Telecommunications 101

Flagg

Limp Gawd
Joined
Mar 12, 2001
Messages
266
I have seen many questions over the years that I have been coming here, regarding T1’s. DSL, SONET, DACS’s and other telecom equipment. Since I am bored at work and need to do something to exercise my brain I thought I would make this FAQ (that will hopefully get stickied :) ) To address some of the common questions I have seen. Please feel free to ask more if you like.

Terminology

LEC - Local Exchange Carrier aka the phone company
ILEC - Incumbent Local Exchange Carrier
CLEC - Competetive Local Exchange Carrier
DACS – Digital Access and Cross Connect System
SONET – Synchronous optical network
BERT - Bit Error Rate Tester

What is an ILEC?
A large telephone company that has been providing local telephone service in the United States since the divestiture of the AT&T telephone monopoly in 1982.
These guys are the old fogeys on the block :D. With all the buyouts, mergers, Chapter 11's going on in telco right now I'm am not entirely sure who has been around since AT&T anymore. But I THINK Qwest(formerly USWest), Verizon (formally GTE) are ILEC's.

What is a CLEC?
A new local telephone company (eg, US ONE, Winstar, TCG, ICG) that is entering the local market in order to compete with the incumbent local telephone company.
There are not too many "new" ones starting up anymore, but this definition works the best.

What is the Telecom Act of 1996

This is a tricky one. I will probably massacre the full definition of what this bill was, but I will do my best. The Telecom Act of 1996 was a bill put into action by congress to de-regulate the telecommunications industry, including telco, radio, wireless and many other aspects of telecommunications. The "official" description of the bill is
An original bill to provide for a pro-competitive, de-regulatory national policy framework designed to accelerate rapidly private sector deployment of advanced telecommunications and information technologies and services to all Americans by opening all telecommunications markets to competition, and for other purposes.
What that meant for the telco market was that ILEC's were forced in to opening up their facilities for third-party companies to use. This gave way to a surge of CLEC's. CLEC's now had the means to deliver POTS, T1,T3 and other services to customers that were normally forced to use ILEC services.

Was the Telecom Act of 1996 good or bad?
This is a touchy subject. Depending on who you talk to, this bill either helped the industry or helped destroy the industry. I happen to believe its the former. I worked for a CLEC for several years, and people were migrating over at an insane rate. Fed up with lousy customer service, and ZERO competition for customers to choose from, people were surprised by what a telco company could be. I watched the change in attitude from companies such as Qwest and Verizon over the course of 2-3 years, going from "CLEC's will never amount to anything" to "Um, why did we lose 10,000 customers last quarter?" To eventually "I guess we need to change our game plan" All of a sudden ILEC's were not the only game in town. have you ever seen the "Spirit of service in action" commercial from Qwest? That whole mantra started when Qwest realized they were losing customers left and right to CLEC's.

All that being said the CLEC business model was fundamentally flawed. Even though CLEC's had access to ILEC facilities they were still customers of the ILEC's. CLEC's could not work on ILEC equipment, and still had to wait for trobule tickets to be resolved. And because CLEC's had to buy facilities from ILEC's they typically had to undersell their products in order for their service to be attractice to customers.

Eventually just like the dot com bubble, the telco bubble popped. There are still a few around (Yipes!, COVAD and XO come to mind) but its nothing compared to 2 or 3 years ago. In the end many CLEC's were bought up by LEC's. One thing the CLEC era helped in, was to prove to LEC's that you cannot get too happy sitting on your high hoarse. Given the right options people will leave LEC's in search of better service. I have dealt with telco's since then, and I can tell you its nothing like what it used to be, and I am sure it is for the better.

Today, we are slowly migrating back to the way things were. With mergers and buyouts happening on a regular basis, we will soon have no competition to choose from.

What is the difference between a bit and a byte
1.544MBps is completely different than 1.544Mbps.
MB = Megabytes
Mb = Megabits
A bit is single numerical value. It's either a 1 or a 0, that encodes a single unit of digital information. A byte is 8 bits. So when you see 1.544Mbps that is 1,544,000 bits per second. Thats a big difference from MBps.

What is a T1?

A dedicated point-to-point digital connection supporting data rates of 1.544Mbits per second. A T-1 line actually consists of 24individual channels, each of which supports 64Kbits per second. Each 64Kbit/second channel can be configured to carry voice or data traffic. Most telephone companies allow you to buy just some of these individual channels, known as fractional T-1.

What is a T3?

A dedicated point to point, high capacity digital service capable of transmitting data at speeds up to 45 Mbps. Comprised of 28 T1's.

Why do T1/T3's cost so much?

When you get T1/T3 service you are paying for an SLA. (Service level agreement)
When a T1/T3 drops it is treated differently than a DSL/phone line. T1/T3's get
worked / dispatched much quicker than normal POTS lines. Response times for T1/T3's
typically is 1-4 hours, where POTS lines have a response time of 1-4 days.

What is the difference between a DS1 and a T1?

Excellent question. Thanks to unhappy_mage for bringing it up. T1 / DS1 and T3 / DS3 are often used interchangeably. But there is a distinct difference between T1/T3 and DS1/DS3. (In the example I use T1/DS1 for simplicity. But the rule applies to T3/DS3 as well)

You can get DS1 service without a T1 line but if you have a T1 line, you've got DS1.
DS1 is a digital signal level that is part of a standardized hierarchy or building blocks of digital services for TDM (Time Division Multiplexing) digital lines. A DS1 is a DS1 regardless of whether it is delivered on a T1 line, a microwave relay, a T3 line or a fiber optic carrier. So basically a T1 line is the physical copper pair which the "digital signal" (DS1) rides on. For a little more detail try this link.


