# Difference between revisions of "Pendulum (mathematics)"

en>Krishnavedala m (→Power series solution for the elliptic integral: superseded vector image) |
|||

(2 intermediate revisions by 2 users not shown) | |||

Line 1: | Line 1: | ||

{{Dynamics}} | {{Dynamics}} | ||

The mathematics of [[pendulum]]s are in general quite complicated. Simplifying assumptions can be made, which in the case of a simple | {{seeintro}} | ||

The mathematics of [[pendulum]]s are in general quite complicated. Simplifying assumptions can be made, which in the case of a [[#Simple gravity pendulum|simple pendulum]] allows the equations of motion to be solved analytically for small-angle oscillations. | |||

== Simple gravity pendulum == | == Simple gravity pendulum == | ||

A simple pendulum is an idealization of a real pendulum using the following assumptions: | [[File:Oscillating pendulum.gif|right|300px|thumb|Animation of a pendulum showing the [[Equations of motion|velocity and acceleration vectors]].]] | ||

A so-called "simple pendulum" is an idealization of a "real pendulum" but in an [[isolated system]] using the following assumptions: | |||

* The rod or cord on which the bob swings is massless, inextensible and always remains taut; | * The rod or cord on which the bob swings is massless, inextensible and always remains taut; | ||

* The bob is a point mass; | |||

* Motion occurs only in [[two dimensions]], i.e. the bob does not trace an [[ellipse]] but an [[Arc (geometry)|arc]]. | * Motion occurs only in [[two dimensions]], i.e. the bob does not trace an [[ellipse]] but an [[Arc (geometry)|arc]]. | ||

* The motion does not lose energy to [[friction]] or [[air resistance]]. | * The motion does not lose energy to [[friction]] or [[air resistance]]. | ||

Line 10: | Line 14: | ||

The [[differential equation]] which represents the motion of a simple pendulum is | The [[differential equation]] which represents the motion of a simple pendulum is | ||

:<math>{d^2\theta\over dt^2}+{g\over \ell} \sin\theta=0 | {{NumBlk|:|<math>{d^2\theta\over dt^2}+{g\over \ell} \sin\theta=0</math>|{{EquationRef|Eq. 1}}|LnSty=1px dashed black}} | ||

where g is acceleration due to gravity, <math>\ell</math> is the length of the pendulum, and | where <math>g</math> is acceleration due to gravity, <math>\ell</math> is the length of the pendulum, and <math>\theta</math> is the angular displacement. | ||

{| class="wikitable" width=100%; | {| class="wikitable" width=100%; | ||

Line 17: | Line 21: | ||

| | | | ||

{{show | {{show | ||

|"Force" derivation of (1) | |"Force" derivation of ({{EquationNote|Eq. 1}}) | ||

| | | | ||

[[File:Pendulum | [[File:Pendulum gravity.svg|thumb|'''Figure 1.''' Force diagram of a simple gravity pendulum.|right|300px]] | ||

Please take the time to consider Figure 1 on the right, showing the forces acting on a simple pendulum. Note that the path of the pendulum sweeps out an arc of a circle. The angle <math>\theta</math> is measured in [[radian]]s, and this is crucial for this formula. The blue arrow is the [[gravitational force]] acting on the bob, and the violet arrows are that same force resolved into components parallel and perpendicular to the bob's instantaneous motion. The direction of the bob's instantaneous [[velocity]] always points along the red axis, which is considered the tangential axis because its direction is always tangent to the circle. Consider Newton's second law, | |||

:<math>F=ma\,</math> | :<math>F=ma\,</math> | ||

Line 30: | Line 34: | ||

:<math>a = -g \sin\theta\,</math> | :<math>a = -g \sin\theta\,</math> | ||

where | where | ||

: | :<math>g</math> is the acceleration due to gravity near the surface of the earth. The negative sign on the right hand side implies that <math>\theta</math> and <math>a</math> always point in opposite directions. This makes sense because when a pendulum swings further to the left, we would expect it to accelerate back toward the right. | ||

