axial tension) ... but i think this might be unconservative (consider von mises)

using von mises stress is probably conserative (this will be bigger than max principal)

either way, you've got a tension stress to apply to your SN diagram (i wouldn't factor by 0.577, i don't think the rationalisation applies)

i wonder what you'll do for Kt ? maybe Peterson has something ?

Quando Omni Flunkus Moritati

 

[bxtsafe]

Hi redroab!

Your question is very interesting because it's unusual in the industrial products.

The pin has been loaded uniquely by a pure shear load. There is no bending, axial nor torsional loading. In all the fatigue history, fatigue tests with pure shear never were done, because this situation rarely happens in the day-by-day analyses.

So, you should generate a S-N curve for your pin. This curve should be a S-N curve for shear loading. As there is no modifying factor to pure shear, I recommend you to use the torsional S-N curve in your analysis. The reason of this is because this is the situation that most approximate yours. So, you shold multiply the Fatigue Limit (found in the standard bending fatigue test) by the loading modification factor C_L= 0,577 , as shown in many fatigue books (for exemple Yung-Li Lee, Jaap Schijve, Shigley, Norton, etc).

After generate the shear S-N curve, you should compare the shear stress applied in the pin with the shear S-N curve and assess the number of cicles that the pin would last.

I'm at your disposal.

bxtguard

 

[redroab]

rb1957, bxtguard, thank you both very much for your responses. I've been reassured that this is indeed a rare problem. I'll go through the processes you both have described and in all liklihood come back with more questions!

Thanks again,

redroab.