Talk:Category 6 cable

From Wikipedia, the free encyclopedia

Contents

[edit] Intermixed cable fallacy?

I've recently completed a job with a contractor that forcefully asserted that Cat 5e and Cat 6 cabling cannot be intermixed. This went against my belief, however, I've known this guy for years and he has ne'er led me astray on purpose. Whats the deal?

Both the situation and the assertion you cite are vague, so allow me to hypothesize a few installations that intermix these cable types and describe the probable results:
  • A facility in which backbone, horizontal, and patch cables randomly use either cat-5e or cat-6 cables, but each individual channel is exclusively one cable type end-to-end. This would result in a facility in which any individual channel could be certified to either cat-6 or cat-5e performance requirements. Very suitable for gigabit ethernet, but only the cat-6 runs would be able to support 10GBASE-T (or other communications systems designed for cat-6). The end result is a system in which only a subset of work areas support advanced communication standards.
  • A facility in which every channel contains a random mix of cat-5e and cat-6 cables. A very poor idea. In all likelihood, these channels could be certified to cat-5e performance standards (although with the increased cost of the cat-6 components), but it is possible that small impedance mismatches, NVP differences, skin effect variances or other differences between the properties of the cables could cause the channel to fail cat-5e certification. I consider this rather unlikely, but certainly I would not expect any of these channels to pass cat-6 performance testing, so one is left with a higher cost, higher risk, cable system that delivers no benefit over pure cat-5e.
Bottom line: they *can* be intermixed, but doing so increases the risk of certification troubles and delivers no performance benefit. Thus, any justification for mixing them should be based on other factors, like logistics (cable availability, colors, etc), financial, etc., and with a full understanding of the risks.
dpotter 23:44, 10 October 2006 (UTC)
I can think of two main reasons to mix them.
The first and most obvious is extending or modifying an existing installation. You certainly don't want to rip out existing cable that is doing its current job perfectly well but you also want the new parts of the installation to remain usable for as long as possible.
The second is patch cords, patch cords are easy to replace and likely to have a much shorter lifetime (due to being handled more) than fixed cabling so there seems to be a lot less reason to choose future-proof patch cords than future proof fixed cabling. Plugwash 20:18, 3 March 2007 (UTC)

[edit] What maximum bandwidth is supported?

I'm pretty sure the bandwidth is 550 mhz not 250 mhz. —Preceding unsigned comment added by 66.92.160.28 (talk • contribs)

Although some cable manufacturers may provide (or claim to provide) cables capable of transmitting very high-frequency signals, the Cat 6 specification (TIA/EIA-568.B.2-1) specifies transmission requirements only within the range of 1MHz-250MHz. dpotter 20:16, 11 May 2006 (UTC)

Could someone translate this into layman's terms? How many Mbps is 250Mhz, and how does this relate to carrying Gigabit Ethernet (such as 1000-Base-T) signals? 4.239.48.201 (talk) 14:36, 27 January 2008 (UTC)

4.349.48.201 Mbps and bandwidth are not the same. The cable has a bandwidth, which is the maximum operating frequencies of the signals. To send a number of digital bits, it will need to be converted into a signal. This is why we have 'CAT 5', CAT 5e and CAT 6 cable, and 10BASE-T, 100BASE-T and 1000BASE-T. See Ethernet 62.190.112.2 (talk) 11:31, 4 April 2008 (UTC)

[edit] What is "horizontal cable length"?

65.27.174.174 00:06, 8 June 2006 (UTC)

