Jump to content


Photo
- - - - -

IP/Ethernet twisted pair polarity


  • Please log in to reply
22 replies to this topic

#1 mikefamig

mikefamig

    Cocoonut

  • Registered
  • PipPipPipPip
  • 2575 posts
  • Experience:novice
  • Hardware:Elk M1
  • Tech:UPB
  • CCTV:ip

Posted 28 December 2016 - 11:28 AM

I am neatening up some ehternet wiring in my house and found that the ATT installer that terminated the Uverse VDSL service cable at my gateway reversed the blue/blue-white wires according to the wiring diagram that came with an rj25 socket that I bought. They occupy the middle two conductors in a six conductor plug.

 

My question - Is there polarity to an ethernet twisted pair? I understand that the twisted pair is a balanced or differential signal pair and it seems to me that the order of the wires shouldn't matter. I connected the pair to agree with the diagram that came with the connector which is opposite of how the installer had them and everything is working well but inquiring minds want to know.

 

Mike.


Edited by mikefamig, 28 December 2016 - 11:28 AM.


#2 pete_c

pete_c

    Cocoonut

  • -=Gold Supporter=-
  • 7395 posts
  • Location:House
  • Experience:average
  • Software:Main Lobby, Open Source Automation
  • Hardware:HAI OmniPro II, Mi Casa Verde Vera, Ocelot
  • Tech:X10-PLC, X10-RF, UPB, INSTEON, Z-Wave, ZigBee, 1-Wire, xAP, xPL, ALC
  • Audio:Russound
  • Video:MythTV
  • CCTV:analog, ip, dvr
  • Phone:Asterisk, FreePBX, Ooma, POTS, VoIP via ISP

Posted 28 December 2016 - 11:46 AM

Relating to a POTs line.

 

Yeah here converting or trying to convert multiple analog phone lines to catXX.  For phone lines it starts in the middle with blue / white.   I was using catxx but just hooked up any pairs of wires in a willy nilly fashion for it to work but not following any of the documented standards.

 

Just last week I messed up here with first two lines and swapped the two white striped leads and didn't get a dial tone until I tested each line with a VOM. It is 48VDC off hook and 3VDC-9VDC on hook.

 

Wall mounted my 2 line Panasonic DECT system and didn't want any wires to show.  Originally wired the power with the catxx cables.  Redid it such that I ran one separate two 16/2 wire to the wall mounted telephone jack.  In the process though I did burn up a Panasonic transformer switching the leads as the barrel connector is opposite the norm (like some of the Digi stuff I use).  I do have another pots line device in series which seizes the line similar to the alarm panel that I use today.  I can use it to send CID automation messages to my phones (rings plus message) or reroute the line if I choose to.

 

The above said there is polarity on the old POTs phone lines (similar to reversing leads for a touch tone phone).  Telephone lines running on catxx still utilize polarity.   Thinking with POE (47-54VDC) that polarity is used with 4 wires (2 + 2). 

 

Maximum voltage for VDSL is 7VDC with average being 6VDC.

 

So the standard is really relating to what traverses the cable versus what the standard would be for the cable.



#3 mikefamig

mikefamig

    Cocoonut

  • Registered
  • PipPipPipPip
  • 2575 posts
  • Experience:novice
  • Hardware:Elk M1
  • Tech:UPB
  • CCTV:ip

Posted 28 December 2016 - 01:51 PM

I remember the terms ring and tip relating to pots but I don't believe that applies to ethernet twisted pair. I think that the blue pair is just a standard and so I used it but I think that the ethernet signal is +5v -5v differential signal which shouldn't matter for polarity, no?
 

Mike


Edited by mikefamig, 28 December 2016 - 03:38 PM.


#4 Frunple

Frunple

    Dedicated Cocooner

  • Registered
  • PipPipPip
  • 657 posts

Posted 28 December 2016 - 03:18 PM

An RJ25 is not ethernet, that would be the DSL side. Polarity wouldn't matter.

You don't say how it is exactly wired, just that the pair is reversed, but if they are reversed on both sides, they're not reversed polarity wise.

Gigabit ethernet, polarity wouldn't matter. 100M and below, might depend on the device being used but I can almost guarantee newer devices it wouldn't matter.



