There are basically two large camps on the issue of a possible 300-meter fiber horizontal, divided mostly along party lines-copper cabling manufacturers and fiber cabling manufacturers.
And then there are the rest of us who try to discern the “why” so we can explain it to the architects, space planners and building owners who are our customers.
The precedent for use of a 300-meter fiber horizontal has already been set. TIA-942 Data Centers currently allows a 300-meter fiber horizontal, but TIA-568-B.1 Commercial and TIA-570-B Residential do not. We already know that it works; the question is within which of the “unique premises environments” should it be considered as the “minimum standard allowed,” and why?
I decided to ask members of each of these camps for some of the pros and cons regarding a ubiquitous 300-meter horizontal for fiber cabling.
The copper camp is quick to quote studies, which are well over 30 years old, on typical lengths of horizontal cabling that show most “existing cables” are about 150 feet long. Oddly enough, that is about the same as the length as a typical electrical branch circuit. You see, during the era “under study,” most of the “telephone” cabling was terminated on the “other wall” in the electrical rooms.
Given the choice, would we design systems that way today? Probably not. It is interesting to note that there have not been any “new” studies presented to the committee for consideration, so the original 30-year-old study still stands as “typical.”
The fiber camp wants to reduce the number of telecommunications rooms within the building as a “cost saving measure,” which realistically is sort of the “all your eggs (or, in this case, network ports) in one basket” approach.
This idea was first introduced as a Centralized Optical Fiber Cabling System. But once TSB-72-1997 finally made its way through the committee process, the result (maximum horizontal length of 90 meters) was quite different than what was originally intended (maximum horizontal length of 300 meters) because it also included copper cabling.
Yes, I know what the TSB’s title says, but both media types were covered in the original TSB-72, and later when the material was incorporated into TIA-568-B.1.
Fewer telecommunications rooms would mean less building space to power, cool, ground, etc., and few network maintenance points. But too many cables concentrated into one space-for example all the ports in a commercial office building-and you have built yourself a data center.
It is true that dense concentrations of network ports within fewer telecommunications rooms will much improve the port utilization rate; however, the optical network equipment has historically been more expensive than its copper counterpart. This is where the fiber camp argues that an increase in use of optical network equipment will encourage reductions in pricing due to higher volume purchases.
I don’t know that I would “buy” into that argument, but I do believe that over time, because network speeds continue to increase, the cost of network equipment for optical fiber and copper cabling will begin to approach parity, while the cabling distances supported will become even further divided.
But is this a “commercial building”? When TIA began writing premises cabling standards in the mid-1980s, there were only two types of buildings on their radar: residential and commercial. The method of determining which was which was fairly simple. If it is not someone’s home, then it is a commercial building.
Using this classification method, hospitals, schools, factories, power plants, data centers, office buildings, etc., are all commercial buildings.
Today, we are seeing more granularity within the standards, and many of these will someday have their own “unique premises environment” standard.
But the title of TIA-568-C.1 is still “Commercial Building,” and I believe that this is going to cause a lot of confusion as to which standard will take precedence. If TIA-568-C.1 is actually meant to address commercial office buildings, then a simple change in the title should correct the problem, and if not, I am certain that the discussions will be long and interesting.
It is my personal opinion that the 300-meter fiber horizontal should be included in TIA-568-C.0, which would allow use in all “unique premises environment” standards where it was not specifically prohibited by an exception. For example, there would not be an exception within data center or industrial standards but there would likely be one within the commercial (office) building standard where the copper cabling camp has a strong presence.
So, what would happen in the case of an airport or hospital for which there are no “unique premises environment” standards? In my opinion, TIA-568-C.0 would apply, and it would then be the responsibility of the designer to determine if the 300-meter fiber horizontal should be allowed.
What is seen as the limiting factor in using fiber in the horizontal today? How do you power the Power over Ethernet (PoE) and PoE Plus devices with optical fiber?
What we need is “just a little DC.” Power over fiber?
The fiber-optic powering system consists of a high-power laser diode, an optical fiber for transmission, and a photovoltaic cell. Photovoltaic cells have been used for years in solar panels to convert sunlight into heat or electrical energy.
The high-power laser diode is the device that converts electrical energy (DC) into light energy, transmitted through a medium. At the far end of the fiber-optic cable, the photovoltaic cell converts the light energy back into electrical energy (DC) where it is used to power an electronic device like a WiFi antenna or a VoIP telephone. And the same optical fiber that is used to power the device can also be used to communicate with it. Now it is not only possible, but actually in production.
On May 5, JDS Uniphase Corporation announced that its Photonic Power Business Unit has achieved optical-to-electrical conversion efficiency greater than 50% on their 3-volt and 5-volt gallium arsenide (GaAs) Photovoltaic Power Converter (PPC). The company reports that more than 10,000 units have been deployed, serving more than 50 customers that include Siemens, Raytheon, ETS Lindgren, and NEC.
Yes, things are about to get interesting.
Next month, I plan to discuss why, if 150 feet was typical, 100 meters was chosen as the maximum channel length.
DONNA BALLAST is BICSI’s standards representative, and a BICSI registered communications distribution designer (RCDD). Send your question to Donna at: firstname.lastname@example.org
Unfortunately, we goofed a few times in the past several issues, and want to take this opportunity to clarify some points and acknowledge some oversights.
In the March issue, a sidebar that accompanied the article “Fiber-optic adapters: More than just a good mate” (p. 52) failed to list Panduit as a manufacturer of fiber-optic adapters. Panduit does indeed manufacture fiber-optic adapters, and information on its products is available at the company’s Web site, www.panduit.com
Similarly, Sumitomo Electric Lightwave was omitted from the listing of fiber-cleaver manufacturers that accompanied the article “Fiber cleavers couple simplicity with accuracy” in the May issue (p. 46). Information on the company’s fiber cleavers is available at www.sumitomoelectric.com.
Finally, several typographical errors showed up in the article “Wi-Fi means business for cabling installers,” authored by Fluke Networks’ Mark Johnston, RCDD, which appeared in our April issue (p. 18). Johnston authored a clean and technically accurate article; the errors were introduced in the editing stage and were the work of Cabling Installation & Maintenance magazine, not Johnston. CI&M apologizes for providing inaccurate information, and to Mark Johnston for subtracting value from, rather than adding value to, his work.
It is our aim to provide you with useful, practical, and accurate information about structured cabling systems, standards, and technologies. We encourage all of you to let us know if we fall short of your expectations. E-mail chief editor Patrick McLaughlin at email@example.com.