Harmonic Oscillator with Additional Repulsive Cubic Force: Solutions and Study

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
2 replies · 2K views
Miesvama
Messages
2
Reaction score
0
Hi all,

this is my first time on PF.

I do not know English but I have a problem of a harmonic oscillator.
I have rather large head, help me please , I do not know what else to do ...
I have this problem:

Consider the harmonic oscillator with an additional repulsive
cubic force, whose potential is U(q1)=[itex]\frac{k}{2}[/itex]*[itex]q1^{2}[/itex] - k'[itex]q1^{4}[/itex], (k, k > 0), and study all
possible solutions, periodic and non-periodic.

I do know the Hamiltonian and the equation solution of the system, giving

q1*=[itex]\sqrt{1/2}[/itex][itex]\int(dq1/\sqrt{(E/m)-U(q1)})[/itex]

I tried to do it by trigonometric substitution but does not work, i do not know if anyone could give me some idea of how I can solve, I'll be very grateful.
 
Physics news on Phys.org
Miesvama said:
Hi all,

this is my first time on PF.

I do not know English but I have a problem of a harmonic oscillator.
I have rather large head, help me please , I do not know what else to do ...
I have this problem:

Consider the harmonic oscillator with an additional repulsive
cubic force, whose potential is U(q1)=[itex]\frac{k}{2}[/itex]*[itex]q1^{2}[/itex] - k'[itex]q1^{4}[/itex], (k, k > 0), and study all
possible solutions, periodic and non-periodic.

I do know the Hamiltonian and the equation solution of the system, giving

q1*=[itex]\sqrt{1/2}[/itex][itex]\int(dq1/\sqrt{(E/m)-U(q1)})[/itex]

I tried to do it by trigonometric substitution but does not work, i do not know if anyone could give me some idea of how I can solve, I'll be very grateful.

Hi. Nobody else has responded to your question, so I guess I'll give it a try.

The equation that I think you meant is this:

[itex]t = \sqrt{m/2}\int(dq1/\sqrt{E-U(q1)})[/itex]

If we change to new variables [itex]x[/itex] and [itex]s[/itex] where
[itex]x = A q1[/itex] and [itex]s = B t[/itex], where [itex]A[/itex] and
[itex]B[/itex] are constants, we can choose the constants to make the
equation look like this:

[itex]t = \int(dx/\sqrt{(1-x^2)(1- \lambda^2 x^2)}[/itex]

where [itex]\lambda[/itex] is yet another constant.

That's called the "Jacobi form of the incomplete elliptic integral of the first kind", [itex]F(s;\lambda)[/itex].

http://en.wikipedia.org/wiki/Jacobi..._of_nonlinear_ordinary_differential_equations
 
Thank you.. :D
 
Last edited: