Consider the case of objects rolling down an inclined plane at different speeds depending on their moment of inertia. Some basic equations are shown in this screenshot from Michel van Biezen. He is considering the specific case of a solid cylinder but my question is more general:

Physics - Application of the Moment of Inertia

Let's say we are comparing a metal solid sphere with a shell of the same radius made with the same metal. We know from demonstrations that the objects do not have the same speed as they move down the plane and don't arrive at the bottom at the same time.

Focusing on the equation of motion along the surface of the plane,

g sinπœƒ - gπœ‡ cosπœƒ = a

we see that for the acceleration to be different for the two objects the coefficient of friction, πœ‡, has to be different. If πœ‡ were the same then both objects's center of mass would reach the bottom of the plane at the same time.

So that leads me to ask why is πœ‡ different for a solid sphere and a shell, as it appears it must be?

I have looked at What factors determine the coefficient of friction? and don't think it answers my question. In the case I'm asking about, the materials of the sphere and shell are the same, as are the total mass and radius.

  • $\begingroup$ I have to questions: (1) What is a "solid shell"? Is it a solid cylinder? (2) By "focussing" on the linear motion, you are missing the key point: the rotation. It's an equation which does not describe the physical phenomena. So why should we be able to deduce anything from this equation? $\endgroup$
    – Semoi
    Commented Jun 12, 2020 at 16:48
  • $\begingroup$ Why does your link lead to a screen shot? $\endgroup$
    – Gert
    Commented Jun 12, 2020 at 17:09
  • $\begingroup$ I edited my post: replaced "solid shell" with "shell" and hopefully fixed the link. As for the comment on the equation - it is a valid equation describing the motion of the center of mass of an object based on the net force on the object. $\endgroup$ Commented Jun 12, 2020 at 18:27
  • $\begingroup$ So let's consider the case of a solid metal sphere and a shell of the same radius and metal but smaller mass as I think in all equations the mass drops out. How does πœ‡ "know" what's inside each object in order to have a different value, as it appears it must. $\endgroup$ Commented Jun 12, 2020 at 18:49
  • $\begingroup$ The equation you're asking about applies to objects sliding down the wedge without rolling. The coefficient of friction cannot change unless the objects are made of different materials. $\endgroup$
    – Triatticus
    Commented Jun 12, 2020 at 22:45

2 Answers 2


So that leads me to ask why is $\mu$ different for a solid sphere and a shell, as it appears it must be?

The $\mu$ don't have to be different or the same. Depending on the inclination $\theta$ only a part of the potentially available friction force is used to prevent sliding. Assuming roughness is sufficient, neither rolling objects will start slipping, even if the friction force will in both cases be smaller than the potential friction force.

Consider the simpler case: a block that might slide down an incline:


For a stationary block the net force would appear to be:

$$mg\sin\theta-\mu mg\cos\theta=0$$

But of course it cannot be negative! That would mean the block would start moving to the left!

In reality, only the useful part of the friction force is used:

$$F_f \leq mg\sin\theta$$

Assuming there's enough friction to prevent slipping, then inertia comes into play. As potential energy $U$ is converted to kinetic energy $K$, we can write:

$$\Delta U=\Delta K$$ $$mgh=\Delta K_{trans}+\Delta K_{rot}$$ $$mgh=\frac12 mv^2+\frac12 \gamma mv^2$$ where $\gamma$ is a factor, dependent on the inertial moment $I$.

We get:


For a full cylinder $\gamma$ is higher than for a shell-type cylinder, so $v$ is higher for the latter case.

  • $\begingroup$ If the πœ‡ are not different, then both objects would have the same equation of motion of their centers of mass along the plane and arrive at the bottom at the same time, no? $\endgroup$ Commented Jun 12, 2020 at 18:46
  • $\begingroup$ No. See my edit, below the page break. $\endgroup$
    – Gert
    Commented Jun 13, 2020 at 5:49
  • $\begingroup$ I don't question what you wrote nor the fact that objects with different moments of inertia roll with different accelerations/speeds. I'm just trying to reconcile the implication of the eq. I wrote in my original post that the πœ‡'s must therefore be different. There is no other parameter in that eq. to distinguish one object from another. $\endgroup$ Commented Jun 13, 2020 at 12:20
  • $\begingroup$ Your E.o.M. $g\sin\theta-\mu g \cos\theta=a$ is not correct (as I outlined in my answer) because it offers the possibility of $a<0$!!, high $\mu$. Not all friction force $F_f=\mu g\cos\theta$ is 'used'. $\endgroup$
    – Gert
    Commented Jun 13, 2020 at 14:30
  • $\begingroup$ This is why it's inertia that determines the velocity (and angular velocity), not the coefficient of friction $\mu$. $\endgroup$
    – Gert
    Commented Jun 13, 2020 at 14:32

While Gert's answer implies this, your main issue is in assuming that static friction is equal to $\mu N$ here. For static friction, $\mu N$ is the maximum value static friction can take, but generally static friction will be less than this value.

Therefore, all we can say for the net force acting on the object along the incline is $$ma=mg\sin\theta-F_s$$

And the net torque acting on the object must be $$I\alpha=RF_s$$

Noting that for simple rolling objects $I=\gamma mR^2$, and for rolling without slippping $a=\alpha R$, one can show that

$$F_s=\frac{\gamma}{1+\gamma}\cdot mg\sin\theta$$

So as we can see, we do get different static friction forces for objects with different moments of inertia, even if they are made of the same material on the same incline.

If you do want to look at slipping conditions, then we see what $\theta=\theta_\text{slip}$ needs to be such that slipping occurs, i.e. where $F_s=\mu N=\mu mg\cos\theta$. This gives us $$\theta_\text{slip}=\tan^{-1}\left(\frac{1+\gamma}{\gamma}\cdot\mu\right)$$

So, we also see that the angle of slippage depends on both $\mu$ and $\gamma$.

  • $\begingroup$ With your answer and Gert's it's starting to make sense. The confusion was due to the use of πœ‡ in the above referenced video. As you did, he should have just kept the frictional force as 𝐹𝑠 and not used πœ‡. Doing so ends up with the same answer he gets, a = 2/3 gsinπœƒ. $\endgroup$ Commented Jun 13, 2020 at 19:08
  • $\begingroup$ @Not_Einstein If the objects were slipping with the same $\mu_k$ then they would have the same linear acceleration. But then they would have different angular accelerations. $\endgroup$ Commented Jun 13, 2020 at 19:14

Your Answer

By clicking β€œPost Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.