1. Limited time only! Sign up for a free 30min personal 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!

Homework Help: 1-D Wave equation with mixed boundary conditions

  1. Dec 12, 2009 #1
    1. The problem statement, all variables and given/known data
    Solve, [tex]u_{t} = u_{xx}c^{2}[/tex]

    given the following boundary and initial conditions

    [tex]u_{x}(0,t) = 0, u(L,t) = 0[/tex]

    [tex]u(x,0) = f(x) , u_{t}(x,0) = g(x)[/tex]

    2. Relevant equations

    [tex]u(x,t) = F(x)G(t)[/tex]

    3. The attempt at a solution
    I solved it, I am just not sure if it is right.

    [tex]u(x,t) = \sum_{n=1}^\infty(a_{n}cos(\lambda_{n}t) + b_{n}sin(\lambda_{n}t))cos((n-\frac{1}{2})\frac{\pi}{L}x)

    , \lambda_{n} = (n-\frac{1}{2})\frac{\pi}{L}c [/tex]

    [tex]a_{n} = \frac{2}{L}\int_0^L f(x)cos((n-\frac{1}{2})\frac{\pi}{L}x)dx,

    b_{n} = \frac{4}{(2n-1)c\pi}\int_0^L g(x)cos((n-\frac{1}{2})\frac{\pi}{L}x)dx

    Can someone please verify this for me?

    Thanks in advance,
    Last edited: Dec 12, 2009
  2. jcsd
  3. Dec 13, 2009 #2


    User Avatar
    Science Advisor

    Why does your formula for bn[/b] involve cosine rather than sine?
  4. Dec 13, 2009 #3
    Thanks for replying HallsofIvy.

    Well to get [tex]b_{n}[/tex] I applied the initial condition

    [tex]u_{t}(x,0) = g(x).[/tex]

    So I first find [tex]u_{t}[/tex] which comes to

    [tex]u_{t} = \sum_{n=1}^\infty(-a_{n}\lambda_{n}sin(\lambda_{n}t) + b_{n}\lambda_{n}cos(\lambda_{n}t))cos((n-\frac{1}{2})\frac{\pi}{L}x) [/tex]

    now evaluating [tex]u_{t}[/tex] at t = 0 and setting it equal to g(x) we get,

    [tex]u_{t}(x,0) = \sum_{n=1}^\infty b_{n}\lambda_{n}cos((n-\frac{1}{2})\frac{\pi}{L}x) = g(x) [/tex]

    But that is just the Fourier cosine series of g(x) just with the extra [tex]\lambda_{n}[/tex] and with the zero term equal to zero

    so [tex]b_{n} = \frac{2}{\lambda_{n}}\int_0^L g(x)cos((n-\frac{1}{2})\frac{\pi}{L}x)dx[/tex]

    where [tex]\lambda_{n} = (n-\frac{1}{2})\frac{\pi}{L}c

    [tex] \Rightarrow b_{n} = \frac{4}{(2n-1)c\pi}\int_0^L g(x)cos((n-\frac{1}{2})\frac{\pi}{L}x)dx[/tex]

    Please let me know if I did something wrong.

    Thanks again,
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook