the book of next gen networks

8/8/2019 The Book of Next Gen Networks

1/164

The Book on

The essential inormation you need to knowwhen deploying FTTX, rom the central ofce to

the outside plant to the customer premises

Foreword by Jason MeyersManaging Director, Penton Custom MediaPenton Media is the publisher o TelephonyMagazine

The eagerly awaitedollow-up to ADCs

The Book on FTTX


2/164

www.adc.com

The Book on Next Generation Networksii


3/164

www.adc.com

iiiThe Book on Next Generation Networks

Foreword

The Problem with Innovation

By Jason Meyers, Managing Director, Penton Custom MediaPenton Media is the publisher o TelephonyMagazine

The above is a title most people probably would not expect to see on a

oreword to a book about next generation networks. But there is a reason

behind it and a point to it, both o which I will get to in a moment.

First, though, what is that problem? What could be problematic about

innovationin particular, about the network technology innovation that

drives communication networks into the next generation, driven by the

need and demand or advanced services and increasingly ubiquitous and

continuous and instantaneous communications capabilities?

The problem can be summed up in two words: expectation and execution.

Innovation creates expectation in droves. Industries like telecom live and

die by the expectation that is created by innovation. Companies get put

on the map because o it. Whole market sectors are created around that

innovation and the accompanying marketing buzz it generates. Its electric.

Industry associations and alliances are ormed around those expectations.

The media thrives on the expectation and multiplies it. (Some might say its

the medias ault.) Promises are made.


4/164

www.adc.com

iv Foreword

Then comes the executionor lack o it. This is where the rubber hits the

road (or skids o into the ditch). Its one thing to make promises, to build up

expectations. Its another to deliver on the expectations created, regardless

o how technologically promising the innovation may be. Those markets

and buzz created by the expectation? Without proper execution, they are

more than likely to zzle.

So the problem with innovation, quite simply, is one o ollow-through. The

problem is an inadequate attention to the detail required to turn innovationinto a market.

So why did I choose this phrase as a title to the oreword o The Book on

Next Gen Networks? Because I contend that this book goes a long way

toward solving the problem. This is a book about executionnamely, the

execution required to leverage next generation network innovation and use

it to build markets.

How does one volume accomplish that which whole market sectors have

at times tried and ailed to accomplish? By concentrating on the details.

This book doesnt speak in broad strokes about what various technologies

can potentially accomplish, the services they can potentially enable or how

competitively important it is to deploy those technologies in your networks.

Instead, this book is a practical exploration and application o specics.

The Book on Next Gen Networks goes deep, into the central oce, to thedistribution hub, the access network and into the customer premises. It

explores, or example, why a proper ber cable management system is so

critical to network perormancenot only right now, but also in the not-

so-distant uture, when todays will be carrying applications no one has

yet thought o, and expanding because o it. Or where (and why) splitters

should be deployed in a PON environment, and how a decision like that can

help a network accommodate new services.


5/164

www.adc.com

vThe Book on Next Generation Networks

This book analyzes the perormance and cost issues that can occur i the

wrong moves are made, and the benets that can be realized by making

the right ones. To that end, this is a book about preparing or the uture.

In act, it attemptsas much as is possible in this ever-adapting network

environmentto actually predict the uture: What could the long-term

consequences o a deployment decision or process be? How will the role o

the network technicians who deploy the networks evolve, and what training

will be required o them? How will new construction and the changing

architecture o buildings impact how FTTP will be deployed?

The Book on Next Gen Networks is conceived and written to help those

who consume it bridge the gap between expectation and execution. Read

it, apply it, repeat it. Industry associations and alliances and alliances are

ormed around that expectation. It will help you deliver on the promise

o innovation.

Enjoy!


6/164

www.adc.com

The Book on Next Generation Networksvi


7/164

www.adc.com

viiThe Book on Next Generation Networks

Table o Contents

Introduction: The Motivation or GPON Migration ........................................... 3

Central Oice

Chapter 1 The Elements o Fiber Cable Management ................................. 11

Chapter 2 Eective Integration o Reduced Bend Radius Fiberinto the Network ........................................................................19

Chapter 3 Incorporating Passive CWDM Technology vs. DeployingAdditional Optical Fiber ..............................................................25

Chapter 4 Adding New Video Services Warrants New CentralOice Considerations .................................................................31

Distribution

Chapter 5 Its Happening in the Hub ........................................................... 39

