22 replies to this topic

[gbaker]

Is the coefficient of friction between two sliding surfaces a function of velocity?

(No, this has nothing to do with H&N restraints.)

[Moon Tricky]

No. At least not according to a physicist.
It depends (linearly) on the normal force between the two surfaces (i.e. how hard they are pushed together), and that could conceivably vary with velocity (for instance a car with downforce will experience more friction at high speeds because downforce increases with speed and friction increases with downforce).

[rms]

There is no generic answer, you must know the specifics.

[gbaker]

Let's take the example of a race car sliding through a corner. Forget all the vector stuff. The question, in this example, is whether the dynamic coefficient of friction between the tires and track surface is constant. If not, how does it vary with relative velocity?

[Moon Tricky]

It should be constant, all other things being equal. Although the direction of the force will depend on how fast the wheels are spinning.

[Ben]

Originally posted by Moon Tricky
It should be constant, all other things being equal. Although the direction of the force will depend on how fast the wheels are spinning.

It's not a constant. Simple physics books will tell you it's a constant and independent of normal load.

In practice it depends on temperature, pressure, load, sliding speed, track surface roughness, compound properties, and lots of other things.

In broad terms, the relationship with sliding speed is that friction drops because of surface temperature increases, although you can mitigate this to a degree with different compound formulations because you can't eliminate surface temperature rise so you have to try and deal with it.

Ben

[Moon Tricky]

Originally posted by Ben


It's not a constant. Simple physics books will tell you it's a constant and independent of normal load.

In practice it depends on temperature, pressure, load, sliding speed, track surface roughness, compound properties, and lots of other things.

That's why I said "all other things being equal". Although I hadn't taken into account the fact that the temperature will rise during the course of the sliding action. And how fast the temperature rises will depend on the relative velocity of the two surfaces, as Work Done is the integral of force over distance, or, in simple terms, Power=Force × Velocity. How that will affect the coefficient of friction, I have no idea.

The roughness of the track and the properties of the rubber in the tyre are a part of the definition of the coefficient of friction. Neither of those things are going to change, although of course you could drive in a different track with different tyres, in which case the coefficient will be different.

Load of course affects the area of tyre that is in contact with the road.

[J. Edlund]

In general the friction coefficient (as dependand on micro welding) is independant on load and speed. Basicly the friction coefficient is a material/material property where the environment also have effect (mainly due to oxygen, and the production of oxides on the surfaces).

Tires are a bit tricky though as they have adhesive properties like glue or tape. This means that the friction coefficient of tires won't be constant as with for example metals.

[Ben]

Originally posted by Moon Tricky


The roughness of the track and the properties of the rubber in the tyre are a part of the definition of the coefficient of friction. Neither of those things are going to change, although of course you could drive in a different track with different tyres, in which case the coefficient will be different.

Believe me the track can change. Rubbering in anyone... Tends to be worse on low grip rough circuits BTW. Also the ambient (and therefore the track) temperature can also affect things.

If you have no idea how temperature rise due to sliding can affect grip, how can you possibly state that it's a "constant, all things being equal" when clearly things are never equal when considering the friction of race tyres.

Ben

[Moon Tricky]

Originally posted by Ben

Believe me the track can change. Rubbering in anyone...

Yes ok the track can change... we're getting into a silly pedantic argument now. Of course it changes OVER TIME with people constantly driving over it. But that is really an entirely different question. How rough a track is does not depend on how fast you slide over it on any one specific occasion.

And I can state that it's "constant, all things being equal" because... I'm a physicist! LOL! There you go. I said it.

[gbaker]

Okay, you guys have been helpful. I believe we can agree it is constant with the exception of some odd circumstance which affects the material interface. Velocities so high that the tire begins to liquefy due to heat might qualify. A state change, if you will.

Thanks.

[Greg Locock]

As a practical example, the Fx/Fz developed in an ABS stop doesn't change /much/ during one straight line braking event from say 70 mph, as measured by a wheel force transducer. That's with production tires on good coarse asphalt.

But the grip does change substantially depending on the slip ratio, on a dry track, falling off by say 20% as you lock the wheel completely.

I try to avoid mu when it comes to tires, it make the armchair enthusiasts think that the physicists were right!

[gbaker]

Originally posted by Greg Locock
I try to avoid mu when it comes to tires, it make the armchair enthusiasts think that the physicists were right!

Well, we can't have that.;)

But, absent the state change of lock-up, I'm concluding the coefficient is constant.

Yea, nea?

[Ben]

Originally posted by gbaker

Well, we can't have that.;)

But, absent the state change of lock-up, I'm concluding the coefficient is constant.

Yea, nea?

It's probably pretty close on a road tyre. But soft race stuff is another matter.

Ben

[Fat Boy]

Originally posted by J. Edlund
Tires are a bit tricky though as they have adhesive properties like glue or tape. This means that the friction coefficient of tires won't be constant as with for example metals.

Bingo. Racing tires don't follow the newtonian friction model. The 2 main grip mechanisms of a racing tire is due to the chemical adhesive grip in the tire (thing tape) and the mechanical interaction of the rubber filling voids in the pavement aggregate and then being sheared off (think velcro).

Both of these mechanisms are significantly different than the newtonia physics model of 2 smooth sliding against each other (did that sound sexual to anyone else?). Anyway, you get the point.

