Cat5, 4 pair crossover (for gigabit uplink)

Discussion in 'Computer Support' started by Keme, Sep 18, 2006.

  1. Keme

    Keme Guest

    So what happened to polarity?

    With common crossover cables, pair 2 and 3 switch places, but color goes
    to color (ring to ring, or minus to minus) and white goes to white (tip
    to tip, or plus to plus). Pair 1 and 4 go straight through (not used in
    10/100 Mb signalling).

    With gigabit on copper, you use all four pairs, so (using coloring from
    a common four pair cable) pair 2 and 3, green and orange, switch places,
    and pair 1 and 4, blue and brown, switch places. But with pair 1 and 4,
    color goes to white (i.e. "ring to tip", or minus to plus) according to
    every pinout scheme I've seen.

    I'm having some trouble making this work, and suspect the cabling in the
    wall, but the above is a potential source of error (and due to
    availability demands I'm reluctant to move the switches right now, to
    bypass the wall cabling). Is the "ring to tip" crossing correct for this
    kind of cable? (Most cables I made are conformant to normal polarity
    spec, i.e. color to color and white to white, but I've made some that
    follow published pinout schemes too. None work reliably, though
    sometimes I get momentary gigabit indications.)
    Keme, Sep 18, 2006
    #1
    1. Advertising

  2. Keme

    why? Guest

    On Mon, 18 Sep 2006 22:07:03 +0200, Keme wrote:

    >So what happened to polarity?


    It's still there and you are just getting totally confused.

    >
    >With common crossover cables, pair 2 and 3 switch places, but color goes


    They are all common :) , Up to 100Mbps requires only pairs 2/3 swapped.
    A few odd cases have needed 7<->8 and 4<->5 as well.

    <snip the unneeded ring/tip stuff>

    >With gigabit on copper, you use all four pairs, so (using coloring from
    >a common four pair cable) pair 2 and 3, green and orange, switch places,
    >and pair 1 and 4, blue and brown, switch places. But with pair 1 and 4,
    >color goes to white (i.e. "ring to tip", or minus to plus) according to
    >every pinout scheme I've seen.


    http://www.pccables.com/02303.htm

    Save time and hassle, as the crimping rules are more strict and just buy
    a premade cable.

    <snip>

    Me
    why?, Sep 18, 2006
    #2
    1. Advertising

  3. Keme

    Keme Guest

    why? wrote:
    > On Mon, 18 Sep 2006 22:07:03 +0200, Keme wrote:
    >
    >
    >>So what happened to polarity?

    >
    >
    > It's still there and you are just getting totally confused.
    >
    >
    >>With common crossover cables, pair 2 and 3 switch places, but color goes

    >
    >
    > They are all common :) , Up to 100Mbps requires only pairs 2/3 swapped.
    > A few odd cases have needed 7<->8 and 4<->5 as well.
    >
    > <snip the unneeded ring/tip stuff>
    >
    >
    >>With gigabit on copper, you use all four pairs, so (using coloring from
    >>a common four pair cable) pair 2 and 3, green and orange, switch places,
    >>and pair 1 and 4, blue and brown, switch places. But with pair 1 and 4,
    >>color goes to white (i.e. "ring to tip", or minus to plus) according to
    >>every pinout scheme I've seen.

    >
    >
    > http://www.pccables.com/02303.htm
    >
    > Save time and hassle, as the crimping rules are more strict and just buy
    > a premade cable.
    >
    > <snip>
    >
    > Me

    Thanks for nothing...
    I won't order from the other side of the earth, but I have a supplier
    that'll probably supply the patches. Only trouble is, they're not in
    store, so delivery may take a week. (Delivery from the US takes longer,
    longer than any other source/region I've tried. Thanks anyway...)

    As for crimping requirements, I've made custom lengths for server-switch
    Gb connection, and get at least the same performance as with premade
    cables.

    What I really want to know is if/when/why polarity on pair 1 was
    reversed. If nothing else, it's a useful and interesting fact for my
    trainees, who are also supposed to know the facts about cabling.
    Keme, Sep 18, 2006
    #3
  4. Keme

    Frosty Guest

    As the sun rose on Mon, 18 Sep 2006 22:07:03 +0200, the distinguished
    and oh so talented Keme <> climbed up to the
    podium, shuffled some papers, took a sip of ice water and shouted in a
    loud voice:

    >So what happened to polarity?


    We voted it out.
    Frosty, Sep 19, 2006
    #4
  5. Keme

    Frosty Guest

    As the sun rose on Tue, 19 Sep 2006 00:59:17 +0200, the distinguished
    and oh so talented Keme <> climbed up to the
    podium, shuffled some papers, took a sip of ice water and shouted in a
    loud voice:

    <snip>
    >Thanks for nothing...


