Studies on the use of lock washers? 6

[bumpjones]

Does any one know of any engineering studies performed to determine what if any affect the use of lock washers has on the integerity of s structural joint?

 

[IRstuff]

This has already been covered here before. The bottonline is that lock washers generally do nothing.

TTFN

FAQ731-376

 

[bumpjones]

Yes the subject of lockwasher performance has been covered many times in this forum. I reviewed the FAQ section and understand that the predominant opinion is that lockwashers don't help. I was asking if any one knew of any studies that supported this positon. It helps to have supporting data when making an engineering arguement.

 

[Bribyk]

I found NASA Reference 1228 (1990) "Fastener Design Manual" one day in my Google travels. Direct quote "its locking ability is nonexistent... [and] is useless".

 

[John2004]

Here is the link to the NASA document 1228...

http://gltrs.grc.nasa.gov/reports/1990/RP-1228.pdf

Once a split lock washer is made flat by fastener tension, it's basically just a plain flat washer at that point.

I think

http://www.boltscience.com

has a video of the split lock washer coming loose under vibration, and they compare it to other locking methods on video. You might contact them for references to other studies, I think they have a section on their web site that lists studies and references.

I think the lock washers from

http://www.nord-lock.com/

are supposed to actually work.

Loc-tite works well.

John

 

[DesignerGuy16]

I've used the nord-lock washers in a robotics environment with high vibration and they worked great.

But yea, I read that NASA article and it's changed my view. Then again my industry (food), any exposed split lock washers are just another food trap so it didn't take much to sway me.

James Spisich
Design Engineer, CSWP

 

[MintJulep]

There are several papers and sources cited in faq404-1257.

 

[vims]

Nord-lock has a nice video on this.

http://www.nord-lock.com/default.asp?url=51.16.37

 

[unclesyd]

Without comment.

http://itwmc.com/content/default.aspx?cID=78

 

[MintJulep]

From unclesyd's link, with my bolds.

Helical Spring (split) lock washers are a time tested design preference in many assembly applications.

Improving assembly integrity is accomplished with the introduction of the helical spring lock washer in the following ways.

The split in the lock washer absorbs initial driving torque and visually closes under nominal bolt loading. When tension in the assembly is reduced and loosening occurs, it provides resistance to the back-off rotation of the bolt.

 

[Bribyk]

I can vouch for their loosening resistance...sometimes. They seem to relax in use and become ineffective though. If you tighten a nut down with a new split lock washer and then undo it shortly after there is some resistance to loosening and it usually gouges the clamped surface. Loosening such an assembly that has been in service for a while, the washer acts like a regular flat washer because they seem to lose their temper.

 

[John2004]

I think the main use of a split lock washer is just to add a little axial tension to the system as the nut or bolt is being tightened. This is useful if you cannot get a wrench on both the nut and bolt head, and it helps to get things tightened down in those situations.

A split lock washer probably does bite or gouge into the mounting and/or bolt/nut surfaces a little, but I'm not sure how much it would help with locking and gouging of the surface does not seem a dependable locking method.

If the split lock washer had knurled surfaces or a surface that could "bite into" the mounting surface and fastener faces, I would think that might help a little with preventing loosening. Maybe rough them up a little with a file or something ?

From Unclesyd's link...

"When tension in the assembly is reduced and loosening occurs, it provides resistance to the back-off rotation of the bolt".

I think the statement above is true, but the problem I see is that the vibration, movement, or force that initially overcomes the friction of a properly tightened fastener causing it to loosen, would probably very easily overcome the friction resulting from the axial tension of the split lock washer, which only decreases as the joint gets looser. I guess it depends on the application and what forces are at play but it seems to me that as soon as loosening occurs, the game is basically over that point.

"Technology University have shown that the typical helical spring lock washer exhibits a spring rate after flattening which is approximately 70% more effective than a flat washer of the same thickness".

70% is significant but I would be interested to see how the test was conducted. There can be more friction variation in identical fasteners than 70%, so I'm not sure how they accounted for that in their tests. Did they measure actual fastener tension or stretch in their comparison or did they just torque the fastener down ? I would think they should have tested many fasteners from different production lots and manufacturers.

