What Wires Are Used in an Ethernet Cable?

An Ethernet cable holds eight copper conductors arranged as four twisted pairs: orange, green, blue, and brown, each paired with a white-striped mate. The twists help reject electrical noise, letting a modest-looking cable carry reliable data between a router, switch, wall jack, or device.

The pairs do not all do the same job at every speed. 10BASE-T and 100BASE-TX use two pairs; Gigabit Ethernet and faster standards use all four. A Structured Cabling Contractor Pasadena can help ensure the correct cable type, pair arrangement, and termination are used for the required network speed. Wire order matters too: T568A and T568B place the same pairs on different pins. Here’s how the conductors, twists, and pinouts fit together.

⚡ Key takeaways

Ethernet cables contain eight copper wires arranged as four twisted pairs to reduce interference.
Use the same T568A or T568B standard at both ends for normal straight-through cables.
Gigabit and faster Ethernet require all four pairs; missing pairs often force links down to 100 Mbps.
Choose solid copper cable for permanent runs and stranded cable for flexible patch leads.
Avoid CCA cable, especially for PoE; pure copper reduces voltage drop, heat and reliability problems.
Cat6 suits most new installations; choose Cat6A for reliable 10Gbps over full-length runs.

What wires are used in an Ethernet cable?

An Ethernet cable carries network data through eight individual wires arranged as four pairs. Each pair is twisted to limit electrical noise and preserve signal quality along the cable’s length. Most Ethernet cables use copper conductors, with insulation colors identifying each wire and guiding termination. At either end, the eight wires sit in an RJ45-style connector. Their exact order determines which connections carry transmit and receive signals, so a simple-looking cable still needs precise wiring for dependable links between devices.

Ethernet wire colors and pair identification

The jacket colors are not decoration: they identify the four wire pairs inside the cable. A standard set includes blue and blue-white, orange and orange-white, green and green-white, plus brown and brown-white. The white conductor in each wire pair carries a colored stripe, making it easier to distinguish from the solid-color mate under good light.

At the connector, the pin sequence matters more than the apparent color grouping. On a common T568B termination, the orange-white pin is pin 1 and the solid orange wire is pin 2; the green-white pin is pin 3 and solid green sits at pin 6. Blue occupies pins 4 and 5, and brown occupies pins 7 and 8. The orange and green pairs split around the blue pair, which can seem odd during hand termination.

Keep each wire pair twisted as close to the plug as possible. Untwisting too much cable can introduce interference and reduce performance, particularly on faster links. A basic cable tester will confirm continuity and reveal swapped, crossed, or open pins before the cable is put into service.

T568A and T568B wiring at the connector

When considering “I choose the right network cable for my business,” it is important to understand the wiring standards used for Ethernet connections. T568A and T568B are the two recognized wiring standards for eight-position Ethernet connectors. T568A places green-white on pin 1, green on pin 2, orange-white on pin 3, and orange on pin 6. T568B reverses those green and orange pair positions: orange-white and orange use pins 1 and 2, with green-white and green on pins 3 and 6. Blue remains on pins 4 and 5, and brown remains on pins 7 and 8 in both layouts.

For a normal straight-through patch cable, use the same standard at both ends. T568B is common in many existing commercial installations, yet T568A is equally valid where it matches the site’s documented network wiring. Consistency matters more than selecting one standard over the other.

Older Ethernet equipment used MDI and MDI-X port arrangements. An MDI device would transmit on pins 1 and 2 and receive, or RX, on pins 3 and 6; an MDI-X port used the opposite arrangement. A crossover cable swapped those pairs for direct connections between like devices. Most current network gear supports auto-MDI-X, so it detects the pair relationship automatically, though a correctly terminated straight-through cable remains the usual choice.

Ethernet cables contain eight colour-coded copper wires arranged as four twisted pairs, with precise T568A or T568B termination ensuring reliable connections. Older 10/100 Mbps links use two pairs, while gigabit and faster Ethernet use all four. Choose solid pure-copper cable for permanent runs, stranded leads for flexibility, and suitable Cat6 or Cat6A cabling for speed, PoE and installation conditions.

Why each twisted pair carries Ethernet data

Ethernet cable is built around a simple but disciplined idea: send a signal over two conductors that work as one. Each twisted pair carries the same electrical information in opposite polarity, and the receiving equipment reads the difference between them. Interference picked up along the cable tends to affect both conductors much the same way, so the receiver can reject much of that unwanted noise and recover the intended data.

The twist matters. By repeatedly swapping the physical position of the two wires, a twisted pair reduces its exposure to interference from nearby power cables, lighting, motors, and the other wire pairs inside the jacket. Different twist rates help keep pairs from coupling too strongly with one another. It is a practical piece of cable engineering that supports dependable connectivity in homes, offices, retail units, and larger network installations.

Not every Ethernet system uses every pair in the same fashion. Older 10BASE-T and 100BASE-TX connections use two pairs: one pair to transmit and one to receive. Gigabit Ethernet uses all four wire pairs at once, with each pair sending and receiving data simultaneously. Faster standards place even greater demands on all four pairs, treating the cable as a coordinated path rather than a set of spare wires.

