The following is a suggested sequence for testing twisted pair cabling installed under the proposed UCSB standards for communications cabling.
Check for the following.
This test measures attenuation of the wire by injecting a signal of known strength at one end and measuring the strength of the signal at the opposite end. This measurement over known distance is compared to the manufacturer's design specification for loss over distance (typically xx dB per km).
Signal loss should be measured at multiple defined frequencies and compared with manufacturer's specifications for loss at each of those frequencies.
NOTE: This test should be performed with the full length of cable off of the shipping and/or package reel.
Near-End Crosstalk (NEXT) is measured to determine a signal-to-noise ratio for a cable run and normally will be a measurement of the noise level in a receive pair generated by an adjacent transmit pair.
A common source of NEXT is the existence of a split pair in which one wire of a transmit pair is reversed with one wire of a receive pair.
Interference is measured, or identified, as a source of noise external to the twisted pair structure of the cable under evaluation. Typical interference sources include electric motors and flourescent lighting, as well as any high-voltage circuit running parallel to the signal cable path. Increasing the distance between the signal and the electric cables is required. Shielding on the cable, or for the termination may also be required.
In a large wiring system, with multiple sub-terminals and backboards, a ground loop may be formed by inadvertent use of two ground references for shielding or equipment. Such a group loop may result in both a rise in crosstalk signals or loss of attenuation.
A single grounding system should be established for each building, using a single copper ground rod (8' earth driven) and running a bus circuit (#6 wire) to each closet with a mounted ground bar. All equipment and cable shielding should be grounded as required to this bar.