When Cable Fails Part 3

We walk through the cable test results that look “fine” until you understand what the graphs are really saying, from split pairs that still link up to length readings that don’t match reality. Along the way, we share practical ways to spot water intrusion, validate extended distance copper, and avoid patch cord mistakes that can quietly break PoE.
• split pairs and transposed pairs causing crosstalk and data loss even with a link light
• cable length vs electrical length plus propagation delay and delay skew for gig speeds
• why NVP settings matter and how faked settings show up in project-wide test reviews
• return loss patterns that point to water in the cable and using HDTDR to locate it
• pulling lubricant vs water symptoms and how insertion loss can improve as lube dries
• extended distance cabling and selecting vendor-specific test limits in the tester
• alien crosstalk testing workflow using saved results, disturbed links, and disturbers
• thin patch cords, PoE risk from rising resistance, and using the correct adapters for patch cord tests
So make sure you join us for this conversation and let me know in the comments below if you enjoyed this episode and should I do more like this?
Email Chuck at advertising at letstalkcabling.com and let's connect your brand to the right audience today.

Transcript

Copy Transcript Copied

Welcome And Testing Focus

SPEAKER_03 0:00

Hey, this is Edding Talk again, week three for the testing series with Mike Pinocchio. This week we talk about alien crosstalk testing, we talk about water on the cable, we talk about extended distance cabling. So make sure you join us for this conversation and let me know in the comments below if you enjoyed this episode and should I do more like this? Let's get to the show.

Split Pairs And Pair Swaps

SPEAKER_03 0:49

Here's the next one.

SPEAKER_04 0:50

Oh, split pair. Yeah, that I love this one because people go, what's happened to my cable? It's all shorted together. It's well, it's not really shorted together. The tester just doesn't know what to do with it. And uh this is a case where one is going to one and two is going to two, but they are not on the same pair. And if you really want to mess with somebody, wire this cable up for them and let them try and get to the internet or something like that. Because they may get a link light. Yeah, but they're gonna see a lot of data loss. And if we were to look at the performance graphs on that one, we would see uh a massive amount of crosstalk. Right.

SPEAKER_03 1:32

And this one I think is the easiest one of all, but yet so many people have problems with it. Yeah, A on one end, B on the other. Exactly, right. They got what's called a transposal. They flip, they flipped entire pairs. Alrighty, let me flip back over to my notes because I lost track of where we were at. Length,

Length Delay Skew And NVP Tricks

SPEAKER_03 1:48

length of propagation delay problems. So we know what length is, right? Yeah. So what propagation and what's what's delay skew?

SPEAKER_04 1:59

Well, let's take a look at an example here.

SPEAKER_03 2:02

Um it's your turn to put up some pictures.

SPEAKER_04 2:08

Yeah, let's see if I've got oh, here's a good one. All right. And this is this is a fun thing, is I get to sit out here and make bad cables, make things fail so that I can have examples to show. So here, uh, I'm gonna use this one as an example, and that is uh, first off, when we go into length, it gives us our it gives us three measurements. We've got length, we've got delay skew, we've got propagation delay. Now, the interesting thing with this is let's say I was sitting there and I was using the tester and I got somebody that's looking over my shoulder, kind of like I am when the Comcast guy shows up here, right? Right, right, right. I think they have to draw straws to see who's going to come into my house because I got lots of questions for them. I do the same thing. And the last time uh they ended up have they brought out a bunch of the facilities guys because they found things that were wrong along the path because I kept asking questions. But I got someone looking over my shoulder and they say, Mike, this says 314 feet and the limit's 295. Why did it pass? And what we find is that ANSI TI 1152A gives us some wiggle room when it comes to length. If length is the only thing that fails, because we could screw up the nominal velocity of propagation. And the fact that we are allowed to control this value that determines the length of the cable, I think is kind of an interesting one.

SPEAKER_03 3:35

Right.

SPEAKER_04 3:35

And it is why selecting the manufacturer's cable from the list of manufacturers is good because you're using the MVP that they've specified that they've given Fluke, the Fluke is put into the firmware.

