I would like to solve the equations of motion with the Lagrangian function for two point-bodies that interact gravitationally via the potential $$V= {-Gm_1m_2 \over r_{12}} $$ where $$r_{12} = **r_1 - r_2** $$ the distance between them with the beginning of the axes on a point different than the bodies.

I work on polar coordinates.With $$L=T-V$$and $$T= {{1} \over {2}} m_1 (x_1 '+y_1') + {{1} \over {2}} m_2 (x_2 '+y_2') .$$

I have $$x_1=r_1 \cos u_1 ~~y_1=r_1\sin u_1 $$ and respectively for the second body. Also, for convenience $$θ=u_1 - u_2$$

If I haven't any mistakes then the Lagrangian is $$ L={1 \over 2 } m_1 ( {r_1 '}^2 + {r_1}^2 {u_1 '}^2 ) + {1 \over 2 } m_2 ( {r_2 '}^2 + {r_2}^2 {u_2 '}^2 ) + {Gm_1 m_2 \over [{r_1}^2 +{r_2}^2 + 2r_1 r_2 \cos θ ]^{1/2} }$$

If the above is correct, then for the first body we have $$r_1 '' - r_1 {u_1 ' } ^2 + {Gm_2 (r_1 + r_2 cosθ) \over [{r_1}^2 + {r_2}^2 + 2r_1 r_2 cosθ]^{3/2}}=0 ~~~~(1) $$ and $$r_1 ' u_1 ' +{r_1}^2 u_1 '' - {Gm_2 sinθ \over 4[{r_1}^2 + {r_2}^2 + 2r_1 r_2 \cos θ]^{3/2}}=0 ~~~ (2)$$Respectively we can find the equation for the second body

Even with the assumption that $u_i '' =0$ (which eliminates the second term in (2) ) the system is very difficult.

The question is this (I don't ask the solution of this problem):

  1. Is there an analytical solution to this system? If yes:

  2. Is there a way to make it simpler? (without changing the beginning of the axons)

  3. Shouldn’t this system have at least angular momentum as a constant of motion?

I'm dealing with this problem because I would like to understand not only in principle or theoretically, but to see , through the equations how the motion is dependable on the reference frame(for example that the earth doesn't move around the sun but that in another coordinate system - like that of the example- we have a different but equivalent understanding of the system's motion - like a geocentric system in another example).

Note: The tones are for time derivatives

  • 2
    $\begingroup$ You'll save yourself a lot of time and energy by replacing ${\vec r_{1}}$ and ${\vec r_{2}}$ with ${\vec r} = {\vec r_{1}} - {\vec r_{2}}$ and ${\vec R} = \frac{1}{m_{1}+m_{2}}\left(m_{1}{\vec r_{1}} + m_{2}{\vec r_{2}}\right)$ $\endgroup$ Commented Apr 17, 2015 at 16:58
  • $\begingroup$ If you do this, then you'll see that your answers are yes, yes, and yes $\endgroup$ Commented Apr 17, 2015 at 16:58
  • $\begingroup$ @Jerry Schirmer I should do this before deriving the equations or is it irrelevant? $\endgroup$ Commented Apr 17, 2015 at 17:09
  • $\begingroup$ It will make deriving the equations a lot easier if you do the substitution first. $\endgroup$ Commented Apr 17, 2015 at 17:38

1 Answer 1


The two body problem can be made equivalent to the one body problem. Say you have a mass $m_1$ at $\vec{r}_1$ and a mass $m_2$ at $\vec{r}_2$ interacting gravitationally. Now focus your attention on the center of mass $\vec{r}$ and the difference vector $\vec{r}_{12}=\vec{r}_2-\vec{r}_1$. Then the kinetic energy of the system can be written as $$K=\frac{1}{2}(m_1+m_2)\dot{\vec{r}}^2+\frac{1}{2}\frac{m_1m_2}{m_1+m_2}\dot{\vec{r}_{12}}^2$$. Since the potential energy doesn't depend on the the center of mass, the $\vec{r}$ coordinates will be cyclic and you can drop it from the Lagrangian. Then you will only have a Lagrangian in terms of $\vec{r}_{12}$ which is the same as that of a body under a central potential with mass $\frac{m_1m_2}{m_1+m_2}$. See the treatment given in Goldstein's Classical Mechanics


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.