Friction and contact area 2

[corus]

Is the force required to slide apart two surfaces related only to the reaction force and coefficient of friction or does the amount of contact surface area play a part perhaps by 'ploughing' effects or by elastic deformation of the surfaces?

corus

 

[PeterCharles]

I was always taught that the frictional force was proportional to the normal load, the coefficient of friction and was independant of surface area.

 

[desertfox]

Hi corus

Assuming were talking dry friction here then yes the force required for one surface to slide over the other surface is
related only by the coefficient of friction and the reaction force.However according to "Applied Mechanics by
Hannah & Hillier this is only true for average loads because the area of true contact is proportional to the load applied and is almost independent of the apparent area of contact. Sadly they don't define what they class as an average load, however in all my experience of design, friction as always been treated as defined earlier that it is independent of apparent contact area.

desertfox

 

[GregLocock]

In general, for engineering type materials, if you increase the surface area you will increase the apparent coefficient of /static/ friction.

On the other hand I can think of one obvious case where reducing the surface area reduces the apparent coefficient of /dynamic/ friction.

The problem is that there are several different mechanisms at play, trying to capture them all in one single number is ridiculous.



Cheers

Greg Locock

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

 

[MintJulep]

Normal force and contact area define an average contact pressure. However in real world materials, the contact pressure is almost always nonuniform.

If you break the contact surface into infinitesimal elements you can cause mu to be just about anything you like depending upon the assumptions you use in your calculation.

This is why you should never rely on handbook values for mu when designing things that rely on friction. You need to test your application.

 

[corus]

In this particular case the area in contact is not the same as the 'average' area as perceived, and within that contact area the contact pressure varies considerably. Primarily it's the effect on the static coefficient I'd be interested in, but the effect on the dynamic value would be of interest too. The materials involved are ceramics with a relatively smooth surface.

At the macro level of the surfaces does localised yielding of peaks and troughs in the surface cause a non-linear effect to the value of the apparent coefficient of friction as the load increases?

Greg doesn't give any examples or references but I presume that the obvious case of reducing the apparent coefficient of dynamic friction with area is in the case of ice skating, ie. through melting at the ice surface, perhaps an example of macro changes at the surface referred to above.

Any references would be useful.

corus

 

[GregLocock]

How about a flat knife in butter? flat side to the butter is very draggy, compared with sharp side down.

The references I have are concerned with tires and road surfaces, probably not very helpful, here's the best: John Dixon : Tires, Suspension, and Handling

Cheers

Greg Locock

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

 

[BigInch]

Greglocock,

The butter effect is viscosity related, which is not to say it doesn't involve friction, but it is a friction of a different... "flavor" shall we say?

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[GregLocock]

Sure, and that illustrates my point. There are at least 4 mechanisms I can think of that generate "friction", hence the difficulty with assigning a one size fits all factor to them.

The worst case I remember was polyurethane tubing against itself. Depending on the lubrication ( a mixture of seawater and kerosene in arbitrary proportions) and other grot that was hanging around on the deck, I got a factor of about 3 in stiction, maybe 1.5 in sliding contact. All figures from memory.

Cheers

Greg Locock

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

 

[vonlueke]

corus: For dry, unlubricated surfaces, the following applies. For low surface pressure, the coefficient of friction (COF) is approximately independent of contact area, over wide limits, regardless of whether the contact pressure is uniform or nonuniform. There is always elastic and plastic deformation of asperities (peaks), at least to some extent; but at low surface pressure, friction is mainly a shearing of the surface film. For very high surface pressure, the surface oxide apparently becomes ineffective, or is pushed aside, and the parent materials begin to cold weld (gall), such that the COF quickly increases. If this continues, the surfaces seize.

Surface roughness surprisingly has relatively little effect on COF (provided the surfaces are not so extremely rough that you enter the realm of what could be termed visible mechanical interlocking, instead of friction). COF is highly sensitive to what the parent materials are made of (intermolecular electromagnetic attraction), and highly sensitive to surface films and oxides.

 

[25362]

Engineering Tribology by Stachowiak and Batchelor, Chapter 16: Wear and Friction of Ceramics. Tribology Series, 24. Elsevier.

 

[FACS]

"Is the force required to slide apart two surfaces related only to the reaction force and coefficient of friction or does the amount of contact surface area play a part perhaps by 'ploughing' effects or by elastic deformation of the surfaces?
corus"

Everyone seems to be using "Pressure" and "force" as if they are the same thing.

Under constant load, the friction does not increase or decrease as the area increases or decrease.

The frictional coefficient will always remain the same.

The required force to move the load will only change if the load on the surface area changes.

Deforming the surface, as a result of the load/friction damage, will change the coefficient of friction and thus change the required force to move.

