# The Naked Scientists Forum

### Author Topic: Is the coefficient of sliding friction constant, or not?  (Read 5302 times)

#### Geezer

• Neilep Level Member
• Posts: 8328
• "Vive la résistance!"
##### Is the coefficient of sliding friction constant, or not?
« on: 04/03/2012 20:13:16 »
I was always taught that the coefficient of sliding friction remains constant regardless of speed, or at least until so much heat is generated that surfaces start to melt and act as a lubricant.

I've found data on gears (worm gears more specifically) that indicate the coefficient of sliding friction (bronze on hard steel) actually reduces with greater sliding speeds. Presumably this has been confirmed by lots of measurements over a large timespan.

Any suggestions as to why this might be? Is it a fluid dynamic effect associated with lubricant?

#### yor_on

• Naked Science Forum GOD!
• Posts: 11999
• Thanked: 4 times
• (Ah, yes:) *a table is always good to hide under*
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #1 on: 04/03/2012 20:43:36 »
If it is correct I would ask what differs between a slower and a faster 'gliding speed'. Humidity (lubrication)? heat? and what more? The extent of 'touching' at each point as it slide through?

#### yor_on

• Naked Science Forum GOD!
• Posts: 11999
• Thanked: 4 times
• (Ah, yes:) *a table is always good to hide under*
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #2 on: 04/03/2012 20:47:29 »
if it was two rails laid against each other I would expect that the friction would grow with what speed they glided against each other, as well as what surface area that touched. But you're not referring to that kind of friction, right? You are thinking 'gears' that constantly are in touch but with each point meeting only at a instant, well, more or less?
==

But humidity/some sort of lubrication must be the thing, and possibly that each 'point' touch at shorter intervals?
Well?? Maybe :)

Ah, and heat.. Wouldn't that be able to act as a 'lubrication' between the layers meeting?
And that should also mean the 'pressure' created per time interval. It should be largest as you start up but as you build a momentum it should go down, unless you accelerate?
« Last Edit: 04/03/2012 21:00:27 by yor_on »

#### yor_on

• Naked Science Forum GOD!
• Posts: 11999
• Thanked: 4 times
• (Ah, yes:) *a table is always good to hide under*
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #3 on: 04/03/2012 21:04:49 »
And thinking that way the friction between the rails should go down until they interconnect, melting as you said. But wild guesses :)
Or you could state it as the friction doesn't go down, but the momentum minimize its effects as it builds up.
And then there is the question if heat can 'lubricate' between the rails too.
« Last Edit: 04/03/2012 21:07:38 by yor_on »

#### graham.d

• Neilep Level Member
• Posts: 2208
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #4 on: 04/03/2012 21:41:22 »
I suspect the friction would be only approximately constant anyway because of the practicality of trying to get perfect surfaces. My guess is that lubrication would definitely affect the friction vs speed because of a degree of compressibility and variation with applied pressure and temperature and also how the lubricant was flowing (lamina or turbulent). Wouldn't the friction be more dependent on the lubricant than the materials up to a point. I guess the bronze/steel combination allows the bronze to deform and/or wear until effectively run-in. Usually this involves deformation to align of the surface granularity.

These are all guesses though!

#### Geezer

• Neilep Level Member
• Posts: 8328
• "Vive la résistance!"
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #5 on: 04/03/2012 22:02:30 »
Some of the confusion may come from "friction" and "coefficient of friction". The friction (heat) produced by sliding is linear with speed as long as the coefficient remains constant.

A constant coefficient means that the force required to overcome the friction between two sliding surfaces is constant as long as the pressure between the two surfaces remains constant, regardless of the sliding speed.

The brakes on your car are a good example. For a given brake pressure, the force that is (negatively) accelerating your car is constant. Although, when the brakes get really hot, the force reduces because the surface conditions of the materials change quite a bit (at least I think that's how it works).

Friction is a notoriously trixy subject. The Real World frequently ignores the theory.

This is where I found the information. It's in a table about half-way down.

http://www.roymech.co.uk/Useful_Tables/Drive/Worm_Gears.html

I suspect it's because, at high speeds, the lubricant "can't get out of the way" fast enough, so it forms some sort of fluid boundary.

