Cabling & Connectors

Signals are only as good as the cable carrying them. This guide covers the wire and cable a technician works with daily — conductor sizing, coaxial cable and its impedance, twisted-pair categories, the common connectors, and the rules for choosing and installing cable safely. It also covers gauge by purpose, TIA-568 and the T568A and T568B pin-outs, the three cable types compared, keeping a cable's characteristics, grounding and bonding, splicing, testing, connector defects, prep tools and crimping.

Conductors & wire gauge

Wire size is given by the American Wire Gauge (AWG) number, and the counterintuitive rule is that a smaller gauge number means a larger diameter. A #8 wire is far thicker than #20. A larger conductor carries more current with less voltage drop. The properties that matter for a conductor are its diameter/area, resistance, current capacity, and insulation rating — notably not its capacitance, which is a cable property, not a single-conductor spec.

Competency 7.1 asks you to list wire types and their construction. Inside the insulation, a conductor is built one of three ways:

Coaxial cable

Coax carries a signal on a center conductor surrounded by a dielectric, a shield (braid/foil), and a jacket. The shield confines the field and rejects interference, which is why coax is used for RF and video. Its defining spec is characteristic impedance — a value (commonly 50 Ω for radio/RF and 75 Ω for video/CATV) set by the cable's geometry, not something you measure with an ohmmeter.

Two more rules the exam likes: loss rises with frequency — the same coax attenuates a 1 GHz signal far more than a 100 MHz one — and you must match impedances. Connecting cables or loads of different impedance causes reflections and standing waves, wasting power and potentially damaging equipment.

Twisted pair & categories

Twisted-pair cable twists each signal pair to cancel interference, and comes in performance categories for data networking. Higher category = higher bandwidth and data rate:

STP vs. UTP: shielded twisted pair (STP) adds a foil/braid shield for noisy environments; unshielded (UTP) is cheaper and most common. Twisted pair terminates in the familiar RJ-45 8-pin connector for Ethernet.

Common connectors

Cable selection & installation safety

Choosing cable isn't just about the conductor — the jacket rating matters for where it can be installed. Cable run in a building's air-handling spaces (plenums) must use plenum-rated jacketing that limits flame spread and smoke in a fire.

Wire gauge by purpose

Competency 7.2 asks you to list the wire gauges used for various purposes. The gauge follows from the job the wire does: how much current it carries, how much voltage drop the load can stand, and what a standard or code requires.

ANSI/TIA-568: twisted-pair cable, T568A and T568B

Competency 7.4 asks you to list the common identifications for copper cables in standards, such as #18 and #24 AWG and UTP cable in ANSI/TIA-568. ANSI/TIA-568 defines the cable types, connectors and topologies of structured cabling. Its UTP is four twisted pairs with a characteristic impedance of 100 Ω (±15 Ω from 1 to 100 MHz), usually 24 AWG solid copper; the category sets the frequency it is specified to: Cat 5e to 100 MHz, Cat 6 to 250 MHz and Cat 6A to 500 MHz (Wikipedia, Category 5 cable). A channel, from one active device to the other with its patch and equipment cords, is at most 100 m; the fixed permanent link within it is at most 90 m (Fluke Networks, via Accu-Tech).

The 8-position modular jack and plug (commonly called RJ-45) is wired to one of two pin-outs, T568A or T568B. They work equally well; the rule is to use the same one at both ends. T568A puts the orange pair on pins 3 and 6, T568B the green pair; the blue and brown pairs do not move (Fluke Networks). A cable with T568A at one end and T568B at the other has a crossed pair and fails a wire map test.

TIA-568 also limits the installation: untwist no more than 0.5 inch (about 12.7 mm) of any pair at a termination (Cabling Installation & Maintenance, on TIA-568-A); bend 4-pair UTP no tighter than four times the cable's diameter, so a 0.25 inch cable needs at least a 1 inch radius; and pull 4-pair 24 AWG UTP with no more than 110 N (25 lbf) (TIA-568, as quoted by Elliott Electric and in Cisco's cabling course notes).

Coax, twisted pair and fiber compared

Competency 7.5 asks you to explain the major differences between copper, coaxial and fiber optic cables.

Keeping a cable's characteristics

Competency 7.6.1 asks you to explain the reasons for maintaining a cable's characteristics. A coax's impedance is set by its construction: the size of the center conductor and its exact spacing from the shield, with the dielectric between them (section 2). Change that geometry anywhere and the impedance changes at that point (Electrical Connection). Kinks, crushing, stretching and too-tight bends push the center conductor to one side, so it is no longer concentric with the shield; the result is a local impedance change that reflects part of the signal, raises the SWR and shows up on a TDR as a bump (PIC Wire & Cable). The reflections hurt the high frequencies first.

Bend radius. Every cable has a minimum bend radius: for LMR-400, 1.00 inch for an installation bend and 4.0 inches for repeated bending (Times Microwave); for 4-pair UTP, four times the diameter (TIA-568). The weakest point is where the cable leaves the connector, so do not pull a cable at a right angle to its connector (PIC).

Crushing. Secure cable with straps, staples, ties or hangers designed not to damage it (NEC 820.24, Mike Holt), and fasten cable ties loosely (Cabling Installation & Maintenance).

Water. Weatherproof every outdoor connector, splice and jumper, and form a drip loop where the line turns from vertical to horizontal (Motorola R56).

Mismatched parts. Use connectors and adapters of the cable's own impedance (the trap in section 4); in twisted pair, untwisting more than needed at a termination and kinked or damaged cable worsen the return loss (Fluke Networks).