[Joe Bosworth]

FB

With your last you said, "Both of these mechanisms are significantly different than the newtonia physics model of 2 smooth sliding against each other (did that sound sexual to anyone else?). Anyway, you get the point".

Smooth(ness) is all relative and I believe that you will find that the coefficient of friction for tyres is found exactly as with other "smoother" bodies.

Even these "'smoother" bodies have Cf that change with temperature just like tyres. Think in terms of smooth brake pucks against smooth brake disks and they don't work very well while cold or very hot depending on the materials.

Basic physics is very simple until people try to complicate it.

Regards

[Bloggsworth]

All you need to know about friction ( well, almost)

http://www.saburchil...pters/0008.html

[Catalina Park]

Slightly OT, One of the commentators for the Australian V8 racing (a former driver) always claims that when a car runs off the track onto grass it will not slow down at all and that it will accelerate until it hits something. Must be some strange force of physics that I never learned.

[Moon Tricky]

Originally posted by Joe Bosworth

Basic physics is very simple until people try to complicate it.

Basic physics is very simple, it just doesn't always correspond with reality.

[Bloggsworth]

Originally posted by Catalina Park
Slightly OT, One of the commentators for the Australian V8 racing (a former driver) always claims that when a car runs off the track onto grass it will not slow down at all and that it will accelerate until it hits something. Must be some strange force of physics that I never learned.

Its called the Coefficient of Fear

[J. Edlund]

Originally posted by Joe Bosworth
FB

With your last you said, "Both of these mechanisms are significantly different than the newtonia physics model of 2 smooth sliding against each other (did that sound sexual to anyone else?). Anyway, you get the point".

Smooth(ness) is all relative and I believe that you will find that the coefficient of friction for tyres is found exactly as with other "smoother" bodies.

Even these "'smoother" bodies have Cf that change with temperature just like tyres. Think in terms of smooth brake pucks against smooth brake disks and they don't work very well while cold or very hot depending on the materials.

Basic physics is very simple until people try to complicate it.

Regards

No, as already pointed out tires ARE different, and it do get a lot more complicated when you go from simple physics about friction to actually look what mechanisms are causing that friction.

Temperature dependacy is one thing, so is environmental dependacy (friction coefficients can for example be higher in vacuum) but the friction generated by tires is strongly load dependant. Increase the load on the tire and the friction coefficient goes down (this is why we want big fat tires on racing vehicles). Note that the friction coefficient between to metals also can be dependant on load, but this is different, as it is related to the breakdown of the friction reducing oxide film that cover most metals at a certain load.

Friction between two smooth surfaces, such as two pieces of metal is generally dependant on microwelding at small "contact zones". The area of these zones are dependant on the youngs module of the softer piece and the load. Increase the load or reduce the youngs module and you have increased the area of the contact zones. At these contact zones welding and shearing will occur. Given the knowledge of the contact area and and the shear strength of the material we can in theory calculate the friction coefficient for a material combination. Notice that we need to know the material properties at a microscale which can be different compared to the properties of the bulk material.

If the surfaces are rough additional shearing will occur as the peaks of one surface try to shear though the peaks of the other surface. This will typically increase the friction coefficient somewhat.

But tires are different, they won't follow Coulomb friction, and they can provide very high friction coefficients at low loads. Friction coefficients up to 5 isn't impossible under certain conditions. The reason for this is the adhesive properties of the tires. The "adhesive friction" isn't dependant on the common microwelding, but instead mechanical and chemical mechanisms.

[Paolo]

Coefficient of friction is almost fixed.
Coefficient of adherence is not.

Coefficient of friction is a local property.
It has basically two values: a standing value, good when there is no sliding between surface, and a sliding value.
Transition between these two is not totally abrupt, but it is quite sharp anyway.

When a wheel is moving, it will touch the road through an area of a certain extension, which is called contact patch. For several reasons, too long to be discussed now, speeds inside the contact patch are different from point to point.
Some areas of the contact patch will not move relative to the road surface; they will then experience the standing friction coefficient.
Other areas will slide, and will esperience the sliding friction coefficient.

The force a wheel is able to exert depends on the distribution of these two different coefficients and on the distribution of load in the contact patch.

Dividing this force by the total load you have a coefficient of adherence, which varies a lot with speed because speed causes changes in load distribution and in the distribution of friction coefficients inside the patch.

These phenomena are not exclusive of rubber: railway wheels will behave in exactly the same way.

[phantom II]

I was at Oshkosh last month. If you want to attend next years convention, pm me.

I did a course in Materials Science at Uni in the 60s and one of the things we did at a lab tut was to determine friction coefficients. We had a 10" dia. rotating drum that had various materials stuck to its surface. A lever was used to bring a test sample into contact with the drum at the same PSI as the other samples and the torque was then measured. There were materials even at that time that had the same drag as melting ice sliding on melting ice. It also took the same force to begin the drum rotating as it required to maintain a constant RPM. Pure rubber was the most intriguing substance. It required 20 times more force to get the drum rotating than it did to run it at a constant rpm which was about 200 rpm. As it got hotter the torque required to turn the drum got higher and higher as the rubber melted onto the drum. Each material pair had completely different results.
I actually remember something from college.

Originally posted by Paolo
Coefficient of friction is almost fixed.
Coefficient of adherence is not.

Coefficient of friction is a local property.
It has basically two values: a standing value, good when there is no sliding between surface, and a sliding value.
Transition between these two is not totally abrupt, but it is quite sharp anyway.