#### yor_on

• Naked Science Forum GOD!
• Posts: 11999
• Thanked: 4 times
• (Ah, yes:) *a table is always good to hide under*
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #6 on: 04/03/2012 22:18:17 »
Could it be the momentum reducing the friction, practically I mean?
The higher your momentum the more 'force' pushing into the direction of motion?

Although then you would still be right as it's not the friction 'going down' but the momentum going up hiding the friction?

You say "the force required to overcome the friction between two sliding surfaces is constant as long as the pressure between the two surfaces remains constant"

But when you've started to move you get a momentum. How exactly do they define that friction there? From practical tests or theory.
Weird.
=

I was thinking of this one when first reading you http://en.wikipedia.org/wiki/File:Face_Worm_Gear.jpg
« Last Edit: 04/03/2012 22:21:13 by yor_on »

#### Geezer

• Neilep Level Member
• Posts: 8328
• "Vive la résistance!"
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #7 on: 05/03/2012 00:46:40 »

You say "the force required to overcome the friction between two sliding surfaces is constant as long as the pressure between the two surfaces remains constant"

But when you've started to move you get a momentum. How exactly do they define that friction there? From practical tests or theory.

I think it's always determined empirically. If you look up different sources for the coefficients for various combinations of materials, there is always a pretty wide range of possible values. The science is very complicated. There doesn't even seem to be much agreement about why skates slide on ice (but let's not start that one again!)

You might assume that the smoother the surfaces are, the less friction there is. Apparently, that's not necessarily true either.

Here's an interesting way of getting around friction.

http://en.wikipedia.org/wiki/Foil_bearing

#### yor_on

• Naked Science Forum GOD!
• Posts: 11999
• Thanked: 4 times
• (Ah, yes:) *a table is always good to hide under*
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #8 on: 05/03/2012 01:33:51 »
Now, that was one smart Geezer :)
Very innovative.

#### peppercorn

• Neilep Level Member
• Posts: 1466
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #9 on: 05/03/2012 14:07:25 »
I suspect it's because, at high speeds, the lubricant "can't get out of the way" fast enough, so it forms some sort of fluid boundary.

Perhaps the topology of the worm gear is more akin to the foil bearing than it might first appear. Maybe something like fluid cavitation is playing a role as the speed increases, causing lubricant to be thinned out between the moving surfaces.

Foil bearings remind me of being taught about the Winchester disk drive design at college - The read/write heads 'flying' a micron above the platters on a tiny bubble of air.

#### Geezer

• Neilep Level Member
• Posts: 8328
• "Vive la résistance!"
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #10 on: 06/03/2012 09:02:55 »
I suspect it's because, at high speeds, the lubricant "can't get out of the way" fast enough, so it forms some sort of fluid boundary.

Perhaps the topology of the worm gear is more akin to the foil bearing than it might first appear. Maybe something like fluid cavitation is playing a role as the speed increases, causing lubricant to be thinned out between the moving surfaces.

Foil bearings remind me of being taught about the Winchester disk drive design at college - The read/write heads 'flying' a micron above the platters on a tiny bubble of air.

I think it must be something like that. The speeds are a lot lower, but oil is lot more viscous than air.

#### damocles

• Hero Member
• Posts: 756
• Thanked: 1 times
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #11 on: 08/03/2012 03:31:41 »
Dynamic friction is only approximately constant at different sliding speeds. The approximation is excellent at relatively low sliding speeds, but the coefficient of friction falls fairly rapidly with speed when you get past a particular point. The reason probably has to do with the fact that in a friction situation adsorbed layers of air and water are involved, and there is also microscopic melting of some of the softer material, so that some form of lubrication will take over.

#### Geezer

• Neilep Level Member
• Posts: 8328
• "Vive la résistance!"
##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #12 on: 08/03/2012 19:08:00 »
Thanks Damocles!

(BTW, you really need to do something about that thing dangling over your head. Somebody could get hurt.)

#### The Naked Scientists Forum

##### Re: Is the coefficient of sliding friction constant, or not?
« Reply #12 on: 08/03/2012 19:08:00 »