bolts that back out...why?

[jimg]

Theres a thread over in the ag forum asking about pulley bolts that backed out unexpectedly. That got me wondering why that happens? Is it vibration only or are there other reasons? Id just like to understand the mechanisms at work...might give some insight into what things to check periodically.

 

[RobJ]

Vibration, metal fatigue, not tightened correctly (under and over), constant stress (moving of parts) under the bolt, wear would be some of mine. On airplanes they safety wire most stuff. On a tractor you can convert nuts to locking nuts or double nut the bolt if you can.

 

[AndyMA]

One of the most effective ways to rpevent bolts from backing out is to use a product like Loctite when you put them it. Loctite US

Andy

 

[jimg]

Interesting...thanx Robj. Do you understand the mechanism(s) ov movement? I can sort of understand vibration but even in that case what causes a bolt to loosen vs tighten?

Andy...thanx...I knew about loctite and use it somewhat often.

 

[3RRL]

Jim,
Don't know for sure but I think a bolt with threads is much like a gutter and water. It will find it's own path of least resistance.
Pieces bolted together twist and turn and flex under stress imposed. May not be seen by the naked eye, but those forces are at work. Things trying to move will move towards the path of least resistance. We see that when we engineer say several connections, and we can identify the weakest link.
In other words, we know stress imposed on all those connections will impact that weakest connection the most. So I think of the bolts being that way. They can stretch due to force and then any movement taking place will the path of least resistance, or the unscrewing of the bolt or nut. Going the other way is the hard way or the path of "most" resistance.
That's my 2 cents...

 

[MJPetersen]

I would think that it probably is a lubrication issue as to the "why?" Bolts subject to rotational forces should be safety wired (as above) or pinned (as in the nuts on front hubs. A cotter key or roll pin will prohibit a nut from loosening. Why it does not tighten? As Rob said, it takes the path of least resistance.

Mike

 

[BTDT]

Lack of maintenance. On some things, trailer lugnuts come to mind, you are supposed to retighten every so often. I think thread pitch also has something to do with it.

 

[RobJ]

Interesting...thanx Robj. Do you understand the mechanism(s) ov movement? I can sort of understand vibration but even in that case what causes a bolt to loosen vs tighten?

Andy...thanx...I knew about loctite and use it somewhat often.

Heck no!! If you break it down I would guess metallurgy is in there somewhere. Tighten a bolt, you get friction, heat, wear, wear equal loose. Then some stretching over time. If you need to hold something down big, you get a bigger bolt. I guess all of the above will be less on like loads with a bigger bolt. Any of us can pretty easily twist off a 1/4" bolt, but not as easy with a 1/2" bolt.

 

[jimg]

neat and interesting...you guys are pretty smart. Well, at least it all sounds plausible but I suspect too its right on the mark. Thanx again!!

 

[Reg]

In a properly tightened bolt/nut thread set there is elastic tension and friction (at least). As long as the elastic tension is holding the friction surfaces together there is "holding". Torque values are an attempt to quantify the desired amount of tension, bolt elongation is measured as a representation of it in critical applications.

Errr, Ummmm, I think that is most of it - its been a long time.
Oh, improper selection of bolt size, material, thread profile, improper tightening, corrosion, etc can all lead to loosening. Overtightening is BAD, but you probably knew that.

EDIT; Consider bicycle pedals. The rotation relative to the crank suggests that they should have a left hand thread on the right hand pedal and a right hand thread on the left hand (foot) pedal. Due to some whacky epicyclic effect they're the other way around. So yes, proper DESIGN counts. ;END EDIT

 

[jimg]

Ummmm....hold on guys. re: path of least resistance. I recall several times having bolts tighten to the point of busting a gut to loosen them. I mean those suckers really wound themselves into the hole to the point of needing a LONG cheater bar and lots of grit to move them. Guess it was induced by the direction of rotation of the part...just a guess on my part. Anyway, there was nothing least resistance at those times!!!

 

[BTDT]

In pre-Loctite days, (so I've been told ), there were little stamped pieces of metal with a tab sticking out, that the bolt would go through, and when you got bolt/nut tight, you would bend the tab up against the flat side to keep it from loosening. I've seen these on axle bolts on trucks, u-joints, etc.

 

[bx23barry]

In pre-Loctite days, (so I've been told ), there were little stamped pieces of metal with a tab sticking out, that the bolt would go through, and when you got bolt/nut tight, you would bend the tab up against the flat side to keep it from loosening. I've seen these on axle bolts on trucks, u-joints, etc.

Go here

Welcome to STAGE 8... Call Us Now! 1.800.843.7836

For the modern system

 

[john_bud]

Bolts loosen up in the inverse proportion to your ability to retighten them. Easy bolts that are right out in the open will nearly never loosen up. Bolts that require major dissasembly just to feel, will practically fall out after the first bump!

That's my story - and I'm sure I didn't forget to torque or add loctite !
(well, pretty sure anyway!)

jb

 

[mlauk]

The threads should be clean when installing a bolt. The bolt should be torqued properly based on the size of the bolt and the material, ie grade 8 bolt requires more torque than a grade 5 bolt. The proper torque keeps the bolt from over stretching and puts tension on the threads so they don't back out. Any kind of locking device also helps such as a lock washer, lock nut or a thread sealer.

 

[shaley]

A good, short presentation on bolt proof stress.
http://www.utm.edu/departments/engin/lemaster/Machine Design/Notes 29.pdf

 

[RobJ]

This thread also reminds me of the "where does the grease go" thread.

 

[tallyho8]

Why is it that when I am changing a tire at my shop and have all my tools handy, the lug nuts seem like they are almost too loose. But when I have a flat on the road, with a stupid manufacturer supplied lug wrench, the lug nuts seemed to have tightened to the point of nearly breaking?

 

[SnowRidge]

Nobody mentioned the "dash of weld" method of locking the bolt or nut to whatever it is holding.

 

[zzvyb6]

Pulleys "walk", meaning they can have out of plane forces acting on the shaft locking point. Since a screw is a form of inclined plane, motion can develop as the thread is cyclicly loaded.

Best way to stop its effect is to put a point on the sheave bolt/screw and a dimple in the shaft. The point will maintain the position in spite of the pulley wobble. Normally a key is used to maintain pully drive torque, so the pointed fastener is only needed to maintain position, not to transmit torque.

And I an unanimous on this.