Chapter 6 Extreme-Environment Perormance Considerationsor FTTX Splitter Modules ...........................................................51

Chapter 7 Plug and Play Splitter Architectures Drive Operational Savings .... 61

Chapter 8 The Economics o FTTN vs. FTTP ................................................. 65

Chapter 9 Resectionalizing the Distribution Area ..........................................71

Access

Chapter 10 Creating a Cost-Eective Plug and Play FTTX Architecture ..........79

Chapter 11 Innovative Installation Techniques or Fiber Drop Terminals .........83

Chapter 12 Above vs. Below Ground Drop Splicing: Considerationsor Drop Cable Connections in the FTTX Network ......................89

Chapter 13 Outside Plant Connections You Can Rely On .............................. 93


8/164

www.adc.com

viii Table o Contents

Customer Premises

Chapter 14 Multiple Solutions or Connecting MultipleDwelling Units (MDUs) .............................................................. 107

Chapter 15 Deploying Reduced Bend Radius Fiber in MDU Environments...119

The Technician

Chapter 16 Properly Training Next-Generation Technicianson Next-Generation Products ....................................................127

Chapter 17 The Technicians Perspectiveon Reduced Bend Radius Fiber .................................................131

Glossary .................................................................................................137


9/164

The Book on Next Generation Networks

Introduction


10/164

www.adc.com

The Book on Next Generation Networks2


11/164

www.adc.com

3The Book on Next Generation Networks

IntroductionThe Motivation or GPON Migration

By December 2007, approximately eight million homes had been passed

with ber or Fiber-to-the-Home (FTTH) or Fiber-to-the-Premises (FTTP)

applications. Included in these numbers are an astonishing ve hundredcommunities that have chosen ber as a means o delivering broadband

applications to homes and businesses. O these numbers, it is estimated

that almost hal, or around 3.5-million o these homes and businesses are

connected using Broadband Passive Optical Networking (BPON), Ethernet

Passive Optical Networking (EPON) or Ethernet-in-the-First-Mile (EFM)1.

Predicting the telecom uture is never easyand it ollows that building

an access network that is uture-prooed against rising bandwidthdemand and next-generation technologies is a major challenge or todays

service providers. But that doesnt mean decisions have to be based on

a coin fip either. There are many practical considerations that can be

examined when selecting an FTTP inrastructure that will not only meet

current demand, but also provide the fexibility or a smooth migration to

next-generation demands.

This is particularly true o the passive optical network (PON) portion o

the network. A close look at several practical considerations, based on

inormed decision making, will provide a rm oundation or designing

a network that can cost-eectively transition between legacy and uture

access technologies. Our own telecommunication history provides many

troubling examples o networks that were built without giving thought

to uture innovation. Building telephone networks with copper, our

predecessors could not have predicted todays broadband revolutioneventhough we seem to have made the most o this legacy inrastructure with

xDSL technologies.


12/164

www.adc.com

4 Introduction

However, through the unpredictable perormance o xDSL and the overall

condition o the legacy copper networknot to mention some very costly

lessons learnedservice providers have realized the importance o network

fexibility. FTTP oers service providers a clean slate or deploying todays

new services to their bandwidth-hungry subscribersand it all begins with

designing the proper PON architecture.

For the access protocols and the movement to Gigabit PON (GPON)

migration, some additional concepts may need to be considered: GPON is the next generation of PON electronics currently being

introduced to the marketplace.

GPON will NOT be the nal technology deployed.

The network design should accommodate exibility for the

current migration and beyond.

In theory, the passive connectivity infrastructure must be

agnostic to the service delivery technology.

GPON is making it easier or PON networks to move to an all-IP ormat

where the external interaces to the core are moving to an all Gigabit

Ethernet network creating a movement away rom the traditional ATM

transport to a pure IP transport. GPON is IP-centric while allowing the

traditional services o voice and video, yet acknowledges the strengthso the service provider to dierentiate themselves on quality o service

(QoS) issues.


13/164

www.adc.com


GPON continues to have the long reach that eectively eliminates active

components in the access network with little or no signicant changes to

the physical architecture that has already been built or BPON and EPON.

Architecture designs should account or a smooth transition between

technologies by accommodating practical considerations or uture

architectures. We do not have a true crystal ball as to what these technologies

will become. I we did, we would simply build or the uture. However, isnt

this exactly what we should be doingbuilding or the uture?

