Ask your own question, for FREE!
Mathematics 11 Online
OpenStudy (anonymous):

Sturm-Louiville problem

OpenStudy (anonymous):

Find eigenvalues and eigenfunctions

OpenStudy (anonymous):

\[Y''+\lambda*Y\] y(0)=0, y'(3)=0

OpenStudy (anonymous):

@Kainui

OpenStudy (anonymous):

I was going to re-post it. But if anyone can help me hear that would be great

OpenStudy (kainui):

Yeah, I'm still pretty fresh with SL problems, I should be able to help you out.

OpenStudy (anonymous):

i had\[\lambda=((\pi+2n \pi)/6)^2\]

OpenStudy (kainui):

Your diffeq up there isn't equal to anything though, so I'm a little confused.

OpenStudy (anonymous):

=0

OpenStudy (anonymous):

sorry

OpenStudy (anonymous):

only deal with homogenous equations

OpenStudy (kainui):

Ok ok, just making sure you didn't mean to put a = sign where the + sign is.

OpenStudy (anonymous):

yea. so i only dealt with the case where\[\lambda>0\]

OpenStudy (anonymous):

the other two cases would mean the eigenfunctions would be 0 which weren't really relevant. Thats what i thought

OpenStudy (kainui):

Yeah, you need it to be positive for the theorem to work I believe.

OpenStudy (anonymous):

yep!

OpenStudy (kainui):

When I see this though, I immediately see that you can just solve the ODE. \[\large y=A \sin( \sqrt{\lambda}x)+B \cos( \sqrt{\lambda}x)\] Then plug y(0)=0 and y'(3)=0 to find the coefficients. From there we sum up all the solutions and then use the orthogonality to solve for the eigenvalues I believe. Maybe I'm missing something, it's been a while since I've done these and I'm still working it out.

OpenStudy (anonymous):

pellet. i didn't do sqrt lambda

OpenStudy (kainui):

Haha yeah I've done that more than I care to admit too.

OpenStudy (anonymous):

wait isn't it the same if i let \[\lambda=\]

OpenStudy (anonymous):

i just have\[y(x)=Acos(\mu x)+Bsin(\mu x)\]

OpenStudy (anonymous):

\[\lambda=\mu^2\]

OpenStudy (kainui):

Yeah that's fine to do it that way, that's what I usually do.

OpenStudy (anonymous):

yea. so then with y(0)=0 it implies A=0. so \[y(x)=Bsin(\mu x)\]

OpenStudy (anonymous):

and differentiating that you get \[y(x) \prime=B \mu \cos(\mu x)\]

OpenStudy (kainui):

Yep you got it.

OpenStudy (anonymous):

and then applying the boundary condition \[0=B \mu \cos(3\mu)\]

OpenStudy (anonymous):

and thus \[B \mu=0\] OR\[\cos(3 \mu)=0\]

OpenStudy (anonymous):

but we deal with the second equation since it would be silly if B=0

OpenStudy (anonymous):

so \[3\mu=\pi/2+n \pi \]

OpenStudy (kainui):

Yeah, exactly. You've found your eigenvalue mu_n, you're golden now.

OpenStudy (anonymous):

but when i type it in to the online quiz it says its still wrong....

OpenStudy (anonymous):

so \[\lambda=\mu^2\]

OpenStudy (anonymous):

and you just square mu

OpenStudy (kainui):

Hmm well can you take a screen shot or something? Maybe you're just plugging it in wrong. This is what I'm getting: \[y= \sum_{n=0}^\infty A_n \sin (\sqrt{\lambda_n} x)= \sum_{n=0}^\infty A_n \sin (\frac{\pi(2n+1)}{6}x)\]

OpenStudy (anonymous):

OpenStudy (kainui):

Hmm, maybe if you rewrite it as: \[\large \lambda_n=\frac{\pi^2(2n+1)^2}{36}\] it will be better? I don't know, that's really weird to me. I'm gonna take a bathroom break and relook at our work to see if we've messed up somewhere.

OpenStudy (anonymous):

No worries! I'll try that.

OpenStudy (anonymous):

Yea. Still wrong. Sure we didn't do it wrong if we both got the same answer

OpenStudy (kainui):

Maybe this is where we went wrong, with the cosine. \[3x=\pi(2n-1)\] If they're starting their index at n=1 instead of n=0, then this will make up for the term we're leaving out.

OpenStudy (anonymous):

at n=0 \[\mu=\pi/6 \] right

OpenStudy (kainui):

Yeah, but they start out with n=1 then we're essentially starting at (2n+1) which gives 3. Since they're using n=1 with (2n-1) then we have 1 for the first number. I hope that makes sense of why. Just put a negative sign in there and you'll be right I promise.

OpenStudy (anonymous):

SUCCESS

OpenStudy (anonymous):

let me think about this

OpenStudy (anonymous):

ah so our index shift will be n--->n-1 so that we start from n=1 right? AH MAKES SENSE

OpenStudy (anonymous):

so we end up have in the numerator 2(n-1)+1 2n-2+1 2n-1

OpenStudy (anonymous):

ah awesome

OpenStudy (anonymous):

so your eigenfunction would therefore be \[F _{n}=\cos(\lambda x)\]

OpenStudy (anonymous):

is that correct?

OpenStudy (anonymous):

wait no thats wrong

OpenStudy (anonymous):

\[F _{n}=\sin(\sqrt{\lambda}x)\]

OpenStudy (anonymous):

ah awesome. i got it out! Thanks alot!

OpenStudy (kainui):

Haha awesome. Yeah, just a subtle little thing always has to mess it up haha.

OpenStudy (anonymous):

No kidding right! Thanks alot

OpenStudy (anonymous):

could you help me with this question?

OpenStudy (anonymous):

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!