For my first post on Stack Exchange I have a situation that seems to be impossible: fast internet on cables that show no continuity.

I just installed 16 Cat6 ethernet runs in an office. All of them are under 100 feet and most are under 50 feet. The 10 shortest runs have no troubles at all. All 16 runs tested good for continuity when installation was completed. Four days later I re-tested all of them because one wasn't working right (fixed by re-terminating, one wire was a little loose). With the second testing, the 6 longest runs have the oddest results I've ever seen:

  • No continuity on any of the 8 wires. Four different testers agree. (Two NETcat NC-100 from Greenlee and two IPC testers.)
  • Excellent internet speeds. > 500/500 Mbps after hours, roughly what the ISP provides. Measured from my laptop with wifi turned off.

My internet searching only turns up the opposite problem in which continuity is good but there's no internet. That's not my situation.

Any ideas what could be going on?

That situation and possibly solving it is my top concern. ("Possibly" solving it because the cables are working, after all.) The discussion could easily turn to getting better testing equipment. I am open to that and interested in suggestions. Part of the context is that my shop is a small, rural, jack-of-all-trades shop which is primarily a managed service provider for small businesses and residential customers. We rarely pull more than 5,000 feet of ethernet cable per year, and none of our customers (so far) ask for cable certification results. A high-end tester for $2k or more would be excellent and might be what we get, but the return on investment would be much less than it would for a tech whose main job is pulling data cables.

1 Answer 1


Without even passing continuity tests, there's no way a twisted-pair cable could work. Chances are you've mixed up runs. Checking just continuity for any twisted-pair deployment isn't near enough.

Gauging cable quality by achievable Internet rates is very inaccurate. You need to check each run individually. Lacking a decent tester I used to run two-way iperf tests using two laptops and monitoring error counts.

Higher cable categories require increasingly sophisticated measuring for certification - a decent cable tester can be rented.

  • You were correct, Zac67. My assistant punching down at the panel told me he'd changed the order of the wires when he punched them down and I labeled them based on what he told me... but he hadn't actually changed the order. The greatest problem with communication is the illusion that it has occurred. But my bad not double-checking nor thinking of that possibility. He was at the patch panel moving the connector for the first round of testing, which is why everything passed then, but not for the second round of testing, for which I followed the labels. Now all I have to do is re-label. I'll star
    – WolfeWires
    Mar 16 at 15:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.