Fastener Axial Load Exceeds Preload Question

[Transient1]

The situation is that I have a fastener that may exceed its estimated preload by 50%. This axial load is not enough to damage the threads. What is the consequence of exceeding the preload if the pull out load is not surpassed and the fastener is staked around the fastener head to prevent rotation? I suppose I can expect the joint to shift, but
as long as no yielding occurs I would expect after the axial load is removed that the joint would go back to its original preloaded condition. Is this fauly logic?

 

[rb1957]

if the applied load exceeds the preload, the joint gaps, and the fastener reacts all of the tension load. in the gapped condition, the joint becomes a lot looser and maybe this could affect the rest of the structure. if P/A < fty then i wouldn't expect significant permanent deformation. if this happens a lot, you should do a fatigue analysis of the fastener.

 

[desertfox]

Hi Transient1

I agree with rb1957 in short either the fastener comes loose
and fall apart or comes loose and fails by fatigue.

http://www.boltscience.com/pages/vibloose.htm

Have a look at this site

Regards

desertfox

 

[HeviiGuy]

Transient1:

The other thing which is important in this situation is the original preload of the fastener. How close to yield is it? A 50% excursion may take it in to yield. Consequently, no, the fastener will not return to its original condition.

Ciao,

HevïGuy

http://www.heviitech.com

 

[drawoh]

Transient1,

What about metal fatigue?

If the force on the joint stays below the preload, you do not have metal fatigue.

JHG

 

[Transient1]

Thanks for your input. I've decided to just redesign the part rather than justify using it as is. All of you confirmed what I was thinking. However, I think I might be missing something conceptually about the fastener fatigue based on some of the comments that were made.

It seems to me that if the cyclic axial load does not exceed the preload that fatigue of the fastener threads is still possible.

(1) The fastener is under tension, applying an axial load (0 to < preload) still reduces the resultant compression of the fastener. Therefore, there is a cyclic load and unless the fastener preload is below the endurance limit of the fastener eventually you would see a fatigue failure.

(2) If the material the fastener is thread into (a nut or internal thread) is compressed. An axial load adds to the load on those threads. Therefore, they could fail in fatigue if the cyclic preload+axial load is above the endurance limit.

 

[drawoh]

Transient1,

I have been totally confused by this, in the past. Here is how I visualize it now.

[ol]
[li]Your external load varies from 250lb to 750lb. [/li]
[li]You torque your fastener to a preload of 500lb.[/li]
[li]When the load exceeds 500lb, the load on the clamp face drops to zero, and the clamped part is lifted off the face. The bolt strains due to the increase of force from 500lb to 750lb. [/li]
[li]Your bolt fatigues due to the varying stress.[/li]
[li]You replace the fastener and torque the new one down to 1000lb. [/li]
[li]The clamped part does not lift off the face.[/li]
[li]The bolt's strain does not vary.[/li]
[li]The stress on the bolt is constant, and there is no fatigue.[/li]
[/ol]

JHG

 

[desertfox]

Hi Transient1

drawoh beat me to it but he is spot on, provide your preload is high enough fatigue in the fastener should not occur.

regards

desertfox

 

[rb1957]

point 8 is not Really Correct ... there is a much smaller stress cycle in the bolt (not zero), the load inside the bolt increases slightly as the external load is applied depending on the stiffness of the joint.

 

[drawoh]

rb1957,

I would think that would be way less significant than the difficulties in imposing exactly 1000lbf on the bolt. If the joint flexibility is significant, the joint design becomes way more difficult.

JHG

 

[Transient1]

From drawoh's post it made sense to me that the stress in the bolt is not changing appreciably because the strain is roughly constant. However, rb1957's comment fills that idea out more. I agree with his comment that the stress cycle is dependent upon the joint stiffness. This is what the equation for axial bolt load shows:

P_b = Preload_max + Loading Plane Factor*Load Factor*(Safety Factor * External Axial Load)

- AXIAL LOAD MUST BE LESS THAN PRELOAD.

-The Load factor is equal to the bolt stiffness divided by
the sum of bolt stiffness and joint stiffness.

-The loading plane factor is the thickness of joint between
force application points divided by the total thickness
of the joint.

Therefore, the load on the internal threads will vary between preload max and P_b. Conservatively, the internal threads have to withstand preload_max+SF*external axial load. That is closest to reality when the joint is less
stiff than the fastener. If the joint is stiffer than the fastener then the fastener picks up the load.

However, this equation suggests that if the bolt and joint stiffness is the same the bolt load will vary between preload and preload+50%*axial load. Suggesting, fatigue failure is possible if the preload is all ready 80% of yield.


This equation is in Shigley (p 421) in a slightly different form (no loading plane factor). I got this particular formula from a NASA docuement, NSTS 08307 Criteria for preloaded bolts.