This linear acceleration <math>a</math> along the red axis can be related to the change in angle <math>\theta</math> by the arc length formulas; | This linear acceleration <math>a</math> along the red axis can be related to the change in angle <math>\theta</math> by the arc length formulas; <math>s</math> is arc length: | ||

:<math> s = \ell\theta\,</math> | :<math> s = \ell\theta\,</math> | ||

:<math> v = {ds\over dt} = \ell{d\theta\over dt}</math> | :<math> v = {ds\over dt} = \ell{d\theta\over dt}</math> | ||

Line 38: | Line 42: | ||

thus: | thus: | ||

:<math>\ell{d^2\theta\over dt^2} = -g \sin\theta</math> | |||

:<math>{d^2\theta\over dt^2} + {g\over\ell}\sin\theta = 0</math> | :<math>{d^2\theta\over dt^2} + {g\over\ell}\sin\theta = 0</math> | ||

}} | }} | ||

Line 46: | Line 52: | ||

| | | | ||

{{show | {{show | ||

|"Torque" derivation of (1) | |"Torque" derivation of ({{EquationNote|Eq. 1}}) | ||

| | | | ||

Equation (1) can be obtained using two definitions for torque. | |||

:<math>\mathbf{\tau} = \mathbf{r} \times \mathbf{F} = {{d\mathbf{ L}} \over {dt}}</math> | |||

First start by defining the torque on the pendulum bob using the force due to gravity. | |||

:<math>\mathbf{ \tau } = \mathbf{ l \times F_g },</math> | |||

where <math>\mathbf{l}</math> is the length vector of the pendulum and <math>\mathbf{F_g}</math> is the force due to gravity. | |||

For now just consider the magnitude of the torque on the pendulum. | |||

:<math>\mathbf{ |\tau| } = -mg l \sin\theta,</math> | |||

where <math>m</math> is the mass of the pendulum, <math>g</math> is the acceleration due to gravity, <math>l</math> is the length of the pendulum | |||

and <math>\theta</math> is the angle between the length vector and the force due to gravity. | |||

:<math>\mathbf{ \ | Next rewrite the angular momentum. | ||

:<math>\mathbf{ L } = \mathbf{ r \times p } = m\mathbf{ r \times (\omega \times r) }</math>. | |||

Again just consider the magnitude of the angular momentum. | |||

:<math> \mathbf{ |L| } = mr^2 \omega = m l^2 {d\theta \over dt} </math>. | |||

and its time derivative | |||

:<math> | :<math> {d \over dt}\mathbf{|L|} = m l^2 {d^2\theta \over dt^2} </math>, | ||

According to <math>{ \mathbf{ \tau } = d \mathbf{ L} | According to <math>{ \mathbf{ \tau } = {d \mathbf{ L} \over dt} }</math>, we can get by comparing the magnitudes | ||

:<math>- | :<math>-mgl \sin\theta = m l^2 {d^2\theta \over dt^2} </math>, | ||

thus: | thus: | ||

:<math>{d^2\theta\over dt^2} + {g\over | :<math>{d^2\theta\over dt^2} + {g\over l}\sin\theta = 0,</math> | ||

which is the same result as obtained through force analysis. | which is the same result as obtained through force analysis. | ||

}} | }} | ||

|} | |} | ||

Line 72: | Line 90: | ||

| | | | ||

{{show | {{show | ||

|"Energy" derivation of (1) | |"Energy" derivation of ({{EquationNote|Eq. 1}}) | ||

| | | | ||

[[Image:simple pendulum height.svg|thumb|'''Figure 2.''' Trigonometry of a simple gravity pendulum.|right|300px]] | [[Image:simple pendulum height.svg|thumb|'''Figure 2.''' Trigonometry of a simple gravity pendulum.|right|300px]] | ||

Line 109: | Line 127: | ||

in terms of <math>{d\theta\over dt}</math> gives | in terms of <math>{d\theta\over dt}</math> gives | ||

