Counterfeit CAT6 ethernet cable? Or something else?

[Ranayna]

The colors look way off to me, that could be the camera though.

There is not supposed to be a red wire in network cabling. The blue looks like purple, and the brown is off as well.
And the complementary conductors are totally off color as well. They are usually white, on some older cable i have seen striped, but that is likely more expensive.

The picture of Psi shows how the colors are supposed to look.

That, in combination with the loose twist makes me very suspicious of the cable.
I suspect there is more wrong with the cable than just the twists though. Ethernet over copper is surprisingly robust, and i have seen gigabit over old phone wires a couple of times. Though not at 50 meters, more like 20 to 30.
Another thing to test would be if the cable is even copper, and not just copper clad aluminium, and which AWG it is.

[Randy222]

The days of analog punch-down wiring.
I suspect maybe it's also not stranded wire.

[Ranayna]

LSA punch downs are still in use for gigabit wiring. Sure, nowadays usually keystone modules are used and not every wire needs to be punched down individually anymore, but the inside of a keystone is not much different: The wires are shoved into a blade connector that cuts through the insulation.

Wire type and connector need to match. A connector for solid wire won't work well for stranded wire, and vice versa. Same for AWG, this needs to match between connector and cable.
For stationary cabling, solid core is almost always used, at least here in Germany i have never seen anything else in a commercial and even most home setups. We usually go all out with CAT cable and use S/FTP.

EDIT: This is a typical CAT7 cable used hereabouts:
https://www.kabelscheune.de/Netzwerkkabel-Verlegekabel/Cat-7-Netzwerkkabel-Verlegekabel-1000-MHz-S-FTP-PIMF-orange.html
Interestingly this has way fewer twists than the CAT6 cable on Psi's screenshot.

And another EDIT: It never occured to me that LSA is a german abbreviation

[MLXXXp]

Another thing about proper CAT type ethernet cable is that each pair will have a slightly different twist pitch to reduce crosstalk between the pairs. From this Wikipedia article:

When nearby pairs have equal twist rates, the same conductors of the different pairs may repeatedly lie next to each other, partially undoing the benefits of twisting. For this reason, it is commonly specified that, at least for cables containing small numbers of pairs, the twist rates must differ.

Although hard to tell in the photo, all pairs appear to have the same pitch.

[hans]

Is the copper used that 102% Oxygen free stuff?

Even better question is, is it copper at all? The cheaper cabling often isn't.

A lot of new ethernet cables are not made from copper anymore but aluminium. The idea is that due to skin effect, HF signals only travel in the outer part of a cable, so that copper can be coated on while aluminium conducts main current for PoE.
Bottomline is: the datasheet should mark them as "CCA" - Copper Clad Aluminium.

https://en.wikipedia.org/wiki/Copper-clad_aluminium_wire

The theory sounds great. But I've seen many people complain about these kinds of cable on a Dutch IT forum for CAT6 and above. They were planning to run 10Gbps at 10m+ lengths, but that failed and some couldn't even establish a stable 1Gbps link.
I went out of my way to find proper copper cables when pulling cables through the house. I paid something like 1 euro per meter for CAT7, which is grossly overkill, since all my wall plates are only Cat6. But hey, buy once, should last for decades.

I'm not sure what is going on with OP's installation, but to me it sounds like something that should be looked into.
You can test whether a conductor is pure copper by scraping the coating off or make a side cut into the conductor.

[Ranayna]

Anything CCA is not a CAT5/6/7/8 cable.
The standards are very clear there: Network cables need copper conductors.
https://www.flukenetworks.com/content/application-note-copper-clad-aluminum-cables

[m k]

I've met RJ45 plugs with shorter connection fingers.
When mechanical connection is not stiff the plug can move a bit and disrupt at least an EtherCAT connection.

I've also bought plugs that can't fit, but you can't "miss" with them very easily.

[Randy222]

The colors on pic #1 might just be poorly died insulation, but can be Green Blue Orange Brown
Probably cat1 or cat2 UTP.

[wraper]

The colors on pic #1 might just be poorly died insulation, but can be Green Blue Orange Brown
Probably cat1 or cat2 UTP.