#5 pete_c

pete_c

    Cocoonut

  • -=Gold Supporter=-
  • 7395 posts
  • Location:House
  • Experience:average
  • Software:Main Lobby, Open Source Automation
  • Hardware:HAI OmniPro II, Mi Casa Verde Vera, Ocelot
  • Tech:X10-PLC, X10-RF, UPB, INSTEON, Z-Wave, ZigBee, 1-Wire, xAP, xPL, ALC
  • Audio:Russound
  • Video:MythTV
  • CCTV:analog, ip, dvr
  • Phone:Asterisk, FreePBX, Ooma, POTS, VoIP via ISP

Posted 28 December 2016 - 03:32 PM

You mentioned Ethernet twisted pair.   Yup it is different for Ethernet voltage and signal wise.

 

Found a ring and tip drawing standards using catxx.

 

Attached File  ringtip.jpg   84.02K   8 downloads

 

A 100BASE-TX transmitter sends three differential voltages, +1 V, 0 V, or −1 V.

 

1000BASE-T uses all four pairs bi-directionally and the standard includes auto MDI-X; however, implementation is optional. With the way that 1000BASE-T implements signaling, how the cable is wired is immaterial in actual usage. The standard on copper twisted pair is IEEE 802.3ab for Cat 5e UTP, or 4D-PAM5; four dimensions using PAM (pulse amplitude modulation) with five voltages, −2 V, −1 V, 0 V, +1 V, and +2 V. While +2 V to −2 V voltage may appear at the pins of the line driver, the voltage on the cable is nominally +1 V, +0.5 V, 0 V, −0.5 V and −1 V.

 

Thinking polarity starts to matter more with POE.

 

House here when built the telephone man ran CatXX cable from the d mark box to the inside around 2001 or so.  Guessing telco's just wanted to upgrade their wiring from cat3 that I think they were using back then. 

 

When I was doing the phone thing used an old ma bell telephone with alligator clips on it to test the line in the kitchen; wife laughed when she saw the old phone.  This is how I would test prior to know the color coordination of the cat5e.  IE: got my OmniPro working this way using 2-3 lines on one wire.

 

Thinking there is still polarity using an RJ-25.  Gee what is the difference then between an RJ-11 and an RJ-12?

 

Notice that blue is always pair #1. 

 

Attached File  polarity.jpg   36.88K   6 downloads

 

I used RJ-12's for my 1-wire network way back before going to RJ-45's.

 

In the 1990's someone gave me a vampire clamp (well they called it that) and I used it as a paperweight on my desk.



#6 mikefamig

mikefamig

    Cocoonut

  • Registered
  • PipPipPipPip
  • 2575 posts
  • Experience:novice
  • Hardware:Elk M1
  • Tech:UPB
  • CCTV:ip

Posted 28 December 2016 - 03:46 PM

This is simply one twisted pair carrying the signal from outside which I think is called VDSL or VDSL2 or something like that? It was ATT Uverse and is now Frontier Communications in Connecticut.

 

The plug was an rj25 and the installer used the two center conductors. I used the same two center conductors but reversed the pair. The original installer terminated the incoming cable with rj25 6-pin male connector that plugged into the gateway. I reversed the order that they connect to the gateway and there is no apparent difference.

 

Mike.



#7 pete_c

pete_c

    Cocoonut

  • -=Gold Supporter=-
  • 7395 posts
  • Location:House
  • Experience:average
  • Software:Main Lobby, Open Source Automation
  • Hardware:HAI OmniPro II, Mi Casa Verde Vera, Ocelot
  • Tech:X10-PLC, X10-RF, UPB, INSTEON, Z-Wave, ZigBee, 1-Wire, xAP, xPL, ALC
  • Audio:Russound
  • Video:MythTV
  • CCTV:analog, ip, dvr
  • Phone:Asterisk, FreePBX, Ooma, POTS, VoIP via ISP

Posted 28 December 2016 - 04:17 PM

Ahh...personally thought that a POTs line was similiar to an ADSL and a VDSL line...well and here it was two line pairs that I switched wires to.

 

Unrelated to OP very first Internet connection at home here was ISDN at a whopping 128 kbit/s (guess it was two channels?).

 

Curious what the difference is between ISDN and DSL..

 

Regardless of color it's always the two center wires in an RJ11,12,25,45 that are considered pair #1 in the telco world and for whatever reason I thought that the wires (ring tip) were polarized.

 

I am getting old and

 

Mike the best proof is in the pudding eh?  (IE the leads were flipped and they worked anyhow).