Charlie

http://www.facsco.com

 

[electricpete]

FACS - scroll down to the bottom of page 1 for the Stribeck curve:

http://www.me.utexas.edu/~bryant/courses/me383s/DownloadFiles/LectureNotes/BoundaryLubrication.pdf

Wouldn't you agree this represents friction coefficient changing as a function of pressure (among other things).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[corus]

Just to add to previous comments, this particular application involves ceramic type materials at very high temperatures so the comments regarding lubrication/surface films/oxides/welding in relation to pressure are very relevant with regard to chemical releases at the surfaces. I'll refer to the book recommended by 25362 to see if there is more information there.



corus

 

[EnglishMuffin]

For rigid materials, friction force is often said to be largely independent of the apparent contact surface area. Friction force is mainly dependent on the actual contact area, which is usually much less, and is heavily influenced by the normal force, having little relationship with the apparent area of contact. But for other materials such as elastomers, for example, this is not so true, perhaps because the actual and apparent contact areas are becoming similar (they also change greatly with time and temperature). Everyone knows that large tires have more grip than small ones. And Teflon behaves strangely too - I recently made some special hydraulic cylinders with all Teflon-to-metal contact and sealing surfaces, and the friction force was considerable, but totally independent of hydraulic pressure up to 2000 psi.

 

[blfarrar]

The coefficiaent is independant of surface area, the actual efficiency of device is dependant on the surface area - hence bigger brakes on faster cars.

Bruce L Farrar.
Works Engineering Manager
Marshalls Mono PLC.Brookfoot Works.
Halifax W.Yorks UK

 

[corus]

I would describe efficiency as the ratio of the force required to overcome friction in relation to the applied force, ie. the coefficient of friction. Whether this is due to other physical phenomena occuring is unsure, but it does seem that area does play a part whether it's actual or apparent surfaces in contact.

In this application I've found that a change in temperature has a significant affect on friction. It's well known that two surfaces in frction will generate heat. Can an increase in contact area between the two surfaces (or vice versa), for the same applied reaction force, affect the temperature increase between two sliding surfaces?

corus

 

[EnglishMuffin]

I did not notice the two posts after my last one until now. Even today, friction is a controversial subject, but the following quotes are taken from a relatively old book I possess entitled : "The Friction and Lubrication of Elastomers". I think they basically support the comments I have made in posts above.

Beginning of quote :

The classic laws of friction as they evolved from the early work of da Vinci, Amonton and Coulomb may be summarized as follows :

1. Friction force is proportional to load

2. Coefficient of friction is independent of apparent contact area

3. Static coefficient is greater than the kinetic coefficient

4. Coefficient of friction is independent of sliding speed

...... the classic laws have survived the years ....... most of them have been found to be incorrect.

The first law is correct except at high pressure when the actual contact area approaches the apparent area in magnitude. (emphasis mine)

The second law appears to be valid only for materials possessing a definite yield point such as metals and it does not apply to elastic and viscoelastic materials (such as rubber). (emphasis not mine)

The third law is not obeyed by any viscoelastic material - indeed, a controversy exists today as to whether viscoelastic materials possess any coefficient of friction at all. (emphasis not mine)

The fourth law is not valid for any material, and it is now well establihed that elastomer friction exhibits distinct viscoelastic properties. (emphasis not mine)

End of quote.

Personally, I think the concept of a "coefficient of friction" is a somewhat misleading and outdated one in many cases, if applied without broader knowlege. But like many other historical yet admittedly insightful concepts, it will probably never die.

 

[zifos]

Probably enough technical info so I'm going to not be technical.

I remember when I first heard that firction was independant of contact area, I couldn't believe it. I grew up around cars and I know that bigger tires = more traction.

Then I came to the realization (or I made it up to help justify) that "traction" is different then "friction." Friction, in my definition, is based on two "smooth" surfaces. Tires have tread lugs and asphalt has relatively large (macroscopic) bumps that dig into the rubber and catch on the edges of the lugs, thus a worn tire (no/rounded lugs) has less traction. Tires rely on surfaces interacting with perpendicular components (not unlike pushing against a wall). Hold your hands flat and rub them together, then interlock your fingers and try to rub them together.

Probably not a very good technical definition and the line between friction and traction may be somewhat ethereal but it helps me feel better about friction.

Also I believe brakes are larger on some cars to help deal with cooling and wear. For heavier cars with more energy to dissipate, it heats the brakes less if you spread it out over larger areas. For a given friction force the normal force is the same but if you spread it out over a larger surface area you have a smaller pressure which means less heat and wear. Breaks are a good example of friction because contact surfaces are quite smooth. i.e. machined steel against compressed asbestos with binders and what not.

 

[EdDanzer]

Coefficient of friction is dependant on normal force until it becomes a bearing or exceeds the contact force exceeds the yield strength of the material.
Examples, when a tire spins there is material being torn from the surface among other things.
Brakes require that heat be dissipated before the materials fail.
In most cases it seems the two materials are required to slide enough to qualify as plain bearings. This when design becomes difficult, as the structural integrity of the parts, the projected area pressure, velocity and lubrication have a major affect on wear/life. The experimentation I’ve done so far qualifies coefficient of friction is independent of contact area, but since the two parts must slide for a reasonable distance, the heat and P/V become the primary design constraints. Most of the published material P/V values are suspect.