Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

[Phys-L] Re: fiber optic cables and waveguides



I agree its boundary condix, etc. However, reflection due to TIR (bound
electrons) and quasi free (conduction) electrons seems different to me.
For example, are the phase shifts on reflection different? True, normal
reflection at the interface of two different indices is different for
the two directions, and no one would think it's a different mechanism.

On the easy vs. tricky. My criterion is much more severe. If a HS lab.
(typical ones equipment -- e.g. stuff from RAFT, Radio Shack, Pasco,
Vernier, etc.) can't do, it its too tricky.

Yes, but co-ax. does exhibit dispersion, however, its due to the
dielectric and the bumpiness of the cable. Engineers include in
dispersion, absorption as a function of freq. as well as speed. [The
effect on a pulse is similar.]

bc

p.s. I find very interesting the use of a fiber as a spatial filter. Is
this another example of identity of eqs. with different? mechanisms?
Does a multimode beam have greater angular spreading? Does an iris in a
waveguide remove higher modes. [Large guide, obviously.] Can one
characterize a noisy beam as a multimode bean?

John S. Denker wrote:

On 03/04/05 01:08, Bernard Cleyet wrote:


"Well, in that sense a hall of mirrors could be called a waveguide.
I suppose that it is etymologically true, but at some point it stops
being a useful physical model. (Here I am referring to the "solve
Maxwell's Eq. with boundary conditions" model, as opposed to the
"light ricochet" model.)"

I think tho the mechanisms are different the equations are the same



Hmmmm.

To my way of looking at it, the mechanism is the same. It's
just Maxwell equations with boundary conditions. There are
some simplifying assumptions you can make in certain limiting
cases -- such as restricting attention to the 0,0 mode -- but
we are not required to make such assumptions.



So the hall of mirrors is the
limit where group v is ~= C.



Yes.

One could equally well characterize the hall of mirrors as
the case where the mode-number is huge, a long ways from
0,0 ... and derive v ~ C as a consequence.




"Huh? Perhaps we have different ideas of what is
"easily-observable"?"

I came to this conclusion ("back of envelope" calc.) and then
realized JD was assuming the fiber was kilometers long (original
expt. description). However, I am a bit lost on how to separate the
index effect from the waveguide one.



Second issue answered first: study the propagation speed
as a function of mode number. A mode with one node will be
twice as scrunched as one with none.

As for the first issue: Build an interferometer with fiber
in one arm and something else (free space, or just a different
fiber with a larger core) in the other arm. The fringes
will shift as a function of mode number. You will need to
use *less than one meter* of fiber because the effet is so
large compared to the sensitivity of the method.

A key part of the experiment is controlling which mode you're
injecting into. There are standard techniques for doing this.

I know somebody who likes to pipe her lasers through a
single-mode fiber ... to achieve the best possible spatial
filtering. This is for atom-trapping experiments; she's
not interested in measuring the fiber, just using it as an
easy trick of the trade.

If you want to know my idea of what's easy and what's hard:
atom trapping is hard. The single-mode fiber trick is
less than 0.01 percent of the trickiness required to make
the atom trap work.




Returning to my early suggestion of using co-ax., RG/62U


....


co-ax doesn't exhibit waveguide dispersion.



Indeed! Coax physics is different from waveguide physics.
No cutoff frequency, for starters.



_______________________________________________
Phys-L mailing list
Phys-L@electron.physics.buffalo.edu
https://www.physics.buffalo.edu/mailman/listinfo/phys-l