That's neither, even not considering marking. X shaped plastic separator in the middle means it was built to imitate CAT-6.

[m k]

I have a recollection of something from the past, possibly with a spine, possibly mixing many things.
Darker blue jacket, twisted pairs, shielded with a foil and solid plain metal wire for grounding, stiff compared to regular 10baseT and possibly for multipurpose connection system, possibly for Open DECconnect of some sort.
Must be from late '80s, from time when DEC launched their future office with blue dropped from ceiling spirals for cabling.
Data cabling had a general wall socket and many different adapter plugs for final output style, video was there for sure, maybe some king of RF also.
Search results are limited, only some minimal text references.

[ve7xen]

100M uses only two pairs, 1-2 and 3-6, both ends pin to pin is a norm today.
(only very old stuff can't do pair crossing internally)
If PoE is not needed you can try different cable pairs, doubling pairs is more bad than good, issue is more noise than signal strength.

Equipment that does not support Auto MDI-X is probably sitting in a landfill by now, but AFAIK that does not mean you can use cables that are 'straight' in the sense that they have their pairs next to each other in the RJ45 connectors. You still need to make either the 568A or 568B crossover,  the same on both sides of the cable. Failing to do so might give you problems with longer cables. Like described in the OP.

Auto MDI-X isn't so much the issue, but if you use a crossover cable, you'll be limited to 100 Mbps link speed. For 1 Gbps and higher, you need all 4 pairs to be identical. Then of course there is PoE. I've ran issue some issues with some Gigabit gear and crossover cables, even though technically it should work?

That is not the case. 1000base-T runs all 4 pairs bidirectionally, and supports Auto MDIX independently on every pair. A cable 568A<->568B will work fine at 1Gbps, despite only crossing half the pairs, as long as the PHY supports Auto-MDIX which is mandated by 802.3 for 1000base-T. Some legacy, non-compliant devices might fail to link up at all in that situation, but in practice you will not likely run into it.

Regarding the OP, it seems likely that one or both of the auxiliary pairs is not terminated correctly or is broken. With 1000base-T transceivers on both sides, the FLP-based autonegotiation will agree on 1000base-T, but link training will fail, and link will not come up. What happens next is implementation-dependent. Some implementations will fail 1000base-T link a few times, and fall back to 100base-TX. Others will just keep trying 1000base-T, which will keep failing.

Poor signal quality due to low quality cable will very rarely cause link to fail entirely, because link negotiation is low bandwidth. Link will come up, usually at the fastest negotiated rate, but with a high bit error rate. That leads to packet loss and poor performance.

The cable here looks low quality and probably not Cat6 compliant, but I would be very surprised if it can't carry 1000base-T. The standard is quite resilient against poor conditions. Unless you are trying to push 100+m, in a bundle of many similar cables, in a noisy electrical environment, this cable is probably 'good enough'. Problems are almost certainly caused by termination issues (wrong connector type / AWG, crossed pairs, etc.).

[Psi]

The cable here looks low quality and probably not Cat6 compliant, but I would be very surprised if it can't carry 1000base-T. The standard is quite resilient against poor conditions.

Probably?  haha  It's definitely no where near Cat6 compliant with those twists.

But yes, we had 50pair phone cables in walls of our house growing up. Dad ran some cable he got from work when the house was being renovated i think.
I had a 25meter 1000base-T link working fine over it for years, and it was either cat1 cable or what came before that.

[Randy222]

It's not eth data cable

More likely one of these.
alarm panel cable
telephone station cable
access control cable
low-speed RS-485/control wiring

[sausage]

My apologies to the late closing of this thread while I got a photo of the markings. This is the cable clearly marked as CAT6E, whether it was genuine or not.

The second length of cable that I ran from the switch to the shed was fine and didn't need replacing.
I can only conclude that the original cable I used which I made was poorly made by me.

Thank you everyone for your input and information, it was really helpful for me to understand how crucial or not crucial some parameters of ethernet cables are.

[ve7xen]

Well there ya go, there's no such thing as Cat6E. It is probably a way for the manufacturer to try and get away with claiming it is not Cat6 and shouldn't be held to the same standards. Glad it's working for you anyway, though.

