19
$\begingroup$

I am fairly new to Mathematica, and I cannot figure out how to plot an epicycloid. I have plotted some neat looking things in my attempts, but I still can't make one. I am not looking to make an animation; I just want the plot of an epicycloid.

Would I use PolarPlot or ParametricPlot? How do I get a static picture of an epicycloid?

Improve this question
$\endgroup$
4
  • $\begingroup$ so what did you try then? where did you fail? $\endgroup$ – Pinguin Dirk Sep 16 '13 at 14:53
  • 5
    $\begingroup$ Welcome to Mathematica.SE! Have a look at demonstrations.wolfram.com/search.html?query=epicycloid $\endgroup$ – Yves Klett Sep 16 '13 at 14:56
  • $\begingroup$ Mostly a lot of fiddling with the epicycloid equations in a PolarPlot (and some ParametricPlots). $\endgroup$ – btalbot Sep 16 '13 at 14:56
  • 4
    $\begingroup$ @PinguinDirk is asking you to post the code you have tried thus far, so we can see where there might be a problem. $\endgroup$ – bobthechemist Sep 16 '13 at 15:20
38
$\begingroup$

I recreated the animation on Wikipedia:

enter image description here

Here is the Manipulate version:

R = 3; r = 1;

fx[θ_, a_: 1] := (R + r) Cos[θ] - a r Cos[(R + r) θ/r];
fy[θ_, a_: 1] := (R + r) Sin[θ] - a r Sin[(R + r) θ/r];

gridlines = Table[{x, GrayLevel[0.9]}, {x, -6, 6, 0.5}];

plot[max_] := ParametricPlot[
  {fx[θ], fy[θ]}, {θ, 0, max},
  PlotStyle -> {Red, Thick},
  Epilog -> {
    Thick,
    Blue, Circle[{0, 0}, R],
    Black, Circle[{fx[max, 0], fy[max, 0]}, r],
    Red, PointSize[Large], Point[{fx[max], fy[max]}],
    Black, Line[{{fx[max, 0], fy[max, 0]}, {fx[max], fy[max]}}]
    },
  GridLines -> {gridlines, gridlines},
  PlotRange -> {-6, 6}
  ]
Manipulate[plot[max], {max, 0.01, 2 Pi}]
Improve this answer
$\endgroup$
2
  • 3
    $\begingroup$ I don't know how many hours I've spent (wasted) on playing with roulettes, but it sure was fun. :) $\endgroup$ – J. M.'s ennui♦ May 3 '15 at 6:54
  • $\begingroup$ @J.M. Upvote for that masterpiece. :) $\endgroup$ – yode Jul 24 '16 at 6:52
18
$\begingroup$

It is easiest to use ParametricPlot and RotationTransform.

ParametricPlot[
  RotationTransform[a][{1, 0}] + RotationTransform[4 a][{1/4, 0}],
  {a, 0, 2 Pi}, Evaluated -> True]

enter image description here

Evaluated -> True is not strictly necessary; it's there just to construct the equation once instead of for every point in the graph separately (... which is slow).

EDIT: Improved per comment from Mr.Wizard.

Improve this answer
$\endgroup$
1
  • 2
    $\begingroup$ +1 -- however, I recommend using the Evaluated -> True option rather than Evaluate to properly localize a. Please see: (7832), (5220), (7561) $\endgroup$ – Mr.Wizard Sep 16 '13 at 17:34
17
$\begingroup$

Epicycloids (and other roulette curves) are fun to play with using Manipulate. Here is one of many possible implementations.

Manipulate[
  ParametricPlot[{
    (1 + r)  Cos[theta] + a Cos[(1 + r) theta], 
    (1 + r) Sin[theta] + a Sin[(1 + r) theta]}, 
    {theta, 0, 2 Pi}],
  {{r, 1}, 1, 20, 1, Appearance -> "Labeled"},
  {{a, 1}, -10, 10, .2, Appearance -> "Labeled"}]

epicycloid.png

Edit

Although the question only calls for epicycloids, it is very easy to make interactive panel that allows the user to play with hypocycloids as well. Because there is a simple relationship between epicycloids and hypocycloids, doing so adds very little code.

Manipulate[
  ParametricPlot[{
      type r Cos[theta] + a Cos[type r theta], 
      type r Sin[theta] + a Sin[type  r theta]},
    {theta, 0, 2 Pi},
    PlotStyle -> {Red, Thick},
    ImageSize -> {400, 400}],
  {{type, 1}, {1 -> "epicycloid", -1 -> "hypocycloid"}},
  {{r, 1}, 1, 20, 1, Appearance -> "Labeled"},
  {{a, 1}, -10, 10, .1, Appearance -> "Labeled"}]

roulettes.png

Improve this answer
$\endgroup$
13
$\begingroup$

Looking up epicycloid we get the parametric equations describing it and then ParametricPlot does the rest of our work.

