How can I replicate conditions on pin on disc test

[knight92]

Hello,

I need to replicate the conditions of a clutch on a pin on disc tribometer to measure the coefficient of friction. However the tribometer I have access to can't spin faster than 1000 RPM. The clutch will reach speeds from 3000 RPM to 9000 RPM. I calculated the normal force required on the pin area to replicate pressure conditions on the clutch faceing. Shall I increase or decrease my normal force to replicate conditions at 3000 RPM as my rig can only do 1000 RPM?

Thanks

 

[Kishan Majethia]

Hello,

I need to replicate the conditions of a clutch on a pin on disc tribometer to measure the coefficient of friction. However the tribometer I have access to can't spin faster than 1000 RPM. The clutch will reach speeds from 3000 RPM to 9000 RPM. I calculated the normal force required on the pin area to replicate pressure conditions on the clutch faceing. Shall I increase or decrease my normal force to replicate conditions at 3000 RPM as my rig can only do 1000 RPM?

Thanks

Can you please elaborate your question. As you are saying your rig can reach maximum speed of 1000 rpm so how you want to replicate?

 

[knight92]

Can you please elaborate your question. As you are saying your rig can reach maximum speed of 1000 rpm so how you want to replicate?

Friction coefficient depends on sliding velocity and sliding velocity depends on RPM right? So since I have no way of altering the sliding velocity can I alter the force is such a way that it replicates the friction coefficient at higher sliding velocities? Thanks

 

[mfb]

You want to change the mass to simulate friction forces at higher rpm?
Do you have a model that predicts an rpm-dependent coefficient of friction? And did you test how it behaves up to 1000 rpm?

 

[knight92]

You want to change the mass to simulate friction forces at higher rpm?
Do you have a model that predicts an rpm-dependent coefficient of friction? And did you test how it behaves up to 1000 rpm?

Well yes I want to change other factors to match the higher RPMs. I can go upto 5 kg in mass on the Pin on Disc tribometer.

No I don't have a model that predicts an RPM-dependent coefficient of friction as it is something I am trying to make.

Yes up to 1000 RPM it gives me a friction coefficient of 0.18 and as I increase the normal force this coefficient of friction decreases.

 

[Kishan Majethia]

Yes up to 1000 RPM it gives me a friction coefficient of 0.18 and as I increase the normal force this coefficient of friction decreases.

Does it really decrease with increase in normal force? As per me co-efficient of kinetic or static friction will increase with corresponding increase in normal force applied but ultimately it won't affect the co-efficient of friction which is a constant quantity for any two given surfaces.

Co-efficient of friction = F_lim/N

Where F_lim = Maximum frictional force applicable between surfaces
N = Normal force

(I suppose this is for static friction and am not perfectly sure about kinetic friction if anybody know's please suggest corrections.)

So with increase in normal force F_lim or the required force to rotate will also increase ultimately leaving co-efficient to constant value.

 

[knight92]

Does it really decrease with increase in normal force? As per me co-efficient of kinetic or static friction will increase with corresponding increase in normal force applied but ultimately it won't affect the co-efficient of friction which is a constant quantity for any two given surfaces.

Co-efficient of friction = F_lim/N

Where F_lim = Maximum frictional force applicable between surfaces
N = Normal force

(I suppose this is for static friction and am not perfectly sure about kinetic friction if anybody know's please suggest corrections.)

So with increase in normal force F_lim or the required force to rotate will also increase ultimately leaving co-efficient to constant value.

I thought that the coefficient of friction would be constant or increase with increase in normal force but my experiment shows different. Some research also shows that the coefficient of friction reduces as the normal force is increased.

See figure 5 on page 47:
http://www.ijens.org/vol 11 i 01/111701-6868 ijmme-ijens.pdf

 

[mfb]

The coefficient of friction / rpm curve looks surprisingly linear. If you can measure it up to 1000 rpm, it could be interesting to extrapolate it to larger values. If a linear fit is good in your measured range, if it looks similar to the graph in the pdf and if the extrapolated value still makes sense, it might be a reasonable estimate.

There are different things you could try to match - coefficient of friction, friction force (normal force * coefficient of friction) or energy dissipation (proportional to normal force * coefficient of friction * rpm). As you cannot match rpm you cannot get all right at the same time, so you'll have to see what you want to test.

 

[knight92]

The coefficient of friction / rpm curve looks surprisingly linear. If you can measure it up to 1000 rpm, it could be interesting to extrapolate it to larger values. If a linear fit is good in your measured range, if it looks similar to the graph in the pdf and if the extrapolated value still makes sense, it might be a reasonable estimate.

There are different things you could try to match - coefficient of friction, friction force (normal force * coefficient of friction) or energy dissipation (proportional to normal force * coefficient of friction * rpm). As you cannot match rpm you cannot get all right at the same time, so you'll have to see what you want to test.

I think I will try to match the energy dissipated.

Energy/Work Done = J or Nm (Because Force * Distance)

Using units and what you have told me I came up with an equation:

Energy/Work Done = COF * Normal Force * v * sample time = (N * m/s * s) = (N*m)
where v = radius of friction surfaces * Angular Velocity (RPM)
COF = Coefficient of Friction
My sample time will be 10 minutes (Why 10 minutes? I don't know I just randomly chose it)

Outer friction disk radius is 0.0725 m.
Inner friction disk radius is 0.056 m
Overall normal force applied on friction plate face is 372 N.
Area of friction plate = PI * (0.0725² - 0.056²) = 6.67 x 10^-3

My sample time will be 10 minutes

Energy Dissipated through friction in the clutch at 3000 RPM (314 rad/s) = COF * 372 * (0.0725*314) * (10*60) = 5.08 x 10⁶ * COF

Now I am calculating how much force should be altered to keep COF and energy at 3000 RPM equal to 1000 RPM (104 rad/s) on the friction disk:

5.08 x 10⁶ * COF = Force Altered * 104 * 0.0725 * 60 * 10 * COF

(COF cancel out)

Force Altered = 1115 N
Pressure applied on clutch friction disk using the force of 1115 N = 1115/6.67 x 10^-3 = 166417 Pa

Force required to replicate pressure on the pin on disk (pin of radius 0.003 m) = 166417 * (PI * 0.003²) = 4.7 N

This means I only need to attach a mass of 0.5 kg to the pin on disc machine and I will get an estimate of the COF at 3000 RPM without going above 1000 RPM on the tribometer.

The problem I see with this is that the disc on the pin on disc machine has a radius of 0.021 m which I haven't taken into account but does it matter?

Sorry for the long post but does it sound like I am going in the right direction? Thanks.

 

[mfb]

Keep it the same size and it should be fine. The actual relative velocity of the surfaces will vary over the area of the pin but that ratio is the same in both setups.

Another thought: can you move the pin outwards? That would give a higher velocity at the same rotation speed.

 

[knight92]

Keep it the same size and it should be fine. The actual relative velocity of the surfaces will vary over the area of the pin but that ratio is the same in both setups.

Another thought: can you move the pin outwards? That would give a higher velocity at the same rotation speed.

So do I not need to consider the 0.021 m radius of the little disc in the pin on disc tribometer?

And no, the position of the pin is fixed.

 

[mfb]

So do I not need to consider the 0.021 m radius of the little disc in the pin on disc tribometer?

If it is the same for every rpm value, I would not expect effects from the size of the pin.

 

[knight92]

If it is the same for every rpm value, I would not expect effects from the size of the pin.

Yes the pin and disc both have the constant diameter throughout the tests.