    <snip>
    >What I really want to know is if/when/why polarity on pair 1 was
    >reversed.


    It wasn't, you're just on the wrong side of the earth.
    Look it it from *this* side & you'll see it's still the same!
    --

    Estne volumen in toga, an solum tibi libet me videre?
    Frosty, Sep 19, 2006
    #5
  6. Keme

    why? Guest

    On Tue, 19 Sep 2006 00:59:17 +0200, Keme wrote:

    >why? wrote:
    >> On Mon, 18 Sep 2006 22:07:03 +0200, Keme wrote:
    >>
    >>
    >>>So what happened to polarity?

    >>
    >>
    >> It's still there and you are just getting totally confused.
    >>
    >>
    >>>With common crossover cables, pair 2 and 3 switch places, but color goes

    >>
    >>
    >> They are all common :) , Up to 100Mbps requires only pairs 2/3 swapped.
    >> A few odd cases have needed 7<->8 and 4<->5 as well.
    >>
    >> <snip the unneeded ring/tip stuff>
    >>
    >>
    >>>With gigabit on copper, you use all four pairs, so (using coloring from
    >>>a common four pair cable) pair 2 and 3, green and orange, switch places,


    So that's common depending on 568A or 568B starting position?

    >>>and pair 1 and 4, blue and brown, switch places. But with pair 1 and 4,
    >>>color goes to white (i.e. "ring to tip", or minus to plus) according to
    >>>every pinout scheme I've seen.

    >>
    >>
    >> http://www.pccables.com/02303.htm
    >>
    >> Save time and hassle, as the crimping rules are more strict and just buy
    >> a premade cable.
    >>
    >> <snip>
    >>
    >> Me

    >Thanks for nothing...


    That's your problem, the site has a wiring scheme for the connections
    and the signals marked.

    >I won't order from the other side of the earth, but I have a supplier


    So you assume without saying the server you are posting from .no isn't
    just where you post from but the country you are in. I used to have a
    posting account on a .no server but am 100's of miles away.

    <snip the rest of the post>

    Me
    why?, Sep 19, 2006
    #6
  7. Keme

    why? Guest

    On Tue, 19 Sep 2006 00:59:17 +0200, Keme wrote:

    >why? wrote:
    >> On Mon, 18 Sep 2006 22:07:03 +0200, Keme wrote:
    >>
    >>
    >>>So what happened to polarity?

    >>
    >>
    >> It's still there and you are just getting totally confused.
    >>
    >>
    >>>With common crossover cables, pair 2 and 3 switch places, but color goes


    <snip>


    >What I really want to know is if/when/why polarity on pair 1 was
    >reversed. If nothing else, it's a useful and interesting fact for my
    >trainees, who are also supposed to know the facts about cabling.


    Try asking in this ng,
    comp.dcom.lans.ethernet
    or searching
    http://groups.google.com/group/comp.dcom.lans.ethernet?
    or the group
    FAQ
    http://groups.google.com/group/comp.dcom.lans.ethernet?lnk=gschg
    or
    www.cisco.com
    it suppose really depends if 1000Base-T or PoE came first, you can chek
    for that.
    why?, Sep 19, 2006
    #7
  8. "Keme" <> wrote in message
    news:450f24c6$...

    > What I really want to know is if/when/why polarity on pair 1 was
    > reversed. If nothing else, it's a useful and interesting fact for my
    > trainees, who are also supposed to know the facts about cabling.


    If you mean the white-color-white-color-white-color-white-color sequence, it
    hasn't.

    It was that way so that tip and ring on audio phone circuits for pairs 1 and
    2 were out of phase, thereby allowing for better crosstalk rejection.

    That's all.

    RwP
    Ralph Wade Phillips, Sep 23, 2006
    #8
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Bill Evans
    Replies:
    5
    Views:
    4,370
    Beverly Howard [Ms-MVP/MobileDev]
    Feb 3, 2005
  2. Replies:
    5
    Views:
    1,118
  3. richard

    instructions needed on cat5 crossover networking

    richard, Jun 7, 2007, in forum: Computer Support
    Replies:
    2
    Views:
    555
  4. Joey

    Gigabit over 2 pair?

    Joey, Jul 4, 2007, in forum: Cisco
    Replies:
    5
    Views:
    808
    Peter Van Epp
    Jul 6, 2007
  5. Giuen
    Replies:
    0
    Views:
    552
    Giuen
    Sep 12, 2008
Loading...

Share This Page