Wheres the motivation?

As predicted, GPON, a culmination o the best in BPON and EPON, is poised

to dominate the access market by oering a much-needed bandwidth boost.

We can all agree that eventually everythingvoice, video, and datawill

be moving to IP and the quadruple-play applications, including network

appliances, security, video surveillance, etc. The advantages o GPON are a

key driver or gaining the commitment o the large-volume carriers toward

the GPON standard.

GPON is emerging on queue with higher split ratios that can deal with

the challenges o delivering high-speed, high-bandwidth packaged services

to business and residential customers. This is putting pressure on service

providers to make decisions or ramping up their networks or GPON rom

the central oce (CO) to the outside plant (OSP).

Ensuring FTTP networks can easily migrate to GPON promises to pay huge

dividends to service providers in the coming years. As GPON develops

as the standard o choice or FTTP networks, both cost reductions and

interoperability will be accelerated. Those providers who make inormed

choices in deploying fexible, interoperable, recongurable networks will

reap substantial benets in the move to GPON and beyond. They will be

able to quickly oer new and improved services as they evolve, without theneed or major network overhauls.


14/164

www.adc.com

6 Introduction

Standards bodies

I service providers arent already convinced by GPONs ability to provide

uture enhanced services, maximize interoperability, utilize enhancement

bands, and provide increased capacity with the promise o higher split

ratios, the International Telecommunication Union (ITU) provides urther

motivation. The ITU points out that we can expect a signicant increase in

demand or dedicated Gigabit Ethernet (GigE) and 10GigE services to both

businesses and residential customers.

This means every service provider must decide how to best integrate all

types o services onto a single backhaul ber network. A smooth and easy

migration capability to GPON is the most viable solution. GPON enables

PON networks to easily move to an all-IP ormat while external interaces to

the core move to an all-gigabit ethernet ormata movement away rom

the traditional ATM transport to pure IP transport.

The ITUs ratication o the GPON standard in 2003 has also helped putelectronics vendors on the same page in terms o getting behind one

standard. This standard will enable the major cost challenges associated with

optical network terminals (ONTs) at the customer premise to be addressed

and, in time, will bring those costs down signicantly.

GPON combines the best o BPONs quality-o-service attributes with the

best o EPONs ability to transport and interace on an all-IP network. It

also addresses the higher application bandwidth needs by providing

2.4 Gbits/sec downstream and 1.2 Gbits/sec upstream.

The transition to GPON

Making the move rom BPON or EPON to GPON involves three key

architectural components. Addressing the bers loss characteristics in

terms o spectral attenuation, using the appropriate class o optics, andconsidering the advantages oered by greater split ratio capability will all


15/164

www.adc.com


aect the networks migration to GPON. Each o these considerations will

be addressed in greater detail within this book.

Connectorization also plays a role in creating a migration-ready FTTP

network, particularly when considering the single ber requirements o

next-generation video applications in GPON architectures. The use o APC

connectors that oer the lowest return loss characteristics o all current

connectors will optimize high bandwidth and allow or longer reach.

Splitter conguration in the optical distribution portion o the network

between customers and the COhas been a hot topic over the last ew

years. We believe a centralized splitter approach oers the best fexibility

advantages. It maximizes the eciency o OLT PON ports, and unlike the

cascaded approach, does not risk stranding unused ports in areas o low

take rates. There will also be urther advantages when it comes to testing

and troubleshooting the network.

With the GPON standard already revolving around centralized 1x32 splitter

architectures in the OSP, GPONs promise o a 1x64 splitter ratio oers

even more incentive to service providers by doubling the number o homes

serviced rom a single splitter.

Moving to the CO, fexibility becomes the pathway to easy migration

capability. A network must always be built as a fexible long-term entity

that adapts to inevitable changes in both equipment and technology. Across-connect network oers excellent fexibility or conguration points

and should include high-quality APC connectors or handling the higher

power necessary or any analog video application.

Cable management in the CO is also an issue worth consideration. In act,

the considerations or GPON within the CO can be summed up in just three

wordsfexibility, quality, and protection.


16/164

www.adc.com

8 Introduction

A fnal word

Weve covered a lot o ground in a short time, but these and other topics

are covered in greater detail as you read through this book. Suce it to

say that network architects owe it to themselves to careully plan ahead to

avoid having to re-build the network to accommodate each new application

or technology.

