A single press of TEST runs a full Cat 6A autotest on the TestPro in ~6 seconds: wire map, length, DC resistance, insertion loss, return loss, crosstalk and the timing parameters. Here we break down each one: what it means, where a FAIL comes from, and how to read the tester's screen.
| Group | Parameters | Why it matters |
|---|---|---|
| Basic | Wire Map · Length (TDR) · Propagation Delay · Delay Skew | Whether the link is wired correctly at all, and how many meters it has |
| Transmission | Insertion Loss (IL) · Return Loss (RL) | How much signal gets through and how much is reflected along the way |
| Crosstalk | NEXT, PS-NEXT · ACR-N · ACR-F, PS-ACR-F | Whether the pairs interfere with each other — critical for 10GBASE-T |
| Resistance / PoE | DC Loop Resistance ≤ 21 Ω · DC Resistance Unbalance ≤ 3% | Whether the link can safely carry PoE power |
| EMI immunity | TCL · ELTCL | Cable balance in noisy environments — required with the ++ limit |
The set of parameters that actually counts toward the PASS/FAIL criteria depends on the limit modifier (none / + / ++) — details in module 3. The rule: a network carrying PoE gets certified with the ++ limit.
The tester measures length by TDR separately for each pair — which is why the results for pairs 12/36/45/78 differ by a few dozen centimeters: each pair has a different twist rate, so its conductors are physically longer or shorter. Propagation Delay (≤ ~555 ns) is the time a signal takes to travel down the pair; Delay Skew (≤ 45 ns) is the difference between the fastest and the slowest pair.
Why skew is dangerous: 1000BASE-T and 10GBASE-T transmit in parallel over all 4 pairs. Too much spread in the delays desynchronizes the streams — bit errors and retransmissions appear, even though "the cable works".
The tester screen: the "Length (m)" table shows the result for each pair individually against the link limit. Differences between pairs are normal — the alarm is a value over the limit, or a result completely at odds with the design (suspect a wrong NVP).
The drop in signal power along the link. It grows with frequency and cable length; the Cat 6A limit extends to 500 MHz. On screen: the pair curves plotted against frequency, with a red dashed limit line — the curves must stay below it, and the tester reports the critical pair, the result and the margin in dB.
Energy reflected from points where the impedance deviates from the nominal 100 Ω. The reflected signal superimposes on the useful one and distorts the data. Here, the higher the value in dB, the better: RL > 20 dB means negligible reflections, RL ≈ 0 dB means a short or an open.
After an RL FAIL, go straight to the HDTDR locator (RL Locator) — the peak on the trace pinpoints the fault location in meters. We break down troubleshooting in module 5.
Picture two pairs of people talking side by side: pair A speaks loudly and part of their voice "leaks" into pair B's conversation. The leak is loudest right next to the speaker — that is, at the near end. That is how NEXT (Near-End Crosstalk) works: a signal transmitted on one pair couples into a neighboring one and is strongest at the transmitter. The twist is how each pair "keeps its voice down" and stays in its own conversation — a tighter twist means less leakage and a higher (better) result in dB.
NEXT is the most sensitive indicator of termination quality. The standard allows a maximum of 13 mm of pair untwist at the jack — in practice: the less, the better. PS-NEXT sums the crosstalk from the other three pairs at once, simulating real 10GBASE-T transmission.
The same analogy, but we listen at the other end of the room: pair A's voice has faded along the way (attenuation), and the leak is only heard at the far end — that is FEXT. Since raw FEXT depends on the link length, it is hard to set a direct limit for it. That is why the criterion is ACR-F = FEXT − IL (formerly ELFEXT): it compares the leakage against the signal attenuation, which makes it length-independent. The higher the ACR-F, the "cleaner" the far end; PS-ACR-F is the headroom for 10GBASE-T with all 4 pairs active.
The ACR-N / ACR-F screen: the result curve should stay above the limit curve across the entire band. ACR-N = NEXT − IL answers the question "is the useful signal stronger than the interference" — a positive margin across the whole band means the link will sustain 10GBASE-T.
The opposition the conductors present to direct current — like a narrow pipe restricting water flow. The longer and thinner the conductor (or the cheaper the material, like CCA), the higher the resistance. Two parameters:
The "DC Resistance" screen: Loop against the limit, unbalance within the pair [%], P2P unbalance. DC resistance only counts toward PASS/FAIL with the + / ++ limit — which is why for PoE installations the ++ limit is not an option, it is a must.
The cable balance (symmetry) parameters — the better balanced the cable, the less it radiates and the more effectively it rejects radio noise. Critical in smart buildings, industrial installations and anywhere twisted pair runs next to power lines and drives. TCL measures the asymmetry at the near end, ELTCL — with attenuation compensation, along the entire line.
After you press TEST, the tester runs in one pass: wire map, length (TDR), DC resistance (loop + unbalances), IL, RL, NEXT/PS-NEXT, ACR-N, ACR-F, the Power Sum values, Delay & Skew and the headroom. It is the standard — the selected limit — that decides which of them enter the PASS/FAIL criteria and which are informational only. PASS (green) = every parameter with margin to spare; FAIL (red) = at least one outside the limit; margin = the difference in dB between the result and the limit, and the verdict is decided by the worst margin across all frequencies and pairs.
You now know what the tester computes under the hood. Time to configure it — so it computes the right things against the right limit.