ParametricPlot[
     {3 3.1 Cos[θ] - 3 Cos[3.1 θ], 3 3.1 Sin[θ] - 3 Sin[3.1 θ]}, 
     {θ, 0, 20 π},
     ColorFunction -> "AtlanticColors"]

Plot

Improve this answer
$\endgroup$
9
$\begingroup$

Just for fun, here is a variation of C. E.'s animation, which demonstrates that an epicycloid can be constructed as an envelope of the diameter of a rolling circle:

With[{n = 3, r = 1, m = 31},
     Animate[ParametricPlot[ReIm[(n + 1) r E^(I t) - r E^(I (n + 1) t)],
                            {t, 0, 2 Denominator[n] π}, Axes -> None, 
                            Epilog -> {Circle[ReIm[(n + 2) r E^(I u)], 2 r],
                                       {Red, Line[{ReIm[(n + 2) r E^(I u) -
                                                        2 r E^(I (1 + n/2) u)], 
                                                   ReIm[(n + 2) r E^(I u) +
                                                        2 r E^(I (1 + n/2) u)]}]}}, 
                            Frame -> True, PlotRange -> (n + 4) r, 
                            PlotStyle -> Red, Prolog -> {Blue, Circle[{0, 0}, n r]}],
             {u, 0, 2 Denominator[n] π, (2 Denominator[n] π)/(m - 1)}]]

epicycloid as envelope

(Note the use of the complex form of the epicycloid.)

A slight modification of the code given above can be done to produce the multiple exposure version: multiple exposure

If one just wants to see the lines enveloping the epicycloid, the code is much simpler:

With[{n = 3, r = 1, m = 101}, 
     Graphics[{Opacity[1/5], 
               Table[InfiniteLine[{ReIm[(n + 2) r E^(I u) - 2 r E^(I (1 + n/2) u)], 
                                   ReIm[(n + 2) r E^(I u) + 2 r E^(I (1 + n/2) u)]}],
                     {u, 0, 2 Denominator[n] π, (2 Denominator[n] π)/(m - 1)}]}]]

just the envelope please, Louise

Still another neat demonstration is the Bernoulli-Euler double generation theorem: an epicycloid is equivalent to a pericycloid (which can be thought of as the locus of a point mounted on a "hula hoop").

With[{n = 3, r = 1, m = 31},
     Animate[{ParametricPlot[ReIm[(n + 1) r E^(I t) - r E^(I (n + 1) t)],
                             {t, -$MachineEpsilon, u}, Axes -> None, 
                             Epilog -> {Thick, Circle[ReIm[(n + 1) r E^(I u)], r], 
                                        Line[{ReIm[(n + 1) r E^(I u)], 
                                              ReIm[(n + 1) r E^(I u) -
                                                   r E^(I (n + 1) u)]}],
                                        {Directive[Red, PointSize[Large]], 
                                         Point[ReIm[(n + 1) r E^(I u) -
                                                    r E^(I (n + 1) u)]]}}, 
                             Frame -> True, PlotRange -> (n + 2) r, 
                             PlotStyle -> Directive[Red, Thick], 
                             Prolog -> {Directive[Blue, Thick], Circle[{0, 0}, n r]}], 
              ParametricPlot[ReIm[(n + 1) r E^(I t) - r E^(I (n + 1) t)],
                             {t, -$MachineEpsilon, u}, Axes -> None, 
                             Epilog -> {Thick, Circle[ReIm[-r E^(I (n + 1) u)],
                                                      (n + 1) r], 
                                        Line[{ReIm[-r E^(I (n + 1) u)], 
                                              ReIm[(n + 1) r E^(I u) -
                                                   r E^(I (n + 1) u)]}],
                                        {Directive[Red, PointSize[Large]], 
                                         Point[ReIm[(n + 1) r E^(I u) -
                                                    r E^(I (n + 1) u)]]}}, 
                             Frame -> True, PlotRange -> (n + 2) r, 
                             PlotStyle -> Directive[Red, Thick], 
                             Prolog -> {Directive[Blue, Thick], Circle[{0, 0}, n r]}]}
             // GraphicsRow, {u, 0, 2 Denominator[n] π, (2 Denominator[n] π)/(m - 1)}]]

double generation

Improve this answer
$\endgroup$

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.