I am looking at simple harmonic oscillators. The maximum potential energy is equal to the maximum kinetic energy:

$k {x_{max}}^2 = m {v_{max}}^2 \rightarrow x_{max}=v_{max}\sqrt{\frac{m}{k}} = \frac{v_{max}}{\omega}$

Also, the time to find the average speed, calculate the total distance traveled over one oscillation by the total time it takes.

$v_{ave} = \frac{4 {x_{max}}}{\tau} = 4 {x_{max}} \omega$

Now if I plug in the first result into the second result I get:

$v_{ave} = 4 v_{max}$

But surely this is not right. The average speed is equal to 4 times the maximum? How can this be?

  • 3
    $\begingroup$ $1/ \tau = f = \omega /2 \pi $. $\endgroup$
    – Omar Nagib
    Jul 9, 2020 at 21:56
  • $\begingroup$ Depends on the units. If frequency is in Hertz instead of radians... In any case, this has no impact on the problem that I can see. $\endgroup$
    – Karlton
    Jul 9, 2020 at 22:40
  • $\begingroup$ It does have an impact. It gives the right answer, in which the average speed is less than the maximum speed. It doesn’t depend on the units. $\endgroup$
    – G. Smith
    Jul 10, 2020 at 5:42
  • $\begingroup$ The title should refer to average speed over a period as the average velocity of shm over a period is zero. $\endgroup$
    – Farcher
    Jul 14, 2020 at 20:28

1 Answer 1


You made a simple mistake converting time period to angular frequency: $\omega = \frac{2\pi}{T}$, so $\frac{1}{T} = \frac{W}{2\pi}$.

I will not work it through for you, but you should get the answers you are looking for.

If you are interested in working out max velocity and average velocity in general, i.e. not just a mass on a spring, then you could use the general SHO wave equation:

$$y = A\cos(wt).$$

Where $y$ is the displacement and $A$ is the amplitude (your $x_\text{max}$).

We can differentiate to find the velocity,

$$\frac{dy}{dt} = -Aw\sin(wt).$$

From which we can calculate the average velocity and maximum velocity.

The maximum velocity is easiest as we know the largest value of $\sin(wt)$ is one, so

$$v_\text{max} = \text{max } \left|\frac{dy}{dt}\right| = Aw.$$

Which is what you got: $v_\text{max} = x_\text{max}w$.

The average velocity is a little harder. The average value of a function $g(x)$ in the interval $[a,b]$ is defined as

$$\frac{1}{b-a}\int_a^bg(x) \, dx.$$

So we could take the interval to be one time period (because any more and it just repeats itself!), but the problem with this is that we will just get zero for the average velocity ($\frac{dy}{dt}$). This is because the average value of $\sin$ from $0$ to $2\pi$ is zero and $w$ just stretches this graph.

The mistake we have made is that we actually want to find the average value of the speed, $\left|\frac{dy}{dt}\right|$, given by the integral

$$\frac{1}{T}\int_0^T\left|\frac{dy}{dt}\right|\, dt = \frac{Aw}{T}\int_0^T|\sin(wt)|\, dt$$

There are two ways I can see of evaluating this, either write $|\sin (wt)|$ as $\sqrt{\sin^2(wt)}$ and use a double-angle substitution, or we can use the symmetry of the sine function, seen below.

Graph of sine and abs sine

From this symmetry, we can see that

$$\int_0^{2\pi}|\sin x| \, dx = 2\int_0^{\pi}\sin x \, dx$$.

Or in our case,

\begin{align} \frac{Aw}{T}\int_0^T|\sin(wt)|\, dt &= \frac{2Aw}{T}\int_0^{T/2} \sin(wt)\, dt \\ &= \left. -\frac{2A}{T}\cos(wt)\right|_0^{T/2} \\ &= -\frac{2A}{T}\left[\cos\left(\frac{wT}{2}\right) - \cos(0)\right] \\ &= -\frac{2A}{T}\left[\cos(\pi) - \cos(0)\right] \\ &= -\frac{2A}{T}\left[(-1) - 1\right] \\ &= \frac{4A}{T} = v_\text{ave} \end{align}

which is of course what you got with the simpler $\text{speed} = \frac{\text{distance}}{\text{time}}$ approach: $v_\text{ave} = \frac{4x_\text{max}}{T}$.

So although your inkling that the average speed shouldn't be greater than the max speed was correct and solved by the simple calculation error I pointed out at the start, I hope you can learn something extra from this answer in the form of an alternative, more mathematical approach, that you could have taken.

I will conclude with the final results.

\begin{align} v_\text{max} = Aw && v_\text{ave} = \frac{4A}{T} = \frac{2Aw}{\pi} \end{align}

  • 1
    $\begingroup$ Note that in step 1: $1/2 kx^2 =1/2 mv^2$ or $kx^2 =mv^2$ as stated by the OP. $\endgroup$
    – Omar Nagib
    Jul 9, 2020 at 22:04
  • $\begingroup$ Yes, the factors of 1/2 cancel... Also, I am free to consider non-angular frequency. But at any rate, this has NO impact on the conclusion... $\endgroup$
    – Karlton
    Jul 9, 2020 at 22:12
  • $\begingroup$ @OmarNagib I realised my mistake mid-edit, but thank you! $\endgroup$
    – Joe Iddon
    Jul 9, 2020 at 22:54
  • $\begingroup$ @Karlton Please see my answer, the result does come out as less. What do you mean by non-angular frequency? Although $\omega$ is called "angular frequency" it is merely a scaling factor; it maps the period of $\sin t$, which is $2\pi$, to the period of your oscillation, as $\sin(wt) = \sin(\frac{2\pi}{Tt})$ has period $T$. If you say $\omega = \frac{1}{T}$, as you seem to claim you can, then $\sqrt\frac{k}{m}$ is not equal to $\omega$ anymore! $\endgroup$
    – Joe Iddon
    Jul 9, 2020 at 23:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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