Information Technology Reference
In-Depth Information
TABLE 12.5
Category 5 Extended Performance Levels for Cable*
Parameter
Units
Level 5
Level 6
Level 7
Test range
MHz
1-200
1-350
1-400
0dB
†
PS-ACR
MHz
140
165
280
PS-ACR
10dB
MHz
80
100
180
Attenuation
33 dB
MHz
200
200
240
Connecting hardware
Test range MHz 100 100 100
Attenuation dB 0.4 0.4 .02
PS-NEXT dB N/A 40 50
NEXT dB 40 42 54
Return loss dB N/A 18 20
*These cable and connecting hardware performance levels were promulgated by Anixter Inc. to define advanced
quality levels of Cat 5 components in structured cabling.
†The 0 dB level of ACR is generally considered the bandwidth of the cable.
signal. Some of the energy may be lost through pure resistance, and some may be
coupled out of the cable pair to adjacent pairs or to other conductive objects. In
addition, any variation in the magnitude of resistance, capacitance, or inductance
along the pair or at a termination point can cause a mismatch that serves to further
attenuate the signal (among other things).
Attenuation in a cable is frequency dependent. The higher the frequency, the
more the attenuation. This effect is particularly noticeable at the extremely high fre-
quencies that make up the gigabit signal. Attenuation is measured in dB in 1 MHz
increments up to the maximum range of the cable.
Advanced cables maintain very tight control on the geometry of the cable to
minimize attenuation up to 100 MHz, 250 MHz, and beyond. Accordingly, the
attenuation at 100 MHz progressively improves for categories above Cat 5, as it
must to achieve reasonable levels at the higher frequencies.
Near- and Far-End Crosstalk for Gigabit.
In reality, it is the ratio of signal to inter-
ference (presumably the most significant component of noise) at the far end of a link
that is critical. At the far end, however, signal components exist of both the local
near-end crosstalk and the distributed far-end crosstalk (including the transmitting
connector's NEXT and the pair-to-pair coupling along the cable).
In gigabit signaling, all the pairs are in use, so the situation is really complex.
It turns out that it is particularly complex to test the parameters that cause perform-
ance to degrade at gigabit rates. This testing is described in detail in Chapter 15, but
it seems a little like a day at the races.