:<math>{d\theta\over dt} = \sqrt{{2g\over \ell}\left(\cos\theta-\cos\theta_0\right)} | {{NumBlk|:|<math>{d\theta\over dt} = \sqrt{{2g\over \ell}\left(\cos\theta-\cos\theta_0\right)}</math>|{{EquationRef|Eq. 2}}|LnSty=1px dashed black}} | ||

This equation is known as the ''first integral of motion'', it gives the velocity in terms of the location and includes an integration constant related to the initial displacement (<math>\theta_0</math>). We can differentiate, by applying the [[chain rule]], with respect to time to get the acceleration | This equation is known as the ''first integral of motion'', it gives the velocity in terms of the location and includes an integration constant related to the initial displacement (<math>\theta_0</math>). We can differentiate, by applying the [[chain rule]], with respect to time to get the acceleration | ||

Line 129: | Line 147: | ||

==Small-angle approximation== | ==Small-angle approximation== | ||

The differential equation given above is not easily solved. However adding a restriction to the size of the oscillation's amplitude gives a form whose solution can be easily obtained. If it is assumed that the angle is much less than 1 radian, or | The differential equation given above is not easily solved, and there is no solution that can be written in terms of elementary functions. However adding a restriction to the size of the oscillation's amplitude gives a form whose solution can be easily obtained. If it is assumed that the angle is much less than 1 radian, or | ||

:<math>\theta \ll 1\,</math>, | :<math>\theta \ll 1\,</math>, | ||

then substituting for sin θ into | then substituting for sin θ into {{EquationNote|Eq. 1}} using the [[small-angle approximation]], | ||

:<math>\sin\theta\approx\theta\,</math>, | :<math>\sin\theta\approx\theta\,</math>, | ||

yields the equation for a [[harmonic oscillator]] | yields the equation for a [[harmonic oscillator]], | ||

:<math>{d^2\theta\over dt^2}+{g\over \ell}\theta=0.</math> | :<math>{d^2\theta\over dt^2}+{g\over \ell}\theta=0.</math> | ||

Line 161: | Line 179: | ||

:<math>T_0 = 2\pi\sqrt{\frac{\ell}{g}}</math> can be expressed as <math>\ell = {\frac{g}{\pi^2}}\times{\frac{T_0^2}{4}}.</math> | :<math>T_0 = 2\pi\sqrt{\frac{\ell}{g}}</math> can be expressed as <math>\ell = {\frac{g}{\pi^2}}\times{\frac{T_0^2}{4}}.</math> | ||

If [[International System of Units|SI units]] are used (i.e. measure in metres and seconds), and assuming the measurement is taking place on the Earth's surface, then <math> | If [[International System of Units|SI units]] are used (i.e. measure in metres and seconds), and assuming the measurement is taking place on the Earth's surface, then <math>g\approx9.81</math> m/s<sup>2</sup>, and <math>g/\pi^2\approx{1}</math> (0.994 is the approximation to 3 decimal places). | ||

Therefore a relatively reasonable approximation for the length and period are, | Therefore a relatively reasonable approximation for the length and period are, | ||

Line 169: | Line 187: | ||

== Arbitrary-amplitude period == | == Arbitrary-amplitude period == | ||

For amplitudes beyond the [[small angle approximation]], one can compute the exact period by first inverting the equation for the angular velocity obtained from the energy method ({{EquationNote|Eq. 2}}), | |||

[[Image:Pendulum period.svg|thumb|'''Figure 3.''' Deviation of the "true" period of a pendulum from the small-angle approximation of the period. "True" value was obtained using Matlab to numerically evaluate the elliptic integral. |right|300px]] | |||

[[Image:Pendulum period.svg|thumb|'''Figure 3.''' Deviation of the period from small-angle approximation.|right|300px]] | |||

[[Image:Pendulum Rel Error90a.png|thumb|'''Figure 4.''' Relative errors using the power series.|right|300px]] | [[Image:Pendulum Rel Error90a.png|thumb|'''Figure 4.''' Relative errors using the power series.|right|300px]] | ||