Two-pair Ethernet: 10BASE-T and 100BASE-TX use one pair to transmit and one to receive.
Four-pair Ethernet: Gigabit Ethernet uses all four wire pairs simultaneously.
Signal protection: Twisting helps limit noise and coupling between wire pairs.

That distinction helps explain why a cable can look intact yet fail at higher speeds. A damaged or poorly terminated pair may leave a basic link working, yet limit the connection to 100 Mbps or cause dropouts under load. Correct pin order, tight terminations, and cable rated for the intended run all protect the signal quality Ethernet systems rely on.

Twisted pair signal essentials

Each pair carries matching electrical signals in opposite polarity, allowing equipment to read their difference.
Noise affecting both wires similarly is largely rejected by the receiver, preserving the intended Ethernet data.
Twisting repeatedly changes wire positions, reducing interference from power cables, motors, lighting, and adjacent pairs.
Different twist rates help prevent pairs inside the cable from interfering too strongly with one another.
10BASE-T and 100BASE-TX use two pairs: one transmitting and one receiving.
Gigabit Ethernet uses all four pairs simultaneously, with every pair both sending and receiving data.
Damaged pairs or poor terminations may support basic links but reduce speed, stability, or performance under load.

Which wire pairs are used at each Ethernet speed

An Ethernet cable has four pairs, or eight wires, arranged as twisted pairs to control interference. The number of pairs a link actively uses changes with its Ethernet speed, yet a properly terminated modern cable normally carries all eight wires from end to end. That distinction matters when tracing a slow connection, repairing a damaged run, or deciding whether an older cable can serve newer devices on home and office networks.

10 Mbps and 100 Mbps: two pairs
1 Gbps, 2.5 Gbps, and 5 Gbps: all four pairs
10 Gbps: all four pairs, with stricter cable and installation requirements

At 10 Mbps Ethernet (10BASE-T) and 100 Mbps Fast Ethernet (100BASE-TX), data travels on two pairs: pins 1 and 2 form one pair, and pins 3 and 6 form the other. One pair sends and the other receives. This is why an old four-wire telephone-style lead can sometimes link equipment at 100 Mbps, provided its conductors land on the right pins. It cannot support gigabit service.

Gigabit Ethernet, rated at 1 Gbps, uses all four pairs at once. Each pair carries traffic in both directions, with the interface separating transmit and receive signals. The same four-pair approach applies to 2.5Gbps and 5Gbps Ethernet, often used to connect faster access points, switches, and storage devices without replacing sound Cat5e or Cat6 cabling.

10Gbps Ethernet over twisted-pair copper, known as 10GBASE-T, uses every pair too, but is more demanding about cable category, installation quality, length, and nearby electrical noise. Cat6 can handle 10Gbps on shorter runs; Cat6A is the usual choice for a full 100-metre channel.

For homes and small offices, the practical rule is simple: connect routers, switches, and other wired equipment with all eight wires present and correctly terminated. A wireless router may deliver Wi-Fi to phones and laptops, yet its wired uplink and LAN ports still rely on Ethernet pairs. If a gigabit-capable port negotiates only 100 Mbps, a missing, split, or poorly punched-down pair is a common suspect.

Ethernet pair usage guide

10BASE-T uses two pairs: pins 1–2 and pins 3–6, supporting 10 Mbps.
100BASE-TX also uses two pairs, allowing Fast Ethernet speeds up to 100 Mbps.
Gigabit Ethernet uses all four twisted pairs, requiring all eight conductors to be properly connected.
2.5Gbps and 5Gbps Ethernet likewise use all four pairs and can often run over sound Cat5e or Cat6.
10GBASE-T uses every pair but needs higher-quality cabling, careful termination, and low electrical interference.
Cat6 supports 10Gbps over shorter runs; Cat6A is preferred for full 100-metre 10Gbps links.
If a gigabit port connects at 100 Mbps, inspect for missing, split, damaged, or poorly terminated pairs.

Ethernet cable conductors: solid, stranded, copper and CCA

The conductor hidden beneath a cable’s coloured insulation has a direct effect on speed stability, PoE delivery and how well an installation survives daily use. Standard Ethernet cables contain eight conductors arranged as four twisted pairs. In a solid-core design, each conductor is one solid copper wire: a single, rigid strand that holds its shape and offers low electrical resistance. That makes solid cable the usual choice for permanent runs through walls, lofts, conduits and patch panels in a home. Once installed, it stays put, and its consistent geometry supports reliable long-distance links.

Stranded cable uses many fine copper strands for each conductor. It bends repeatedly without fatigue and is far better suited to patch leads between a wall outlet, switch, router, access point or desk equipment. Its flexibility comes with a small trade-off: stranded wire has slightly higher resistance and is less suited to long fixed runs. Use the right connector type, too; plugs and jacks designed for solid conductors do not always make a dependable termination on stranded cable.

