I was given a problem in a Classical Mechanics course that went somehow like the following:

"Consider a particle of mass $m$ moving under the presence of a force $\vec{F} = k\hat{x}\times\vec{v}$, where $\hat{x}$ is an unit vector in the positive direction of the $x$ axis and $k$ any constant. Prove that the movement of a particle under this force is restricted to a circular motion with angular velocity $\vec{\omega}=(k/m)\hat{x}$ or, in a more general case, a spiral movement parallel to the direction of $\hat{x}$."

In an Electrodynamics elementary college course you can see and solve that a magnetic force sort of as:


with a magnetic field, say, $\vec{B}=B_0\hat{z}$ can drive a particle through a spiral movement in the precise direction of that magnetic field you customize, involving a cyclotron frequency and so, if and only if you input further initial conditions to the movement in x, y and z.

My inquiry then is, how can you prove the relation given above for the angular velocity and conclude a spiral movement, from a classical mechanics perspective? I can see there's a link between both procedures, but I cannot try solving the first one without giving a glimpse to the latter.

  • 1
    $\begingroup$ Did you mean helical movement? Spiral has changes in curvature whereas helix has constant curvature and torsion. $\endgroup$ – Ng Chung Tak Mar 8 '17 at 10:04
  • $\begingroup$ Helical was the word, I kind of get lost into translation (my teacher stated spiral movement as well when the problem was given). Thanks for clearing it out. $\endgroup$ – chandrasekhar17 Mar 9 '17 at 8:13

Write $\vec{v} = \dot{x}\hat{x} + \dot{y}\hat{y} + \dot{z}\hat{z}$, where $\hat{x}, \hat{y}, \hat{z}$ denote unit vectors in the positive direction of the axis $x$, $y$, $z$ respectively. Then,

$$k \hat{x}\times\vec{v} = -k \dot{z}\hat{y} + k\dot{y}\hat{z}.$$

Equating this expression with the analogous for the force, we get a system of three differential equations, which, after dividing by $m$, are as follows:

$$\begin{align*} \ddot{x} &= 0,\\ \ddot{y} &= -\frac{k}{m}\dot{z},\\ \ddot{z} &= \frac{k}{m}\dot{y}. \end{align*}$$

The first equation is readily solved to get the description of a uniform rectilinear movement along the $x$ direction,

$$x = at + b,$$ with $a$, $b$ constants. The remaining pair of equations correspond to the movement of a particle in a central field in the $yz$ plane. To prove that it corresponds to a circular motion, just substitute $v_y = \dot{y}$ and $v_z = \dot{z}$, so that the second and third equations become

$$\begin{align*} \dot{v}_y &= -\frac{k}{m} v_z,\\ \dot{v}_z &= \frac{k}{m} v_y. \end{align*}$$

These equations can be solved to obtain $v_y = A\cos(\frac{k}{m}t + \phi)$, $v_z = A\sin(\frac{k}{m}t + \phi)$. Since $v_y = \dot{y}$, $v_z = \dot{z}$, one last pair of integrals give the final solution for $y$ and $z$:

$$\begin{align*} y &= A\,\frac{m}{k}\,\sin\left(\frac{k}{m} t + \phi\right),\\ z &= -A\,\frac{m}{k}\,\cos\left(\frac{k}{m} t + \phi\right). \end{align*}$$

The equations for $x$, $y$ and $z$ are the parametric equations for a spiral movement, parallel to the direcion of $\hat{x}$.

  • $\begingroup$ How did yoi get the equation of velocity from the equation of accleration? $\endgroup$ – Mockingbird Mar 8 '17 at 6:55
  • $\begingroup$ @Mockingbird Differentiate $\dot{v}_y = -(k/m)v_z$ to get $\ddot{v}_y = -(k/m)\dot{v}_z$ and then substitute in the expression for $\dot{v}_z = (k/m)v_y$ from the next line. $\endgroup$ – Mark H Mar 8 '17 at 7:45
  • $\begingroup$ Neat and clear, pretty close to what I solved once in electrodynamics. Thanks for your answer :) $\endgroup$ – chandrasekhar17 Mar 9 '17 at 8:16

With an aid of differential geometry, velocity, acceleration and jerk can be written as:

\begin{align*} \mathbf{v} &= \dot{s} \, \mathbf{T} \\ &= v \, \mathbf{T} \\ \mathbf{a} &= \ddot{s} \, \mathbf{T}+ \kappa \, \dot{s}^2\mathbf{N} \\ \mathbf{b} &= (\dddot{s}-\kappa^2 \dot{s}^3) \mathbf{T}+ (3\kappa \dot{s} \ddot{s}+\dot{\kappa} \dot{s}^2) \mathbf{N}+ \kappa \tau \dot{s}^3 \mathbf{B} \\ \end{align*}

Now $$ \mathbf{a}=\boldsymbol{\omega} \times \mathbf{v} \implies \mathbf{a} \perp \mathbf{T} \implies \ddot{s}=0 \implies \dot{s}=\text{constant} \implies v=u$$

and $$ \mathbf{b}= \dot{\mathbf{a}}=\boldsymbol{\omega} \times \mathbf{a} \implies \mathbf{b} \perp \mathbf{a} \implies \mathbf{b} \perp \mathbf{N} \implies \dot{\kappa}\dot{s}^2=0 \implies \kappa=\text{constant}$$