Line 178: | Line 195: | ||

:<math>{dt\over d\theta} = \sqrt{\ell\over 2g}{1\over\sqrt{\cos\theta-\cos\theta_0}}</math> | :<math>{dt\over d\theta} = \sqrt{\ell\over 2g}{1\over\sqrt{\cos\theta-\cos\theta_0}}</math> | ||

and then integrating over one complete cycle, | and then integrating over one complete cycle, | ||

:<math>T = t(\theta_0\rightarrow0\rightarrow-\theta_0\rightarrow0\rightarrow\theta_0),</math> | :<math>T = t(\theta_0\rightarrow0\rightarrow-\theta_0\rightarrow0\rightarrow\theta_0),</math> | ||

Line 194: | Line 211: | ||

:<math>T = 4\sqrt{\ell\over 2g}\int^{\theta_0}_0 {1\over\sqrt{\cos\theta-\cos\theta_0}}\,d\theta.</math> | :<math>T = 4\sqrt{\ell\over 2g}\int^{\theta_0}_0 {1\over\sqrt{\cos\theta-\cos\theta_0}}\,d\theta.</math> | ||

Note that this integral diverges as <math> \theta_0 </math> approaches the vertical | |||

:<math> \lim_{\theta_0 \rightarrow \pi} T = \infty </math>, | |||

so that a pendulum with just the right energy to go vertical will never actually get there. (Conversely, a pendulum close to its maximum can take an arbitrarily long time to fall down.) | |||

This integral can be re-written in terms of [[elliptic integral]]s as | |||

: <math>F(\ | :<math>T = 4\sqrt{\ell\over g}F\left( {\theta_0}, \csc{\theta_0\over2}\right)\csc {\theta_0\over 2}</math> | ||

where <math>F</math> is the [[elliptic integral#Incomplete elliptic integral of the first kind|incomplete elliptic integral of the first kind]] defined by | |||

:<math> | : <math>F(\varphi , k) = \int_0^\varphi {1\over\sqrt{1-k^2\sin^2{u}}}\,du\,.</math> | ||

Or more concisely by the [[integration by substitution|substitution]] <math>\sin{u} = \frac{\sin{\theta\over 2}}{\sin{\theta_0\over 2}}</math> expressing <math>\theta</math> in terms of <math>u</math>, | |||

: <math> | {{NumBlk|:|<math>T = 4\sqrt{\ell\over g}\,K\left( \sin{\theta_0\over 2} \right)</math>|{{EquationRef|Eq. 3}}|LnSty=1px dashed black}} | ||

where <math>K</math> is the [[elliptic integral#Complete elliptic integral of the first kind|complete elliptic integral of the first kind]] defined by | |||

: <math>K(k) = F \left( {\pi\over 2}, k \right) = \int_0^{\pi/2} {1\over\sqrt{1-k^2\sin^2{u}}}\,du\,.</math> | |||

For comparison of the approximation to the full solution, consider the period of a pendulum of length 1 m on Earth (''g'' = 9.80665 m/s<sup>2</sup>) at initial angle 10 degrees is <math>4\sqrt{1\ \mathrm{m}\over g}K\left( {\sin {10^\circ\over 2}} \right) \approx 2.0102\ \mathrm{s}</math>. The linear approximation gives <math>2\pi \sqrt{1\ \mathrm{m}\over g} \approx 2.0064\ \mathrm{s}</math>. The difference between the two values, less than 0.2%, is much less than that caused by the variation of g with geographical location. | |||

From here there are many ways to proceed to calculate the elliptic integral: | |||

===Legendre polynomial solution for the elliptic integral=== | |||

Given {{EquationNote|Eq. 3}} and the [[Legendre polynomial]] solution for the elliptic integral: | |||

:<math>K(k) = \frac{\pi}{2}\left\{1 + \left(\frac{1}{2}\right)^2 k^{2} + \left(\frac{1 \cdot 3}{2 \cdot 4}\right)^2 k^{4} + \cdots + \left[\frac{\left(2n - 1\right)!!}{\left(2n\right)!!}\right]^2 k^{2n} + \cdots \right\},</math> | |||

where {{math|''n''!!}} denotes the [[double factorial]], an exact solution to the period of a pendulum is: | |||