Summing it all up, the inevitable need to migrate to GPON is already upon

us, and the uture generations o PON are already on the drawing board.Making inormed network decisions today will not only make a migration

process less painul, but it is also good business sense. GPON not only

supports TDM voice today, it has a true migration platorm to an all-IP

network. But most importantly, it guarantees that existing architectures will

migrate to uture technologies without requiring orklit upgrades.

I hope youll see this latest edition o The Book on Next Generation

Networks as a tool or helping you make good decisions or upgradingyour access network. It represents the experience and know-how o many

ne architects, planners, and design technicians. I wish you the best o luck

in meeting the unique challenges o your network and hope youll consider

our ADC team as you work towards making your network plans a reality.

Enjoy!

Patrick J. Sims, RCDDPrincipal Systems Engineer, [email protected]

1. Source: RVA LLC, Market Research & Consulting, Fiber to the Home: Advanced

Broadband 2007


17/164

Central Oice



18/164

www.adc.com



19/164

www.adc.com


Chapter 1The Elements o Cable Management

As service providers continue upgrading their networks to transport

high-bandwidth broadband services, an increase in ber usage is essential

to meet both bandwidth and cost requirements. But just deploying this ad-ditional ber is not enougha successul, well-built network must also be

based on a strong ber cable management system.

Proper ber management has a direct impact on the networks reliabil-

ity, perormance, and cost. Additionally, it aects network maintenance,

operations, expansion, restoration, and the rapid implementation o new

services. A strong ber cable management system provides bend radius

protection, cable routing paths, cable accessibility, and physical protectiono the ber network. Executing these concepts correctly will enable the

network to realize its ull competitive potential.

Introduction

With demand steadily increasing or broadband services that will include

several bandwidth-hungry technologies like high-denition television(HDTV) and higher Internet speeds or handling larger le sharing re-

quirements, ber is being pushed closer and closer to the customer

premises. This, in turn, creates a need or both additional ber in the

central oce /data center and the active equipment that must be managed

to accommodate uture network growth.


20/164

www.adc.com

12 Central Ofce

Any new broadband network inrastructure must have the inherent

capability to easily migrate to the next generation o technologies and

services. This is a key consideration or service providers beginning to

deploy triple-play broadband serviceswhether its rom a multiple

service operator (MSO) headend, a central oce (CO), or wireless mobile

switching center (MSC). As the amount o ber dramatically increases,

the importance o properly managing the ber cables becomes a more cru-

cial issue.

The manner in which ber cables are connected, terminated, routed, spliced,

stored, and handled will directly and substantially impact the networks per-

ormance and, more importantly, its protability. New technologies and

products have been developed in the last ew years to improve bend radius

protection, cable routing paths, accessibility, and physical protection.

Bend radius protection

There are two types o bends in bermicrobends and macrobendsthat

can aect the ber networks long-term reliability and perormance.

The microbend is a small, microscopic bend that may be caused by the

cabling process itsel, packaging, installation, or mechanical stress due to

water in the cable during repeated reeze and thaw cycles. External orc-

es are also a source o microbends. An external orce deorms the cabledjacket surrounding the ber, but causes only a small bend in the ber. A

microbend typically changes the path that propagating modes take, result-

ing in loss rom increased attenuation as low-order modes become coupled

with high-order modes that are naturally lossy.

A macrobend is a larger cable bend that can be seen with the unaided eye

and is oten reversible. As the macrobend occurs, the radius can become

too small and allow light to escape the core and enter the cladding. Theresult is insertion loss at best and, in worse cases, the signal is decreased


21/164

www.adc.com


or completely lost. Both microbends and macrobends can, however, be re-

duced and even prevented through proper ber handling and routing.

The minimum bend radius will vary depending on the specic ber cable.

However, in general, the minimum bend radius o a ber should not be less

than ten times its outer diameter. Thus, a 3 mm cable should not have any

bends less than 30 mm in radius. Telcordia recommends a minimum 38 mm

bend radius or 3 mm patch cords. Also, i a tensile load is applied to a ber

cable, such as the weight o a cable in a long vertical run or a cable pulledtightly between two points, the minimum bend radius is increased due to

the added stress.

The advent o bend insensitive or reduced bend radius ber is an example

o how technology has addressed the bend radius issue. Whereas the mini-

mum bend radius should not be less than ten times the outer diameter

o the ber cable in typical ber, reduced bend radius ber provides more

leeway. However, service providers must understand that these new bersdo not diminish the need or solid ber cable management. On the con-

trary, the increase in the sheer number o bers being added to the system

to accommodate broadband upgrades makes bend radius protection as

important as ever.