Also, \begin{align*} \int \mathbf{a} \, dt &= \boldsymbol{\omega} \times \int \mathbf{v} \, dt \\ \mathbf{v} &= \mathbf{u}+\boldsymbol{\omega} \times \mathbf{r} \\ \mathbf{r}(0) &= \mathbf{0} \\ \dot{\mathbf{r}}(0) &= \mathbf{u} \\ \boldsymbol{\omega} \cdot \mathbf{v} &= \boldsymbol{\omega} \cdot \mathbf{u} \\ &= \text{constant} \end{align*}

We have

\begin{align*} \mathbf{v} \times \mathbf{a} &= \mathbf{v} \times (\boldsymbol{\omega} \times \mathbf{v}) \\ &= v^2 \boldsymbol{\omega}-(\boldsymbol{\omega} \cdot \mathbf{v})\mathbf{v} \\ \kappa &= \frac{|v^2 \boldsymbol{\omega}-(\boldsymbol{\omega} \cdot \mathbf{v})\mathbf{v}|} {v^3} \\ &= \frac{|\boldsymbol{\omega} \times \mathbf{v}|}{v^2} \\ &= \frac{|\boldsymbol{\omega} \times \mathbf{u}|}{u^2} \\ |\mathbf{a}| &= |\boldsymbol{\omega} \times \mathbf{u}| \\ &= \text{constant} \\ \mathbf{a} \times \mathbf{b} &= a^2 \boldsymbol{\omega}-(\boldsymbol{\omega} \cdot \mathbf{a}) \mathbf{a} \\ &= a^2 \boldsymbol{\omega} \\ \tau &= \frac{\mathbf{v} \cdot a^2\boldsymbol{\omega}} {(\mathbf{v} \times \mathbf{a})^2} \\ &= \frac{\boldsymbol{\omega} \cdot \mathbf{v}}{v^2} \\ &= \frac{\boldsymbol{\omega} \cdot \mathbf{u}}{u^2} \\ &= \text{constant} \end{align*}

Both $\kappa$ and $\tau$ are constants implying the path is helical. If $\mathbf{u} \cdot \boldsymbol{\omega}=0 \implies \tau=0$, then it'll be a circle. While $\mathbf{u} \times \boldsymbol{\omega}=\mathbf{0} \implies \kappa=0$, that'll be a straight line.

Fitting with initial conditions:

$$\fbox{$\quad \mathbf{r}=\mathbf{u}t+ \frac{\mathbf{u} \times \boldsymbol{\omega}}{\omega^2}(\cos \omega t-1)+ \frac{\boldsymbol{\omega} \times (\mathbf{u} \times \boldsymbol{\omega})}{\omega^{3}}(\sin \omega t-\omega t) \quad \\$}$$

Some facts from differential geometry \begin{align*} s &= \int |\mathbf{v}| \, dt \tag{arclength} \\ \dot{s} &= |\mathbf{v}| \tag{speed} \\ &= v \\ \mathbf{T} &= \frac{\mathbf{v}}{v} \tag{tangent vector}\\ \mathbf{B} &= \frac{\mathbf{v} \times \mathbf{a}}{|\mathbf{v} \times \mathbf{a}|} \tag{binormal vector} \\ \mathbf{N} &= \mathbf{B} \times \mathbf{T} \tag{normal vector} \\ \kappa &= \frac{|\mathbf{v} \times \mathbf{a}|}{v^3} \tag{curvature} \\ \tau &= \frac{\mathbf{v} \cdot \mathbf{a} \times \mathbf{b}} {(\mathbf{v} \times \mathbf{a})^2} \tag{torsion} \end{align*}

  • $\begingroup$ Nice! How did you get that keen on differential geometry? That is definitely an approach I would like to go in further similar problems $\endgroup$ – chandrasekhar17 Mar 9 '17 at 8:17

Take the equations of motion $$ \ddot{\mathbf{r}} = \lambda \;( \hat{k} \times \dot{\mathbf{r}}) $$ and express them in cylindrical coordinates $(r,\theta,z)$

$$\begin{pmatrix} \ddot{r} - r \dot{\theta}^2 \\ r \ddot{\theta} + 2 \dot{r} \dot{\theta} \\ \ddot{z} \end{pmatrix} = \lambda \left( \begin{pmatrix}0\\0\\1 \end{pmatrix} \times \begin{pmatrix} \dot{r} \\ r \dot{\theta} \\ \dot{z} \end{pmatrix} \right) = \lambda \begin{pmatrix} -r \dot\theta \\ \dot{r} \\ 0 \end{pmatrix} $$

which is solved by

$$\begin{align} \ddot{r} & = r \dot{\theta} ( \dot{\theta}-\lambda ) \\ \ddot{\theta} & = \frac{\dot{r} }{r} ( \lambda - 2 \dot{\theta} ) \\ \ddot{z} & =0 \end{align} $$

I see two special cases in the solutions here

  • Set $\dot{\theta} = \lambda$ and $\ddot{\theta}=0$ to yield $$ \begin{align} \ddot{r} & = 0 & \dot{r} & = 0 \\ \ddot{z} & = 0 & \dot{z} & = \mbox{const} \end{align} $$

  • Set $\dot{\theta} = \frac{\lambda}{2}$ and $\ddot{\theta}=0$ to yield $$ \begin{align} \ddot{r} & = -\frac{\lambda^2}{4} \,r & r & = \mbox{harmonic} \\ \ddot{z} & = 0 & \dot{z} & = \mbox{const} \end{align} $$


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.