:<math>\begin{alignat}{2} | :<math>\begin{alignat}{2} | ||

T & = 2\pi \sqrt{\ell\over g} \left( 1+ \left( \frac{1}{2} \right)^2 \sin^2\left(\frac{\theta_0}{2}\right) + \left( \frac{1 \cdot 3}{2 \cdot 4} \right)^2 \sin^4\left(\frac{\theta_0}{2}\right) + \left( \frac {1 \cdot 3 \cdot 5}{2 \cdot 4 \cdot 6} \right)^2 \sin^6\left(\frac{\theta_0}{2}\right) + \cdots \right) \\ | T & = 2\pi \sqrt{\ell\over g} \left( 1+ \left( \frac{1}{2} \right)^2 \sin^2\left(\frac{\theta_0}{2}\right) + \left( \frac{1 \cdot 3}{2 \cdot 4} \right)^2 \sin^4\left(\frac{\theta_0}{2}\right) + \left( \frac {1 \cdot 3 \cdot 5}{2 \cdot 4 \cdot 6} \right)^2 \sin^6\left(\frac{\theta_0}{2}\right) + \cdots \right) \\ | ||

Line 220: | Line 246: | ||

Figure 4 shows the relative errors using the power series. ''T''<sub>0</sub> is the linear approximation, and ''T''<sub>2</sub> to ''T''<sub>10</sub> include respectively the terms up to the 2nd to the 10th powers. | Figure 4 shows the relative errors using the power series. ''T''<sub>0</sub> is the linear approximation, and ''T''<sub>2</sub> to ''T''<sub>10</sub> include respectively the terms up to the 2nd to the 10th powers. | ||

[[File:Pendulum phase portrait.svg|thumb|312x312px|'''Figure 5.''' Potential energy and phase portrait of a simple pendulum. Note that the ''x''-axis, being angle, wraps onto itself after every 2π radians.]] | |||

=== Power series solution for the elliptic integral === | |||

Another formulation of the above solution can be found if the following Maclaurin series: | |||

[[ | :<math>\sin {\theta_0 \over 2}=\frac{1}{2}\theta_0 - \frac{1}{48}\theta_0^3 + \frac{1}{3840}\theta_0^5 - \frac{1}{645120}\theta_0^7 + \cdots.</math> | ||

is used in the Legendre polynomial solution above. | |||

The resulting power series is:<ref>{{cite journal | |||

:<math> | |||

\sin | |||

</math> | |||

The | |||

| last = Nelson | | last = Nelson | ||

| first = Robert | | first = Robert | ||

| authorlink = | | authorlink = | ||

| | |author2=M. G. Olsson | ||

| title = The pendulum — Rich physics from a simple system | | title = The pendulum — Rich physics from a simple system | ||

| journal = American Journal of Physics | | journal = American Journal of Physics | ||

Line 254: | Line 267: | ||

| doi = | | doi = | ||

10.1119/1.14703| id = | 10.1119/1.14703| id = | ||

| accessdate = 2012- | | accessdate = 2012-04-30}}</ref> | ||

:<math>\begin{alignat}{2} | :<math>\begin{alignat}{2} | ||

T & = 2\pi \sqrt{\ell\over g} \left( 1+ \frac{1}{16}\theta_0^2 + \frac{11}{3072}\theta_0^4 + \frac{173}{737280}\theta_0^6 + \frac{22931}{1321205760}\theta_0^8 + \frac{1319183}{951268147200}\theta_0^{10} + \frac{233526463}{2009078326886400}\theta_0^{12} + . . . \right) | T & = 2\pi \sqrt{\ell\over g} \left( 1+ \frac{1}{16}\theta_0^2 + \frac{11}{3072}\theta_0^4 + \frac{173}{737280}\theta_0^6 + \frac{22931}{1321205760}\theta_0^8 + \frac{1319183}{951268147200}\theta_0^{10} + \frac{233526463}{2009078326886400}\theta_0^{12} + . . . \right) | ||

