Orrery Designs using LEGO® Elements

Since 2007 I have worked on Orrery designs using LEGO elements. This began when I looked at the NASA Kepler Mission website and found their educational material including a miniature tabletop "solar system" designed (in 2004) to demonstrate how the Kepler mission will work.

I liked their model so much I decided to work on a few improvements, then began developing my own completely new designs.

As this "page" has gotten rather long, I have broken up the material into a few pages, organized chronologically (oldest first).

Table of Contents

History

3-Planet Version (Jovilabe)

Stacked Turntables method 2

Constrained models (using parts from only two brand-new sets)

Model F: "Constrained" 5-Planet Design

G: Adding Moons and Refinements (Off-Road Truck rev. 2)

H: Finishing the Inner Planets (Off-Road Truck rev. 3)

More Fancy Ideas

Stacked Turntables method 3

The Hailfire Model: Extreme Co-planar Gears

Small Model 2C

Miniature 3-moon Jovilabe

Robust and Modular Designs

Stacked Turntables: Two-Module Prototype

Stacked Turntables: Four-Module Prototype

Appendices

Pictures of Special Parts

Script for the model A movie

Script for model B2 movie

Script for model D YouTube video

Script for Geartrain Module video

References, Links, etc.

History: Previous Versions of the Orrery

2- and 3-planet versions

20040915

My 2004 designs are descended from versions by Allan Ayres and Dave Koch. The first two were 2-planet and 3-planet designs by Ayres. I know little of the 2-planet version, but I suspect it was like its successor but without the middle planet.

Here is a picture of the 3-planet design; the building instructions for it are here: transitdemo.pdf.

A1. Ayres' 3-planet orrery

A large Technic turntable, whose base remains fixed, is made to turn through epicyclic gearing. 4 rotations of the inner shaft makes the turntable top rotate once. The 3^rd planet is attached to the turntable, and the 1^st planet to the central shaft. Another gear on the turntable makes the 2^nd planet turn at a rate twice that of the 3^rd.

The orbit period ratios can be measured easily just by turning the crank a few times. Here are the periods; I have also shown a "radius" computed by the formulas discussed here.

planet	velocity	period	radius
1^st	1	1	6 (datum)
2^nd	1/2	2	9.5 ({~=} 6 × 2^(2/3))
3^rd	1/4	4	15 ({~=} 6 × 4^(2/3))

3-Moon Jovilabe

These orbital periods might seem a little artificial, but it does actually happen in real life — Jupiter's moons Io, Europa and Ganymede have exactly this ratio of periods, due to orbital resonance. You can approximate their orbits by setting up the model so that the 1^st and 3^rd planets are aligned and the 2^nd is all the way around on the other side:

A2: Representing Io, Jupiter, Europa and Ganymede

Diameter of Jupiter and distances of moons are to scale. The yellow dot on the base indicates the direction that was towards the Sun when the photo was taken (early 2008); the black dot is 2007 and the gray dots are 2009, 2010, etc.

Ayres/Koch 4-planet version

20050729

You may notice in the above 3-planet designs, there is a gear on the upper part of the differential housing that doesn't come into contact with any other gear. That gear provides an opportunity to add a fourth planet, and Dave Koch did exactly that. He took Ayres' 3-planet design and modified it to look roughly like this:

B1. My Rendering of the Ayres/Koch 4-Planet Orrery (without moon)

(Here, I have added half-stud adjustments to make the planet distances more closely conform with Kepler's 3^rd law (discussed below).)

Then he added a moon:

Koch's version (with moon on 2^nd planet)

B2. My Version, With Moon

These 4-planet designs incorporate two modifications over the 3-planet design:

A new planet is added, carried by an arm with epicyclic gears that compute a weighted average of the rotations of its neighbors (details below). What were the 2^nd and 3^rd planets are now the 3^rd and 4^th, respectively.

The differential housing is turned upside down, and gearing changed so that its rotation is slower (but still faster than that of the 4^th planet).

The instructions to build this model are here: The Kepler Transit Demonstration

{Updated instructions for the Kepler project's 4-planet orrery are here: LEGOorrery2011.pdf -RPM 20121223}

Kepler's Third Law

To make the model more accurate, one should adjust the lengths of the arms so that the planets' distances are related to their orbital periods by the formula

(^P/_2π)² = ^a³/_(G(M+m))

where

P is the orbital period,
a is the semimajor axis of the orbit (the radius, for a circle)
G is the universal gravitational constant
M and m are the masses of the sun and planet

for our purposes we can simplify it to

a = K P^2/3

where K is an arbitrary constant, which can be 1 if you express P and a as multiples of one planet's period and radius. So for example, knowing that Mars has an orbital period 1.88 times that of Earth, the formula tells us that its semimajor axis is 1.88^2/3 {~=} 1.523 times that of Earth.

All of the models above (A and my versions of B) are built with the proper distances to within ±2mm.

20071212

B's orbit periods are a bit more complex than A's. The turntable turns the same way as before, through epicyclic gearing. 12 turns of the crank make it go around once. This is one orbit of the 4^th planet. 36 turns of the crank is the time it takes the 4^th planet to orbit 3 times. In this period the 3^rd planet orbits 4 times, the 2^nd planet orbits 7 times, and the innermost planet orbits 12 times. I call this a syzygy cycle because if you start the orrery with all planets aligned, it takes this long for them all to become aligned again.

