Light in an optical fiber - variational calculus

Click For Summary

Homework Help Overview

The discussion revolves around a variational calculus problem related to light propagation in optical fibers. Participants are exploring the setup of the problem, particularly focusing on the functional and the Euler-Lagrange equations involved in minimizing the path length of light.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning, Problem interpretation

Approaches and Questions Raised

  • Participants are attempting to set up the functional for the problem and apply the Euler-Lagrange equation. There are questions about the treatment of variables and the implications of certain assumptions, such as the behavior of the refractive index and the independence of the variable "y".

Discussion Status

The discussion is active, with participants providing insights and clarifications on the formulation of the problem. Some have offered alternative perspectives on the functional and its derivatives, while others are questioning the assumptions made regarding the variables involved.

Contextual Notes

There are indications that certain approximations are being considered, such as assuming a small ratio of p/a, which may influence the choice of the refractive index function. Additionally, the role of "y" as the principal variable is being emphasized, which affects how the functional is expressed.

physman55
Messages
13
Reaction score
0

Homework Statement



e5fkb5.jpg

Homework Equations



[tex]\frac{\delta{F}}{\delta{y}} - \frac{d}{dx}\frac{\delta{F}}{\delta{y'}} = 0[/tex]

The Attempt at a Solution



I'm having trouble setting this one up. If I let the functional be

[tex]F(x,x',y) = n(y)\sqrt{1+(x')^2}[/tex]

Applying the LE equation I obtain:

[tex]\frac{d}{dy}\frac{x'}{\sqrt{1+(x')^2}}n_0(1-\frac{y^2}{a^2})=0[/tex]

And this is where I stopped (I haven't used either of the hints). Any hints?
 
Physics news on Phys.org
Any ideas please? If p/a is small then I could write n(y)=n_0; but then I'm minimizing path length... so the answer is a straight line. Obviously this problem couldn't be that easy.
 
it tells you y should be the principal variable so your functional should read

[tex]F(y,y')[/tex]= [tex]n(y)[/tex][tex]\sqrt{1+y'^{2}}[/tex] and then calculate the EL equations

by my calculations you should get

[tex]\frac{\partial}{\partial x}[/tex][tex](\frac{y'}{\sqrt{1+y'^{2}}})[/tex]= [tex]-2\frac{y}{a^{2}}\sqrt{1+y'^{2}}[/tex]

disregarding the (y/a)^2 term
 
Last edited:
By "y" being the principal value my prof means that "y" is the independent variable.

For the last line, why did you ignore n(y) when you took the partial with respect to y' (on the left)?
 
yeah Y is the degree of freedom and the Lagrangian are always expressed in term of the degrees of freedom L(q,q') and this makes more sense since you are asked to find y(x)

I didn't ignore the n(y) its just that y is comparible to [tex]\rho[/tex] and since [tex]\frac{\rho}{a}[/tex] is small n(y) = n0
 
Ok thanks for the explanation, that makes sense. But shouldn't the d/dx be a total derivative and not a partial; and secondly; how the hell do you solve that ODE?
 
i'm not sure about the entire thing but since dy is comparable to rho you can deduce that y' is approximate to alpha and thus all the y'^2 terms are negligible and you get the resulting ODE

[tex]y''=\frac{-2y}{a^{2}}[/tex]

which has solution

[tex]Acos(\frac{\sqrt{2}x}{a}) +Bsin(\frac{\sqrt{2}x}{a})[/tex] which kind of makes sense since you expect it to bounce off the walls in a periodic fashion
 

Similar threads

  • · Replies 19 ·
Replies
19
Views
3K
Replies
19
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 3 ·
Replies
3
Views
2K
Replies
1
Views
1K
Replies
1
Views
2K
Replies
4
Views
2K
Replies
8
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K