\end{alignat}</math> | \end{alignat}.</math> | ||

:<math> | ===Arithmetic-geometric mean solution for elliptic integral=== | ||

Given {{EquationNote|Eq. 3}} and the [[Arithmetic-geometric mean]] solution of the elliptic integral: | |||

:<math>K(k) = \frac {\pi /2}{M(1-k,1+k)},</math> | |||

where <math>M(x,y)</math> is the arithmetic-geometric mean of <math>x</math> and <math>y</math>. | |||

This yields an alternative and faster-converging formula for the period:<ref>{{Citation |title=Approximations for the period of the simple pendulum based on the arithmetic-geometric mean |url=http://suppes-corpus.stanford.edu/articles/physics/431.pdf |first1=Claudio G. |last1=Carvalhaes |first2=Patrick |last2=Suppes |journal=[[American Journal of Physics|Am. J. Phys.]] |issn=0002-9505 |volume=76 |issue=12͒ |date=December 2008 |pages=1150–1154 |doi=10.1119/1.2968864͔ |accessdate=2013-12-14}}</ref><ref>{{Citation |title=An eloquent formula for the perimeter of an ellipse | url=http://www.ams.org/notices/201208/rtx120801094p.pdf |first=Semjon | last=Adlaj |journal=[[Notices of the American Mathematial Society |Notices of the AMS]] |issn=1088-9477 |volume=76 |issue=8 |date=September 2012 |pages=1094・099 | doi=}}</ref> | |||

:<math>T = \frac{2\pi}{M(1, \cos(\theta_0/2))} \sqrt\frac{\ell}{g}.</math> | |||

== Examples == | == Examples == | ||

Line 276: | Line 292: | ||

File:Pendulum_135deg.gif|Initial angle of 135° | File:Pendulum_135deg.gif|Initial angle of 135° | ||

File:Pendulum_170deg.gif|Initial angle of 170° | File:Pendulum_170deg.gif|Initial angle of 170° | ||

File:Pendulum_180deg.gif|Initial angle of 180°, unstable equilibrium. | |||

File:Pendulum_190deg.gif|Pendulum with just barely enough energy for a full swing | File:Pendulum_190deg.gif|Pendulum with just barely enough energy for a full swing | ||

File:Pendulum_220deg.gif|Pendulum with enough energy for a full swing | File:Pendulum_220deg.gif|Pendulum with enough energy for a full swing | ||

</gallery> | </gallery> | ||

Line 317: | Line 333: | ||

== Physical interpretation of the imaginary period == | == Physical interpretation of the imaginary period == | ||

The [[Jacobi's elliptic functions|Jacobian elliptic function]] that expresses the position of a pendulum as a function of time is a [[doubly periodic function]] with a [[real number|real]] period and an [[imaginary number|imaginary]] period. The real period is of course the time it takes the pendulum to go through one full cycle. [[Paul Appell]] pointed out a physical interpretation of the imaginary period:<ref>Paul Appell, "Sur une interprétation des valeurs imaginaires du temps en Mécanique", ''[[Comptes Rendus]] Hebdomadaires des Scéances de l'Académie des Sciences'', volume 87, number 1, July, 1878</ref> if ''θ''<sub>0</sub> is the maximum angle of one pendulum and 180° − ''θ''<sub>0</sub> is the maximum angle of another, then the real period of each is the magnitude of the imaginary period of the other. | The [[Jacobi's elliptic functions|Jacobian elliptic function]] that expresses the position of a pendulum as a function of time is a [[doubly periodic function]] with a [[real number|real]] period and an [[imaginary number|imaginary]] period. The real period is of course the time it takes the pendulum to go through one full cycle. [[Paul Appell]] pointed out a physical interpretation of the imaginary period:<ref>Paul Appell, "Sur une interprétation des valeurs imaginaires du temps en Mécanique", ''[[Comptes Rendus]] Hebdomadaires des Scéances de l'Académie des Sciences'', volume 87, number 1, July, 1878</ref> if ''θ''<sub>0</sub> is the maximum angle of one pendulum and 180° − ''θ''<sub>0</sub> is the maximum angle of another, then the real period of each is the magnitude of the imaginary period of the other. This interpretation, involving dual forces in opposite directions, might be further clarified and generalized to other classical problems in mechanics with dual solutions.<ref>Adlaj, S. ''Mechanical interpretation of negative and imaginary tension of a tether in a linear parallel force field '', Selected papers of the International Scientific Conference on Mechanics "SIXTH POLYAKHOV READINGS", January 31 - February 3, 2012, Saint-Petersburg, Russia, pp. 13-18.</ref> | ||

