Unraveling the Mystery: How Your Network Cable Tester Really Works

A network LAN cable tester works by sending a signal down each wire of an Ethernet cable from a main unit and checking, with a remote unit at the far end, that the signal arrives on the correct pin. A simple tester lights up LEDs 1 through 8 in order to show the wiring is right, and a skipped, doubled or out-of-order light points to a broken wire, a short or a miswire. More advanced testers go further: they time reflected pulses to measure length and find faults, and certification testers measure how well the cable carries high-frequency signals against the ANSI/TIA-568 limits. This guide explains each level, how to read the results, and when a simple tester is not enough.

The three kinds of network cable tester

"Cable tester" covers tools that range from a pocket gadget to a field instrument costing as much as a used car. The industry groups them into three levels, and each one answers a different question.

Level	Question it answers	What it measures
Verification (continuity or wiremap tester)	Is every wire connected to the right pin?	Opens, shorts, crossed and reversed wires; some models add length and split pairs
Qualification	Will this link run a given network speed, such as 1 Gb/s?	Wiremap and length plus signal checks tied to Ethernet speeds; often PoE and live network details
Certification	Does this link meet the standard for its category?	Every ANSI/TIA-568 parameter, including insertion loss, NEXT and return loss, with a pass or fail report

A cable can pass a verification test and still fail certification. Wiring correctness and signal quality are different things, and the rest of this article explains why.

How a basic continuity and wiremap tester works

A basic tester has two parts: a main unit with a battery and a small circuit that drives each pin in turn, and a remote (sometimes called a terminator) that plugs into the other end. A standard Ethernet cable has eight wires arranged as four twisted pairs, terminated on pins 1 through 8 of an RJ45 plug or jack. Many testers also check pin "G" for the shield on shielded cable.

The main unit puts a small voltage on pin 1, then pin 2, and so on. Each wire that is intact completes a circuit to the matching LED on the remote. On a correctly wired straight-through cable, the main and remote LEDs step together: 1 and 1, 2 and 2, through 8 and 8.

Reading the lights

• A light that never comes on means an open: the wire is broken, a conductor did not seat in the jack, or a punch-down missed.
• Two lights at once means a short: two conductors are touching, often from a crushed cable or a bad crimp.
• Lights out of order (for example main 1 lights remote 2) means a crossed or reversed wire. One end may be wired to T568A and the other to T568B.
• The crossover pattern (1 to 3, 2 to 6, 3 to 1, 6 to 2) means the cable was deliberately or accidentally made as a crossover.

T568A and T568B are the two color codes defined in ANSI/TIA-568. Either works as long as both ends of a link use the same one; mixing them produces a crossover.

What a basic tester cannot see: split pairs

A split pair is the classic fault that fools an LED tester. Every pin connects to the matching pin at the far end, so the lights step perfectly, but two wires from different twisted pairs have been swapped at both ends. Because the twist is what cancels interference, the signal on that "pair" picks up heavy crosstalk. The link may work at low speed and fail at gigabit speeds. Finding split pairs takes a tester that measures crosstalk between pairs, which basic continuity testers do not do.

How testers measure length and find the fault: TDR

Better testers can tell you that a cable is 62 meters long and that wire pair 3-6 is open at 41 meters. They do this with time domain reflectometry (TDR), the same idea as sonar.

1. The tester sends a short electrical pulse down a pair.
2. Wherever the cable's impedance changes (an open end, a short, a kink, a bad connector), part of the pulse reflects back.
3. The tester times the round trip and converts it to distance using the cable's nominal velocity of propagation (NVP), the speed of a signal in that cable as a percentage of the speed of light. Twisted-pair cable is commonly around 65 to 70 percent.
4. Distance equals the time multiplied by the signal speed, divided by two (because the pulse travels out and back).

Because NVP varies between cable brands and categories, length readings can be off by a few percent unless the tester is set to the cable's published NVP. Some inexpensive testers estimate length from the cable's capacitance instead, which is less precise.

Length matters because ANSI/TIA-568 limits a twisted-pair channel to 100 meters: up to 90 meters of permanent cabling plus up to 10 meters of patch cords. A tester that reports 105 meters has found a design problem, not just a wiring one.