SPEAKER_03 3:48

Yeah, exactly. And you know, when you you know, I think that's called the 10% rule, I think.

SPEAKER_04 3:53

Yes, it is exactly called the 10% rule.

SPEAKER_03 3:56

So you're like it's gonna be 295 plus or minus 10%. So that's why you can technically have that. And as you as you mentioned earlier, the difference between those pairs is the number of twists. The tighter the twist, the longer it is. Um, the shorter the twist. So there's a there's a there's a physical length to a cable and an electrical length to a cable. So the physical length is physically how long the cable is. And the electrical length is when you actually go look at the length of the pairs, and that's where a lot of technicians um are still getting kind of messed up. Although it's gotten better. You know, when I first started teaching classes 15 years ago, it would just, I mean, it would blow their mind when I go down that conversation. But now, again, that's one of the questions I ask. You know, hey, when you're testing, what do you notice about the lengths of the pairs? They always coming, yeah, it's the they're all different languages because of the twist of the pairs. So the industry's doing good about educating that, right? Yep.

SPEAKER_00 4:49

Um at TextGiving this year. You get to be a part of something bigger than just another industry event. TextGiving brings together technology professionals, manufacturers, vendors, and volunteers from across North America to use their skills for real impact. This year we're heading to Philadelphia thrust on McDonald's house in what is shaping up to be our biggest and most ambitious project yet. Together, we'll help upgrade critical technology infrastructure, including networking, security, access control, and more, so they can stay focused on supporting families when they need it most. But it's not just about giving back, it's also an incredible opportunity to network, learn, and work side by side with some of the best people in the industry. Whether you're brand new to the field or season pro, you'll make connections, share knowledge, and get hands-on experience. And that is what tech knowledge is all about: community over competition. Nearly a million dollars in product and services have already been donated through our Tex Giving initiative. So come join us, bring your skills, meet amazing people, and be a part of something that makes a real difference. We'll see you in Philly at Texgiving, October 23rd through the 25th.

SPEAKER_03 6:08

So, what is the what's the delay skew measurement?

SPEAKER_04 6:12

So, what we have is we're transmitting and receiving on all four pairs at the same time. So that means it's critical that if I put a signal in on two pairs at the same time, that those signals arrive at roughly the same time at the far end. So for gig and above, our delay skew becomes very critical. And in this case, we are allowed up to 44 nanoseconds of difference. And here, uh, because pair seven eighths are shortest pair, that's the one we use to evaluate the length. And it's also zero for the delay skew because it's the shortest. They're going to get there first. And then we see that you know, we've got two, 10, 15 coming in here. And but we're under the 44. Now, there is a document that anybody can download off the Fluke site. If you go out to Fluke Networks, you go to download support downloads and updates, and it's called the limit lines. And the limit lines defines every one of our test limits. And so it says in there, here's the maximum allowable delay skew, the maximum allowable propagation delay. This is the time it takes to get from one end of the cable to the other. So just like we can screw up the NVP and make the cable look longer than it really is, we can change the MVP to make it look shorter. And people have figured that one out. And they have messed with the NVP to try and make a cable that's over length look shorter than it is.

SPEAKER_03 7:46

Manufacturers catch that though.

SPEAKER_04 7:48

They absolutely not only do they catch that, and one thing I've played around with uh is running some of the test reports through Chat GPT and saying, look for any deviations. I never thought about that. In in this, and there's uh a guy that's at a company in California that was showing Jim and Jim and I that, and we've been doing a lot for the fiber side of things. But one thing that I've I want to see if I can do is get it to recognize if two cables look so much alike that if somebody has a cable that fails, we've seen cases where people will back up and test the previous cable again. Right. And so if we can use AI to say, hmm, no, because things like return loss become a fingerprint of the cable, no two cables could look exactly the same. Right. This would identify cases like that. But our propagation delay is just you know how long it took to get from one end to the other. And that we can't fake, we can't change. So if the cable's really too long, we're gonna fail propagation delay or insertion laws. Right.

