Hitachi claims that cable constructions such as 2 x 2 do not offer the necessary robustness and balance
Brian Griggs
Hitachi Cable Manchester Inc.
The drive to produce more Category 5 cable has generated new cable constructions known as 3-and-1 (3 x 1) and 2-and-2 (2 x 2). These constructions are so named because the conductor pairs within the outer cable jacket are coated with different insulators to stretch a material in short supply--fluorinated ethylene propylene (FEP). In the past, FEP was used to coat all four pairs to meet fire ratings for plenum spaces. The new constructions use FEP substitutes to coat one or two pairs within a 4-pair Category 5 cable.
The initial reaction to the development of these new constructions has been enthusiastic because it stretched the limited supply of a critical cabling material. However, they contain many pitfalls that contractors and end-users should be aware of.
To pass the UL-910 fire test of Underwriters Laboratories (Northbrook, IL), these newer constructions require heavier jacket walls. A heavier jacket can increase kinking during packaging and unpacking, create bend-radius problems during installation, lead to a lower density of cabling in conduits and wireways, and make jacket removal more difficult during termination. There is, however, a more important issue of technical performance, particularly with regard to velocity-of-propagation skew, or delay skew.
Velocity of propagation
Velocity of propagation is the speed at which a signal can be transmitted over a cable. Typically, it is stated as a percentage of the speed of light but can also be stated as a time-to-distance measurement--for example, 500 nanoseconds per 100 meters. Velocity of propagation is calculated using the following formula: 1 divided by the square root of k, where k is the effective dielectric constant of the insulating material.
To meet the electrical requirements of Category 5 cable, the primary insulating material should display the proper value for the dielectric constant. FEP and unmodified olefins offer the best values for this type of insulation. Unmodified olefins are used for general-purpose (CM) and riser-rated (CMR) cables; they are not used for plenum-rated cables (CMP), though, because of flammability requirements. For olefins to be used in plenum cable, their composition must be modified to reduce their flammability.
The more olefin used in the cable, the more drastic this modification has to be. The more drastic the modification, the greater the deterioration of the dielectric constant of the material.
The problem is that the dielectric constant of a 4-pair cable becomes skewed if two pairs are insulated with FEP and two with modified olefin. (The dielectric constant of FEP is 2.15, versus 2.45 to 2.6 for modified olefin.)
Improving the dielectric constant
Some manufacturers claim that the effective dielectric constant can be improved by loosening the jacket and allowing more air around the pairs. However, an unacceptable level of skew is still present. And, when the dielectric constant is skewed, so is the velocity of propagation.
With current transmission rates, this skew may not have an adverse effect on system performance. Many existing network topologies, for instance, use only two of the four pairs available. It would be imprudent, though, not to recognize that schemes requiring all four pairs, such as 100Base-T4 and 100Base-T2, are emerging and will most likely become more common as transmission rates increase. Cables with delay skew may still function when these 4-pair schemes are used, but systems using highly skewed cables may burden transmit/receive memory and transmission timing, causing unacceptable bit- error rates--or worse. This could lead to sluggish transmission characteristics and signal errors that would not be encountered in a well-balanced network running on a cable system exhibiting low skew.
There is already evidence that the dissimilar transmission characteristics of differently insulated conductors in a 4-pair cable can lead to problems. When 2 x 2 cables are used, the certifying agent typically finds it necessary to change the nominal velocity-of-propagation calibration on the handheld tester to obtain a "pass" reading. This clearly indicates that any sophisticated electronic device using the same path will also "see" this imbalance.
Those who are currently installing Category 5 cabling systems expect them to last for 10 to 15 years, as the warranties usually promise. For these systems to be able to handle higher data rates and new encoding schemes as they emerge, they must be as robust and balanced as possible.
Hitachi claims that some cable constructions, including most 2 x 2, do not offer the necessary robustness and balance. Because of the possible negative technical impact of such constructions, the company decided more than a year ago not to offer them, despite the commercial benefits.
Soliciting support
Hitachi has used its presence at Electronic Industries Association/Telecommunications Industry Association (EIA/TIA) meetings to solicit support for recognizing this transmission characteristic. Although the TIA/EIA-568A commercial telecommunications cabling standard was published without a delay-skew requirement, a recent EIA/TIA meeting spotlighted the importance of this issue. As a result, additional information on this subject is expected in the near future.
The company says that Category 5 cabling is the best choice for horizontal applications, but believes that the contractor and end-user should be aware that some Category 5 cable constructions are better than others. A highly skewed or unbalanced cable could have an undesirable effect on system transmission characteristics.