# Stability of Orbits

1. May 7, 2012

### DuckAmuck

So I saw a video on youtube: http://youtu.be/uhS8K4gFu4s

And so I thought I'd try to understand the whole stable orbit thing.

So first you have a simple energy equation of E = K + U.

K = (1/2)mv^2 (kinetic energy)
U = -GMm/r^2 (gravitational energy)

r = radius of orbit
v = velocity of orbiting object
M = mass of center object
m = mass of orbiting object

And we want it to be a stable orbit in the first place, so we have:
v=sqrt(GM/r)

Now, let's say we slightly bump the object in orbit. So r -> r + s, where s is much less than r.
(also w is much less than v)

So we can adjust the stable orbit equation:

v=sqrt(GM/r)

v+w = sqrt(GM/(r+s))

approximate for small distances:

v+w = sqrt(GM/r)(1-s/(2r))

So then, you can subtract the original equation out and have:
w = -sqrt(GM/r)s/(2r)
This makes sense directionally, if you bump the orbiting object inwards, the velocity with increase.

So now, back to the energy equation: E = K + U
Since it's a stable orbit being bumped only slightly, you expect it to be able to eventually return to the same state, so E can't change.

So initially you have:

E = (1/2)mv^2 - GMm/r^2

Then apply the bump:

E' = ((1/2)mv^2 )(1+2w/v) - (GMm/r^2 )(1-s/r)

This then reduces to the form:

E' = E + mvw + (GMm/r^2 )s

Since E' must equal E, the two extra terms should add to zero:

E' = E + (mv)sqrt(GM/r)(-s/(2r)) + (GMm/r^2 )s

E' = E + (-1/2)(GMm/r^2 )s + (GMm/r^2 )s

So I am doing something wrong. I have a missing factor of two somewhere, but I don't know why.

If you guys can help, it will be appreciated. Thanks.

2. May 8, 2012

### K^2

Stable orbit need not be circular. So if you bump it a little, it can become slightly elliptic and v=sqrt(GM/r) no longer holds. I think that's where things go wrong with your derivation.

Best check of orbit stability is looking at effective potential.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook