Here is a mathematical proof that any force $F(t)$, which affects a body, so that $\vec{F(t)} \cdot \vec{v(t)} = 0$, where $v(t)$ is its velocity cannot change the amount of this velocity.

Further, there is stated that $\vec{v(t)}$ itself cannot change, what I think is nonsense. Since:

$$\begin{pmatrix} 1 \\ 0 \\ 0 \end{pmatrix} + \begin{pmatrix} 0 \\ 1 \\ 0 \end{pmatrix} \times t = \begin{pmatrix} 1 \\ t \\ 0 \end{pmatrix} $$

But maybe I am just wrong. Now, I am further wondering in how far $\vec{F(t)} \cdot \vec{v(t)} = 0$ leads to a circular motion and how to proof this using Netwons laws and calculus? Can you explain why a circle is created, if you let the time tick?


Let the velocity vector be defined as $v(t) = (s \cos( \theta(t)), s \sin (\theta(t))$, where $\theta(t)$ is a time varying quantity. Note that $s$ is a constant, since we have already proved that the speed of the velocity vector does not change.

Let the force vector be defined as $F(t) = (r(t) \cos(\phi(t)), r(t) \sin(\phi(t))$ where $r(t)$ and $\phi(t)$ are time varying quantities.

Now, we know that $F . v = 0$. Hence:

\begin{align*} &r(t) s \left[\cos(\theta(t))\cos(\phi(t)) + \sin(\theta(t))\sin(\phi(t)) \right] = 0 \\ &r(t)s[\cos(\theta(t) - \phi(t))] = 0 \quad \text{(since $\cos(a - b) = \cos a \cos b + \sin a \sin b)$} \end{align*}

Let us assume that $r(t), s \neq 0$ for the moment. If they are, then the problem reduces to trivial cases. So, this now means that $cos(\theta(t) - \phi(t)) = 0$, or $\theta(t) - \phi(t) = \pi /2$. Hence, $ \theta(t) = \pi/2 + \phi(t) $.

Next, let us assume the mass of the body is 1 (otherwise, we will need to carry a factor of $m$ everywhere which is annoying and adds no real insight), and hence $F = \frac{dv}{dt}$.

\begin{align*} &F = \frac{dv}{dt} \\ &(r(t) \cos(\phi(t)), r(t) \sin(\phi(t)) = \frac{d \left(s \cos (\theta(t)), s \sin (\theta(t)) \right)}{dt} \\ % &(r(t) \cos(\phi(t)), r(t) \sin(\phi(t)) = \left(-s \sin (\theta(t)) \theta'(t), s \cos (\theta(t)) \theta'(t) \right) \quad \text{(Differentiating with respect to $t$)} \\ % &(r(t) \cos(\phi(t)), r(t) \sin(\phi(t)) = \left(-s \sin (\pi/2 + \phi(t)) \theta'(t), s \cos (\pi/2 + \phi(t)) \theta'(t) \right)\quad \text{($\theta(t) = \pi/2 + \phi(t)$)} \\ % &(r(t) \cos(\phi(t)), r(t) \sin(\phi(t)) = \left(-s\theta'(t) \cos (\phi(t)) , -s\theta'(t) \sin (\phi(t)) \right)\quad \\ % \end{align*} Comparing the LHS and the RHs, we conclude that $r(t) = -s \theta'(t)$.

If we are interested in uniform circular motion, then we would set $\theta'(t)$ to a constant and proceed to solve the system as described on wikipedia

For non-uniform circular motion, I don't know off-hand how to solve the system of equations, but I presume it is possible. I'll update the answer once I go look it up


Since $$ \vec{F} = m{d\vec{v}\over dt} =m\cdot \vec{v}'$$

so $$\vec{v}'\cdot \vec{v} =0\implies (\vec{v}^2)' = 0 \implies \vec{v}^2 = constant$$

So $$|\vec{v}|^2 = constant\implies |\vec{v}| = constant_2$$

So the magnitude of velocity is constant, but not the velocity it self.

  • $\begingroup$ Sure, but this does not explain the circle which is created by $\vec{s}(t)$, does it? $\endgroup$ – TVSuchty Mar 30 at 10:44
  • $\begingroup$ In the other prove s is defined to be the velocity squared. In physics generally the position is meant... $\endgroup$ – TVSuchty Mar 30 at 11:10
  • $\begingroup$ Well, I'm not sure, but I think that what there was said s^2=vv is not true. It sholud be v =s ′, no? $\endgroup$ – Maria Mazur Mar 30 at 11:14
  • $\begingroup$ I think I answered this already in my last comment. I assume a merge-conflict here... $\endgroup$ – TVSuchty Mar 30 at 11:16
  • $\begingroup$ Any curve can be walked through by constant speed. $\endgroup$ – Berci Mar 30 at 11:34

Try to think like this: the component of the force parallel to the velocity changes the module of the velocity, while component perpendicular to the velocity changes its direction. Since you have only a perpendicular component, the velocity will stay constant in module while changing its direction. Now, if your force rotationally symmetric and is parallel to $\vec{s}$, you also have that $\vec{s}$ and $\vec{v}$ are perpendicular. Therefore by the same euristic argument as before we can say that $\vec{s}$ will only change its direction and not its module. This is exactly like saying that the trajectory is a circle


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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