DComTalk.com

I went to new customer's office yesterday to identify and label the
data cabling installed (but never used) by the previous occupant.
EVERY cable was miswired, with ALL pairs reversed at the patch panel.

The problem was easy enough to identify and correct, however, I'm
baffeled by the fact that the (5) PCs they had already connected using
the installed cabling were working! Out of curiosity, I check some
(but not all) of the patch cables they were using, and they were all
passed a wiremap test.

I know that current hardware will detect and compensate for
straight-through and cross-over wiring, but I've never heard of
auto-correcting for reversed pairs.

Anyone got an explaination for this one!?

Michael Quinlan <my02explorer@yahoo.com> wrote:

I went to new customer's office yesterday to identify and
label the data cabling installed (but never used) by the
previous occupant. EVERY cable was miswired, with ALL
pairs reversed at the patch panel.

Electrons are color blind, but they do know who their dance
(twist) partners are. TIA-568A & B are side-to-side symmetrical,
so it will work just fine because no pairs are split.

I'm a little surprised because Cat5+ jacks have color codes
on them.

-- Robert

Michael Quinlan wrote:

I went to new customer's office yesterday to identify and label the
data cabling installed (but never used) by the previous occupant.
EVERY cable was miswired, with ALL pairs reversed at the patch panel.

The problem was easy enough to identify and correct, however, I'm
baffeled by the fact that the (5) PCs they had already connected using
the installed cabling were working! Out of curiosity, I check some
(but not all) of the patch cables they were using, and they were all
passed a wiremap test.

I know that current hardware will detect and compensate for
straight-through and cross-over wiring, but I've never heard of
auto-correcting for reversed pairs.

Anyone got an explaination for this one!?

It is a balanced (symmetrical) signal. The hardware sees reversed signal
exactly like normal. It would not be a good idea to do for an un-balanced
application like RS-232, but those are EXTEMELY rare these days. Well,
maybe video without a balun would be another possible un-balanced
application, but I presume they do not send video over those cables either.

But I agree, sloppy work indeed.

--
Dmitri Abaimov, RCDD
http://www.cabling-design.com
Cabling Forum, color codes, pinouts and other useful resources for
premises cabling users and pros
http://www.cabling-design.com/homecabling
Residential Cabling Guide
-------------------------------------



##-----------------------------------------------#
Article posted with Cabling-Design.com Newsgroup Archiv
http://www.cabling-design.com/forum
no-spam read and post WWW interface to your favorite newsgroup -
comp.dcom.cabling - 1984 messages and counting
##-----------------------------------------------##

Michael Quinlan wrote:

I know that current hardware will detect and compensate for
straight-through and cross-over wiring, but I've never heard of
auto-correcting for reversed pairs.

Anyone got an explaination for this one!?

What do you mean by "reversed pairs", if not cross over? It doesn't matter
what colour the wires are, so long as proper pairing is maintained.

My guess is that it worked because it was punched down as 568A and
he 'saw' the colors as 568B. As long as it is 568A or 568B at each end, it
will work. Just my 2 cents.

...Bob

"James Knott" <james.knott@rogers.com> wrote in message
news:zamdnbPE29GeUh3fRVn-gQ@rogers.com...

Michael Quinlan wrote:

I know that current hardware will detect and compensate for
straight-through and cross-over wiring, but I've never heard of
auto-correcting for reversed pairs.

Anyone got an explaination for this one!?

What do you mean by "reversed pairs", if not cross over? It doesn't
matter
what colour the wires are, so long as proper pairing is maintained.

James Knott <james.knott@rogers.com> writes:

What do you mean by "reversed pairs", if not cross over? It doesn't matter
what colour the wires are, so long as proper pairing is maintained.

At one end, the blue-white is punched down where the white-blue
should be and vice versa.



--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

On Wed, 11 May 2005 01:02:30 +0000 (UTC), David Lesher
<wb8foz@panix.com> wrote:

James Knott <james.knott@rogers.com> writes:


What do you mean by "reversed pairs", if not cross over? It doesn't matter
what colour the wires are, so long as proper pairing is maintained.

At one end, the blue-white is punched down where the white-blue
should be and vice versa.

I've seen it done - one customer had the jacks wired properly, but
polarity of every pair was reversed at the patch panel.

Must have had two people working at each end when they terminated the
jacks and backboard.

Worked fine for data, totally choked when they needed to use a data
run for another Legend MLX telephone.

Robert Redelmeier <redelm@ev1.net.invalid> wrote:

Michael Quinlan <my02explorer@yahoo.com> wrote:
However, at the patch panel, the white-blue and blue-white
were reversed, with the same problem appearing on EVERY pair
on EVERY cable.