==See also== | ==See also== | ||

Line 325: | Line 341: | ||

*[[Double pendulum]] | *[[Double pendulum]] | ||

*[[Inverted pendulum]] | *[[Inverted pendulum]] | ||

*[[Kapitza's pendulum]] | |||

*[[Spring pendulum]] | *[[Spring pendulum]] | ||

*[[Mathieu function]] | *[[Mathieu function]] | ||

Line 340: | Line 357: | ||

*[http://mathworld.wolfram.com/MathieuFunction.html Mathworld article on Mathieu Function] | *[http://mathworld.wolfram.com/MathieuFunction.html Mathworld article on Mathieu Function] | ||

[[Category:Horology]] | |||

[[Category:Mathematical physics]] | [[Category:Mathematical physics]] | ||

[[Category:Pendulums]] | [[Category:Pendulums]] | ||

## Latest revision as of 20:37, 29 November 2014

Template:Dynamics Template:Seeintro The mathematics of pendulums are in general quite complicated. Simplifying assumptions can be made, which in the case of a simple pendulum allows the equations of motion to be solved analytically for small-angle oscillations.

## Simple gravity pendulum

A so-called "simple pendulum" is an idealization of a "real pendulum" but in an isolated system using the following assumptions:

- The rod or cord on which the bob swings is massless, inextensible and always remains taut;
- The bob is a point mass;
- Motion occurs only in two dimensions, i.e. the bob does not trace an ellipse but an arc.
- The motion does not lose energy to friction or air resistance.

The differential equation which represents the motion of a simple pendulum is

Template:NumBlk where is acceleration due to gravity, is the length of the pendulum, and is the angular displacement.

## Small-angle approximation

The differential equation given above is not easily solved, and there is no solution that can be written in terms of elementary functions. However adding a restriction to the size of the oscillation's amplitude gives a form whose solution can be easily obtained. If it is assumed that the angle is much less than 1 radian, or

then substituting for sin θ into Template:EquationNote using the small-angle approximation,

yields the equation for a harmonic oscillator,

The error due to the approximation is of order *θ*^{ 3} (from the Maclaurin series for sin *θ*).

Given the initial conditions *θ*(0) = *θ*_{0} and *dθ*/*dt*(0) = 0, the solution becomes,

The motion is simple harmonic motion where *θ*_{0} is the semi-amplitude of the oscillation (that is, the maximum angle between the rod of the pendulum and the vertical). The period of the motion, the time for a complete oscillation (outward and return) is

which is known as Christiaan Huygens's law for the period. Note that under the small-angle approximation, the period is independent of the amplitude *θ*_{0}; this is the property of isochronism that Galileo discovered.

### Rule of thumb for pendulum length

If SI units are used (i.e. measure in metres and seconds), and assuming the measurement is taking place on the Earth's surface, then m/s^{2}, and (0.994 is the approximation to 3 decimal places).

Therefore a relatively reasonable approximation for the length and period are,

## Arbitrary-amplitude period

For amplitudes beyond the small angle approximation, one can compute the exact period by first inverting the equation for the angular velocity obtained from the energy method (Template:EquationNote),

and then integrating over one complete cycle,

or twice the half-cycle

or 4 times the quarter-cycle

which leads to

Note that this integral diverges as approaches the vertical

so that a pendulum with just the right energy to go vertical will never actually get there. (Conversely, a pendulum close to its maximum can take an arbitrarily long time to fall down.)

