Bolt Loosening 4

[MechEngIntern]

Hello All,

I am a mechanical engineering intern. I have been tasked with analyzing bolted connections and their associated stresses on some of our products. I am moslty workng with single shear bolted connections. All my bolts are preloaded to tabulated values, I am working with M8X1.25 property class 10.9. My biggest obstacle has been coming up with a numerical value that will indicate at what stress points by bolts will start loosening. I would also like to know where to look in an ANSYS simulation to determine whether or not it will loosen (im thinkin it has to do with shear stress). I know that it has to do with the relative movement between the threads and the joint but I just cannot figure out how to put a value to it.

Thank you,

Kevin

 

[flash3780]

I'd recommend that you look into using locknuts, Loctite, or some other anti-rotation feature. If the joint is subjected to vibrations, you can assume that it will loosen over time.

 

[desertfox]

Hi

Tart ith this site for some information

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

Now you mention the bolts are tightened to tabulated values, I presume you are looking at bolt torques for a given size and grade?
If so you need to study the strength of the materials your clamping and check that the clamping force derived from the bolt torque isn't crushing them beyond their yield stress.
The torque value gives you a clamping force the bolt material can withstand but doesn't consider the strength of the materials being clamped.

 

[MechEngIntern]

Yes it is subject to vibration - seismic and shipping. Can I assume that any bolt, no matter how big will loosen over time due to vibration?


And I will make sure to check for the clamping force/yield stress of the material.

Thank you

 

[MechEngIntern]

What happens if the clamping force is greater than the yield stress? I just want a tight connection., will that lead to loosening over time?

 

[CoryPad]

Please search Forum725 with the keyword loosening. You will find a wealth of information.

 

[Tmoose]

"Can I assume that any bolt, no matter how big will loosen over time due to vibration?"

///// If //////// the bolt is tightened/preloaded to provide a clamping force comfortably greater than ( not targeted to barely exceed ) the loads in service

////////// AND ////////////// the bolted components are solid metal, with faying surfaces machined smooth and accurate enough that embedment (and resulting loss of preload) will not occur

//////// AND /////////////// the bearing stress of the nut and bolt faces is well below the yield strength of the nicely finished components

/////////ANd/////////////// lockwashers are forbidden in the assembly

/////////////AND ////////////// there are no temperature gradients or extremes to hinder the fastener maintaining consistent preload

///////////// AND /////////// the grip length/diamter ratio of the fastener > 7

then I would say the bolted joint should be quite reliable.

Adding loctite or safety wire or prevailing torque nuts or other anti-rotation devices would at best buy a few hours of fail safe run time if by chance the bolts were not tightened correctly and thus loosened. In all other circumstances they are superfluous.

Passenger car main bearing cap and con rod boits don't loosen very often in service with pretty severe loads, and even very severe loading. I submit it is because those joints are designed to satisfy the 6 guidelines above.

Conversely, there are plenty of stories of crankshaft snouts being progressively destroyed when the damper bolt is not torqued/tightened (sufficiently) correctly after a timing belt change, so insufficient clamping is applied. Loctiting the bolt will not substitute for insufficient clamping.
Check out the bolt with the about 80% of the way down this page

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

 

[Screwman1]

Moose has it right.
You can't just look at a torque table to decide if a fastener has the necessary clamp load to prevent loosening. first off, you have to know what the service load is and how it will effect the joint and components and then you have to look at the friction in the joint and fastener to decide if the applied torque created the required preload in the joint. this is not a trivial exercise and people make careers out of joint design (an example being CoryP).
A properly designed and installed joint will not loosen in service, but an improperly designed or installed joint will likely loosen and fail.

 

[Tmoose]

Can you post a few example pics?

 

[3DDave]

The best defense is a shear load path that doesn't involve the bolts or their friction component.

And everything TMoose and Screwman1 wrote.

Anti-turn materials can cover the symptoms, but not for long.

30 years and seeing hundreds of different successful and failed fastening applications.

 

[flash3780]

I agree with everything that Tmoose said. You'll want to make the flange as stiff as possible and the fasteners as soft as possible to maintain a constant preload on your joint. Torque the bolt to as near as yield as possible while maintaining adequate safety margin to prevent the nut from backing off. Rabbeted joints will also help prevent shear motion, as was mentioned. Also, I recognize that locking features certainly add a bit of cost.

That being said, I'd still recommend some sort of a locking feature to prevent nuts from backing off in vibratory environments (esp. very high frequency vibrations). The only reason not to use them is if cost drivers prevent you from doing so. Maintaining proper preload is essential for a reliable joint. It's a safer to assume that any joint without secondary retention will eventually loosen if subjected to vibratory loads.

As far as secondary retention goes, some good options are Nylock nuts and Loctite. For higher temperature applications, distorted thread locknuts are a good option. If access is limited, consider locking Keensert or Rosan studs/inserts.

One advantage of Loctite is that it provides lubrication during assembly for consistent torque-tension behavior. On the other hand, it's not possible to recheck torque on a joint with Loctite applied and dries.

Interestingly, some industries rely on rust for secondary retention. I've read that the US Marine Corps runs their tanks through salt water after assembling the treads to rust the bolts in place.

If using a mechanical secondary retention system (not Loctite), assembly oil or anti-seize compound is always good idea. Anti-seize is usually a good idea when using stainless steel bolts. One interesting anti-seize is (unflavored) milk of magnesia (magnesium hydroxide); I can attest that it is used in some very high performance applications.

One thing to note is that secondary retention devices often increase run-on torque: This must be accounted for in your bolt torque specification. Getting the proper preload on your joint is really paramount. If you don't, your joint will separate and/or your bolt will fail in fatigue.

 

[Tunalover]

I'd like to add that helical split lock washers DO NOT help prevent loosening. They actually make the joint more likely to loosen.

Tunalover