# B Coefficient of friction of various materials not listed in standard references

1. May 15, 2017

### R_Rose

I'm interested in finding out what the coefficient of friction would be for a few different materials such as:

Copper on aluminum
Copper on graphite
graphite on aluminum

I'm not sure if it matters if the order is reversed such as Aluminum on copper vs copper on aluminum..

also I've seen that there is a difference between static and kinetic which I'm not sure what the difference is - obviously kinetic means moving but I'm not sure which I'd need b/c the material will be static until it starts moving - then it will speed up and I'm trying to calculate how quickly it may speed up with respect to the friction & energy input. This page shows how much difference there is between some static and kinetic coefficients - http://www.physlink.com/Reference/FrictionCoefficients.cfm

This page gives a lot of values for static friction but it doesn't have the ones I listed above - http://www.engineeringtoolbox.com/friction-coefficients-d_778.html

Is there any way I can find out either static or kinetic of those I listed?

2. May 16, 2017

### sophiecentaur

I spent the usual 'PF vs other things' allocated time and. like you, I could find nothing in the usual lists of friction coefficients. Surprising but perhaps not, when you think of the number of combinations of materials possible.
Why have you chosen those particular combinations? Is there some particular application?
Perhaps it might be best to forget about it and quote some different, well known pairs of materials. Several of the Google hits about friction coefficient make the point that the details of the measurement method can affect the result. So you would probably need more than one independent source to get a reliable answer. I get the feeling that many of the published values are pinched from common sources.

3. May 16, 2017

### R_Rose

I agree with you about the numbers probably being taken from a common source or a chain of references leading back to one source - THIS is one thing which I fear the Internet has allowed to happen all too often and it could lead to very bad consequences.

As for the materials listed, I need to know those specifically as they are the only materials available, moderately priced, with the physical characteristics and electrical conductivity/resistivity which would work in my application.

Basically I'm looking for low friction with high electrical conductivity and is easily affordable.

What I find odd about graphite and copper (or graphite & aluminum) not being listed is that they are often used in motors with the graphite being the brushes for the rotor. I would think that knowing the friction of these two would be imperative to know the heat generated, the wear time and the reduction of power to the motor.

In my application, there may be a point where the two substances are static but they will most likely be constantly accelerating up to a point where they reach peak velocity and then sustained at that point - but it is a long linear acceleration period - think of a motor starting from 10rpm going up to 500,000 rpm with an increase of 300rpm per minute until 500k is reached. I need to know the frictions of the materials to see how much heat will be generated, energy lost, etc.

4. May 16, 2017

### Nidum

5. May 16, 2017

### R_Rose

I was wondering if the values could be estimated from comparing similar materials such as:
from table: http://www.engineeringtoolbox.com/friction-coefficients-d_778.html
aluminium - mild steel = .61
Copper - Mild Steel = .53, .36 (IDK why 2 values are given...)
Graphite - steel = .1
steel - steel = .5 - .8

IDK if any values can be extrapolated from these values.

The problem with needing the different values is that as I said, they need to be very electronically conductive and operate within specific electromagnetism properties (eddy currents, flux, becoming slightly magnetic, etc)

Alternatively, I'd be able to do some testing as I have Cu, Al and graphite and I guess I could make an area 1"^2 and place a weight on it and pull it and see how much force is required. If I place 1000g on the area, and it requires 250g of force to pull, then it should be .25, is that correct? There should be an amount to break free (static) and then amount to keep moving - kinetic. My only worry is if the materials I have will be adequate for the test.

I have:
Copper flashing (thin sheet) as well as various copper pipe which can be flattened as well as a round copper plug (1" diam x 1" high CPU heat sink center)
Al - tons of heat sinks, 6160 bar stock, flashing, etc.
Graphite - 3/4" x 6" x 1/8" rods - don't know what "kind" of graphite.

I'm wondering if I should sand the contact surface with 1000-4000 grit sand paper or try it with non-sanded and then different states like sanded with 100, 220, 320, 400, 800, 1000, 2000, 4000 and see if it makes much difference.

6. May 16, 2017

### Nidum

For a real world project a friction test would ideally be conducted in a way that replicated the conditions that will exist in the actual machine being designed .

7. May 17, 2017

### sophiecentaur

In my reading around, I came across numerous links about composites for low friction. But that could be an expensive route for you.

8. May 17, 2017

### Nidum

Those sort of numbers are not realistic for practical devices .

Please tell us more about what you are actually trying to do .