Information Technology Reference
In-Depth Information
TABLE 15.6
Attenuation Limits for Installed Cable Links, in dB
Permanent Link
Channel
Frequency, MHz
Cat 3
Cat 5e
Cat 6
Cat 3
Cat 5e
Cat 6
1.0
3.2
2.1
1.9
4.2
2.5
2.1
4.0
6.1
4.0
3.5
7.3
4.5
4.0
8.0
8.8
5.7
5.0
10.2
6.3
5.7
10.0
10.0
6.3
5.5
11.5
7.0
6.3
16.0
13.2
8.2
7.0
14.9
9.2
8.0
20.0
9.2
7.9
10.3
9.0
25.0
10.3
8.9
11.4
10.1
31.25
11.5
10.0
12.8
11.4
62.5
16.7
14.4
18.5
16.5
100.0
21.6
18.8
24.0
21.3
200
27.4
31.5
250
31.1
35.9
Source: TIA-568-C.
uation. The actual measurements should be made at intervals of 1 MHz or less in
the appropriate frequency band, but the tester will take care of this automatically.
The attenuation test limit is for a worst-case link. The actual attenuation of a meas-
ured link will be less if the cable run is shorter than allowed. However, in most testers,
the measured loss is compared only to the limit for worst case. This means that the test
will not directly catch cable that has greater than expected loss (based on the uninstalled
cable limits of TIA-568-C), but the extra attenuation per meter should have no effect on
link performance. If you wish to catch substandard cable, you will have to compare the
actual loss measurement with the expected loss for that particular length of cable.
As with the specification for uninstalled horizontal cable, there is an allowance
for the effect of temperature on the attenuation of the cable. The allowance is 1.5%
(of the value in dB) per degree Celsius for Category 3 and 0.4% for Categories 4
and 5. To illustrate this, consider a cable installed in an attic, where the actual tem-
perature is 30°C, rather than the nominal 20°C of the standard. The attenuation
limit for a Permanent Link at 10 MHz would be raised from 10.0 dB to 11.5 dB (an
increase of 10.0 dB
1.5%/°C
10°C) for Category 3, but only to 10.4 dB (10.0
dB
10°C) for Category 5e.
Two other environmental factors are known to cause changes in attenuation:
metallic conduit and humidity. The difference expected for cable in conduit is up to
a 3% increase, but the standard allows no latitude in test limits. It should be
expected that long runs in close proximity to other metallic objects might also cause
0.4%/°C
