MHB Orthogonality of Eigenfunctions of Mixed Boundary Conditions

Click For Summary
The discussion focuses on proving the orthogonality of eigenfunctions under mixed boundary conditions defined by constants a, b, c, and d. The key equation to analyze is derived from the boundary conditions and the eigenvalue problem, leading to the relationship involving the integral of the product of eigenfunctions. The participants express uncertainty about how to proceed with the constants affecting the boundary conditions but note that if the constants were absent, the orthogonality could be shown directly. The conclusion drawn is that the eigenfunctions are orthogonal if the integral of their product over the specified domain equals zero, which is contingent on the distinct eigenvalues. The conversation emphasizes the mathematical manipulation required to handle the boundary conditions effectively.
Dustinsfl
Messages
2,217
Reaction score
5
$$
\left.(\phi_n\phi_m' - \phi_m\phi_n')\right|_0^L + (\lambda_m^2 - \lambda_n^2)\int_0^L\phi_n\phi_m dx = 0
$$
where $\phi_{n,m}$ and $\lambda_{n,m}$ represent distinct modal eigenfunctions which satisfy mixed boundary conditions at $x = 0,L$ of the form
\begin{alignat*}{3}
a\phi(0) + b\phi'(0) & = & 0\\
c\phi(L) + d\phi'(L) & = & 0
\end{alignat*}
Show that the eigenfunctions are orthogonal.
$$
\int_0^L\phi_m\phi_m dx = 0.
$$
I not sure how to proceed since there are constants a,b,c,d.

If they weren't there, I would proceed as
$$
(\lambda_m^2 - \lambda_n^2)\int_0^L\phi_n\phi_m dx = -\left.(\phi_n\phi_m' - \phi_m\phi_n')\right|_0^L
$$
\begin{alignat*}{3}
\phi(0) &= &-\phi'(0) \\
\phi(L) &=& -\phi'(L)
\end{alignat*}
Therefore,
$$
(\lambda_m^2 - \lambda_n^2)\int_0^L\phi_n\phi_m dx = 0\iff \int_0^L\phi_n\phi_m dx = 0
$$
 
Physics news on Phys.org
dwsmith said:
$$
\left.(\phi_n\phi_m' - \phi_m\phi_n')\right|_0^L + (\lambda_m^2 - \lambda_n^2)\int_0^L\phi_n\phi_m dx = 0
$$
where $\phi_{n,m}$ and $\lambda_{n,m}$ represent distinct modal eigenfunctions which satisfy mixed boundary conditions at $x = 0,L$ of the form
\begin{alignat*}{3}
a\phi(0) + b\phi'(0) & = & 0\\
c\phi(L) + d\phi'(L) & = & 0
\end{alignat*}
Show that the eigenfunctions are orthogonal.
$$
\int_0^L\phi_m\phi_m dx = 0.
$$
I not sure how to proceed since there are constants a,b,c,d.

If they weren't there, I would proceed as
$$
(\lambda_m^2 - \lambda_n^2)\int_0^L\phi_n\phi_m dx = -\left.(\phi_n\phi_m' - \phi_m\phi_n')\right|_0^L
$$
\begin{alignat*}{3}
\phi(0) &= &-\phi'(0) \\
\phi(L) &=& -\phi'(L)
\end{alignat*}
Therefore,
$$
(\lambda_m^2 - \lambda_n^2)\int_0^L\phi_n\phi_m dx = 0\iff \int_0^L\phi_n\phi_m dx = 0
$$
You know that $a\phi_m(0) + b\phi'_m(0) = 0$ and $a\phi_n(0) + b\phi'_n(0) = 0$. Therefore $\dfrac{\phi'_m(0)}{\phi_m(0)} = \dfrac{\phi'_n(0)}{\phi_n(0)} = -\dfrac ab.$ It follows that $\phi_n(0)\phi_m'(0) - \phi_m(0)\phi_n'(0) = \phi_n(0)\phi_m(0)\Bigl(\dfrac{\phi'_m(0)}{\phi_m(0)} - \dfrac{\phi'_n(0)}{\phi_n(0)}\Bigr) = \phi_n(0)\phi_m(0)\Bigl(-\dfrac ab + \dfrac ab\Bigr) = 0.$ Similarly at the other endpoint $L$. I leave you to fret about what happens if there are any zeros on the denominators of those fractions.
 
Opalg said:
It follows that $\phi_n(0)\phi_m'(0) - \phi_m(0)\phi_n'(0) = \phi_n(0)\phi_m(0)\Bigl(\dfrac{\phi'_m(0)}{\phi_m(0)} - \dfrac{\phi'_n(0)}{\phi_n(0)}\Bigr) $

How did this come about?
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 8 ·
Replies
8
Views
4K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K