When to Use TE, TM, & TEM Modes for Modeling Optical Fields

[J77]

Can anyone point me to an online resource which gives details - and applications - as to when one should model an optical field using TE and TM modes, and when it's OK to use TEM modes.

Cheers.

edit: should've stuck this in the normal physics forum... mods please move...

More specifics - TE and TM modes can exist in a bounded waveguide, TEM modes can't - however, they can exist in cables... but isn't the latter bounded ?

 

[Claude Bile]

Wikipedia sums it up nicely;

http://en.wikipedia.org/wiki/Transverse_mode

Basically, if you are dealing with plane waves (free space optics), TEM modes are ok, in any time of waveguide (guided-wave optics), you need to stick with TE, TM (for weakly-guiding waveguides) or hybrid (HE, EH) modes.

Claude.

 

[dingpud]

Mw

I thought that I had some old links that were still availabe, but the only one that I was able to come across is:

http://www.amanogawa.com/index.html

It may be helpful. It has some interesting applets which allow you to do some quick test guesses. May be useful for other applications as well.

Let me know if they are of any help, and I will continue looking for my old microwave notebook.

-pud-

 

[Claude Bile]

More specifics - TE and TM modes can exist in a bounded waveguide, TEM modes can't - however, they can exist in cables... but isn't the latter bounded ?

I had a thought...

My knowledge of electronics is not that vast, but it is my understanding that cables are NOT bounded waveguides because they radiate energy, particularly at micro-wave frequencies where these types of definitions are likely to apply. A bounded waveguide would not radiate energy in this manner.

Claude.

 

[J77]

Thanks for the info - and those Java applets look fun.

I'm starting to get my head around them now - my missing slight missing link is complete understanding into how the TE, TM and hybrid HE, EH modes can be approximated by linearly polarized LP modes.

I'm getting there tho...

 

[Claude Bile]

As long as your waveguides are weakly guiding you can reduce the vector wave-equation into a series of scalar wave-equations. This essentially de-couples each orthogonal component (x, y, z or r, theta, phi depending on symmetry) from one another. I don't remember the exact details, but this is the first step toward approximating TE, TM and hybrid modes into LP modes.

Just out of interest, what exactly are you trying to model?

Claude.