SPEAKER_03 8:57

And when you run the reports, um, and this is one of the things I used to do when I used to evaluate test results, is I would look at, I mentioned earlier, I would look at through different views, and if it's a 100-cable project and 98 of them said the NVP were 0.71, and two of them said five, why did you change the MVP? I I mean I'm I already know why. I just want to see how honest you are. And from a manufacturer's perspective, if you are one of their trusted contractors and you're willing to fake test results, that's not a good thing. That's not a good thing. So that's why I tell people they think they can get away with it. I'm telling you, there's too many, these things are way too sophisticated that you know you can still you can still kind of figure out what's going on with it, right? Bad or damaged, bad terminations or damaged cable. So you kind of already touch based on that. Um so you look at the you look at the the high definition HDTX. I can never remember the acronym. The acronym person, you think I'd remember HDTX. Um just one of those that in my memory. So you can actually go in there and you find, but let's talk about one of the things that shows up quite often. And I think we actually have a I think we actually even have a uh one forward in the thing here. Hold on. Let me oh hold on. I gotta put on my uh my my glasses so I can read the uh read these these things. I know we have one. I don't see it. Did I forget to put you have one for water?

Water In Cable Return Loss Clues

SPEAKER_04 10:31

Oh yeah, I love that one. Yeah, because I I actually went in and created that one, and that was uh let me bring this up. So what I did was I took uh a 10-foot piece of electrical PVC and I put sweeps on both ends of it, and I um oh, there it is, and so I filled it full of water. Well, actually, I tested the cable first and it passed. I filled it full of water, and then I said, okay, I'm gonna test this every hour and I'm going to see when it starts failing. And after five hours, I got bored. I poked a little hole in it with an X-acto knife, and this is what I got. Now, what we see is everything passes except return loss. Now, when I look at return loss, this is what I see, and it almost looks like waves. And this is very characteristic of water in the cable, and that is all of our pairs are following each other at the same frequency together at the lower frequencies. They're not spread out. And I had some exactly. We would expect them kind of all over the place above the limit line, and this isn't. And uh, someone sent me a linkware report one time, and I saw this in the return loss graph. I said, Okay, you got to tell me a story about where is this cable running and what's happened recently? And they said, runs through a conduit underneath the parking lot. I said, Okay. I said, has it rained recently? Well, we had some monsoons come through here. I said, Well, you have water in that conduit. And they said, but not all the cables are failing. I said, not all the cables are underwater. So it doesn't mean the conduit's full of water, right? But some of them are underwater, and this is what we see.

SPEAKER_03 12:20

Let me ask you this. If it does that for water, pulling lubricant's 97% water. So if you use pulling lubricant, too much pulling lubricant, it's often told to us that it'll fail because of this. Would it look the exact same way?

SPEAKER_04 12:35

No. What we're gonna see with we will fail, we can fail with pulling lubricant. And if you search the fluke knowledge base and you search for LUB, uh, there's a great little video from Adrian Young talking about this.

SPEAKER_03 12:48

Love Adrian, he's a great guy.

SPEAKER_04 12:51

And uh what they found is the lubricant absorbs the signal that's used for insertion loss. And so oftentimes, if we fail insertion loss and we know the cable is shorter than the maximum length, a possibility if we use lubricant is that that lube is that it is absorbing some of that signal. Now, as that lubricant dries or cures, we will see our insertion loss will get progressively better. Now, in this case, because I failed return loss, and if I click on here, it says, look, dude, you're never going to dry this cable out. You're gonna need to put new cable in. What I need to know now is where the water is, because when I pull that cable out, I don't want to drag that water further into that conduit. So I come into HDTDR, and we were looking for this not to exceed 0.8%. And here I've got this spike at 19.76. So one of the cool things that I can do with this graph is I can come in here. Let's see if it'll drag, we'll reach over here to the tester and I'll put the marker right there. So I can tell you that that water is at 31.1 feet in from the main. So that is where the water begins. And if I tap on this and I say mark, I can now drag that yellow dot over to the where it begins at the remote. I can tell you that there is about 15 feet of water in there. So not only can I tell you there's water, I can tell you where the water begins from the main side, and I can tell you how much water, how much of that is wet.

