Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

When I am using a single PC to switch environment with a VLAN function, what is longest Cat 6 cable I can use?

I have read about a 100 m length limitation, but I am not talking about daisy chaining or any other user being on same line. The PC connects directly to switch, and we are using it in a VLAN configuration.

share|improve this question
    
I'm curious, how do you intend to use this single long link? Where is it running? That is a long indoor distance. – Ron Maupin Mar 17 at 3:30
3  
As long as will work. Anything over 90 meters solid core + 10 meters stranded core (patch cables) is outside of spec. That is not to say it won't work. Whether it works will depend on the hardware, cable and environment. (You could go further with fibre cables). I've seen sites which claim up to 700 foot (213 meters) but this is WAY out of spec and a bad idea. You could use Ethenet extenders or jerry rig a small switch powered by POE to break to 100m barrier and stay in spec. – davidgo Mar 17 at 4:04
2  
Additionally: the 100 metres is the minimum that is supposed to work - in theory, 100 metres will always work. 105 metres is quite likely to work but might not. It's not like the switch measures the length of your cable and goes "100.1 metres, I'm going to turn this port off now and not even bother trying to use it." – user20574 Mar 17 at 23:14

The UTP standard, to which the various ethernet standards adhere, is 100 meters, but that assumes 90 meters is solid-core (better performance, more fragile), with 5 meters on each end as stranded (worse performance, more flexible).

It has nothing to do with daisy-chaining, but with several measurements, such as frequency, insertion loss, NEXT, PSNEXT, FEXT, ELFEXT, PSELFEXT, return loss, propagation delay, delay skew, balance, longitudinal conversion transfer loss, etc.

The 100 meters also assumes that the cable is pre-built or professionally installed with all the same rated components, and tested to pass the full test suite. Even experienced installers have problems when installing Category-6 cabling, although Category-5E cabling can do 1000BASE-T at the same distance. Category-6 cabling can do 10GBASE-T at 55 meters, but Category-6A can do 10GBASE-T at the full 100 meters.

share|improve this answer
1  
Solid vs. Stranded can be a critical oversight. I had to replace a long run of stranded with solid-core in one interesting case... – Brian Knoblauch Mar 17 at 18:55
    
I would mention that Cat6 isn't actually rated for 10G @ 55m and no one will make guarantees about it or support it. Cat6 is somewhat of an odd cable as it isn't called for by any standards. – Dev Mar 17 at 20:57
1  
@Dev, there are some (big) equipment vendors (e.g. Cisco) which support Category-6 cable with 10GBASE-T for 55 meters. Also, the participant in the new NBASE-T Alliance support 5 gigabit on Category-5 cable at 100 meters. You can llok at the documents, such as: cisco.com/c/dam/en/us/products/collateral/switches/… – Ron Maupin Mar 17 at 21:04
    
@RonMaupin I stand corrected on the Cat 6, I wasn't aware that Cisco would support it other than saying it might or might not work. – Dev Mar 18 at 3:15

If you want to use the common copper 10, 100, or 1000 Mb/s Ethernet, 100 meters is the standard; any longer and nothing is guaranteed to work right.

However, as far as the longest cat6 cable between a computer and a switch, if you add something like a pair of G.hn bridges you can then use 2700 feet of cat6 between the two bridges, and enjoy one full megabit of bandwidth.

share|improve this answer
3  
The google-fu to find other such bridges is "ethernet extender". – Dan Pritts Mar 17 at 4:07
1  
A full megabit, wow! – chicks Mar 17 at 14:32
2  
"one full megabit" or 1 gigabit? the linked ad mentions only 1gbit and 2700 feet, but not that when using 2700 feet only 1mbit will remain available – Ben Schwehn Mar 17 at 14:55

I am new here, so please forgive me if I am out of line but... (try 2)

To directly answer your question, TIA-568-D.1, section 10.2, Length states:

“The horizontal cabling extends from the termination of the media at the HC in the TR or, when used, the TE to the telecommunications outlet/connector or multi-user telecommunications outlet assembly in the work area. The maximum horizontal cabling length shall be 90 m (295 ft), independent of media type.”

In your specific circumstance, TIA-862-A: Building Automation Systems Cabling is the only ratified standard, I am aware of, that allows direct attach to equipment. This may increase your distance to the full 100M, but I do not own a copy of this standard.

As for Mr. Maupin’s comment,

Cicso does not regulate the low voltage cabling industry, i.e their statements, although an industry leader in network equipment, are just opinions. At no point do they claim any of it will work on Category 5 cable (Category 5e, yes) Even NBASE-T states:

NBASE-T Alliance is pursuing specifications related to:

PHY-MAC system interface PHY- magnetics system interface Channel characteristics (http://www.nbaset.org/technology/specifications/)

TIA does (for the US at least), specifically TIA-568 (current revision D), for performance and test characteristics.

IEEE standards produced by the working group defining the physical layer and data link layer's media access control (MAC) of wired Ethernet. (https://en.wikipedia.org/wiki/IEEE_802.3)

Any other standard being used would be considered proprietary and, typically, leads to issues if/when you change active equipment manufacturers.

As for 10GBASE-T (IEEE 802.3an) on Category 6, TIA's TSB-155 is the test parameter and can only be tested for after the installation. So if someone tells you "I can guarantee a 55M link on Category 6", they are not being honest.

Dev is dead on about Category 6 cable. TIA (and manufacturers) pushed it to the market prior to IEEE ratifying IEEE 802.3ab (1000BASE-T), believing 1000BASE-T would only work on a 250Mhz rated cable. IEEE managed to get it to work on Category 5e also and relegated Category 6 to an odd existence. I am unaware of an IEEE standard protocol that will work on Category 6 and not Category 5e.

The other issue, and the most frightening and dangerous one, is PoE (as governed by IEEE) is going to start butting heads with the NEC in the near future. It should get interesting. If you research heat-loads on the cable plant, specifically for PoE+ (IEEE 802.3at) and PoE++ (IEEE 802.3bt) you will find that there can be significant heat issues with bundled cables (that group of wires in your cable tray and neatly bundled on the back of your rack, See TIA-TSB-184), especially for Cables at and below Category 5e or smaller AWG cables.

Remember, Cisco's original non-IEEE standard PoE implementation did not require an Ack/Nack and because of this, toasted countless multi-thousand dollar Cable Certifiers.

Bottom line, for accurate cabling information, look to TIA and IEEE, not network equipment or cabling manufacturers.

I believe this second attempt has met the factual requirements for a response directly to the OP question.

And not I shall take my soapbox and go home.

Respectfully, BICSI RCDD with 20+ Years in the industry.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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