Scribd
Upload
7 views2 pages

Computational Problem Set 1

This document outlines Computational Problem Set 1 for the Astronomy course at MIT, focusing on the Kepler problem and its equations of motion. It provides a suggested procedure for solving the problem using a 4th-order Runge-Kutta integration scheme and discusses the integration of orbits for both single and two-body systems. The assignment consists of three parts, including the integration of circular and eccentric orbits, the computation of orbits for two objects of unequal mass, and the analysis of precessing orbits using a post-Newtonian potential.

Uploaded by

vhghuy0511
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views2 pages

Computational Problem Set 1

This document outlines Computational Problem Set 1 for the Astronomy course at MIT, focusing on the Kepler problem and its equations of motion. It provides a suggested procedure for solving the problem using a 4th-order Runge-Kutta integration scheme and discusses the integration of orbits for both single and two-body systems. The assignment consists of three parts, including the integration of circular and eccentric orbits, the computation of orbits for two objects of unequal mass, and the analysis of precessing orbits using a post-Newtonian potential.

Uploaded by

vhghuy0511
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

Physics Department
Earth, Atmospheric, and Planetary Sciences Department

Astronomy 8.282J­12.402J February 28, 2006

Computational Problem Set 1


The Kepler Problem

Optional

Suggested Procedure:
The differential equation of motion for the Kepler problem, in vector form, is:

d2�r GM r̂ GM�r
=− 2 =− 3 (1)
dt2 r r
In Cartesian coordinates we have:
d2 x GM x
2
=− 3 (2)
dt r
d2 y GM y
=− 3 (3)
dt2 r
As four, first­order, coupled equations:
dx
= vx (4)
dt
dy
= vy (5)
dt
dvx GM x
= − (6)
dt r3
dvy GM y
= − 3 (7)
dt r
Equations (4) through (7) can be solved by a simple 4th­order Runge­Kutta integration scheme.
A few pages of text describing this method, taken from the Numerical Recipes book, are attached.
In the language of the RK4 program (Numerical Recipes’ Runge­Kutta program):

Y = the four independent variables (x, y, vx , vy )


X = the time t
N = 4 (8 for the case of the two­body problem)
H = the time step Δt
DERIVS = user supplied analytic derivatives, [right hand sides of (4) through (7)]
YOUT = are the incremented Y variables after time step H

Or, you can use the canned integrators in mathematical packages such as Matlab, Mathematica,
IDL, etc.

1
Before carrying out any integrations, it is convenient to cast the equations in dimensionless
form. One possible choice of scaling parameters is:
[length] → a (semimajor axis)
[mass] → M
[speed] → ΩK a
[time] → 1/Ω�K
where ΩK ≡ GM/r3 .

The time steps can be made either fixed (e.g., some small fraction of 1/ΩK ), or variable,
depending on how rapidly the motion is changing. One possible variable time­step scheme is:

|�r| |�r1 − �r2 |


Δt = ξ or ξ , (8)
|v�| |�v1 − �v2 |

depending on whether you are integrating the one­ or two­body problem. The choice of ξ basically
determines the approximate number of integration steps that will be taken around an orbit. A
choice of ξ = 0.01 should produce rather accurate results.
For all three parts, you should display your results graphically.
Part 1
For the case of an infinite central mass and an orbiting test mass, integrate a circular orbit and
an eccentric orbit. Carry out the integration for several orbits, in both the circular and eccentric
orbit cases, to verify that the orbits close. Check that energy and angular momentum are conserved
around the orbit.
Part 2
Compute the orbits of two objects of comparable (but unequal) mass orbiting their common
center of mass. Carry out the calculations for both circular and eccentric orbits.
Part 3
Compute the precessing orbit of the mass in Part 1, if the potential is of the form
GM
φ=− (9)
(r − Rg )

where Rg ≡ 2GM/c2 , is the Schwarzschild radius associated with mass M . This is a post­Newtonian
potential suggested by Paczynski (1980; A&A, 88, 23) that approximates General Relativistic dy­
namics in strong gravitational fields. Treat every other aspect of the problem with simple Newtonian
mechanics.
Consider cases where a/Rg = 5, 10, and 50. Make the orbit eccentric enough to be able to
discern the precession, but avoid periastron distances near 3 Rg .
Expand the potential (or the force) in a series involving terms in Rg ; retain only the first
post­Newtonian term. Use this form of the potential to find an analytic expression for the angular
precession per orbit. Compare the analytic and numerical results.
Finally, for fun, try to produce a circular orbit with r � 3 Rg , and with r � 3 Rg , and contrast
the results.

You might also like