Does bolt torque changes if using locknut

[Bouing]

I have an assembly using a bolt and a stover type locknut. In theory, friction is higher because of the locking mechanism so required torque for a specific preload should be higher too.

When doing the torque calculation (or using a supplied table) should I consider the locknut effect on applied torque or it is just too small compared to the other components? It seem to be difficult to get data of friction of those locking devices.

I am asking the question because I've found a table somewhere stating that torque on bolt should be lower when using a Stover nut and it does not seem logical to me.

Robin

 

[Tmoose]

Prevailing torque of the lock nut should be determined by testing an actual nut and fastener and that number should be added to the desired torque to arrive at a torque spec.

That type of locknut can and will tear up the fastener and gall and even seize unless lubricated with something nice

 

[MintJulep]

Yes, the prevailing torque should be considered, as it does not contribute to producing preload.

Some prevailing torque fasteners are supplied with a wax or other lubricating coating - which is why the "published" torque may be lower than for a standard nut.

As with any fastener, if your joint is sufficiently important that you need to maintain the preload within a known range you should test your actual application and develop a torque-tension relation for the application.

Do not rely on published values from handbooks or manufacturers' literature.

 

[israelkk]

A common approach (at least for military use) is to turn the bolt in the nut and read the torque just before the last turn to clamping. Now add this value to the clamping torque. This way you do not need any prior testing.

 

[Bouing]

Thank Everyone, I will torque the Stover nut and add this value to my equation of torque; really nice tips!. This is my first post here by the way and I've been reading this forum for years...

They are used only one for a permanent assembly so I don't mind with the tearing factor.

Robin

 

[Tmoose]

I suspect the "tearing factor" can mean the assembly is not fully preloaded.

We had 2 large housings attached to a large bore tube. 120 stover type locknuts on each end, conscienscously lubricated with a graphite alcohol mixture as specified by some MIL spec. The actual adjustment was to a specified length, and even the prevailing torque varied considerably amongst all the fasteners.

 

[Bouing]

Ok I did a test...

I measured the torque needed to rotate a bolt in 3 differents Stover nuts 5/8. I got something around 3 to 6 lb-ft. Considering that right now they torque their 5/8-11 bolt at 128lb-ft and this is quite low for a grade 8 fasteners, I don't understant why I should bother with the stover enhancing torque effect?

May be this effect is more important for smaller nuts and should be taken into account.

Robin

 

[Bolteng]

To my opinion you should only add a torque for the locking nut for critical applications where you also know the exact values for the friction coefficients. Otherwise it isn't necessary.

Let me give an example in metric thread (I have these values). Let's take an M16 with property class 10 (diameter and tensile strength near 5/8 grade 8).

The stover locknut according to DIN 980 refers to DIN 267-15 for the locking properties. The 'locking' torque should, according to the standard, be between 6 Nm and 42 Nm.

For tightening of the bolt to 90% of the yield strength you need approx. 333 Nm (if threadfriction = headfriction = 0.14).

If you would find that for one batch the mean 'locking' torque is 20 Nm, you could add this and apply 353 Nm.

When in the application the friction coefficient is lower, for instance thread friction = head friction = 0,12 the correct torque is 298 Nm. If the locking torque of that specific nut is not 20 but 6 Nm, the required torque in that case would be 304 Nm in stead of 353.
In this example the bolt will be tightened into the plastic deformation zone.

Therefore I think that it is only necessary for critical connections where all friction coefficients are known.

Adjo

 

[Tmoose]

Hi Bolteng,

Have you found friction coefficients to be "knowable" to an accuracy closer than 0.XX ?

Thanks,

Dan T

 

[Bolteng]

I have never seen tolerances below 1% on friction coefficient. Even if you lubricate with a very consistant lubricant there are a lot of other influences.
Standard bolts and nuts have tolerances that make it possible to produce in mass production.

Aspects like thread tolerance, surface roughness, angle of the washerface, alignment of the head to the thread, etc. all have an inluence on the torque.

Up to now I can only measure k values. But in the near future I will also be able to measure threadfriction and headfriction separately.