I have seen you mention B8ZS/ESF and AMI/D4. Does this mean there are different types of T1's?

There are different types of T1's but the speed remains the same. B8ZS/ESF and AMI/D4 refer to line encoding and framing. I was going to make another section dedicated to this subject but I found a website that has an excellent tutorial on the inner workings of T1's. This gets fairly advanced but is extremely thorough. Click Me

What is a "smart jack?"

A smart jack is also referred to as a RJ-48x. When nothing is plugged into this smart jack
the smart jack creates a loopback back to the CO. This loopback keeps the LEC's equipment from alarming and indicating a dropped T1. When you plug in to the smart check it breaks this loop and allows for traffic to flow into whatever device you are using. i.e router, pbx, etc.


What is ATM?
This is the international standard for cell relay in which multiple service types (such as voice, video, or data) are conveyed in fixed-length (53-byte) cells. Fixed-length cells allow cell processing to occur in hardware, thereby reducing transit delays. ATM is designed to take advantage of high-speed transmission media. I have never worked with ATM so my knowledge is limited. If anyone cares to add to this, please let me know and I will add it to the FAQ.

What is a BERT
Bit Error Rate Tester is a piece of equipment that is used to test the integrity of a circuit. There are many different models of BERT's and each is typically used for a specific function. BERT's interact with a DS-x circuit at the most basic level and can simualte different conditions on a circuit. They simulat these conditions by sending different bit patterns and reading the responses. Some of the more common patterns are QRSS, 0000, 1111, and 0101. QRSS is quasirandom. Meaning it is used to simulate live traffic. Since voice/data/video are just 1's and 0's to a T1 QRSS can randomly shoot 1's and 0's down a DS1 to see how it will react. 0000 and 1111 is all zeros and all ones respectively. These patterns are used to test the integrity of repeaters that the T1 rides across. Repeaters jobs are to "repeat" the 1's and 0's that travel accross them. By sending all ones or all zeros you are essentially stress testing the repeater. 0101 is alternating 1's and 0's. I think this is used as another QRSS type pattern but have never used it much. There are many more pattern than what I have listed, and the list gets even more confusing when you get up to the DS3 level. But I have tried to cover the most basic patterns and typically these patterns are what the LEC uses when troubleshooting a circuit problem. If you would like some more info on different test patterns check out this link

What is a demarc?

Demarc is short for Demarcation Point. This is the point where the responsibility of the LEc ends. Typically this is at the smart jack. LEC's are responsible for everything in between the smart jack and the originating point. (Commonly referred to as the A and Z location). You can pay a LEC to extend this demarc for you, which isn’t a bad thing to do. When the LEC extends the demarcation point they are now responsible for everything except your CSU/DSU. Example.

Example 1. Acme Inc. order’s T1 from Qwest and elects to not have the demarc extended. The techs that work for Acme Inc. extend the demarc themselves. Acme Inc is located on the 12th floor of a 12 story building. The demarc of said building is in the basement. Acme Inc's tech is now responsible for everything from the basement up to where the T1 connects to their equipment. Any problem with the cabling in between the basement and 12th floor is Acme's Inc's responsibility. If there is an accidental cable cut on the 6th floor, which drops the T1 Acme Inc. must take care of it.

Example 2. Acme Inc. orders a T1 from Qwest and elects to have Qwest extend the demarc for them, up to the 12th floor. Qwest is now responsible for the entire path of the circuit. From the CO to the 12th floor smart jack. If a cable cut occurs on the 6th floor and brings the T1 down, Acme Inc. can contact qwest and have them move the T1 to a different copper pair within the building.

IMHO Example 2 makes life a little easier at the expense of a little more money. LEC's WILL NOT TOUCH ANYTHING PAST THE DEMARC! Nothing, zilch, nadda. In fact they may actually give a little chuckle at your expense if you ask this of them.

What is a MPOE
*THANKS TO SHADE91 for the correction*
MPOE - Short for minimum point of entry, the closest practical point to where the cables of a telecommunications service carrier (i.e., a phone or cable company) cross a property line or where its wiring enters a multi-unit building. The MPOE of a multi-unit building is typically 12 inches inside the building’s foundation.

I have been on the phone with a LEC and they keep making reference to a DACS / Repeater / MUX. What are these pieces of equipment and what do they do.

DACS - Digital Access and Cross Connect System. Think of a DACS as a router for circuits. DAC's take all type of circuits, DS0, DS1, DS3 and even some Optical Carriers. Essentially this allow for the connection of circuit without any physical wiring. If a T1 is ordered from Los Angeles to Seattle, that T1 could pass through a number of DACS's along the way until it arrives at your prem. DACS basically connects one circuit to another. T1-A comes in at ones point and needs to be handed off to T1-B. The DACS performs this cross connect electronically removing the need to hardwire these connections within the CO. DACS are also known as DCS and DXS.

Repeater - There are many different repeaters but I'll stick with the ones for T1's. T1's like any communications medium have a distance limit. To overcome this limitation repeaters are put along the path of the circuit which cleans and boosts the signal as it travels the network. This is one of the reason that T1's are not as limited by distance as a DSL line.

MUX - Multiplexer/Demultiplexer. Muxing is taking in several small circuits and outputting a larger one. For example
28 T1's -> 1 T3. Demux is the opposite. Taking in a large circuit and outputting smaller ones. T3->28T1s. Muxing/Demux is at the very heart of the telco industry. Everything is built on the concept of muxing. DS0's-> Ds1'. DS1s->Ds3's etc. Once you learn how things are muxed/demuxed everything else is easy.


