Re: Melted coax (was: ANI vs. Caller ID) [Telecom]

For many of us in the optics trade, it's fairly common to speak of infrared wavelengths from the red edge of the visible, around 7000 Angstroms or 0.7 microns, out to a few microns, maybe 2 or 3 microns, as the "near IR"; and the longer infrared wavelengths, out around 10 or 20 microns and beyond, as the "far IR" or the long-wave infrared region.

***** Moderator's Note *****

Which wavelength(s) of light are used for fiber-optic transmission? Do single-mode and multimode fibers require different wavelengths?

Sorry if this is Optical Sci 101.

Bill Horne Temporary Moderator

*******************************************************************************************************

• The address for email submissions has changed: if you submit posts via email, send them to *

• telecomdigestmoderator atsign telecom-digest.org. If you submit posts via a newsreader or *

• Google groups, you don't need to change anything. *

*******************************************************************************************************

Whatever works best for the particular fiber deployed.

As I recall, something around 640nm is very common for fiber in the 'visible' spectrum. Easy to get laser diodes for, and easy to get _safety_ gear (eye protctors) for.

In theory, "no"; in practice, because of other differing optical characteristics between the two types of fiber, "yes".

Moderator asked:

If by "fiber-optical transmission" you mean for _telecom_ applications, the answer is, wavelengths from around 8000 A (0.8 microns) in the near IR out to about 1.5 microns, with heavy preference for the 1.3 to 1.5 micron bands.

Reason is, even if you can get rid of all kinds of attenuation associated with absorbing impurities in glass (and current fiber technology is really just astoundingly good at doing that), you are left with an unavoidable (though pretty small) scattering loss in glass fibers, which decreases rapidly with increasing wavelength out to just beyond 1.5 microns, beyond which some other absorption mechanisms in the glass rapidly turn on.

This inherent scattering loss is associated with the fact that the local index of refraction in glass (or any transparent material) has minute, but inherent,and unavoidable, randomly time-varying local variations associated with thermal vibrations of the molecules in the glass; and this produces a correspondingly minute but unavoidable scattering loss.

So, with very good sources (diode lasers), photodetectors, and amplifiers (EDFAs: Erbium Doped Fiber Amplifiers) now available in the

1.3 to 1.5 micron range (and plenty of bandwidth available just within that range), that's the region of choice, at least for high data rate, long distance transmission.

Whether a fiber at any wavelength is single-mode or multimode depends entirely on the diameter of the fiber core relative to the wavelength, and the index difference between core and cladding. Small enough core diameter (down around a few wavelengths): single mode. Larger core diameter (many wavelengths): multimode, with number of modes increasing rapidly with increasing diameter beyond that.

Once you get into longer wavelengths than the mid-infrared region, and especially when you get into the far infrared region, transmission losses in ordinary glass and silica fibers gets so high as to make these fibers unusable. Chalcogenide glasses must be used in these regions to even get reasonable signals over a few meters in length. These specialty fibers sell for thousands of dollars per meter. When I worked in the Research Labs of Eastman Chemical Company, my group developed a process Raman spectrometer whereby we could remotely study the mid- to far-infrared spectra of process streams while using communications grade fibers.

While Stokes and anti-Stokes Raman scattering may be unfamiliar to Telecom readers, the process is easily described in communications terms. A laser in the visible region is used to excite molecules of the unknown material. Most of the light is reflected back at the same wavelength by Rayleigh scattering (elastic collisions of photons). But a small portion of the light interacts with the unknown material via Raman scattering (inelastic collisions) to produce sum and difference frequencies (anti-Stokes and Stokes scattering respectively) around the laser line. The process is akin to amplitude modulation where the infrared spectrum is the modulating component and the laser provides the carrier. But the modulation index is quite low. Only about 1 in 10 to 50 million photons interacts inelastically to replicate the infrared spectra in the visible region. But now the information can be easily sent long distances over inexpensive communication grade fibers. With sensitive detectors and a good monochromator along with a good laser, it is possible to perform infrared spectroscopy remotely without the need for Chalcogenide fibers.

The process is akin to using an audio baseband to modulate a carrier. But I never was able to explain this to the chemists!