I mean, the simplest answer is to lay a new cable, and that is definitely what I am going to do - that’s not my question.
But this is a long run, and it would be neat if I could salvage some of that cable. How can I discover where the cable is damaged?
One stupid solution would be to halve the cable and crimp each end, and then test each new cable. Repeat iteratively. I would end up with a few broken cables and a bunch of tested cables, but they might be short.
How do the pro’s do this? (Short of throwing the whole thing away!)
Higher end cable testers can show you where the break is, but it will be far more expensive that a new cable.
My TP link switch can detect faults and cable length. I’m not sure it can do both together but it’s possible. Worth checking if you have a switch with those features
Hmm, interesting! I have a Synology switch, gotta read up on its capabilities.
you can rent a time domain reflectometer for this purpose… it sends out a signal and then listens for the echos back and calculates the distance. problem is they arent accurate the first like 500 feet so you have to add a reel of 500 ft of matching cable to the end first. electrorent.com rents them.
I had an AMD Phenom-II era motherboard that claimed it would be able to do that. OP, you might be able to find an old NIC/mobo that could do this for cheap.
Should also be able to do it with a NanoVNA or similar.
That sounds triggered.
You can only do 100M runs max anyways, just replace the whole thing? 100M of CAT6 is pretty cheap if you already have a box for it.
Or is this an academic question?
Actual, not academic. And I agree that a new cable is cheap, which is what I will do. My question is about avoiding throwing a mostly good cable in the trash.
Pull it, cut into quarters. Add jacks to each end and test the lines.
Nvm, I see you suggested this…
What is an M? Miles? That doesn’t seem right.
M = meters. Mi = miles
m = meters. M = mega (x 1 000 000).
That’s why Km is 1 thousand meters and Mm is 1 million meters.
The actual unit is lower case, the multiplier is uppercase.
It’s also a lower case k in km.
The actual unit is lower case, the multiplier is uppercase.
Wouldn’t agree with that… There are many different units and multipliers. the letter being uppercase or lowercase has nothing to do with it.
Examples:
letters for prefixes/multipliers being uppercase and lowercase: P, T, G, M, k, h, da, d, c, m, u, n (trillion, billion, million, thousand, hundred, ten, one tenth, one hundredth, one thousandth, one millionth, one billionth)
Letters for units being uppercase and lowercase: s, m, g, N, W, J, A, K, V, h, Hz (seconds, meter, gram, Newton, Watt, Joule, Ampere, Kelvin, Volt, hour, Hertz) (just recognised, that most units, which are named after scientists, are written with capital letters…)
km = thousand meters/kilometer
K = Kelvin (unit for temperature)
M = Mega (prefix for one million)
kJ = thousand joules
s = second
ms = millisecond (one thousandth)
S = siemens (electrical conductivity)
mS = milli siemens
mm = millimeter (one thousandth of a meter)
Mm = megameter (one million meters or thousand kilometers)
For reals? Where can I read more about this?
Edit: nevermind, I just googled it. TIL!
If you just want some spare cable. I’d make 5x 20M pieces. The one that doesn’t work becomes 2x 10M. That bad one becomes 2x 5M … As far down as you like.
Isn’t that almost what they suggested except starting at a different size and doing a binary search basically? You’re just starting the binary search after the first step of cutting into 5 lengths instead of 2.
How do you know the cable is damaged? If you have a toner you can hook the generator to the damaged pair and listen along the length of the cable to where the tone changes. It’s not as fool-proof as a tester, but much cheaper.
I wanted to post a photo of my network tester but Lemmy gives me a JSON error.
Anyway, the tester is supposed to light up each strand one at a time, but what I get is disco fever.
I assume the disco fever is all the pairs shorted together. Is this cable buried or run through the house? I would start by redoing your ends. Make sure they are crimped fully and both sides are identical.
I agree. If that doesn’t work, I’d give the whole line a visual inspection for any more obvious signs of damage. If that doesn’t work there are other options to troubleshoot, but I’d probably just replace the line and attempt to recycle the old cord into shorter lines for device connections, patching, etc.
short between pairs probably. tdr (time domain reflectometer) cable tester should help locate fault. might be pinched somewhere or something else
there are “casts” you can get to repair buried cable. usually used in copper telephone plant, i wouldn’t trust it to certify to a gig but you could use it for a backup link or if you play with pots at some point. or “temporary repair”
Searched “tdr” before replying, and was inexplicably happy. :)
Can’t you just use a cheap non-contact voltage tester to find where the cable is damaged? Just run the tester through the cable until it suddenly stop detecting any AC voltage, which is probably where the cable is broken. But if you’re talking about ethernet cable, then I have no idea.
I just learned about these things today when researching for a device to test that there’s no voltage present before installing a ceiling light. Such a great innovation since it works even through insulation, so no risk of getting shocked.
The easiest thing to do would just be to pull it out and look at it. The break might be obvious.
Yeah, that’s not that easy, unfortunately, because each end of the network cable passes through an insulated wall, through a hole equal to the cable width = smaller than the plug. Even if I find the break, it is likely in the outdoors part of the cable where I would want an unbroken cable without a field repair.
Sure, but you’re going to be pulling it out either way. When you do, inspect it.
You only need 2 pairs for 100base-t, try forcing a lower negotiation, see if the pairs you need work? Maybe unbundle the other set of pairs and try them?
Wire tracker maybe? You might want a higher quality version than that particular one if the cable run is long, one of the reviews suggest that the distance is limited.
This one here does time domain reflectiometry which will tell you exactly where the break is:
https://www.amazon.com/Ethernet-NF-8209-Distance-Location-Measurement/dp/B08M3SRB2Q/
I was curious if that would work on ethernet cable! I’ve seen it done on coax, wasn’t sure if it would work well enough on UTP to be useful outside a lab setting. Cheap too. Cool!
Some of the reviews say it didn’t work for them to find a break, but I have that exact one and have had good luck with it. So YMMV. Obviously the ten times the price Fluke will do better.
TDR isn’t exactly rocket science and the speeds needed to do it are surprisingly reasonable. You can build your own for a few bucks if you have an oscilloscope as well.