Here is a table showing the planet orbital rates and periods, and the radii used in Koch's model (K) and my two models (M₁ and M₂):

planet	velocity	period	K	M₁	M₂
1^st	4	1	10.5	6	7 (datum)
2^nd	7/3	12/7	14	8.5	10 ({~=} 7 × (12/7)^(2/3))
3^rd	4/3	3	20.5	12	14.5 ({~=} 7 × 3^(2/3))
4^th	1	4	25.5	15	18 ({~=} 7 × 4^(2/3))

20071213

The number 7 in the above ratios might come as a bit of a surprise, if you consider that the gears (with 8, 16, 24 and 40 teeth) cannot by themselves produce any ratios involving the prime number 7.

The 7 (and other prime factors in the more complicated ratios below) comes from epicyclic gearing formulas that involve addition and subtraction, and division of ratios.

Consider first the drive of the 4^th planet:

B2's Base.

This is a planetary gear mechanism (a type of epicyclic gearing). The base (black) is the annulus, and has 24 teeth. The planet-carrier (clear, and gray) carries the planet gear (blue) with it. The planet gear has 8 teeth. The sun gear (yellow) has 8 teeth. In this particular setup, the base is fixed, the sun is the input, and the planet-carrier is the output.

Let R_p be the rate of rotation of the planet-carrier, and R_s be the rate of rotation of the sun gear. R_b, the rate of rotation of the base, is fixed at zero.

It is easiest to compute the gear ratio from within the rotating reference frame of the planet. Relative to the planet, the base is rotating at -R_p and the sun is rotating at R_s - R_p. The base has 24 teeth, the sun has 8, and the planet acts as an idler gear. Because the base's gear is inside-out (teeth facing in) a direction reversal is added. The gear ratio is -1×24/8, or -3/1. Therefore, the motion of the planet-carrier and sun will fit the following formula:

R_s - R_p = -3 (-R_p)

and therefore (by algebra)

R_s = R_p + 3 R_p = 4 R_p

This shows that the 3^rd planet in model A will revolve once for each 4 times the 1^st planet revolves.

Notice that we get a ratio of 4, even though the gears we used (24 and 8) have a ratio of 3.

Now consider the transfer of motion to the 3^rd planet:

B2 Transfer from 4th to 3rd
B2 Transfer from 4^th to 3^rd

We will use R₄ now to refer to the rotation rate of the 4^th planet arm (R_p above), R₁ instead of R_s, and R₃ for the 3^rd planet arm. Viewed from the rotating reference of the 4^th planet's arm, this is an ordinary gear train. The sun gear (yellow) with 8 teeth drives a 24-tooth gear, which shares an axle with an 8-tooth gear (gray) which drives the output, a 24-tooth gear (dark gray). The input is R₁-R₄, the output is R₃-R₄, and the ratio is (-24/8)×(-24/8) = 9/1. Therefore:

R₁-R₄ = 9 (R₃-R₄)

we already know from above that R₁ = 4 R₄. So we have

3 R₄ = 9 R₃ - 9 R₄

which gives

R₃ = 4 R₄ / 3

Still a fairly simple ratio, no strange prime numbers here. But let's see what happens when we go up to the 2^nd planet:

B2's 2nd Planet Between 1st and 3rd
B2's 2^nd Planet Between 1^st and 3^rd

The 2^nd planet's arm rotates at rate R₂, and has gears mounted on it. The dark gray gear is turning at R₃ — but from the 2^nd planet's point of view, it rotates at R₃-R₂ (which is negative, hence backwards). The yellow gear rotates at R₁, which is R₁-R₂ from the 2^nd planet's point of view. An extra 16-tooth idler ensures that the gear train has only one direction reversal. The gear train will obey the equation

R₁-R₂ = (-40/24)×(16/16)×(R₃-R₂)

Which gear is the input? We already know that R₁ and R₃ are determined as just described. Therefore, they are both inputs, and R₂, the 2^nd planet's arm, is the output. The equation reduces as follows:

R₁-R₂ = -5 (R₃-R₂) / 3

4R₄ - R₂ = -5 (4R₄/3 - R₂) / 3

R₂ - 4R₄ = 20R₄/9 - 15R₂/9

24 R₂ / 9 = 56 R₄ / 9

R₂ = 7 R₄ / 3

There's that surprise prime number 7. Every time the 4^th planet completes 3 orbits, the 2^nd planet completes 7 orbits.

. . . Forward to page 2 . . . Last page (page 7)

LEGO® creations index

The graph paper in my newer photos is ruled at a specing of 1 LSS, which is about 7.99 mm.

This site is not affiliated with the LEGO® group of companies.

LEGO®, Duplo®, QUATRO®, DACTA®, MINDSTORMS®, Constructopedia®, Robotics Invention System® and Lego Technic®, etc. are trademarks or registered trademarks of LEGO Group. LEGO Group does not sponsor, authorize or endorse this site.

All other trademarks, service marks, and copyrights are property of their respective owners.

If you want to visit the official LEGO® site, click here

Parts images are from LUGNET. On this page they explicitly give permission to link to the images:

Note: you may link (as in, Yes, it's OK) directly to these parts images from an off-site web page.

This page was written in the "embarrassingly readable" markup language RHTF, and some sections were last updated on 2015 Jan 11.

s.27