What is a soft loop VS a hard loop. And is there a way I can make my own hard loopback

A soft loop back is typically a loopback that is used when a LEC is testing a circuit. When a LEC test's a T1 they use a piece of eqpt. called a BERT or Bit Error Rate Tester. This device is capable of looping up the remote NIU and can pass different bit patterns to verify the integrity of the circuit.

A hard loop back is usually just a RJ-45 connector with pins 1,4 and 2,5 shorted together.

A loopback cable is very simple to build. A T1 operates on pins 1,2,4, and 5 on a RJ-45 plug. Pins 1 and 2 are the transmit pair, 4 and 5 are the receive pair. Just short pins 1 and 4 together and 2 and 5 and voila you have a loopback plug. I usually take some cross connect wire or use a piece of ethernet cable to make mine.
- thanks digilink!

That’s all I can think of off the top of my head. If you have any questions or corrections for me please let me know.

*EDIT 07/06/2005* Added info about CLEC's, T1/DS1 Differences, ATM, BERT's, Telecom act of 96 and some minor changes*

*EDIT 07/07/2005* The post was getting way to long. So I am splitting it up between this thread and "Sonet Information" which now on page 2.
 
That's freaking awesome! I love this stuff! Keep up the good work, I have never learnt so much in 5 minutes :) :)

 
Where does "DSn" fit into this? You mentioned DS0, DS1, etc. - explanation? Thanks!

 
DS is digital signal.

A DS0 is a 64kbps voice/data channel.
A DS1 is 24 DS0's that are multiplexed. (The whole MUX thing he referred to.) This is your T1.
 
How about: CLEC, ATM, xDSL, ISDN, T1 PRI, and MPOE's.

Research your smart jack definition. Definitely not accurate. Mention tips of building your own loopback cable and how you build it and how you could rule out issues with your router/smartjack with this.

Why? LECs/CLECs are very quick to blame your equipment if they can loop up both ends of the smart jack but not see possibly one end. I've had MANY smartjacks go bad on me even though the LEC was able to loop up both ends. I basically had to threaten the LEC with removable of my entire account with them if they didn't do what I said (this includes A LOT of T1s, A LOT of ISDN lines, and TONS of analog lines).

Explain the difference between a bit and a byte and how it relates to telecom. I've seen too many people tell me their T1s max at 1.554MB/s (if you understand bits/bytes you'll understand why that is incorrect).
 
Research your smart jack definition. Definitely not accurate.

True, not every smartjack has an xjack. It all depends on the vendor that made it. I work for a well known wireless carrier and we have several cell sites that do not have this feature, most of what we have found have been Westell units.

Mention tips of building your own loopback cable and how you build it and how you could rule out issues with your router/smartjack with this.

A loopback cable is very simple to build. A T1 operates on pins 1,2,4, and 5 on a RJ-45 plug. Pins 1 and 2 are the transmit pair, 4 and 5 are the receive pair. Just short pins 1 and 4 together and 2 and 5 and voila you have a loopback plug. I usually take some cross connect wire or use a piece of ethernet cable to make mine.
 
Flagg said:
What is an OC-3?

OC-3's are optical carrier circuits that can transmit 155.52Mb/s or 3 DS3's. There are several other OC levels. OC-12, OC-48, and OC-192. The transmission rate of each are listed below.

OC-12 = 622MB/sec
OC-48 = 2.49Gb/sec
C-192 = 16Gig/sec
As mentioned, OC-192 is 10gbps, not 16. Also, it'd be good to go over the various DS'es (link). Also, the difference between a T-1 and a DS1, a T-3 and a DS3, and why the different terms for what appears to be the same thing.
 
Thanks for the advice. I will work on the thread soon. I was away from the PC during the 4th. As far as the smartjack definition is concerned, ALL of the circuits I have worked with on the west coast have the RJ-48x functionality to them, so that is all I know :D If it vary's depending on manufacturer then I havent worked with one yet that hasnt had the loopback function built into it.

I will correct at OC-192, that was a fat finger mistake :D You are correct it is 10Gig/sec
 
Thanks for the advice. I will work on the thread soon. I was away from the PC during the 4th. As far as the smartjack definition is concerned, ALL of the circuits I have worked with on the west coast have the RJ-48x functionality to them, so that is all I know If it vary's depending on manufacturer then I havent worked with one yet that hasnt had the loopback function built into it.

Yes, most of them in fact do. We have very few older units that do not however... Most of what we have in our network are Adtran units which do have an Xjack, the older westell units do not, so I guess it would be safe to say that this is correct :)

Good write up. You might also want to add in some info about SS7, and the different types of framing and line coding on a T1(B8ZS/ESF, D4/AMI) but, seeing how most people here deal strictly with data, this may or may not be helpful. I do not believe that there are any data circuits that run D4/AMI?
 
Tools of the trade:

Punchdown tool:
Used to push wires into various connectors. Most common connectors/blocks are 66 blocks, 110 blocks, keystone jacks and patch panels.

66 Blocks use a 66 punch tool
Patch panels, keystone jacks, and 110 blocks use a 110 tool

There are Bix/Krone tools as well but they arent seen all to often anymore (except on service calls)

Basically your tool can be impact or non-impact and can be bladed or non bladed.
Impact Punch Tools use a fast forceful motion to "shove" the wires into the connector.
If you have a bladed tool it will also cut the wire for you leaving a nice clean punch. If you get a new punch tool, only for 110's really, find anything that has a 110 slot and punch the hell out of it. New blades tend to "stick" in the connecting slot and sometimes pull the wire back out of the jack or patch panel.

Electricians Scissors:
This has to be my most used tool in teh field. It strips wires, cuts wires, doubles as many other useful things. only like $12 bucks for a good pair of klein scissors too. Invaluable if you do alot of cabling for voice and data.

