static coefficient of friction between two steel milled surfaces 1

[Mr_Curious]

Hello everyone.
I am looking for a static coefficient of friction between two milled dry A 516-55 steel surfaces for calculating a threaded connection. I have found a table of friction coefficients in Nasa's Fasteners design manual (below). There two types of steel I can choose from. Which is hard and mild. And in my opinion, coefficients are too big (0.74 and 0.78). Also, I am not sure if it is about milled surfaces.
Could anyone tell me whether one of those coefficients is right for my case or not? And provide me with a source, please.

 

[desertfox]

Hi Gordievsky

You won’t find a single value for friction coefficients, there are to many variants and the only way coefficients are found is by experiment. If you are making a threaded connection which needs accuracy then the threaded joint needs making by using a bolt tensioner as that eliminates the inaccuracy of friction on the threads. If you are making a thread connection with torque be aware that method is subject to an error +/- 25%. If you visit this site it gives a lot of coefficients for friction on threads.

Coefficients Of Friction - Roy Mech

roymech.org

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[Tmoose]

Hi Gordievsky,

The choice of that A 516-55 for making fasteners is "interesting." I believe it is usually plate, and used for welded pressure vessels of moderate temperature.

Then again dry milled threads are interesting as well.

 

[LittleInch]

Well if you get them really dry and very clean and in a vacuum they will weld themselves together...

https://en.wikipedia.org/wiki/Cold_welding

As it shows in the table there is a big range between "dry" and "greasy" so maybe a little bit greasy is somewhere inbetween.

Of course then a bit of rust or roughening of the surfaces and a different set of results.

As a max value it looks good to me



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[IRstuff]

What's the post-milling surface treatment?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[dhengr]

Gordievsky:
Friction and friction coefficients are horribly undependable and fickle things. When you want and need them, you can’t count on them being there, because, for a dozen reasons they might not be there. When you wish you didn’t have to deal with them, they always have the chance of being higher than you would like. That’s why torque on a nut and bolt is so unreliable as a means of setting the tensile load in a bolt, or the degree of tightness of a connection/joint. If you are in a production situation, where you know the quality and consistency of your nut and bolt suppliers and their products, and your own production consistency, you can test a few hundred of them, and set a workable torque range for a given bolt tension. Then, torque measurement becomes a practical production means. It should still be checked regularly. It is almost never a reliable means of assuring bolt tension for a single bolt.

 

[MintJulep]

What do you want the value to be? I'm sure there is a reference that will support any value you desire.

Alternately, if it's actually important to you then test your parts in your application.

 

[Tmoose]

The OP's subject line requests static coefficient of friction - µ.
I think any torquing / tightening procedure is likely closer to kinetic/dynamic friction.

Thus, as the OP's OP has placed and left me way out-on-a-limb, it seems possible and even likely to me the OP is not really talking about fastener µ, but instead the faying surfaces of the components being bolted together. But then again perhaps he really IS talking about his fasteners. Only the OP can crack this nutty quandary.
As so often is the case, I most sincerely wish the OP had done so by providing sufficient information in the very first post.

Something I find interesting are the results of "torque cycling" fasteners.
These are for nicely finished, lubricated commercial fasteners. Even if assembled dry in somewhat raggedy tapped holes, if seizure and galling can be forestalled I think the concept is correct.

https://arp-bolts.com/i/t/PreloadGraph.gif

So while in my warm comfortable arm chair if I somehow choose (guess) a perfectly appropriate initial µ, if the device is trial assembled a time or two, or perhaps if the tightening is done in a few stages, I think the achieved fastener tension will be significantly higher than cyphered initially.

For decades, the late great Joe Mondello was an advocate of "torque cycling" critical fasteners to achieve reliability in the various bolted joints on the over-achieving automotive engines that he built . I believe the effect is 100% the result of improving the fastener finish and modifying (lowering)µ.
As exemplifed on Page 9 and 10 here -

https://mondelloperformance.com/wp-content/uploads/2017/09/RebuildYourFirstEngine.pdf