1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Laplace's equation on a ring

  1. Jan 16, 2012 #1

    A_B

    User Avatar

    1. The problem statement, all variables and given/known data
    Solve Laplace's equation on a ring in the plane with r_1<r<r_2. And arbitrary functions on the edges of the ring.


    3. The attempt at a solution

    After separation of variables the solution to the radial factor is an often-seen problem. It's eigenvalues are λ=n² the eigenfunctions Θ(θ)~sin(nθ) + cos(nθ).

    The differential equation for the radial part becomes

    r²R" + rR' - n²R = 0

    I don't know how to solve this thing but my book gives solutions:

    n=0: r → 1 and r → ln(r)
    n>0: r → r^n and r → r^(-n)

    In my book the functions ln(r) and r^(-n) are thrown away because they are singular in r=0. But for this problem r=0 is not part of the domain, so I should keep all these solutions.

    Now I wonder, since ln(r) and r^(-n) can both be written as a Taylor series on the interval [r_1,r_2], can I throw them away anyways?

    If so, the solution will be of the form

    u(r,θ) = Ʃ A_n r^n cos(nθ) + Ʃ B_n r^n sin(nθ)

    Boundary condtions

    u(r_1,θ) = Ʃ A_n r_1^n cos(nθ) + Ʃ B_n r_1^n sin(nθ)
    u(r_2,θ) = Ʃ A_n r_2^n cos(nθ) + Ʃ B_n r_2^n sin(nθ)

    But this seems to be fixing A_n and B_n twice, since both are Fourier series.
    Thanks
    A_B
     
  2. jcsd
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Can you offer guidance or do you also need help?