The shakproof metal components site mentions that the split lock washer is twisted in addition to having the split ends at different elevations. However, whether the washer is twisted or simply has ends with different elevations, flat is flat, and at the point at which the washer is flat (ends at same elevation and twist removed) it would seem to have to act and respond like a flat washer. At the point at which the washer is flat, any difference in spring rate would seem to have more to do with differences in washer material, i.e., elastic modulus of the washer material, etc. as opposed to washer design.

NASA is an authoritative source, but like everyone else their not infallible. I would be interested in seeing actual tests from both sides before forming more rigid opinions.

I'm guilty of using split lock washers for certain things even after reading the NASA report, I guess old habits die hard.

John

 

[zifos]

... "Technology University have shown that the typical helical spring lock washer exhibits a spring rate after flattening which is approximately 70% more effective than a flat washer of the same thickness". ...


So how effective is a "helical spring lock washer" that is 70% more effective then zero???

 

[Bribyk]

All I know is I have yet to see a nylock or stover nut lying uselessly beneath the bolt it was formerly intimate with. I've picked up more than my share of lock washers and nuts from around vibrating equipment. I don't need NASA to convince me...

 

[ShaggyPE]

Not a scientific test by any means, but recently we were doing some grinding on a frame that already had some components mounted to it (using lockwashers and nuts). It is amazing to see how the vibration of the grinder will cause nuts to back right off and continue to unscrew until the fall off. So much for those "lock" washers.

Good thread guys and some good links in there.

-Dustin
Professional Engineer
Certified SolidWorks Professional
Certified COSMOSWorks Designer Specialist
Certified SolidWorks Advanced Sheet Metal Specialist

 

[Tunalover]

Reading a "technical report" by ITW SHAKEPROOF, a major mfgr of helical split lockwashers is like listening to a tabacco industry executive testifying under oath that cigarettes don't cause cancer.

It may be true that the split lockwasher may prevent the screw from turning completely out ONCE THE JOINT IS LOOSE but the idea is to prevent the joint from coming loose in the first place. The best way to do this is with one or more Belleville washer in addition to prevailing torque style fasteners (aka self-locking threads via a plastic locking element). Of course if one can figure out the necessary installation torque to produce 75% of yield and can figure out a way to inspect that torque then no locking features would be needed at all. The problem is figuring out what torque to apply. The T=kPD formula is fine if you can only determine what the empirical constand k is for a particular joint.


Tunalover

 

[IRstuff]

I would think that Belleville washers are in the same league as lock washers. The same basic principles are (mis)applied. You're essentially counting on some spring constant generating some friction to keep the nut in place.

TTFN

FAQ731-376

 

[CoryPad]

IRstuff is correct. The force required to flatten helical and conical washers is much smaller than the forces that can (and should) be generated by threaded fasteners.

To prevent loosening, the best option is high preload, the next best is adhesives. Everything else is much worse in terms of performance, cost, space, etc.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[HeviiGuy]

Of course if one can figure out the necessary installation torque to produce 75% of yield and can figure out a way to inspect that torque then no locking features would be needed at all.

Not a problem at all on either point:
[ul]
[li]One really can't practically calculate the required torque necessary to accurately achieve a desired bolt stress (one can never know what the K factor will actually be). An educated stab-in-the dark is necessary. However, the answer to the second wish made by tunalover solves this dilemma...[/li]
[li]Calculate the fastener's elongation at the desired preload. Then, measure the stretch after tightening. Tweak the "torque" either up or down until the correct elongation has been achieved. Simple.[/li]
[/ul]


I've posted it before but, here once again is a small Excel file that'll help to calculate both...

Ciao,

HevïGuy
www.heviitech.com

 

[tbuelna]

CoryPad,

"To prevent loosening, the best option is high preload"

Technically, I would disagree. The mechanism that keeps the thread locked in most bolted joints is friction. High preload will produce high thread friction as long as the preload is maintained. But in a bolted joint with low elasticity subjected to vibration or large CTE mismatch, the preload can increase or decrease substantially. Sometimes enough to relieve the preload and allow the fastener to back out. So I would argue that the best option to prevent loosening is a joint with adequate elasticity to prevent loss of installed preload. And not necessarily a joint with high preload.