T568A or T568B

NewsGroups | Search | Tools

User Password

Lost Password?

comp.dcom.lans.ethernet

get this group's latest topics as an RSS feed

add this group's latest topics to your My MSN content

add this group's latest topics to your My Yahoo content

add this group's latest topics to your Google content

Yahoo!

Google

Windows Live

del.icio.us digg

digg

Netscape

Subject

Author

Date

T568A or T568B

phil-news-nospam

05-12-2007

Re: T568A or T568B

Daniel J McDona ..

05-12-2007

Re: T568A or T568B

phil-news-nospam

05-12-2007

Re: T568A or T568B

glen herrmannsf ..

05-13-2007

Re: T568A or T568B

Robert Redelmei ..

Al Dykes

glen herrmannsf ..

05-13-2007

Posted by on May 12, 2007, 2:43 pm

If you were Registered and logged in, you could reply and use other advanced thread options

Some information sources indicate there is no difference between T568A and
T568B wiring besides what color is used at which pin position. This implies
that as long as association between pins is correct (reversed in the case of
a crossover cable), a patch cable (8P8C on each end) will work fine whether
wired for T568A or T568B. It would be assumed the "big deal" is just when
the color has to be used to identify something, such as when wiring one end
of a cable to something other than an 8P8C plug.

But some other information sources suggest that using T568A when T568B is
in standard use, or the reverse, that the electrical quality is slightly
degraded even though things generally work. If this information is true,
then that would suggest some kind of _electrical_ or _metallic_ difference
in the wires, and not just the insulation color. As far as I am aware, the
additives to insulation to create the color has no effect.

Could the wires in a CAT5/CAT5e/CAT6 cable be _electrically_ different in
some way, such as a different impedance (I thought it was all supposed to
be 100 ohms within a twisted pair, or a different cross sectional area 9and
thus affect the resistive loss)?

I still look forward to the day that computer interconnections begin using
symmetric bidirectional cabling so we can stop having to deal with various
wiring schemes and get rid of crossover cables and such.

--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
| first name lower case at ipal.net / spamtrap-2007-05-12-1334@ipal.net |
|------------------------------------/-------------------------------------|

Posted by Daniel J McDonald on May 12, 2007, 7:04 pm

If you were Registered and logged in, you could reply and use other advanced thread options

>Some information sources indicate there is no difference between T568A and

>T568B wiring besides what color is used at which pin position.

Generally true.
[...]

>But some other information sources suggest that using T568A when T568B is

>in standard use, or the reverse, that the electrical quality is slightly

>degraded even though things generally work.

I imagine in some corner cases this is correct.

> If this information is true,

>then that would suggest some kind of _electrical_ or _metallic_ difference

>in the wires, and not just the insulation color. As far as I am aware, the

>additives to insulation to create the color has no effect.

>Could the wires in a CAT5/CAT5e/CAT6 cable be _electrically_ different in

>some way, such as a different impedance (I thought it was all supposed to

>be 100 ohms within a twisted pair, or a different cross sectional area 9and

>thus affect the resistive loss)?

Yes, but the twists per foot are different for the different colors. If
each pair had the same twists per foot, you would still get cross-talk
between the pairs, where the differing twists per foot provide a level of
immunity.

>I still look forward to the day that computer interconnections begin using

>symmetric bidirectional cabling so we can stop having to deal with various

>wiring schemes and get rid of crossover cables and such.

Like 1000BaseTX, which already does this...

--
Daniel J McDonald CCIE # 2495, CNX
Visit my website: http://www.austinnetworkdesign.com

Posted by on May 12, 2007, 9:46 pm

If you were Registered and logged in, you could reply and use other advanced thread options

wrote:
|>Some information sources indicate there is no difference between T568A and
|>T568B wiring besides what color is used at which pin position.
|
| Generally true.
| [...]
|>But some other information sources suggest that using T568A when T568B is
|>in standard use, or the reverse, that the electrical quality is slightly
|>degraded even though things generally work.
|
| I imagine in some corner cases this is correct.
|
|> If this information is true,
|>then that would suggest some kind of _electrical_ or _metallic_ difference
|>in the wires, and not just the insulation color. As far as I am aware, the
|>additives to insulation to create the color has no effect.
|>
|>Could the wires in a CAT5/CAT5e/CAT6 cable be _electrically_ different in
|>some way, such as a different impedance (I thought it was all supposed to
|>be 100 ohms within a twisted pair, or a different cross sectional area 9and
|>thus affect the resistive loss)?
|
| Yes, but the twists per foot are different for the different colors. If
| each pair had the same twists per foot, you would still get cross-talk
| between the pairs, where the differing twists per foot provide a level of
| immunity.