#8 DELInstallations

DELInstallations

    Cocoonut

  • Professionals
  • PipPipPipPip
  • 5130 posts
  • Software:Custom
  • Hardware:Custom
  • Tech:ZigBee
  • Audio:Custom
  • Video:Custom
  • CCTV:analog, ip, dvr

Posted 28 December 2016 - 07:24 PM

Polarity can cause issues with POTS, but in this area, it has been known for ATT (SBC, Bell, Frontier) to intentionally reverse the color codes or have sloppy install practice. Normally it doesn't cause an issue, however there is equipment connected to POTS that can have polarity sensitivity. Usually appears as ringer issues.

 

VDSL, generally a non-issue, but if there's a doubt, easy enough to verify which is T and which is R using a DMM.



#9 RAL

RAL

    Cocoonut

  • Registered
  • PipPipPipPip
  • 1471 posts
  • Location:Rhinebeck, NY
  • Experience:average
  • Hardware:Elk M1
  • Tech:X10-PLC
  • Phone:POTS

Posted 28 December 2016 - 08:42 PM

My question - Is there polarity to an ethernet twisted pair? I understand that the twisted pair is a balanced or differential signal pair and it seems to me that the order of the wires shouldn't matter. I connected the pair to agree with the diagram that came with the connector which is opposite of how the installer had them and everything is working well but inquiring minds want to know.

 

 

From a purely low level, technical, hardware point of view, yes there is polarity on ethernet twisted pair.  However, on a lot of newer ethernet hardware, the input circuits detect a polarity reversal condition and correct for it. 

 

So from the user's point of view, it may look like it doesn't matter.  Smart hardware is great, but it can leave you scratching your head sometimes when you come across a cable that happens to have reversed polarity on one end, and it works when connected to one device, but not another.


Edited by RAL, 28 December 2016 - 08:47 PM.


#10 mikefamig

mikefamig

    Cocoonut

  • Registered
  • PipPipPipPip
  • 2575 posts
  • Experience:novice
  • Hardware:Elk M1
  • Tech:UPB
  • CCTV:ip

Posted 29 December 2016 - 08:22 AM

OK so now I have to ask what is the correct orientation of the center pair in rj25 vdsl? Looking at the male plug with the cable facing you and the locking clip facing up is the blue wire on the left or right hand side?



#11 mikefamig

mikefamig

    Cocoonut

  • Registered
  • PipPipPipPip
  • 2575 posts
  • Experience:novice
  • Hardware:Elk M1
  • Tech:UPB
  • CCTV:ip

Posted 29 December 2016 - 08:39 AM

Polarity can cause issues with POTS, but in this area, it has been known for ATT (SBC, Bell, Frontier) to intentionally reverse the color codes or have sloppy install practice. Normally it doesn't cause an issue, however there is equipment connected to POTS that can have polarity sensitivity. Usually appears as ringer issues.

 

VDSL, generally a non-issue, but if there's a doubt, easy enough to verify which is T and which is R using a DMM.

 

Why would an installer intentionally reverse the pair?



#12 pete_c

pete_c

    Cocoonut

  • -=Gold Supporter=-
  • 7395 posts
  • Location:House
  • Experience:average
  • Software:Main Lobby, Open Source Automation
  • Hardware:HAI OmniPro II, Mi Casa Verde Vera, Ocelot
  • Tech:X10-PLC, X10-RF, UPB, INSTEON, Z-Wave, ZigBee, 1-Wire, xAP, xPL, ALC
  • Audio:Russound
  • Video:MythTV
  • CCTV:analog, ip, dvr
  • Phone:Asterisk, FreePBX, Ooma, POTS, VoIP via ISP

Posted 29 December 2016 - 09:28 AM

You can just utilize a VOM (DMM) and check the leads from the dmark to your cable. 

 

My dmark here outside just had the leads screwed down rather than punched in on the left compartment. 

 

I am thinking there is some voltage on the incoming wires outside and one wire is positive and one is negative.  I had catxx coming in from the pots dmark here. 

 

The left side with the torx screw is what you want to look at if you have a modern dmark outside.  You can color coordinate it then if you want.

 

Here are example pictures.  Note they are all similiar in design and the left side is typically not touched and that is why there is torx on it.

 

Attached File  dmark.jpg   34.93K   1 downloads

 

Attached File  dmark-2.jpg   12.85K   2 downloads



#13 mikefamig

mikefamig

    Cocoonut

  • Registered
  • PipPipPipPip
  • 2575 posts
  • Experience:novice
  • Hardware:Elk M1
  • Tech:UPB
  • CCTV:ip

Posted 29 December 2016 - 09:43 AM

You can just utilize a VOM (DMM) and check the leads from the dmark to your cable. 