How certification testers work: NEXT, insertion loss and return loss

Ethernet over copper does not send one simple signal. Gigabit and faster links push high-frequency signals in both directions on all four pairs at once, so the cable's behavior across a range of frequencies decides whether the link works. Cat 5e is specified to 100 MHz, Cat 6 to 250 MHz and Cat 6A to 500 MHz. A certification tester sweeps across that range and compares each measurement to the limit line in ANSI/TIA-568 for the category being tested. The main measurements are:

• Insertion loss (formerly called attenuation): how much signal strength is lost from one end to the other. It rises with length, frequency and temperature.
• NEXT (near-end crosstalk): how much signal from one pair leaks into a neighboring pair at the same end. Untwisting wires too far at a jack is a common cause of NEXT failures. PS NEXT (power sum) adds up the leakage from all the other pairs at once.
• ACR-F (formerly ELFEXT): crosstalk measured at the far end, relative to the received signal.
• Return loss: how much signal reflects back toward the sender because of impedance mismatches, often from poor terminations, kinked cable or mismatched components.
• Propagation delay and delay skew: how long the signal takes, and how far apart the four pairs arrive.
• DC resistance and resistance unbalance: important for links that carry Power over Ethernet, since unequal resistance between wires in a pair can disturb data and power delivery.

The tester reports a pass or fail for each parameter, the worst-case margin, and a plot across the frequency range. Accuracy requirements for these field testers are set in TIA-1152 (and IEC 61935-1 internationally), so a Cat 6A certification needs a tester rated to the matching accuracy level.

Permanent link versus channel

Installers usually certify the permanent link: the cable in the walls from the patch panel to the jack, using special adapters so the tester's own cords do not count. A channel test includes the patch cords the equipment will actually use. Permanent link results are the ones normally handed over at the end of an installation, since patch cords are often replaced.

How testers handle Power over Ethernet and live networks

Many devices now take power from the network cable under the IEEE 802.3 PoE standards (802.3af, 802.3at and 802.3bt). That affects testers in two ways.

First, a basic continuity tester is built for dead cable. Plugging it into a powered switch port or an old telephone line can damage it or give a false reading. Better testers detect voltage on the line and warn you before testing.

Second, qualification and network testers can act like a powered device on purpose. They go through the same detection handshake a camera or access point would: the switch checks for a signature resistance, the device reports its power class, and the switch turns power on. The tester then reports which PoE standard and class the port offers, and some apply a load to confirm the port and cable can actually deliver the power under strain. The same testers often show the negotiated link speed and, on managed networks, the switch name and port number using LLDP or CDP.

For tracing an unlabeled cable through a ceiling, a separate tone generator and probe is still the standard tool: the generator puts a tone on the wire and the probe picks it up through the jacket.

Testing fiber links

Fiber testers use light instead of voltage, but the logic is similar. ANSI/TIA-568.3 describes two tiers. Tier 1 uses an optical loss test set (a calibrated light source at one end and a power meter at the other) to measure total loss and compare it to a loss budget built from the length, number of connectors and splices. Tier 2 adds an OTDR (optical time domain reflectometer), which works like copper TDR: it sends pulses of light and maps reflections to show the location and loss of each connector, splice or bend. A visual fault locator, a bright red laser, is the simple tool for spotting a break or tight bend near the end of a cable.

When a basic tester is enough, and when it is not

A basic wiremap tester is worth having in any IT closet. It is the right tool for checking a patch cord, confirming you crimped a plug correctly, or ruling out an open wire before you dig further.

It is not enough when:

• A new installation needs proof that it meets its category, especially if the cabling manufacturer's warranty requires certification reports.
• Links work at 100 Mb/s but drop or slow down at gigabit or 10 gigabit speeds, which points to crosstalk, split pairs or loss.
• Wi-Fi access points or cameras reboot or lose power, which can point to PoE or resistance problems.
• Faults come and go with heat or movement, which needs TDR to locate.

Troubleshooting older networks is often as much detective work as testing. Careful testing before and after a repair is what separates a fix from a guess.

Frequently asked questions

Can a cheap cable tester tell me if my cable supports gigabit?

No. It confirms the wiring is connected correctly, which is necessary but not sufficient. Gigabit performance depends on crosstalk and loss, which only qualification or certification testers measure.

Why does my tester pass a cable that still does not work?

Common causes are a split pair, a run longer than 100 meters, a damaged section that only fails at high frequency, or a problem at the switch or device rather than the cable. A TDR or certification test usually narrows it down.

What do T568A and T568B mean on a tester?

They are the two wiring color codes in ANSI/TIA-568. Either is fine if both ends match. If one end is A and the other is B, the tester will show a crossover pattern.

Do I need certification results for a new install?

For any business installation, it is wise to get them. They prove each link met its category on the day it was installed, give you a baseline for later troubleshooting, and are usually required for a manufacturer's system warranty.

Accutech Communications has installed and serviced business network cabling across Massachusetts, New Hampshire and Rhode Island since 1993, working from our office in Hopkinton, MA. Our network cabling installation services include Cat 5e, Cat 6 and Cat 6A cabling, cable testing and certification, inspection, and troubleshooting and repair. For a free estimate on a new install or help with a cabling problem, call 978-579-0006.