Effect of surface finish on coefficient of friction 3

[mschro]

I am looking to bolt two steel surfaces (A36) together. Most charts I have found show a static coefficient of friction of .74 for clean and dry sufaces. The surfaces will be subjected to heavy loads that would cause the surfaces to slide against each other. I am limited by the number fasteners that can be used to bolt the surfaces together because of space limitations.

The question is, to what degree would sand blasting (making the surfaces rough)increase the coefficint of friction compared to say s 63-125 micro inch machined surface?

 

[CoryPad]

The 0.74 value you quote was measured during testing conducted in the year 1939 for parts in VACUUM. For parts exposed to air, the value is half that. This is for controlled testing in a laboratory.

In reality, the friction coefficient varies widely. In fact, environmental factors (like relative humidity, accidental contamination, etc.) cause variations that are much larger than that caused by surface roughness.

If there will be no lubricants (water, oil, paint, etc.), then you could use 0.2 as a minimum design number. If there are any lubricants, you should use 0.1 as the minimum.

Of course, you can conduct your own testing. It could be as simple as dragging an object with known mass across the other surface with a fish scale. There is an ASTM standard for this, ASTM G 115 Standard Guide for Measuring and Reporting Friction Coefficients, available at:

http://www.astm.org

Regards,

Cory

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

 

[mschro]

Any idea on the percentage gained by the change in surface finish. Are we talking a couple percent difference, or could it be 10% (or greater) difference?

Thanks

 

[CoryPad]

I think the friction coefficient difference between machined and sand blasted surfaces will be less than the differences that you won't/can't control, and therefore, you shouldn't waste your time with it. If you have to rely on friction for force/motion transfer, then assume a low value or work with special interfaces, like EKagrip:

http://www.esk.com/Products/Ekagrip.asp

Regards,

Cory

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

 

[IRstuff]

If the objective is to keep the surfaces from sliding against each other, why couldn't you machine some locking features in to the plates?

TTFN

 

[NickE]

Sometimes a rougher surface will have a lower coefficient of friction than a smoother. Eg. If there is lubricant, and it coats the surfaces it will be trapped in the valleys of the rougher surface, thus not squeezing out, and provide great lubrication to the interface.



Nick
I love materials science!

 

[FACS]

The question is, to what degree would sand blasting (making the surfaces rough)increase the coefficint of friction compared to say s 63-125 micro inch machined surface?

Probably not much. Machined surfaces of those typical values can include Planing, shaping, drilling, chemical milling, EDM, broaching, reaming and turning... to name a few.

So really, what size shot in the blaster being a variable, you are not going to gain anything noticeable.

Charlie

http://www.facsco.com

 

[MintJulep]

A rough surface may initially have higher friction. However, after you clamp the surfaces together, or after the surfaces move relative to each other, the peaks will tend to compress and flow, reducing friction.

This may be even more pronounced on a joint that is subjected to repeated assembly/dissasembly cycles.

If you really need to know, test your application. Do not rely on handbook values. You don't know how they compare to your application.

 

[corus]

If you're limited by the number of fasteners you can have then why not use a higher grade bolt and torque it up more so you gat a greater clamping force and hence resistance to sliding, or use a bigger bolt diameter for the same effect.
As Corypad suggests, use a coefficient of friction that is recognised. For in-service conditions a value of about 0.1 is generally used, though what the source is for that I don't know.

corus

 

[klaas2]

sorry facs

what is edm ?

 

[prex]

Friction grip joints are normally used for structural steel.
The italian code (but I guess that similar rules exist in other codes) allows for taking two levels of friction:
0.3 for not specially treated surfaces and for site assembled joints
0.45 for white metal sandblasted surfaces, exempt of any dirt and well done as far as planarity and parallelism are concerned.
Bolt tightening must be to 80% of yield and a coefficient of safety of 1.25 (only) is applied to the calculated transmissible load.

prex

http://www.xcalcs.com

Online tools for structural design

 

[CoryPad]

klaas2,

Electrical Discharge Machining

Regards,

Cory

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

 

[klaas2]

ok understand we have another word,but i use then also EDM
nice to know

 

[mschro]

So, based on what prex said, and based on the Italian code, the surface finish can have a very large impact. Correct?

 

[CoryPad]

Surface finish can have an impact (maybe large), but prex's post is not the "apples to apples" comparison for which you looking.

The USA equivalent to his data is from the Specification for Structural Joints Using ASTM A325 or A490 Bolts published by the RCSC, and available at:

http://www.boltcouncil.org/download RCSC Specification.htm

It lists three Classes: A, B, and C. Class A is uncoated clean mill scale steel surfaces or surfaces with Class A coatings on blast-cleaned steel, with a mean slip coefficient of 0.33. Class B is uncoated blast-cleaned steel surfaces or surfaces with Class B coatings on blast-cleaned steel, with a mean slip coefficient of 0.50. Class C is roughened hot-dip galvanized surfaces, with a mean slip coefficient of 0.35.

If you want to use surface roughness to improve slip resistance, you really need to test your actual parts and conditions.

Regards,

Cory

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

 

[dimjim]

What kind of finish would you expect to get
taking a 63 or 125 finish and sandblasting it?
Wouldn't it get finer and smoother?

 

[atlarson]

Is it possible to match drill and ream two or more holes through the plates for press-fit dowel pins? This will gaurantee zero slip as long as the dowels are sized to prevent yielding.

-Andy

 

[KENAT]

I know it may sound like swearing but is some kind of adhesive an option?