As bers are added on top o installed bers, macrobends can be induced

on the installed bers i they are routed over an unprotected bend. A ber

that had been working ne or many years can suddenly have an increasedlevel o attenuation, as well as a potentially shorter service lie. The impor-

tance o bend radius protection is critical to avoid operational problems in

the network.


22/164

www.adc.com

14 Central Ofce

Cable routing paths

The second element o ber cable management is cable routing paths and

is related to bend radius protection. Improper routing o bers by techni-

cians is one o the major causes o bend radius violations. Wherever ber

is used, routing paths must be clearly dened and easy to ollowto the

point where the technician has no other option than to route the cables

properly. Leaving cable routing to the technicians imagination leads to an

inconsistently routed, dicult-to-manage ber network.

The quality o the cable routing paths, particularly within a ber distribution

rame system, can be the dierence between congested chaos and neatly

placed, easily accessible patch cords. Its oten said that the best teacher

in ber routing techniques is the rst technician to route it properly. Con-

versely, the worst teacher is the rst to use improper techniques, since sub-

sequent technicians are likely to ollow his lead.

Well-dened routing paths, thereore, reduce technician training time, in-crease the uniormity o the work done, and ensure and maintain bend

radius requirements at all points, thus improving overall network reliability.

It is important to note that, again, the use o bend insensitive ber does not

diminish the need or clear cable routing pathsthere are benets that go

beyond bend radius protection.

Having dened routing paths makes accessing individual bers easier,

quicker, and saerreducing the time required or recongurations. Fi-

ber twists are reduced to make tracing a particular ber or rerouting

much easier. Even with new technologies, such as the use o LEDs at both

ends o patch cords or easy identication, well-dened cable routing

paths still greatly reduce the time required to route and reroute patch

cords. All o this directly aects network operating costs and the time re-

quired to turn up or restore service.


23/164

www.adc.com


Cable access

Cable access is the third element to good ber cable management and

reers to the accessibility o the installed bers. As the number o bers

increases dramatically in both the distribution rame and the active equip-

ment, cable access becomes an increasingly important issue or broadband

service providers. In the past, an active equipment rack might have had

about 50 bers exiting, and managing those bers was much less o an is-

sue. But as that same rack is tted or next generation broadband services,

there may be up to 500 bers involved, making proper management andaccessibility a vitally important matter.

With huge amounts o dataas well as revenuemoving across those -

bers, the ability or technicians to have quick and easy access is critical.

When there are service level agreements in place, particularly or customers

with high priority trac, the last thing any service provider wants is service

interruptions caused by mishandling one ber to gain access to another.

As previously mentioned, there are patch cords designed today with LEDs at

both ends to help technicians identiy particular cable runs with no chance

o error. These innovations can be implemented into a good cable man-

agement system to help minimize problems caused by disconnecting the

wrong patch cord. There are many other tools and techniques or ensuring

that every ber can be installed or removed without bending or disturbing

an adjacent ber.

The accessibility o the bers in the ber cable management system can

mean the dierence between a network reconguration time o 20

minutes per ber and one o over 90 minutes per ber. Since accessibility

is most critical during network reconguration operations, proper cable ac-

cess directly impacts operational costs and network reliability.


24/164

www.adc.com

16 Central Ofce

Physical fber protection

The last element o a ber cable management system addresses the physi-

cal protection o the installed bers. Every ber throughout the network

must be protected against accidental damage by technicians or equipment.

Fibers traversing rom one piece o equipment to another must be routed

with physical protection in mind, such as using raceway systems that pro-

tect rom outside disturbances.

Without proper physical protection, bers are susceptible to damage thatcan critically aect network reliability. The ber cable management system

should always include attention to ensuring every ber is protected rom

physical damage.

A fnal wordplanning

Finally, since many service providers are in the processor soon will beoupgrading networks or delivering high-bandwidth broadband services, it

is important to stress the need or planning in terms o cable management.

Todays network is a living and growing entityand what is enough today

will almost certainly be too little tomorrow. With that in mind, uture-proo-

ing the network wherever possible should be a major considerationand

ber cable management is no dierent.