SPEAKER_03 14:40

Very cool. All right, I got a couple questions that we didn't talk about, I didn't pre-arrange, but I have a feeling you won't have any problems with them. Okay,

Extended Distance Cable Test Limits

SPEAKER_03 14:48

the first one extended distance cabling. Yes. Right? So again, 295 plus or minus 10%, and every manufacturer now has an extended distance cable. Some of them going up to 200 meters, sometimes even a little bit more, right? Yep. The tester is gonna fail you as soon as you get past that 295. So, what additional steps do they have to do in the fluke to be able to pass that extended distance cable?

SPEAKER_04 15:16

That is a great question. I get that in class a lot of times. And so if we come in and let's bring our tester back up again, and let's say I want to create a new test and I'm just gonna game changer, right? It's one that a lot of people ask about. And so they say, Hey, look, I got some game changer cable I need to test, but it's failing Cat 6. Well, yeah, because it's over the 90 meters or it's over the 100 meters. So the first thing I would do is I would come in and set my cable type. I go into manufacturers, and I would run myself down here to Game Changer because Game Changer is a cable that is listed in here. And I could select, you know, which of the game changer cables. Uh so I'm gonna just I'm gonna say, let's say we're using outside plant shielded cable. So shield test is on, sets my NVP for me. So now I go into test limit, and again, there's always more. And I go into more, and I've got TIA, I've got ISO, I've got patch courts, I've got application. If I come down here, I have vendor. And when I get into vendor, we see I've got game changer one gig, I've got game changer 10 meg. Now 10 meg is good up to 850 feet.

SPEAKER_03 16:37

Yeah.

SPEAKER_04 16:38

Uh one gig is 600 and it's two 200 meters, it's 656 feet. And so when I check that, that is going to apply their specific test limit. They have worked with Fluke to come up with a test limit that matches what they expect out of the cable. So if you pass this, you know that you're doing what Game Changer would expect it to do.

SPEAKER_03 17:01

Do you have a table for? Because like I said, um, Game Changer is not the only one out there now. Like they were for the longest time, but now all the major manufacturers have extended distance. Do you have files for all of them?

SPEAKER_04 17:11

So so there are a lot of them in here. And one of the things that we find is that there is a way that I like I can come in. Um we've got some superior access, we've got Systemax, Giga Reach. Um, these are ones that shipped that are shipped in version 7.1 of the firmware. Uh, there are ways we can import test limits as well. So there are cases where if somebody's working on a job and uh the manufacturer will work with Fluke to define the test limit, we can import a one a one-off test limit uh that may not be in the standard revision of the code.

SPEAKER_01 17:58

Let's take a short break. Are you trying to reach the technicians, project managers, and decision makers of the ICT industry? Then why aren't you advertising on Let's Talk Cabling? With over 150,000 impressions a month across podcasts, YouTube, and social media, this isn't just a show, it's the go-to resource for the low voltage industry. We spotlight the tools, training, and technology shaping the future of structured cabling. And your brand could be front and center. Don't just get noticed, get trusted. Email Chuck at advertising at letstalkcabling.com and let's connect your brand to the right audience today. Another thing. Alien

How To Run Alien Crosstalk

SPEAKER_01 18:46

crosstalk testing.

unknown 18:48

Yeah.

SPEAKER_03 18:48

You're testing uh, you know, um how much one cable interferes with another cable. You know, according to the standards, that's not a uh um a required test, but it I've had people tell me that they've had customers request that.

unknown 19:04

Yep.

SPEAKER_03 19:04

You're not gonna be able to do that with a DSX, are you? Yeah. Oh, you can? You can do alien cross-talk testing?

unknown 19:09

Yeah. Yeah.

SPEAKER_03 19:10

Can you explain now?