Butt Set or "Butt-in Set"
AKA Linemans handset. Many various models out there. Basically lets you test phone connection almost anywhere. Look for one with a regular clip and a bed of nails clip as well, then you can connect on a connector or a wire. My butt set allows me to plug a regular phone cord into it. If you get one that doesnt you can get what they call a "banjo adapter" which will let you plug it into an rj11 jack and then have posts to clip onto, in a pinch a biscuit jack and a small patch cord can be used for this.

Tone Generator / Amplifier probe
If you are trying to trace a wire run this is invaluable. Basically you either clip or plug it into a phone/network cable or jack and it creates a beeping noise on the wire that you can trace in a phone closet or telco room with an amplifier probe (sometimes call wand) and it will pickup any noise on a line and play it through a small speaker. It's kind of funny when you are tracing a wire in victorias secret and accidentally stumble on someones phone call with your probe..

Amp Champ
If you are doing enterprise grade phone systems..you already know what this is. But for those who dont it lets you put an amphenol connector on 25pair wire. Which is what 99.9% of current phone system switches use for dispersing lines and such.

Multimeter
Always good to have when doing any kind of wiring that can have voltage on it. Good if you want to do a quick test to see if a line is hot and your butt set isnt with you. You can make a ghetto wire tracer by putting a 9v battery on a pair of wires (that arent hot) and then volt testing each pair of wires in the closet....i would only do this as a last case scenario

Crimp Tool
Ratcheting Tools are my favorive because they stay closed in my tool bag and you know when youre fully crimped. Ideal makes a nice one. I personally dont use the strippers on them because they tend to nick the twisted pairs and i dont like that.

Wire Stripper
You can get a big fancy one to strip wires, i wouldnt though unless youre an electrician.
I use my electricians scissors for almost everything. Radio shack sells a blue cable jacket stripper with a 110 non bladed tool on the end. One of the tools thats always in my small belt pouch. Love that thing.

Cable Tester
A quick way to know if your wire and terminations are good.

Cable certifier
These are really expensive, on a budget pick up a validator and you have a sub $1000 cable certifier, our big clients require this. Basically it will test the speed of you cables and say "hey this is up to par with cat 5e standard" or cat3 etc...

My own personal tools that I have constructed:
Take a keystone jack and a 1 foor piece of cable (if you are good with cables you can make this VERY short *ill post pics of mine later*) jack the keystone end in to 568B standard. now if you need a cross over adapter you will crimp an rj45 mod plug on the end in the 568a pattern, if you need a rollover (console) cable adapter you just flip the pin order (8 is 1, 7 is 2 etc...) so its the exact opposite of a 568b crimped end. Now you can turn any cable into a cross over/rollover cable. I made one of each. Its also good to have the serial port adapter on hand to connect that rollover to.



FOR PULLING CABLE:
String and heavy object
tie heavy object (not too heavy, full roll of electrical tape works good) to twine or string, again they make string just for pulling cable and its good. tie and tape string to cables. voila! I must stress that pulling cable correctly is an art. for voice lines its not too bad...but for data...woah. Some quick pointers, dont pull on the cable too hard, you'll stretch it. STAY AWAY FROM POWER LINES OR CONDUIT OR HVAC MACHINERY OR ANYTHING THAT IS GOING TO HAVE MASSIVE CURRENT.

You can use STP (shielded twisted pair) cable and that will help some (ground the foil!) and if its in a new building make sure everything is routed away from power lines/conduit or has it's own conduit thats a little further off from the power lines. Itll save you work and headache later.

Small but heavy chain
tie cable to chain and use it for going down thick conduit or power poles, makes for fast drop.

pulling rods
same theory, attach cable and make it go where you want it to. a little modification to the rods lets you shoot them from a crossbow or compound bow and they go reallly far reallly quick. Some rods glow in the dark. Small note: there is a cable pulling gun "cable caster" i think and i personally dont like it, it gets tangled quickly and has a 50ft range, i can throw farther than that.

Cable Lube
conduit tubing can chafe the cable while youre pulling it, you can add lube to the cable or are where its chafing and it will ease the cable pulling and also ease the chafing.

GENERIC TOOLS YOU SHOULD HAVE
Screwdrivers:
Precision up to full size

Knives:
I like folding razor knives the best

Drill:
Mounting plywood, 66blocks, 110 blocks. Various uses.

Spare Cable and cross connect wire:
saved me so many times

SHARPIES/LABEL MAKER:
Label everything on both ends. this makes things so much easier. mark the box of cable and the end a few times. and then when you cut it form the box label that end. label your terminations and your patch panels, label your jacks. Not only is this organized its a lifesaver if someone has to troubleshoot later on.

Drywall saw
for cutting holes in drywall.

Fish tape
another wire pulling tool, can be invaluable

Flashlight
I carry two, a maglight for larger lighting situations and an LED hyperlight in my pouch, led lights create more of a white light which is good for reading fine print or while youre in a pitch dark basement trying to punch 3 ses of 25pair wire while youre holding it with your teeth....dont ask.

Cell Phone
Good for calling for backup or calling into a new system to test it out.

Zip Ties
I only use these for voice now. They tend to contract and expend with heat and cold and can choke wires over time. I hate having to cut them to change a wire arrangement in a closet, i really only tend to use them on ceiling runs. if you leave it really loose you can use them as a cable path for large groups of cable.

Roll of double sided velcro
while just a tad more expensive, it provides a cleaner look, easy serviceability for future work no cable choking unless you pull it too tight and makes it look like you give a damn about cable organization.

STUFF TO HAVE EXTRA OF:
Biscuit Jacks, Cat5 keystone (you can use for voice too), surface mount boxes, panduit, cables of many types, your most used tools, splitters etc....