For example, the current upgrades to broadband service delivery takingplace in COs, MSOs, or MSCs require more ber deployment. Four- and

six-inch ber raceway systems are already becoming inadequate to properly

manage these larger amounts o ber. Service providers must plan ahead

or a centralized, high-density ber distribution rame lineup using 24-inch

raceways that can accommodate not only todays ber requirements, but

also those expected in the uture.

Although installing a 24-inch raceway system is more expensive today, hav-ing to go back in and retrot the system in a ew years represents a much

higher cost and signicant risk to the ber. Ignoring uture growth, particu-


25/164

www.adc.com


larly in terms o ber, will result in higher long-term operational costs result-

ing rom poor network perormance or a requirement to retrot products

that can no longer accommodate network demand.

Another consideration in planning or good ber cable management con-

cerns the active equipment rack. Most manuacturers have traditionally

overlooked the need or providing cable management within their equip-

ment. Beore purchasing, service providers should insist that cable manage-

ment is included within every piece o active equipment to ensure theirinvestment will operate at peak eciency over time.

All our elements o a ber cable management systembend radius protec-

tion, cable routing paths, cable access, and physical protectionstrengthen

the networks reliability and unctionality while lowering operational costs

and ensuring smooth upgrades when necessary.


26/164

www.adc.com



27/164

www.adc.com


Chapter 2Eective Integration o Reduced Bend Radius Fiber into

the Network

Introduction

Bending o singlemode ber has everyone talking these days. The idea

that you can bend a ber around a pencil without a dramatic increase

in attenuation is a concept that has everyone considering new ber

applications and design possibilities.

Today, industry standards or traditional singlemode ber typically speciy

a minimum bend radius o ten times the outside diameter o the jacketedcable or 1.5-inches (38 mm), whichever is greater. This new breed o fex-

ible singlemode optical ber has the potential to signicantly reduce these

minimum bend radius requirements to values as low as 0.6-inches (15 mm),

depending on the cable conguration, without increasing attenuation.

There are many names or optical ber that can endure a tighter bend

radius bend insensitive, bend resistant and bend optimized are sever-

al that come to mind. However, some o these terms can be somewhatmisleading. Designers and installers may believe reduced bend radius

optical ber is impervious to all the orces that can increase attenuation

and cause ailure on an optical ber link. Sta and contract technicians can

make alse assumptions on its durability and perormance capabilities as

well. Such belies can have a serious impact on network perormance.


28/164

www.adc.com

20 Central Ofce

For purposes o accuracy, ADC uses the term reduced bend radius, be-

cause this title best describes what the product actually delivers. As with

any optical ber, attention must be paid to how the cable is deployed

and handled throughout the lietime o the network, in order to ensure

optimal perormance.

What is reduced bend radius optical fber?

As mentioned above, reduced bend radius ber is able to withstand tight-er bends within rames, panels, and pathways. To understand how this is

achieved, it is important to understand that all ber types rely on principles

o Total Internal Refection, which allows light signal to travel rom one

end o the ber to another (see Figure 1). By improving the bend radius

o optical ber, light entering the core is eectively refected by the clad-

ding back into the core. Instead o using a matched clad prole, some con-

structions o reduced bend radius optical ber use a depressed clad prole

with a lower index o reraction than the core, causing light to stay within

this core.

n1

n2

Reracted

Reected

Cladding

Core

Figure 1Principle o Total Internal Reection or Optical Fibers

Fiber cladding has a lower Index o Reraction (IOR) than the core,causing light to stay within the core. Depression o the cladding

profle promotes Total Internal Reection

To achieve tighter bend radii, some constructions change the mode eld

diameter (MFD)the area across the core o the ber that lls with light.

Typical MFD or standard singlemode optical ber is about 10.4m; reducedbend radius optical ber may exhibit MFD o between 8.9m and 10.3m.


29/164

www.adc.com


Regardless o the type o construction, all reduced bend radius ber prod-

ucts do one thing very wellthey can perorm under a tighter bend radius

where macrobends occur. Examples include a central oce application,

where ber passes rom a panel into a vertical cable route or in an FTTX

deployment within the connes o an optical network terminal (ONT).

The bers perormance is denitely impressive. For example, in ADC tests

a standard singlemode optical ber with one turn around a 1.26-inch

(32 mm) diameter mandrel shows induced attenuation o less than 0.50 dBat 1550 nm. This same test on a reduced bend radius singlemode 1550 nm

optical ber shows less than 0.02 dB o attenuation.