SPEAKER_04 19:11

So the way we do, yeah, the way we do the alien crosstalk testing with the DSX is first off, we do all of our, let's say we're doing cat 6A. So I do all my Cat 6A permanent link test first. I save those test results to LinkWare PC and I export it as an FLW file. So I've got my Fluke LinkWare file that has all the results for all those cables. The next step is I need to go in and I need to select my disturbed cable and my disturbers. Okay. So when I select my disturbers, I'm typically going to select those jacks to the left and the right, right, and above and below in the patch panel, and all the cables that are bundled together. Now, what I'll do is I connect my permanent link adapter from my main to the to the disturbed link. I connect my permanent link adapter for my remote to the first disturber. There's a terminator I put at the end, and then flip over here to the bench. So if we flip over to the bench, on the side of the DSX module, there is an RJ45, which we don't normally use. But what we do is we use that RJ45 to connect the main and remote together. So now the main can drive the remote for this testing. So when we go to when we do that, we connect the main to the PC using a USB cable. We load up the alien crosstalk software, which is free from fluke. We specify the disturbed, the disturbers, and then we start running the test. And so what it will do is it'll transmit on the remote, it'll listen on the main. And so that way we can do our powersome alien near-end crosstalk and our power-sum alien attenuation crosstalk ratio near end. That's cool.

SPEAKER_03 21:08

That's cool. I thought it was a new test for some reason. Um, I don't know why I thought that. Now here

The Mystery 600 Foot Reading

SPEAKER_03 21:13

comes, here I'm really gonna stretch you on this one because somebody asked me, somebody asked me, I didn't know the answer to this, and I reached out to several of my friends who are engineers, and they didn't know. It took us, it took us like three weeks to figure out what it was. So let me describe the situation to you. A friend of mine called me, he's he's uh I want to say he's in the Denver area, but it doesn't really matter. He was on a project running some cables. He ran a bunch of cables, they got in before everybody else, they ran the cables in the ceiling, and then uh the other trades came in afterwards the electrical, HVAC, the plumbing, they all came in afterwards. He started testing cables. So he had he had a he had a couple phase plates right next to each other, right? I think it was like four or six cables. Two of them read that the cable was 140 something feet. The other one said they were 600 feet. 600 feet. So my first thing is, okay, did you use the same tester on all those?

SPEAKER_04 22:11

Right.

SPEAKER_03 22:11

Did you have the tester set up correctly on all those? And he's like, Chuck, I'm not an idiot. Yes, we test them all the same day, all with the same tester, all with the same settings. I'm gonna give you a break. I can't do this. So there's this, there's this thing called, hold on, I gotta look it up real quick, called the lens law. You know what the lens law is?

SPEAKER_04 22:34

I have no idea.

SPEAKER_03 22:35

So if you take a copper tube, right? Let's say uh uh you know half-inch copper tube, and if you draw a if you drop a ball, steel ball down the center of it, it slows the ball. And then it comes out. So it doesn't keep the same radius drop. It's called the lens law, right? Okay what we found out was the plumbers came in afterwards and ran their copper pipe right through the cable bundles, so the cables were wrapped around the copper tube.

SPEAKER_04 23:04

Wow.

SPEAKER_03 23:06

Yeah, yeah. You know, when he finally found figured that out, I was like, wow, I would have never guessed that.

SPEAKER_04 23:13

I would have Because we're we're looking at, I mean, the way we measure length is we send a 10 megahertz signal down the cable, and when it hits the end of that cable, we see a big impulse come back to us. And so all we're doing is measuring that propagation time. So that would have been interesting to see what the propagation delay was.

SPEAKER_03 23:33

I'll reach out to him, see if he tested it with a fluke. I don't remember what tester he had. And if he did send, if he did test it with a fluke, um, I'll see if I can't get a copy of those test results and send them to you.

SPEAKER_04 23:44

That'd be great.

SPEAKER_03 23:45

Yeah, that would be great.

SPEAKER_04 23:46

You

Thin Patch Cords Adapters And PoE

SPEAKER_04 23:47

know, one of the things that I you uh there was a webcast, uh one of your webcasts I was watching, and you were talking with somebody, and you were talking about the I this idea of the channel and the permanent link. And the permanent link, we call it that because it's the part that stays in the walls for decades.

SPEAKER_03 24:05

Right.

SPEAKER_04 24:06

And the patch cords are consumables.

SPEAKER_03 24:09

Right.

SPEAKER_04 24:09

And one of the things that we've seen a lot of questions about are thin patch cords.