I know I missed some stuff so please feel free to add on or comment on something that you think needs more info or whatever.

Im going to take pictures of some of my equipment (the important stuff at least) and integrate them into the thread when I get home.

-Kaos

PICS OF WHAT I USE IN THE FIELD 99% OF THE TIME:
Basically my tool box:
IMG_2893.jpg


My Toolbelt with Equipment in it:
IMG_2890.jpg


my linemans set "butt set"
IMG_2889.jpg


Each tool from the belt laid out
IMG_2891.jpg


more detailed pics of individual items can be taken if someone has a special request

ADDING WIRING DIAGRAMS :) Enjoi!
568-B Standard: Making a cat5(e) cord with this wiring on each end will create a "straight through" patch cable.
568-B.gif


568-A Often used for older avaya/lucent/bell phone technology. Making a cat5(e) cable with a 568-A on one end and a 568-B on the other creates a "cross over" cable
568-A.gif


New Wiring to Old wiring for phones:
jack-diagram.gif


Bell 25 pair wiring diagram for punching onto 66 blocks:
25pairdetailed.bmp


another one thats a little less cluttered for 25 pair wiring:
25_pair_color_code_chart.png


Will add more later!
 
A couple of things to add:

1) How about including things like E1, E3, X.21, SDH (I see now where this is mentioned in the SONET section), Etc. and how they relate to technologies here in North America for our friends everywhere else in the world since we are the only ones that use T1's, SONET, V.35, etc.

2) Include some of the test patterns that can be run with a bert (i.e., all 1's, all 0's, Quasi)

3) Maybe a section about traditional Frame and ATM service morphing into MPLS including a definition of MPLS. The difference between MPLS (point-to-multipoint) and traditional Frame/ATM (point-to-point).

4) SONET Rings: Include working and protect path, self healing architecture, etc.

5) Multiservice Access Switches: What are they? Who makes them? How do they operate? (I can write something up on this if you would like as I have been working on the Nortel Passport lineup for quite a few years.)
 
MPOE = Minimum Point of Entry, not Main Point of Entry.

One note about patch panels: Certain vendors (Panduit for example) use a proprietary punch down tool for their patch panels. A normal patch panel punchdown tool will not work on theirs.
 
PHUNBALL said:
A couple of things to add:


5) Multiservice Access Switches: What are they? Who makes them? How do they operate? (I can write something up on this if you would like as I have been working on the Nortel Passport lineup for quite a few years.)

which series?
 
SYN ACK said:
which series?

Mainly the 7k and 10k Passports with some DPN-100's thrown in there...

The last company I worked for had approximately 2000 switches deployed globally...
 
Thanks for all the sugggestions. I have never heard of MPOE reffered to as a minimum point of entry. I worked with CLEC's, LEC's and IXC's and never heard it used. I'm not saying your wrong, one thing about telecom is that acronyms seem to change depending on where in world you are ;)

As far as SS7 is concenred I know nothing about it other than its the "brains" of the PSTN. If someone wants to post some info about feel free.

T1 line encoding and framing. Excellent idea and I will be adding that shortly , hopefully with some links to pretty graphs and pictures :D

Regarding Frame/ATM I have very little experience with this technology and would be uncomfortable trying to write about them. If there is someone who understands the inner workings of this technology please post it. I for one would love to learn more about it.

Sonet Protecet paths / self healing - excellent idea. I will work on that as well.

Kaos - excellent post about some of the more common tools associate with telco. Well done.
Out of all those tools I would recommend a fishtape, pullstring, punchdown, and toner as the most important tools for cable pulling.

PHUNBALL - if you could write something up on MPLS and the Nortel Passports that would be great. These are two areas in which I currenlt have very little experience, but have agreat interest in.
 
All of this info is excellent! I like the part about the tools. One thing, be sure to label or mark your tools so they don't get mistaken for someone else's. I also have a couple adapter boxes I made up: RJ48S and RJ48X T1 loopback jacks and some adapters for toning out pairs. These are RJ11 plugs wired either to the 2 center or 2 outer pins, and the opposite pins on an RJ11 keystone. I also carry at least one RJ45 loopback adapter for testing switch/hub ports. Spare batteries are also very important. (for toner, tracer, butt set, flashlight). I have 3 or 4 mini maglites, 4 utility knives, about 10 pairs of pliers, at least 10-15 screwdrivers, and all the other necessary tools. Luckily I can keep mine at home, in a Craftsman tool chest. When I do a job, I put everything into a tool bag. I also swear by Plano adjustable plastic cases. I have several of these for all different parts: Installation parts like screws and plugs, 2 cases for audio connectors and splices, one for heatshrink, one for Neutrik Speakons, one for XLR and 1/4 inch connectors, and another with screws, bolts, nuts, and washers. I also have a nylon drawstring bag that holds all my conduit fittings. Wow too much stuff!
 
Kaos said:
pulling rods
We call these "fish sticks" at my work. I guess it's because they're the same idea as fish tape, but in stick form. Or perhaps because I work with some whackos. The tartar sauce jokes were unbearable for the first few weeks.

When pulling cable through conduit, one can attach a priceless diamond, a wad of newspaper, or a ping-pong ball to the end of the pullstring and either blow it through with a compressor or suck it through with a vacuum cleaner. Works nicely when you have to go around corners. I recommend against the diamond, it gets stuck easily.

Another nice thing to have, while I'm on the subject, is a pair (or more) walkie-talkies. We have the Motorola ones with like 3 miles (claimed) range. It can be a lot easier to pull cable/test existing cable if you don't have to yell all the way down the hallway or whatever. As an added bonus, you can practice sounding like a trucker or a fighter pilot. "Eagle, this is sparrow; we have a Whiskey Tango Foxtrot at this end."
"Sparrow, this is Eagle. I read you and roger that. Over and out."