Ah. Is there any short info on exactly how many twists per foot per color,
or do I have to go download that hundreds pages document if I can even
find it?

|>I still look forward to the day that computer interconnections begin using
|>symmetric bidirectional cabling so we can stop having to deal with various
|>wiring schemes and get rid of crossover cables and such.
|
| Like 1000BaseTX, which already does this...

There is a different crossover wiring for this, too. So no, it isn't really
that way. Of course some devices have been made to detect which lines are
wired which way, so you could communicate fully regardless of which cable
type is connected. But this (Cat5/6 with 8P8C on each end) is nowhere near
being a symmetric bidirectional cable/connector scheme.

--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
| first name lower case at ipal.net / spamtrap-2007-05-12-2040@ipal.net |
|------------------------------------/-------------------------------------|

Posted by glen herrmannsfeldt on May 13, 2007, 12:32 pm

If you were Registered and logged in, you could reply and use other advanced thread options

phil-news-nospam@ipal.net wrote:
(snip, someone wrote)

> | Yes, but the twists per foot are different for the different colors. If

> | each pair had the same twists per foot, you would still get cross-talk

> | between the pairs, where the differing twists per foot provide a level of

> | immunity.

> Ah. Is there any short info on exactly how many twists per foot per color,

> or do I have to go download that hundreds pages document if I can even

> find it?

I don't know if it is in the standard, or just the allowable
crosstalk at different frequencies. Note, though, that as someone
else said the difference between 568A and 568B is the exchange of the
green and orange pairs, such that both are still being used.
If there were any differences, they would cancel out in a round
trip configuration.

-- glen

Posted by Robert Redelmeier on May 13, 2007, 10:52 am

If you were Registered and logged in, you could reply and use other advanced thread options

phil-news-nospam@ipal.net wrote in part:

> But some other information sources suggest that using T568A when

> T568B is in standard use, or the reverse, that the electrical

> quality is slightly degraded even though things generally work.

I have never heard this. The reasoning for -A over -B in
govt work and residential (T-570) is that pair 2 (orange/wt)
is in the USOC position for POTS (and other phone stds).
-B is an AT&T legacy. Legacy lives!

> If this information is true, then that would suggest some kind

> of _electrical_ or _metallic_ difference in the wires, and not

> just the insulation color. As far as I am aware, the additives

> to insulation to create the color has no effect.

I very much doubt it could affect -B vs -A. Both use pairs
2 & 3 (orange & green/wt) for 100baseTX, and all four pairs
for 1000baseT (GigE).

> Could the wires in a CAT5/CAT5e/CAT6 cable be _electrically_

> different in some way, such as a different impedance (I thought

> it was all supposed to be 100 ohms within a twisted pair, or a

> different cross sectional area 9and thus affect the resistive loss)?

Very different question. I've heard some early Cat5 was rather slack
(untwisted) on pair 4 (brown/wt) since only pairs 2&3 were expected
to be heavily used (100baseTX). A cable with one slack pair will
meet spec for XT and actually makes the rest of the cable easier
(less problems with twist slew). It will have problems with alien XT
(from other cables). It will also have problems with skew -- that
pair is shorter and signals will arrive sooner. AFAIR, Cat5e doesn't
have a slew spec, Cat6 does.

> I still look forward to the day that computer interconnections begin

> using symmetric bidirectional cabling so we can stop having to deal

> with various wiring schemes and get rid of crossover cables and such.

Rejoice, for that day is nigh! GigE does autodetect, and some
more recent 100baseTX equipment do MDI/MDI-X autodetect.

-- Robert

other useful resources:
The Federal Communications Commission (FCC)
Telecommunications Industry Association
Electronic and Software Security Products and Services
International Telecommunication Union

Custom CGI Perl and PHP programming by 1-Script.com

T568A or T568B

Home Cabling Guide

Learn More