It's the length of a horizontal cable. The term "horizontal cable" typically refers to the cabling between a horizontal cross-connect and a telecommunication outlet. An example of this would be an unshielded twisted pair cable that runs from an RJ45 termination panel in a wiring closet to an RJ45 jack in a wall plate. dpotter 05:30, 9 June 2006 (UTC)
So, the "horizontal cable length" is the length of the "permanently installed" cable, from the wall jack at one end to the wall jack or patch panel at the other end, typically hidden away inside walls, ceiling, or under the floor. Once you've cut the cable, its length doesn't change, no matter if the cable goes up, down, horizontal, or in a big circle. The "cords" are the visible length, typically one cord plugged into the wall jack and a PC, the other cord plugged between the patch panel and a router.
Could someone confirm that a cable that runs, say, 60 feet straight up then a couple of feet to the West has a "horizontal cable length" of 62 feet, and clarify this in the article? —Preceding unsigned comment added by 65.70.89.241 (talk • contribs)
You've got the right idea, and your first paragraph above is correct. Horizontal cables are defined by their place in the architecture, not the distance they traverse parallel to the ground. The example in your second paragraph is also correct, although unusual/unrealistic - in a commercial premises, horizontal cables would almost never traverse a 60-foot vertical climb. That would typically indicate a 4-5 story building, and a typical cable system architecture would have a cross-connect installed on each floor, with backbone cabling (not horizontal cabling) interconnecting the cross-connects. See Structured cabling for more information.
I'll add a link to the article as you suggested. dpotter 14:29, 10 August 2006 (UTC)
This then brings up the further question are the rules for backbone cabling different from those for horizontal cabling and if so in what way? Plugwash 19:56, 3 March 2007 (UTC)

[edit] RJ-45

Although the RJ-45 link redirects to the 8P8C page, would it not be better to make a more explicit statement of the real name of the connector? Something like: "normally terminated in 8P8C modular connectors (often casually referred to as RJ-45)"

[edit] Needs updating and sources

I couldn't find any sources listed on this page. Also, the external link "Cat 6 FAQ" appears to be dead as of 2007-02-26. —The preceding unsigned comment was added by 216.152.208.1 (talk) 23:22, 26 February 2007 (UTC).

[edit] Popularity?

To what extent are Cat 6 is being installed as a percentage of all Ethernet cable types? Can someone add some statistics regarding its popularity? Thanks. ---Ransom (71.4.51.150 20:05, 9 March 2007 (UTC))

[edit] Augmented category 6 is not CAT 6a

Cat 6a is as noted, a future standard. Using the term CAT 6a and augmented CAT 6 to mean the same thing is incorrect use of CAT 6a.

Offer in reference

http://www.rdm.ch/cmStreamDown/54684/type.pdf

[edit] Is This True or another cat 6 made up scare monger ??

I have recently been asked to install a number of cat 6 cables for a client. He requested that I ran all cables from the cable box's sittuated at the network point, when I asked why ? He replied they have to run this way becuase of the way the system works ( Its an AT+T style 110 system but cat 6 version known as 210 if using Siemon's system) I qeustion this action as I have never heard of any thing so riddiculous and he replied that it wouldn't work if i ran the cables from the LAN room (or the other way). The cables are normal cat-6 out of a box with no connectots on or anything. Could you please shed some light or dispurse this theory as I have installed many Cat 6 installations and they have all worked wich ever way I have run the cables after all its just copper end to end isn't it ??

Cheers Juls

[edit] PC to PC- innacurate and perhaps too informal

"To connect two Ethernet units of the same type (PC to PC, or hub to hub, for example) a cross over cable should be used, though some modern hardware can use either type of cable automatically." I am correcting this as it implies its a hardware issue, when really if one PC is a server and another PC is a client, crossover cable is never needed. Mentioning this use of crossover cables might only save the casual small office/home peer-to-peer network installers some headaches, but I think we could perhaps use the full term "Personal Computer" might be more appropriate for the Encyclopedia.

For the previous poster, I'm not sure why the cat6 "direction" should make a difference. Can you pass along any cost difference (in time spent or money)? The customer doesn't have to be any wiser. Sometimes being right doesn't pay off. Let someone else educate the blowhard. Cuvtixo 19:13, 24 September 2007 (UTC)

To answer your question, the reason is TIA/EIA-568 Ethernet has separate transmit (TX) and receive (RX) lanes. TX of one equipment must be connected to RX of the other. On older equipment they are fixed, so two equipment of the same type need a cross over cable, otherwise TX connects to TX and RX connects to RX. Newer equipment can automatically detect and adjust which pair is which. Ibjoe 01:52, 27 September 2007 (UTC)

If one PC is a server, and another is a client, a crossover cable will still be required as the network interface cards (NIC) in them will be identical. Rather, the as the above reply mentions, it is to do with TX/RX pins matching up between devices. A computer NIC will always transmit on the same pins, regardless of whether it is installed in a client or a server, as it was designed to match up with receive pins on a hub. Hence, if two computers (regardless of client or server designation) want to communicate directly, rather than through a hub or switch, a crossover cable is required. 210.8.150.249 (talk) 04:38, 23 April 2008 (UTC)