 
What is an OC-3?

OC-3's are optical carrier circuits that can transmit 155.52Mb/s or 3 DS3's. There are several other OC levels. OC-12, OC-48, and OC-192. The transmission rate of each are listed below.

OC-12 = 622MB/sec
OC-48 = 2.49Gb/sec
C-192 = 10Gig/sec

What is SONET and how does it work?

SONET - Synchronous Optical Network. SONET is conceptually easy to grasp, but in practice is a royal bitch to learn.
SONET is an international standard (SDH) that allows for high bandwidth fiber lines to communicate across the world. SONET Rings (OC-192, OC-48 etc) generally connect CO's across great distances. Because fiber is not susceptible to attenuation fiber can be run extremely long distances without a repeater.(Called a regenerator in optical world) SONET is usually deployed in a ring fashion and has ADM (Add/drop mux) attached to the ring. ADM's allow for T1'T3's OC'3 to ride a fiber ring and them get "dropped" off at a CO closer to the location of the customer. Example

A T1 from Seattle to El Paso, TX. The T1 would get muxed into a DS3, then muxed into an OC-48. The OC-48 has ADM's in Seattle, Las Vegas, Phoenix and Austin. The DS3 rides the ring and in Austin gets "dropped" and is then demux from OC-48 - > DS3 -> T1 and then is piped into a DACS to go to a CO in El Paso where finally it is delivered to the customer over a copper pair.

You mention "SDH" What is this?

SDH - Synchronous Digital hierarchy. The new standard appeared first as SONET, drafted by Bellcore in the United States, and then went through revisions before it emerged in a new form compatible with the international SDH. Both SDH and SONET emerged between 1988 and 1992.
SONET is an ANSI standard; it can carry as payloads the North American PDH hierarchy of bit rates: 1.5/6/45 Mbps, plus 2 Mbps (known in the United States as E-1). SDH embraces most of SONET and is an international standard, but it is often regarded as a European standard because its suppliers—with one or two exceptions—carry only the ETSI–defined European PDH bit rates of 2/34/140 Mbps (8 Mbps is omitted from SDH). Both ETSI and ANSI have defined, detailed SDH/SONET feature options for use within their geographical spheres of influence. Basically while SDH is an international standard it is primarily used in europe due to manufacturers producing equipment only capable or E1/E3/STM-1. Thus the equipment would not be of much use here in the states. I should say i have had almost zero experience with SDH/STM (other than its SONET variant) and cannot comment on equipment or one being better than the other. SONET is based of a bit-interleved technology where as SDH using byte interleaving. So theoretically SDH is a superior technology.
For a much more in depth look at SDH and how it related to SONET check out this link

A more detailed look at SONET

STS-1 = This is the most basic building block of SONET. Operating at 51.84Mbps this is what OC-3, OC-12, OC-anything is built off of. An OC-48 will have 48 STS-1's flowing withing ring. STS-x are the electrical equivilant of OC-x. OC-1 is also 51.84 but uses light instead of electrical signals to communicate.

Confused yet? It may help to think of a SONET ring as a "tiered" communications medium. At the bottom you have Light or OC-x. At the next level up you have electrical or STS-1. Above that you have electrical DS3's or EC1's. and at the top you finally have T1's or VT1.5 (electrical)

Now let's apply this to a real world situation. A VT1.5(T1) starts at the top "tier" It then gets multiplexed into an EC1 (DS3) one "tier" below it. That EC1 then gets multiplexed into an STS-1(electrical), and then finally the STS-1 gets converted into light (OC-1) which then rides the fiber ring to the destination node and the process is reveresed OC-1->STS-1->EC1->VT1.5.

Why bother converting light into electricity?

Because there's equipment out there that will not "understand" light. For example a 1/0 DACS (1/0 means it takes in DS-1's and outputs DS-0's or visa versa) does not interface at an optical level. There's no point. A 1/0 dacs understands DS-1's and DS0's and thats it. Because SONET equipment can pipe a T1(optically via an OC-x) across the world and then connect (electrically) that T1 to the 1/0 DACS you have reduced overhead. Without SONET capabilities of speaking both optical and electrical signals you would be forced to pass this T1 via satellite (not as reliable and still costly) or the old fashioned way of passing the T1 through 20 or 30 different global telecom providers. Manufacturers are currently developing ways of switching light without having to convert the signal into an electrical one. In basic terms (which is the only way I can understand it :) ) They are using mirrors to "switch" light from one path to another, thereby removing the need to convert the signal into an electrical ones. This removes the current bottleneck of ligh/electrical conversion and basically makes for unlimited bandwidth. I have not kept up with this technology so if someone knows a bit more about this please let me know.

SONET sounds like a very robust system. But how reliable is it?

Oooooh I'm so glad you asked :) SONET is probably one of the most reliable systems available today. This comes from its "self-healing" nature. First a bit of "conduit and fiber" knowledge. When a fiber ring is engineered it is designed to use different fiber pairs that run through independant conduits. For example a typical node in an OC-48 system has 4 fiber interfaces.
1TX,1RX,2TX,2RX. Those 4 fiber cables are run in seperate conduits that all take different paths to get to the other node. That way if there is a fiber cut in conduit #1 it does not cut all 4 fibers but just one. Now this is all assuming the engineer who designed the fiber ring knew what he/she was doing.

In any telecom/datacom environement you typically have a working and protect. The assumption being, if the working circuit/device fails it will switch over to the protect side and not skip a beat.
SONET is no different. It is designed to have a working and protect path to prevent service interuption.

