Results 1 to 6 of 6

Math Help - ODE Problem

  1. #1
    Newbie
    Joined
    Jan 2010
    Posts
    6

    ODE Problem

    I have this ODE:
    d^2t/dx^2 = -S/K where S and K are constants. Boundary conditions are: T=T1 at x=0 T=T2 at x=L.

    It looks so simple, but i'm having some issues. I used the auxiliary equation first to get the complimentary part: m^2=0, real roots at 0,0 means an equation of form: Tc=A+Bx with A,B as unknown constants.

    Next, i'm trying to get the particular solution using the method which I believe is MUC. This is where i'm a bit confused. I assumed a solution in the form of: Tp=-S/K (constant). But then my general solution becomes: T(x)=A+Bx-S/K... and I don't think I can have two constants (both A and S/K) in there since one would be absorbed.

    In this case, would I assume Tp=-S/K*x^2 since a constant already exists? But if I do this, I wind up with:
    T = A + Bx -S/K*x^2
    T(0) = T1 = A
    T(L) = T2 = T1 + B*L - S/K*L^2 -> B = (T2/L - T1/L + S/K*L)
    T(x)=T1 + (T2/L - T1/L + S/K*L)*x - S/K*x^2 which i'm being told is wrong, and that each S/K term should be S/(2K). I can't see why though.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    MHF Contributor
    Joined
    Mar 2010
    From
    Florida
    Posts
    3,093
    Thanks
    5
    You were correct, sorry.

    m^2=0

    t(x)_c=C_1e^{0*x}+C_2xe^{x*0}\Rightarrow t(x)=C_1+C_2x

    The differential operator D annihilates constants.

    D(D^2)=0\Rightarrow D^3=0\Rightarrow m^3=0\Rightarrow m_1=m_2=m_3=0

    t(x)=C_1+C_2x+C_3x^2

    \displaystyle t''(x)=-\frac{s}{k}=2C_3\Rightarrow C_3=-\frac{s}{2k}

    \displaystyle t(x)=C_1+C_2x-\frac{s}{2k}x^2
    Last edited by dwsmith; January 11th 2011 at 05:19 PM. Reason: Added some math.
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Jan 2010
    Posts
    6
    edit: Nevermind, I figured it out.

    What I was doing wrong was putting Tp=-S/K*x^2 instead of Tp=Cx^2 (leaving a constant in.) At first I thought I couldn't solve it because I was only given 2 initial conditions, but then I figured out that I also had the original DE that I could use to solve for the constant C.
    I plugged Tp into the original diff eq after taking the derivative twice, and that gives me the correct answer.

    Thanks
    Follow Math Help Forum on Facebook and Google+

  4. #4
    MHF Contributor
    Joined
    Mar 2010
    From
    Florida
    Posts
    3,093
    Thanks
    5
    What is MUC?
    Follow Math Help Forum on Facebook and Google+

  5. #5
    MHF Contributor
    Prove It's Avatar
    Joined
    Aug 2008
    Posts
    11,570
    Thanks
    1428
    This can be solved using direct integration...

    \displaystyle \frac{d^2t}{dx^2} = -\frac{S}{K}

    \displaystyle \frac{dt}{dx} = \int{-\frac{S}{K}\,dx}

    \displaystyle \frac{dt}{dx} = -\frac{S\,x}{K} + C_1

    \displaystyle t = \int{-\frac{S\,x}{K} + C_1\,dx}

    \displaystyle t = -\frac{S\,x^2}{2K} + C_1\,x + C_2.
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Rhymes with Orange Chris L T521's Avatar
    Joined
    May 2008
    From
    Chicago, IL
    Posts
    2,844
    Thanks
    3
    Quote Originally Posted by dwsmith View Post
    What is MUC?
    The Method of Undetermined Coefficients.

    I noticed you helped the user using the annihilator approach. In essence, the method of the undetermined coefficients and annihilator method are the "same". The thing that makes them different is that in the annihilator method, you get the proper (general) form of the particular solution without having to guess at it. Using MUC, you would have to guess at the general form based off of your solution to the associated homogeneous equation and any nonhomogeneous terms that may appear in the ODE.
    Last edited by Chris L T521; January 11th 2011 at 09:36 PM.
    Follow Math Help Forum on Facebook and Google+

Search Tags


/mathhelpforum @mathhelpforum