Ah, tip & ring were crossed over on every pair!

Not good practice! But it works for ethernet because the
signalling system looks for _changes_, not absolute levels.

But can cause all sorts of interesting problems if the
link is used for for telecoms or V24.

The only explaination I can come up with is that the
pair-separators on the punch down were labled with the
pair colors, but not with individual conductor colors,
and the installer had no idea which conductor went on
which side of the separator.

Very typical for 110 punchdowns.

If the connectors at both ends are from the same
manufacturer all that's necessary is that things
are done consistently.

Plenty of simple network testers will show if this
has happened.

--
Mark Evans
St. Peter's CofE Aided School
Phone: +44 1392 204764 X109
Fax: +44 1392 204763

Michael Quinlan wrote:

That's exactly the way this installation appeared.  All the work area
outlets were wired correctly, with the exception of one that had the
brown pair wired to a separate voice jack.  However, at the patch
panel, the white-blue and blue-white were reversed, with the same
problem appearing on EVERY pair on EVERY cable.

One thing to remember, is that the blue pair is not used for 10 & 100 Mb
ethernet. All 4 pairs are used for gigabit. So you could cut the blue
pair and it shouldn't affect networking.

Michael Quinlan <my02explorer@yahoo.com> wrote:

However, at the patch panel, the white-blue and blue-white
were reversed, with the same problem appearing on EVERY pair
on EVERY cable.

Ah, tip & ring were crossed over on every pair!

Not good practice! But it works for ethernet because the
signalling system looks for _changes_, not absolute levels.

The only explaination I can come up with is that the
pair-separators on the punch down were labled with the
pair colors, but not with individual conductor colors,
and the installer had no idea which conductor went on
which side of the separator.

Very typical for 110 punchdowns.

-- Robert

That's exactly the way this installation appeared. All the work area
outlets were wired correctly, with the exception of one that had the
brown pair wired to a separate voice jack. However, at the patch
panel, the white-blue and blue-white were reversed, with the same
problem appearing on EVERY pair on EVERY cable. The only explaination
I can come up with is that the pair-separators on the punch down were
labled with the pair colors, but not with individual conductor colors,
and the installer had no idea which conductor went on which side of the
separator.

To clarify for anyone not following this mess, the wiremap results
were:

1 - 2
2 - 1
3 - 6
4 - 5
5 - 4
6 - 3
7 - 8
8 - 7

Obviously, the original installer did no testing, with the possible
exception of testing connectivity between two ethernet devices.

Thanks to all for the helpful replies.

David Lesher wrote:

James Knott <james.knott@rogers.com> writes:


What do you mean by "reversed pairs", if not cross over? It doesn't
matter what colour the wires are, so long as proper pairing is maintained.

At one end, the blue-white is punched down where the white-blue
should be and vice versa.

Shouldn't be a problem, as the signal is not polarity sensitive.

David Lesher wrote:

James Knott <james.knott@rogers.com> writes:

What do you mean by "reversed pairs", if not cross over? It doesn't matter
what colour the wires are, so long as proper pairing is maintained.

At one end, the blue-white is punched down where the white-blue
should be and vice versa.

10baseT does need the right polarity, but as far as I know it
is usual for the transceivers to detect and correct it.

I know many separate transceivers with an LED polarity indicator,
the only reason for having one is for marketing value.
(Or for network installers to check the polarity with.)

As far as I know, all UTP ethernet either don't care or detect
and correct polarity.

-- glen

In article <eM-dnR-TXfcDIrLeRVn-rA@comcast.com>, glen herrmannsfeldt
<gah@ugcs.caltech.edu> writes

David Lesher wrote:

James Knott <james.knott@rogers.com> writes:

What do you mean by "reversed pairs", if not cross over? It doesn't matter
what colour the wires are, so long as proper pairing is maintained.

At one end, the blue-white is punched down where the white-blue
should be and vice versa.

10baseT does need the right polarity, but as far as I know it
is usual for the transceivers to detect and correct it.

I know many separate transceivers with an LED polarity indicator,
the only reason for having one is for marketing value.
(Or for network installers to check the polarity with.)

As far as I know, all UTP ethernet either don't care or detect
and correct polarity.

-- glen

Only the Orange and Green pairs are used for 10/100 Ethernet, so you do

not even need the blue or brown to be punched down for it to work.
That does NOT mean it is correct if they are not!
And, to the pedants, I do know about 100VG which uses all the pairs, but
it is *almost* extinct.
Phil Partridge
philp@pebbleGRIT.demon.co.uk
Remove the grit to reply