Now there are two different types of SONET rings. I am going to use OC-48 as an example.

1. UPSR - Unidirectional Path Switched Ring
UPSR network configurations are typically used in loop applications, where most of the traffic is routed to a single hub node. With UPSR, protected traffic is sent in either direction around a ring. The receiving node monitors the quality of both received signals and switches to the higher quality signal if there is a fiber cut, signal degradation or node failure. Switching is performed locally at each node, providing the fastest switch times.
Translation = ALL STS-1's flow around the ring in one direction. When a fiber cut is detected the ring "wraps"
Code:
Lets take a 4 node ring for an example.  
traffic flows 1->2->3->4->1->2->3->4 etc. 

                         1

                  2            4

                         3


Lets say there is a fiber cut between nodes 3 and 4.(denoted by the X)  
traffic now flows 1->2->3->2->1->4->1->2->3 etc.  
                                                  
                         1

                  2            4
                             X
                         3
You see Nodes 3 and 4 "wrap" to isolate the ring from the fiber break in between nodes 3 and 4.
The downside of UPSR is, if you get another fiber cut your screwed. And yes it can and has happened to me. UPSR is extremely fast when switching to protecect but is not a great setup when operating on long haul circuits. Thats where BLSR comes in.

2. BLSR - Bi Directional Line Switched Ring
Two of the four fibers are dedicated to working traffic, with the traffic traveling in opposite directions, while the remaining two fibers are for protection bandwidth, again with the traffic traveling in opposite directions. BLSR rings are much more complicated than UPSR's but as a result are much more versatile. For a very thorough description of BLSR check out Cisco's site They have diagrams and pictures that can describe BLSR much better than I can.
 
Flagg said:
Thanks for all the sugggestions. I have never heard of MPOE reffered to as a minimum point of entry. I worked with CLEC's, LEC's and IXC's and never heard it used. I'm not saying your wrong, one thing about telecom is that acronyms seem to change depending on where in world you are ;)

According to: http://www.webopedia.com/TERM/M/MPOE.html - I can't see it being 'Main Point of Entry'. In a multi-tenant building you can have multiple MPOEs. Because of that it is no longer a 'main point of entry' into the building. Google for it as well. Every reference to MPOE has it as 'minimum'. Even SBC refers to it as minimum point of entry.

Good thread none-the-less.. would be nice to see it stickied.
 
unhappy_mage said:
Another nice thing to have, while I'm on the subject, is a pair (or more) walkie-talkies. We have the Motorola ones with like 3 miles (claimed) range. It can be a lot easier to pull cable/test existing cable if you don't have to yell all the way down the hallway or whatever. As an added bonus, you can practice sounding like a trucker or a fighter pilot. "Eagle, this is sparrow; we have a Whiskey Tango Foxtrot at this end."
"Sparrow, this is Eagle. I read you and roger that. Over and out."
on a side note.....don't do this when a) you happen to work somewhere where every staff person has radios...including the boss. he will get curious. or b) when your girlfriends dad IS your boss. also a bad thing.

BobSutan, thanks for stickying....
Flagg, rock on for a good thread full of good info
 
i spent tuition on this stuff and dont remember none of it.

thanks for the refresher course.
 
shade91 said:
According to: http://www.webopedia.com/TERM/M/MPOE.html - I can't see it being 'Main Point of Entry'. In a multi-tenant building you can have multiple MPOEs. Because of that it is no longer a 'main point of entry' into the building. Google for it as well. Every reference to MPOE has it as 'minimum'. Even SBC refers to it as minimum point of entry.

Good thread none-the-less.. would be nice to see it stickied.
Well I'll be damned. I stand corrected :) Thread updated
 
Also with regards to walkie-talkies, in some CO environment any radio based transmission equipment i.e. cell phones, walkie talkies etc are not allowed. The reason being, is that most of the equipment in a CO is "wire wrapped" for a connection. If a wire is left un-connected it basically acts like an antenae and the cell phone/radio signals can actually interfere with equipment. Just an FYI
 
Excellent diagrams kaos. two acronyms that have helped me remember my colors codes are

Primaries - Why Run Backwards You Varment
Secondaries - Boy On Girl Brings Satisfaction (my favorite :D)

Primary Colors - White Red Black Yellow Violet
Secondaries - Blue Orange Green Brown Slate (grey )
 
Flagg -

I PMed you with some more info. I don't know if you wanted to add it to one of your posts or if i should just post it myself.
 
I tested this today in a drop ceiling environment for pulling cable and it is the best method i have used thus far for drop ceilings...

wrist secured sling shot + tie nut on end of pull string = it will go 50-60ft and I didnt tanlge it at all today. Slingshot and nuts are now in my toolbox.
 
Phone Cable Info

DISCLAIMER: I am a more or less amateur that has picked up a few tidbits of information in his short IT career. Feel free to challenge/correct anything I've posted.

Almost all home and some business phone systems use plain "patch" style phone cable, commonly called satin cable. This is the silver, black, or white oval-shaped cable that connects your phone to the wall. It has one or two pairs of very thin wire that are usually stranded to make the cable easy to handle. This cable is very flexible but the pairs of wire are usually not twisted so it is susceptible to interference and attenuation over long runs. The wires in these cables have four colors: green, red, black, and yellow. Red and green carry the primary line and yellow and black carry the secondary line.