[Psi]

Well there ya go, there's no such thing as Cat6E. It is probably a way for the manufacturer to try and get away with claiming it is not Cat6 and shouldn't be held to the same standards. Glad it's working for you anyway, though.

Yep, if caught they would probably argue that "Cat6E" is their own internal name for what that cable is.

[amyk]

https://en.wikipedia.org/wiki/Category_6_cable#Category_6e

Cat6E is supposed to be better that Cat6, so it should at least pass Cat6 tests.

[Monkeh]

https://en.wikipedia.org/wiki/Category_6_cable#Category_6e

Cat6E is supposed to be better that Cat6, so it should at least pass Cat6 tests.

As the article quite clearly says, there is no such standard. There is absolutely no reason to attribute any level of performance to such a cable.

[Ranayna]

The official "better than CAT 6" standard is CAT 6A (sometime with a low set a that i don't know how to replicate here )

This was developed, because CAT 7 was never specified for RJ45 plugs.
The company i work at, still has a full building with CAT 7 cabling, not using RJ45 wall plugs, but whats called a "TERA connector" https://en.wikipedia.org/wiki/TERA
I hate these things, they are bloody expensive. But the "powers that be" won't shell out for the work redoing all the patches, so i still ahve to deal with this thing

Then, after CAT 7 was finalised, manufacturers realised that they were indeed able to make RJ45 plugs that support 10GBit/s. But it was too late to amend the CAT 7 standard, and thus CAT 6A was defined as an addition to CAT 6.

[Psi]

It's also possible the cable itself was legit at some point but someone stamped CAT6E on it so they could sell it at a higher price. I could see a bad actor doing that if they found a warehouse full of 20 year old cable boxes. The cost to feed it through the machine that inks text on it would be trivial

[u666sa]

What I don't see in the OP cable pic is shielding. That cable I'm sure is okey to run around the room, but if you have 50 meters of it, shielding becomes important. Also we don't know if it's aluminum or copper.
A) You need copper.
B) You need shielding.
C) You need best category your money can buy.
D) You need proper good connectors.
E) Finally, you are running underneath your house, exposed to elements (for example UV radiation) and rodents, you need to enclose your cable in protective shielding, either corrugated tubing or a pipe.

[ve7xen]

A) You need copper.

I mean I wouldn't recommend CCA, but need is a strong word. If it can meet the spec (and such compliant CCA cables exist), it will work.

B) You need shielding.

Almost nobody needs shielded Ethernet cable, it is very much a special case thing, so no, they do not.

C) You need best category your money can buy.

No, you need the lowest category that supports your desired Ethernet standard. Gilding the lily here buys you absolutely nothing. Why waste the money?

D) You need proper good connectors.

Sure. The main point here is that the terminations for solid and stranded wire are different, and you need to use the right type. Quality helps, especially to get a working termination the first time.

E) Finally, you are running underneath your house, exposed to elements (for example UV radiation) and rodents, you need to enclose your cable in protective shielding, either corrugated tubing or a pipe.

Certainly should, or use armoured cable, But stapled to the floor joists in the crawlspace I don't think it's a major problem. I'd say see how it goes and if you have a problem with rodents then upgrade it.

[tooki]

A) You need copper.

I mean I wouldn't recommend CCA, but need is a strong word. If it can meet the spec (and such compliant CCA cables exist), it will work.

The CAT ratings (at least for CAT 5 and up) explicitly require copper wire. So CCA cables are noncompliant per se, even if they otherwise reach the performance ratings.

[u666sa]

Back in 2010 I used to work for this company who sold of all things network cable. So it was always a choice between copper and aluminum, insulated vs none insulated, insulated better vs insulated not so good. Aluminum is okey around the office if you won't be moving or bending it too much. Copper is king in long distance. Insulated is a must in long distance and helps if it runs through electrical wires. When network cable meets sunlight bad things happen, you must enclose it in something. Rodents do chew on cable. It's like a mantra. What cable customer should get? Best one he can afford. That's just the bottom line.

[thm_w]

I got the cheapest copper CAT5E stranded copper cable I could find to run outside, its unsupported across a ~8m length in the sun. So far so good 10 years now. Only did it thinking it was temporary though.

Definitely would not touch aluminum wiring no matter what, even if it sits in the wall.