SPEAKER_03 24:14

Oh, yeah.

SPEAKER_04 24:17

You know, I would never use those. So I I put a little video up. My the rack I've got in here is chock a block full of 28 gauge patch cords. That's all that's in there. And I've been running that. The switch has been up for over a year, and I posted a picture, I've seen zero errors.

SPEAKER_03 24:38

Oh wow.

SPEAKER_04 24:39

I have not dropped a single packet on that switch in over a year. So when it comes down to the practice practically how do they work, I have seen that I'm running PoE across them, they're working. But let me just bring this up if that's okay. Yeah, yeah, absolutely. Just to show you. Uh, now it's going back to my bench. Let me uh get back over here so that I can.

SPEAKER_03 25:04

I gotta get better at OBS. I could do so many camera shots with OBS.

SPEAKER_04 25:08

Oh, it's uh you know, the thing I'm trying not to do is create too many scenes with it. Right. Um, the other one, I don't know. Have you used the Stream Deck?

SPEAKER_03 25:18

I've got a Nogato Stream Deck, yeah.

SPEAKER_04 25:21

Yeah. So they they've got the control that ties directly in with uh OBS so that you can assign a scene to a button and all that. But when I was going in and doing some uh poking around today, uh I grabbed one of the old thin patch cords that I've had that I bought off Amazon, and it's been around forever, and I ran a cat 6A patch chord test on. Now, one of the things that we always need to make sure we're doing is using the right adapters. And uh Lexi and I had done a video recently on Cat 8, and I hate air quotes, but I'm gonna call it Cat 8 patch cords. And people are saying, oh, I buy these things all the time. And we tested them, and they did not perform very well, the ones we tested. And one of the things that we found is that they have a molded plug, so we really can't see inside the plug. So a company on the East Coast that makes industrial CT scanners, Luma field, contacted Lexi and said, if you send us those cables, we'll scan them for you. Oh, cool. And we'll have to we'll have to bring the images up sometime. Yeah. It is really cool to see the inside of that plug.

SPEAKER_03 26:40

Right.

SPEAKER_04 26:40

And we can see exactly how the pairs are twisted in there. But in this case, what I found with this thin cat patch cord that's been around for a long time is even though resistance is not a oh, sorry, I clicked on the new one. I want the old one. Even though resistance is not a pass fail for our patch cord tests, which we use our patch cord adapters for, and that's where I was headed, the patch cord adapters include the RJ45s in the test result. Right. The channel adapters don't.

SPEAKER_03 27:15

Right.

SPEAKER_04 27:15

And what I found is a number of these Cat 8 patch cord manufacturers said in their literature, we use the Fluke Network's channel adapters to test this. And it's like you're telling us right there, you're using the wrong adapters to test the patch cord. They're not including the connectors. Our resistance is way out of whack. Pair one, two is what we expect. These are way out of whack because as those patch cords get moved around, the connections loosen on those crimp-on modular plugs, and we see our resistance goes way up. And now we start running into problems with PoE. And then that resistance unbalanced in there ends up dropping packets. And so that is, you know, the patch cords wear out. And I don't think people think about that. You know, it's like when you take it out, don't throw it in your box of wires, throw it away.

SPEAKER_03 28:14

Right. Well, they think there's there's two certain types of patch cords, PL1 and PL2. PL1 has is good for 650 cycles, 750 cycles. Okay. I don't remember off the top of my head. PL2 is good for 2500. The vast majority of patch cords are PL1 rated, but just because they're rated for that doesn't mean that that's what you're gonna get out of it. Because, you know, like my dad always used to say, don't buy a car built on a Monday or a Friday. Because it may not be they may not be the right kind of quality. Mike,

Final Thoughts And Thanks

SPEAKER_03 28:52

I appreciate you coming on. This is obviously gonna be a three-part episode because uh because again, we get talking about geeky stuff like this. I will you and I both will talk long and forever about it, but I know the audience is gonna appreciate this. I appreciate you coming on today, my friend.

SPEAKER_04 29:06

I appreciate you having me on, Chuck.