The next step up is called Cat 3, or Category 3, cable. The specifications for this cable aren’t that important but you can think of it as a stripped-down predecessor to the Cat 5 we use for Ethernet today. I think it started when Ethernet first migrated from using coax cable. It usually consists of two pairs of solid core wire that are twisted together. This allows the cable to be run over longer distances and with less interference than regular patch cable. This cable can also be found in 1- and 3- pair configurations. Standard use of this cable dictates that it should never be connected directly to a device on either end but rather to a keystone jack or patch panel and a patch cable is used to complete the connection. Wires in a two-pair cable have four colors: blue, orange, white-blue, and white-orange. Blue and white-blue are used to carry the primary line and orange and white-orange are used for the secondary line. Prior to 1980, before the Cat 3 spec was widely accepted, most of these runs were made with “station cable” which was just one or two pairs of wire in a round grey vinyl (or sometimes even cloth) jacket.

A growing trend in newer homes today is the use of Cat 5 (Category 5) Ethernet cable for phone lines. If you’re lucky your builder used different colored cable for phone runs, or at least labeled them differently. This technique is no harder than using Cat 3 and allows for greater expansion in the future. It is also possible (but not suggested nor supported) to run 10/100 Ethernet and one or two phone lines on the same cable (since 100BaseT only uses two of the four pairs in the cable), but this isn’t endorsed by the IEEE specification. Again, connectors to end equipment should never really be used directly on this cable.

Telephone companies have gotten really efficient with their use of twisted pair cables. It is not uncommon these days for all incoming lines to be punched down to a standard 110 or screw-terminal block and then one large 25-pair cable (or more) to be run to another location in a building to be split into smaller and smaller cables. Even if 25 lines aren’t utilized, it makes things much easier for the phone company later. If a client wants another phone line on the 14th floor and the Demarc is in the basement, the telco tech simply connects the new line to a pair in the 25-pair cable in the basement and then runs a small patch from wherever the cable gets split out on the 14th floor to the client’s office/desk. This idea is also used on a larger scale with runs from the CO. Rather than run 5 million copper pairs from the CO to each house, the phone company runs a few fiber optic lines to a box near the middle of the neighborhood and then splits copper from there. The technology involved with this is beyond the scope of this FAQ, but it is efficient for sure. SEE: Muxing/Demuxing

Most telephone applications utilize RJ-11 connectors on the end of the cable going to the equipment. These are similar to RJ-45 (Ethernet) connectors, just smaller and with fewer terminals. When looking at the terminals on the connector with the toggle down, the primary line is carried on the two center terminals and the secondary line is on the outside.

When connecting patch cable to Cat 3 or Cat 5 cable, use this scheme:
Code:
Cat 3/5           Patch
----------        -----
Wh-Or     <->     Black
Blue      <->     Red
Wh-Bl     <->     Green
Or        <->     Yellow
 
In the industry, we refer to the phone cords as silver satin line cords. Using the patch cable terminology could confuse people with data/Ethernet.
 
Crashsector said:
Phone Cable Info

A growing trend in newer homes today is the use of Cat 5 (Category 5) Ethernet cable for phone lines. If you’re lucky your builder used different colored cable for phone runs, or at least labeled them differently. This technique is no harder than using Cat 3 and allows for greater expansion in the future. It is also possible (but not suggested nor supported) to run 10/100 Ethernet and one or two phone lines on the same cable (since 100BaseT only uses two of the four pairs in the cable), but this isn’t endorsed by the IEEE specification. Again, connectors to end equipment should never really be used directly on this cable.


never ever ever ever ever ever ever run a voice line on a network cable's spair pair(s).

Phone lines run so much more voltage than network traffic that the risk of interference would be alot higher than youd think. Ive seen voice circuits right off the demarc carry more than 140 volts at times...96volts more common...around 40 being the norm. I dont have the exacts at the moment but you get the point. Reccomended distance between another power source is 6 inches for anything over 24volts for data networks. put it inside the same jacket and you're asking for trouble.

enter unapproved spec Cat7 cabling: each pair is shielded and the entire cable is shielded as well...shouldnt be an issue although i havnt tried it yet because well....cat7 is exactly readily available.
 
Good info everyone who added to the faq. If anyone else has any questions that you think could be added please don't hesitate to ask.
 
Im going to try this on my own but I figured id ask first...when using lan cable testers will a loopback plug effectively act as a "remote" for the lan tester? is there a special way that the remotes that come with them are wired? Id love to take some spare keystone jacks and mod plugs so I can test a whole closet without going back and forth.
 
Crashsector said:
[cat 3] usually consists of two pairs of solid core wire that are twisted together.
Cat 3 is usually 4 pairs, in my experience; I haven't seen any 2-pair.

Also, gigabit ethernet may use all 4 pairs.

 
unhappy_mage said:
Cat 3 is usually 4 pairs, in my experience; I haven't seen any 2-pair.

Also, gigabit ethernet may use all 4 pairs.


Actually, Cat 3 is usually 3 pairs and you can get it in 2, 3, or 4 pair configurations. I'm sitting right next to a spool of 2 pair that I use here at work.

I know gigabit uses all 4 pairs, that's what I said. 10/100 doesn't (usually).

 
WORKING WITH FIBER

make sure that you do not poke yourself with the exposed fiber once you strip the jacket off or cleave it. The fiber is small enough to fit into your pores and will disolve into your bloodstream, then you will have to go to the ER which is not fun. Ruber gloves (the thin kind) can help prevent against this. Just be extremely careful and youll be okay. Dont forget to wear eye protection as well and ive been stripping the jacket and saw the fiber shoot right out of the strippers.

Most guys use a ball of tape to keep their cleaved fiber remainsfrom getting everywhere, but I like to use a pop bottle or something to that extent.

safety first!
 
great info

I am in network operations and deal with this stuff alot. I admit I am a data guy and know a little bit of this. Thanks for the great info. My next paycheck is coming your way. :p
 
Back
Top