Material matters just as much as construction. Pure copper is the preferred conductor for network cabling, particularly where the cable carries power through Power over Ethernet. Copper-clad aluminium, usually marked CCA, has an aluminium core with a thin copper coating. It costs less, yet its higher resistance can cause voltage drop, heat and unreliable performance, problems that become more likely on longer routes or with cameras, access points and other powered devices. For a short, low-demand lead, CCA may appear to work; for wiring you expect to leave in place, full copper Ethernet cables are the sounder purchase. Check the printing on the cable jacket rather than relying on a product title: “bare copper” or “solid copper” is the wording to look for.

Ethernet conductor selection checklist

Choose solid-core copper for fixed runs through walls, lofts, conduits and patch panels.
Use stranded copper for flexible patch leads that move frequently between outlets, switches and desk equipment.
Match plugs and jacks to conductor type; solid and stranded cable need different termination contacts.
Buy pure copper cable for Power over Ethernet devices such as cameras, access points and doorbells.
Avoid CCA for permanent wiring: its higher resistance can cause voltage drop, heat and unstable links.
Check jacket printing for “bare copper” or “solid copper”, rather than trusting the product title.

Category cable, shielding, gauge, length and PoE

The label on an ethernet cable tells part of the story, not all of it. Category cable sets the ceiling for speed and frequency: Cat5e remains serviceable for modest gigabit network runs, Cat6 is a sensible baseline for most new work, and Cat6A is built for 10 gbps over the full 100-metre channel. That rating only delivers its promise if installation is sound. A tightly kinked run, poor termination, mixed components or an overlong patch lead can undermine otherwise capable cable.

Shielding deserves a site-specific choice. Unshielded twisted pair suits many homes and ordinary offices, where it is easy to handle and usually keeps noise at bay. In workshops, plant rooms, retail ceilings, near fluorescent fittings, motors, lifts or dense electrical routes, shielded cable can offer worthwhile protection from interference. Shielding needs correct grounding from end to end; fitted carelessly, it adds expense and complexity without solving the problem.

Understanding what are network cables commonly used for can help when choosing the right cable for a particular installation. Gauge matters most on longer runs and any circuit carrying power. Thicker conductors, commonly 23 AWG solid copper in permanent Cat6 and Cat6A links, have lower resistance than thinner patch cable. That means less voltage drop and less heat when PoE is feeding access points, cameras, door controllers, or other devices. Avoid copper-clad aluminum for installed links: it is cheaper, yet performs poorly under PoE loads and is unsuitable for standards-based permanent cabling.

Plan the route before choosing a reel. Ethernet standards allow 90 metres of fixed horizontal cable plus up to 10 metres of patch leads, but real buildings impose bends, risers, service loops and future moves. Keep data cable separated from mains power where practical, cross electrical lines at right angles, and leave room in containment rather than packing it tight. Good cable selection supports stable systems and reliable networks today and leaves the network ready for faster equipment later.

Choose the right Ethernet cable for your installation

The right Ethernet cable starts with the route it must take, the equipment it serves and the speed you expect to use over the next few years. For a short link between routers, a television, games console or desktop PC, Cat6 Ethernet cables are a strong everyday choice: they support gigabit networking with comfortable headroom and are widely available in sensible lengths. Cat6a is the better fit for 10Gb connections, particularly on longer runs around a larger home, office or garden studio.

Length matters. Buy a ready-made cable for visible, simple connections, leaving a little slack for moving furniture or servicing equipment. For permanent wiring through walls, lofts or under floors, use solid-core, in-wall-rated cable rather than flexible patch leads. Solid copper conductors give dependable performance; avoid copper-clad aluminium wire, which costs less but is less suitable for powered devices and long cable runs.

Choose the jacket for the location. Standard PVC suits many indoor installations. Low-smoke, zero-halogen cable may be required in shared buildings or commercial spaces, and external-grade cable is built to cope with sun, damp and changing temperatures. Outdoor connections still need weatherproof boxes and protected entry points; a cable jacket alone will not make a poor installation durable.

Shielding is useful near heavy electrical wiring, machinery or dense AV racks, yet unshielded cable is usually easier to install and perfectly capable for ordinary home connectivity. Keep data cable away from mains wire where practical, cross power lines at right angles, and avoid tight bends, crushed sections and sharp staples. Finish permanent runs at wall plates or a patch panel, then use short patch cables to routers, switches and devices. Good cable management makes faults easier to trace and leaves room to expand your network without reopening walls.

Ethernet cable selection checklist

Use Cat6 for short gigabit links to PCs, TVs, consoles, routers and switches.
Choose Cat6a for 10Gb connections, especially across longer home, office or studio runs.
Buy ready-made patch cables for visible links; leave spare length for furniture moves and servicing.
Fit solid-core, in-wall-rated cable for permanent routes through walls, lofts and floors.
Choose solid copper conductors; avoid copper-clad aluminium, particularly for PoE devices and longer runs.
Match the jacket to its location: PVC indoors, external-grade outside, or low-smoke zero-halogen where required.
Keep data runs clear of mains cables, cross power lines at right angles, and avoid tight bends or crushing.