Grounding and bonding cable systems

Competency 7.8 asks you to explain the purposes and types of cable grounding (bonding) and the common conventions. Bonding protects people from shock, gives the equipment a common reference, and drains surges, especially lightning, to earth before they arc across equipment. Motorola's R56 states the aim: bond every grounding system to one point, so that no difference of potential can build up between parts of the system.

At the building entrance (NEC). Each lead-in from an outdoor receiving antenna needs a listed antenna discharge unit near its point of entrance (810.20(A)); at an amateur or CB transmitting station, 810.57 also accepts a lead-in protected by a continuous metallic shield grounded per 810.58. Antenna masts are grounded per 810.21. The outer shield of a coaxial cable entering a building is bonded or grounded as close as practicable to its point of entrance, with a copper conductor of at least 14 AWG, run as short and straight as practicable; in a one- or two-family dwelling it must be no longer than 20 ft, or else a separate ground rod at least 8 ft long is driven and bonded to the power grounding electrode with at least 6 AWG (820.100). It terminates at the building's intersystem bonding termination where there is one (Mike Holt, EC&M).

At a radio site (Motorola R56). Every transmission line is bonded to the tower with a ground kit near the antenna, at the bottom of the tower and near the building entry, and towers taller than 61 m (200 ft) get another kit midway. Ground-kit conductors run downward toward ground, with no drip loops. Inside, a single-point ground, usually a master ground bus bar, is where all the equipment, antenna transmission lines, surge protective devices and utility grounds are bonded. Grounding conductors run short and straight with bends no tighter than 203 mm (8 inches) and no sharper than 90°, and braided grounding conductors are never used.

Audio and signal shields. In balanced audio, the shield is bonded to the chassis where it enters, and a shield that is part of the signal circuit lets ground-loop current into the audio (Interfacing, section 7).

This safety ground is not the same thing as an antenna's RF ground plane, which is part of how the antenna works (RF Guide, section 7).

Splicing copper, coax and fiber

Competency 7.9 asks you to describe the knowledge and ability needed to splice coaxial and copper cable, and 7.9.1 to explain the types of fiber splices.

Copper. NASA's workmanship standard shows the accepted wire splices: crimp splices (butt or inline, end and parallel), the lash splice, the solder sleeve, and the Western Union splice for solid conductors. A soldered splice is fully wetted, smooth and shiny; a crimp splice is deformed only by the tool's indenters, with the indents symmetrical and centered and the strand ends visible; heat-shrink tubing is shrunk tight with no exposed conductor; and the splice sits outside any flex zone, with strain relief. Twisted-pair data and telephone wiring is joined on insulation-displacement (punch-down) blocks instead, keeping each pair's twist to within 0.5 inch of the termination (section 7).

Coax. A coax joint has to keep the cable's geometry, or it becomes the impedance bump of section 9. Splice coax with a connector on each end and a barrel adapter, or a splice connector, of the cable's own impedance, never a twisted or taped joint, and weatherproof every outdoor splice (Motorola R56).

Testing copper, coax and fiber

Competency 7.10 asks you to briefly explain the testing methods for all three types of cable and to compare their dB loss measurements. Two measurements apply to all three. Insertion loss is the signal lost along the link, in dB, and it grows with length and at every connection. Return loss compares the signal sent with the signal reflected back, also in dB, and here a higher number is better: no reflection at all would be an infinite return loss (Fluke Networks).

Return loss and SWR describe the same mismatch (RF Guide, section 8): an SWR of 1.5:1 is a reflection coefficient of 0.2, a return loss of −20 log 0.2 ≈ 14.0 dB, with 4% of the power reflected.

Connector fittings and their defects

Competency 7.11 asks you to compare the fittings and connectors used in cabling and to list their potential defects. Section 4 compares the connectors. Their fittings differ in how they hold the cable: the center contact is crimped, soldered or clamped onto the center conductor, and the body is crimped or clamped onto the braid (Times Microwave). Each fitting has its own faults:

Cable prep tools and crimping

Competency 7.12 asks you to describe proper crimping of communications wiring connectors, and 7.13 to explain how cable prep tools are used and to demonstrate proper and improper crimps.

What a good crimp does. A crimp is a cold, metal-to-metal joint: enough pressure must be applied to break down the oxides on the stripped conductor and the plating in the terminal, or the resistance rises. Too much pressure, an over-crimp, reduces the conductor's cross-section and also raises the resistance (Molex, Quality Crimping Handbook).

On the job: fitting and checking a coax connector at a radio site

Job practice, not exam content. A typical sequence for putting a crimp N connector on LMR-400 at a two-way radio site. The connector maker's instructions and the site's own standard, often Motorola R56, come first.

1. Prepare the cable. Strip it with the prep tool made for that cable and connector, to the maker's dimensions: square cuts, nothing nicked, the braid flared slightly, never combed out (Times Microwave; Amphenol).
2. Fit and crimp. Seat the center contact so it bottoms against the dielectric, crimp it and the ferrule with the specified die, and give the connector a firm pull test (Amphenol; Molex).
3. Weatherproof and dress the line. Weatherproof the outdoor connector, form a drip loop where the line turns from vertical to horizontal, and keep every bend wider than the cable's minimum radius, 1 inch for LMR-400 (Motorola R56; Times Microwave).
4. Ground it. Fit ground kits near the antenna, at the bottom of the tower and at the building entry, with their leads running downward, and bond the entry surge protector to the single-point ground (Motorola R56). This safety ground is separate from the antenna's RF ground plane (RF Guide, section 7).
5. Sweep it and record it. Measure the return loss across the band, or the SWR, and look at the distance-to-fault display or a TDR: a reflection at the connector's distance means a bad termination. An SWR of 1.5:1 is a return loss of about 14.0 dB. Compare the result with the site's acceptance figure and record it.