Fatigue life 3

[paramathma]

Fatigue life

I need to find the life of a bolt (no of cycles)

I have a bolt, which is subjected to fluctuating Tensile load of 0 to 2000 pounds.
The minimum cross section of the bolt i.e. at the root diameter is .095”.

I then calculated the max stress =max stress/Area, and found the max stress.
The stress ratio is 0.
Can we find the bolt life using the SN curve and the max stress?

I have a simple question
Normally our stress group only provides us the max stress.
They apply the load directly and finds the max stress.
In this the fluctuating load does not comes in to picture.
Then how can we use the stress obtained from the FEA to find the life (component) using the S-N curve?

Please advice

 

[paramathma]

The Bolt thread is .1250-44 UNJF.
The min root dia of the thread is .095"
The bolt is custom made.
the bolt material has a yield strength of 340kSi (18 ni maraging steel)

 

[rb1957]

have we helped any ?

you haven't given us a lot to go on ... how often is the load applied ? how many load app's are you looking for ? is the bolt going to be lifed (and regularly replaced in service)?
why can't you torque the bolt (and create a preload) ? maybe the plastic would just squish out of the way ?
be worried about the strength of the plastic.
are you going to test the installation ?

 

[drawoh]

paramathma,

So it is a #5 screw!

You should be working out the tensile stress area of your screw. You can get this from all sorts of handbooks. The Handbook of Bolts and Bolted Joints, Bickford and Nassar, show how to calculate this. The values are different for UNJ and MJ threads, as opposed to regular threads. Machining the thread, as opposed to rolling it, might be an issue too.

This is a weird size. You might have problems finding standard mating fasteners, as well as fabrication and inspection tools. Why so tiny?

JHG

 

[rb1957]

"Machining the thread, as opposed to rolling it, might be an issue too." ... it certainly will, machined threads have a shorter life than rolled

 

[unclesyd]

Here are two papers on bolt fatigue you might want to get.

http://journals.pepublishing.com/content/f0h87n8481233600/

This one has some references to other investigators. I noticed that there is reference to Shigley.

http://www3.interscience.wiley.com/journal/119385509/abstract?CRETRY=1&SRETRY=0

 

[tbuelna]

Wow, 340ksi yield and 350 ksi ultimate? I don't think even MP35 will get you there.

A simple P/A analysis is not adequate unless you are applying a very conservative FoS. There is the stress concentration (Kt)at the thread root and under-head fillet you must add, you must also analyze using the combined tensile and torsional loads due to wrenching at installation. And don't forget to include some bending due to misalignment of the seating faces.

As for fatigue, the surface finish of the parts will have an effect too. Finally, don't forget to account for any thermal mismatch strains due to dissimilar materials.

 

[paramathma]

Thank you All

Now I know the difficulty in calculating the fatigue life of bolt.

Thanks