I understand this but it still doesn't say which is which(+-).

 

I have found on the net that ring is negative and tip is positive so I will take a look at it today with a meter. I just thought that you guys could explain if the polarity matters. It seems to work both ways and my speed tests online seem the same both ways.

 

So am I correct that I should show positive voltage if I connect my VOM with red lead to the tip and black lead to ring?

 

Mike



#14 DELInstallations

DELInstallations

    Cocoonut

  • Professionals
  • PipPipPipPip
  • 5130 posts
  • Software:Custom
  • Hardware:Custom
  • Tech:ZigBee
  • Audio:Custom
  • Video:Custom
  • CCTV:analog, ip, dvr

Posted 29 December 2016 - 10:02 AM

Why would an installer intentionally reverse the pair?

In the Bell days, it was for "job security". Dates back to the old Bell Dmarcs that had the big carbon fuses mounted inside (or the later ones outside). Well before the plastic days. Reverse the polarity, it prevented others from adding to the system....just after the days when all the phone equipment was hardwired and leased or purchased from ma Bell.

 

The old dmarc terminals, when the drop cables had no ID on them, the ring was supposed to be landed on the right per Bell, but I've seen it reversed way too often to be incidental, same as the inside plant cables.  Also in the era of cloth covered inside cabling with a colored thread added between the sheath and copper with the cable ID (red, green, and I swear there was also yellow/black, but it's been a while since I did hardcore residential).



#15 pete_c

pete_c

    Cocoonut

  • -=Gold Supporter=-
  • 7395 posts
  • Location:House
  • Experience:average
  • Software:Main Lobby, Open Source Automation
  • Hardware:HAI OmniPro II, Mi Casa Verde Vera, Ocelot
  • Tech:X10-PLC, X10-RF, UPB, INSTEON, Z-Wave, ZigBee, 1-Wire, xAP, xPL, ALC
  • Audio:Russound
  • Video:MythTV
  • CCTV:analog, ip, dvr
  • Phone:Asterisk, FreePBX, Ooma, POTS, VoIP via ISP

Posted 29 December 2016 - 10:32 AM

Your dmark should show the old pots colors...

 

Line #1 is red and green

Line #2 is yellow  and black

Line #3 if present is white and blue

 

Is there a cat5 cable going from the dmark to the inside of the house?  Use blue for line #1 as depicted in the drawings above.

 

Attached File  potscolors.jpg   23.14K   0 downloads

 

When simple on-premises wiring is color-coded, two-wire telephone plugs or the first pair of a multi-pair connector commonly have the tip wire coded green and ring coded red. In four wire plugs, the second pair has black tip and yellow ring. A third pair consists of white tip and blue ring. For larger cable assemblies more complex schemes such as the 25-pair color code are used.

 

Telephone technicians often used the phrase "red-right-ring-rear" (or "ring-right-red-rough") to remember that the red wire connects to the right-side post in the wall jack and to the ring on the plug and to the rear lug on main distribution frames. Sometimes "rough" or "ridge" was added for jumper wires with a tactile code

 

In the early years of the telephone industry when rotary dial instruments were in use, the correct polarity of tip and ring was important only for properly ringing a telephone, especially in party line service with selective ringing, and for correctly identifying the calling customer on certain party lines for toll calls.

When Touch-Tone service was introduced in the 1960s, the dual-tone multi-frequency signaling (DTMF) tone generator also required correct polarity as it depended on the line power for operation. Later Touch-Tone telephones included a diode bridge that eliminated the polarity sensitivity, so that consumer telephone service is essentially immune to reversal today.

 

A bit of history...

 

Tip and ring are the names of the two conductors or sides of a telephone line. The terms originate in reference to the telephone plugs used for connecting telephone calls in manual switchboards. One side of the line is connected to the metal tip of the plug, and the second is connected to a metal ring behind the tip, separated and insulated from the tip by a non-conducting material. When inserted into a jack, the plug's tip conductor connects first, followed by the ring conductor. In many European countries tip and ring are referred to as the A and B wires.

 

Attached File  tipandring.jpg   4.42K   0 downloads

1: Sleeve, 2: Ring, 3: Tip, 4: Insulators

The ring conductor has a direct current (DC) potential of −48V to −52V with respect to tip conductor when the line is in the on-hook (idle) state. Neither conductor is referenced to ground. Floating both conductors (not referencing either one to ground) minimizes the pickup of hum from any nearby alternating current (AC) power wires.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users