This integral can be re-written in terms of elliptic integrals as

where is the incomplete elliptic integral of the first kind defined by

Or more concisely by the substitution expressing in terms of ,

Template:NumBlk where is the complete elliptic integral of the first kind defined by

For comparison of the approximation to the full solution, consider the period of a pendulum of length 1 m on Earth (*g* = 9.80665 m/s^{2}) at initial angle 10 degrees is . The linear approximation gives . The difference between the two values, less than 0.2%, is much less than that caused by the variation of g with geographical location.

From here there are many ways to proceed to calculate the elliptic integral:

### Legendre polynomial solution for the elliptic integral

Given Template:EquationNote and the Legendre polynomial solution for the elliptic integral:

where *n*!! denotes the double factorial, an exact solution to the period of a pendulum is:

Figure 4 shows the relative errors using the power series. *T*_{0} is the linear approximation, and *T*_{2} to *T*_{10} include respectively the terms up to the 2nd to the 10th powers.

### Power series solution for the elliptic integral

Another formulation of the above solution can be found if the following Maclaurin series:

is used in the Legendre polynomial solution above.
The resulting power series is:^{[1]}

### Arithmetic-geometric mean solution for elliptic integral

Given Template:EquationNote and the Arithmetic-geometric mean solution of the elliptic integral:

where is the arithmetic-geometric mean of and .

This yields an alternative and faster-converging formula for the period:^{[2]}^{[3]}

## Examples

The animations below depict several different modes of oscillation given different initial conditions. The small graph above the pendulums are their phase portraits.

## Compound pendulum

A **compound pendulum** (or **physical pendulum**) is one where the rod is not massless, and may have extended size; that is, an arbitrarily shaped rigid body swinging by a pivot. In this case the pendulum's period depends on its moment of inertia *I* around the pivot point.

The equation of torque gives:

where:

The torque is generated by gravity so:

where:

*L*is the distance from the pivot to the center of mass of the pendulum*θ*is the angle from the vertical

Hence, under the small-angle approximation ,

This is of the same form as the conventional simple pendulum and this gives a period of:

^{[4]}

And a frequency of:

## Physical interpretation of the imaginary period

The Jacobian elliptic function that expresses the position of a pendulum as a function of time is a doubly periodic function with a real period and an imaginary period. The real period is of course the time it takes the pendulum to go through one full cycle. Paul Appell pointed out a physical interpretation of the imaginary period:^{[5]} if *θ*_{0} is the maximum angle of one pendulum and 180° − *θ*_{0} is the maximum angle of another, then the real period of each is the magnitude of the imaginary period of the other. This interpretation, involving dual forces in opposite directions, might be further clarified and generalized to other classical problems in mechanics with dual solutions.^{[6]}

## See also

- Blackburn pendulum
- Conical pendulum
- Double pendulum
- Inverted pendulum
- Kapitza's pendulum
- Spring pendulum
- Mathieu function
- Pendulum equations (software)

## References

- ↑ {{#invoke:Citation/CS1|citation |CitationClass=journal }}
- ↑ {{#invoke:citation/CS1|citation |CitationClass=citation }}
- ↑ {{#invoke:citation/CS1|citation |CitationClass=citation }}
- ↑ Physical Pendulum
- ↑ Paul Appell, "Sur une interprétation des valeurs imaginaires du temps en Mécanique",
*Comptes Rendus Hebdomadaires des Scéances de l'Académie des Sciences*, volume 87, number 1, July, 1878 - ↑ Adlaj, S.
*Mechanical interpretation of negative and imaginary tension of a tether in a linear parallel force field*, Selected papers of the International Scientific Conference on Mechanics "SIXTH POLYAKHOV READINGS", January 31 - February 3, 2012, Saint-Petersburg, Russia, pp. 13-18.

## Further reading

- {{#invoke:citation/CS1|citation

|CitationClass=book }}

- {{#invoke:Citation/CS1|citation

|CitationClass=journal }}

- {{#invoke:Citation/CS1|citation

|CitationClass=journal }}