Dave, Jim,
I believe that what Jim was referring to when he asked "who would even want
single mode now" was prompted by his awareness of the emergence of new forms of
long haul fiber, whose parameters depart from traditional singlemode fiber (SMF) as
we've known it in the past. Some of the new variants are grouped as:
- dispersion shifted fiber
- non-zero dsf
- by Corning, LEAF (covered below)
- by Lucent, AllWave (covered below)
"Standard" SMF, such as the standard ITU G.652 SMF which offers two or three
wavelength windows, if you include 1625 nm for a network management probe,
differs from the newer grades of dispersion shifted glass and those which Corning and
Lucent have come out with in the past two years. Namely, LEAF fiber and AllWave
fiber, respectively (urls at the bottom of this post), to name two of the more popular
products which are now being used to replace what was earlier the exclusive role of
standard SMF in the long haul.
Original single mode referred to fiber with a core field diameter of 9.3 microns, and
with limited permissiveness at first one wavelength (13xx), and then two (15xx). This
was covered in one of your four references in your reply (uplink to this one).
The newer variants of LEAF and All Wave, however, open up substantially beyond
those of standard SMF, ostensibly. LEAF, for example, which stands for large
effective area fiber boasts fifty to sixty percent (I believe) more wavelength
permissiveness than traditional SMF. AllWave makes similar claims. Maybe you or
someone else here can help us out here by posting what the specific advantages are
for those grades.
In short, not so long ago it was safe to think of fiber as being either:
- multimode fiber (MMF) for the LAN, and
- singlemode fiber (SMF) for the WAN,
The number of options, however, have now grown into what sometimes seems like
an endless list of specialized grades of MMF (lately there has been a rebirth in
thinking about what can be achieved over MMF), and a greater number of long haul
grades of fiber, which were previously simply called SMF fiber.
These are designed to meet a growing list of demands:
- reach higher throughputs,
- over greater distances,
- without the need for electrical regeneration,
- using the fewest number of edfa optical amplifiers, possible.
- what did I miss?
I am sometimes guilty, no.. make that often guilty, during highly generalized
discussions of lumping all of the newer (non-standard SMF) grades of glass under the
moniker of "Single Mode," as long as they are not rated as "multimode." But a
distinction does exist, nonetheless, and the differences between these newer grades
could be dramatic, from the perspective of the need to be specific when discussions
turn to specific network designs. Jim asked, Who would want to buy standard single
mode fiber now?
There is still a market for SMF, where the distance*bit-rate *lambda-count formulae
do not call for special dispersion or polarization mode handling, while they are (the
distances are) greater than the reach of standard MMF.
Both SMF and MMF fibers may come from the factory in raw form, and they may
not enter into a final cable construction until they reach the next two or three stages in
the supply chain, whereupon they are packaged into any number of types of
constructions. These may range from patch cords, to jumpers, to campus grade
all-weather-indoor- outdoor single-mode, SMF or MMF riser cable, to composite
single- AND multimode-under-the-same-jacket cable, to submarine grade, and so on.
Many campus area networks (CANs) and metropolitan area networks (MANs) will
continue to utilize standard SMF, IMO. For greater distances they may elect for
another type, based on their design and service delivery criteria. This should not be
construed to mean that the selection of a fiber type by emerging dark fiber companies
is a no brainer. On the contrary. They will need to be extremely careful to select the
right grades of fiber to meet the changing requirements throughout their serving
territories, over time. Clairvoyance often helps..
Sometimes a carrier must make a selection which is going to satisfy many different
topographies, including
- intramural (very short distance), which uses______
- metro (up to 25 mile radii), which uses______
- regional (several hundred miles), which uses_______,
and,
- the greater WAN (either interfacing to, or extending to) which uses _______,
_______, and _______.
Are the blanks above always filled out with the same answers?
Or, do different Layer 1/2 protocols and throughput rate objectives, combined with
WDM philosophies and in house policies governing vendor interoperability dictate
what those blanks will be on a case by case basis, or will it be a case where one
could legitimately say, "it depends?" Hey! I'm not going to do "all" the work, here. -s-
At the same time they must ensure that the fiber they use meets the specifications of
the optical dwdm's, optical switches and terabit routers, and vice versa, which, for
the most part, haven't even been selected yet. It is for these reasons, and then some
(availability, construction phasing, etc.) that the selection of the right type of fiber by
startup dark fiber carriers (Fiber CLECs, or FLECs) is an increasingly perplexing
task, today, often compounded by the need to rent scarce and expensive-to-re-enter
ducts, in the event you change your mind, from others.
I've compiled a list of articles from Lightwave Magazine that might be helpful, below:
Optical Fiber Designs Evolve:
lw.pennwellnet.com
Non-zero Dispersion Shifted Fiber:
lw.pennwellnet.com
Some AllWave parameters:
lw.pennwellnet.com
One that speaks more to LEAF:
lw.pennwellnet.com
Now, will someone fill in the blanks above, while I get my Ben and Jerry's?
Comments and corrections welcome, as always.
FAC |