In general, reduced bend radius optical ber is designed to perorm with

low loss across the spectrum o wavelengths, rom 1285 nm to

1650 nm, using all the channels available on those wavelengths to

maximize bandwidth. Current designs include low water peak or zero

water peak so that high attenuation is avoided at 1383 nm. Many re-duced bend radius optical ber products meet ITU-T Recommendation

G.657, meaning they work well at 1550 nm or long distance and voice

applications and at 1625 nm or video applications.

Does it improve perormance?

Despite the improved bend radius, the reality o this ber is that bend ra-dius protection is still a concernjust not to the extent that it is in standard

ber. There is still a mechanical limit on how tightly any optical ber can

be routed beore the structural integrity o the glass is violated.

The assumptions about improved perormance are not accurate either, at

least beyond the exceptional bend radius perormance. In reality, the peror-

mance o reduced bend radius optical beror any optical berdepends

upon many actors, not just bend radius properties.


30/164

www.adc.com

22 Central Ofce

By itsel, reduced bend radius optical ber does not oer improvements in

attenuation. True, it bends more tightly without causing additional attenu-

ation. Yet laid out on a long, straight run next to a standard optical ber,

there is no dierence in perormance that can be attributed to the cables

construction. It is inaccurate to believe that reduced bend radius optical

ber is the end-all solution when, in act, there are many other actors that

determine optical ber link perormance.

Durability Reduced bend radius optical ber oers the same crush resis-tance and tensile strength as the same cable with standard singlemode -

ber. As with standard optical ber, excessive weight will crush reduced bend

radius optical ber and excessive pulling tension will damage the cable,

both o which aect attenuation.

Connector pull-o resistance Cable assemblies and connectors must

meet Telcordia (GR-326) requirements or strength o the ber termination

connector. Reduced bend radius optical ber does not improve connectorpull-o resistance. Connectors that are easily loosened or disconnected in-

crease attenuation and cause ailures.

Connector perormance When it comes to connector perormance,

endace characteristics determines loss rom the connector. Reduced

bend radius optical ber does not impact insertion loss rom connectors,

making termination and quality o connectors an important consideration

in link perormance.

Proper applications or reduced bend radius optical fber

Singlemode reduced bend radius optical ber oers benets or

applications that including the central oce, FTTX deployments, data cen-

ter, and OEM solutions. Singlemode reduced bend radius optical ber is

best suited or environments where little or no bend radius protection isavailable. It is also ideal or applications where space is an issue. Specic ap-

plications that make sense or this type o ber include places in which:


31/164

www.adc.com


Space is tight For drop cable or termination o pigtails in multiple dwell-

ing unit (MDU) and optical network terminal (ONT) boxes or FTTX deploy-

mentswhere there is no space and oten no cable managementreduced

bend radius optical ber oers less chance o increased attenuation during

eld installation and maintenance.

No fber management is available The ront o rames and routers

where moves/adds/changes occuris ideal or use o reduced bend ra-

dius patch cords and multiber breakout assemblies. Many OEM activecomponents do not have bend radius limiters or protection on the ront

o the equipment.

Space is at a premium Patch cords and multiber breakout assemblies

that can bend more tightly enable increasing density o active equipment

in racks and cabinets without sacricing access. For manuacturers o ac-

tive equipment, reduced bend radius optical ber can help reduce size

o electronics, improving density and airfow. However, in theseapplications, even more consideration must be paid to the elements

o proper cable management. Tighter bend radius also oers OEMs the

chance to increase the unctionality o active equipment by utilizing less

chassis space.

O course, a key advantage o reduced bend radius optical ber is use in

high bandwidth applications. For standard optical ber, the 1625 nm to

1550 nm wavelengths are the rst to go when the cable is wrapped arounda mandrel. Preserving these wavelengths around tighter bends oers ben-

ets or OEMs seeking to improve unctionality o network equipment or

network managers looking or the eciency o having all wavelengths

available on a given optical link.


32/164

www.adc.com

24 Central Ofce

Conclusion

Singlemode reduced bend radius optical ber has generated quite

a buzz, and it is a great step orward in optical ber construction. It makes

much-handled patch cords and multiber assemblies less susceptible

to macrobends that aect attenuation and limit bandwidth o optical

ber links.

It is crucial or the health and perormance o the network to be aware that

reduced bend radius ber does not, in any case, mean that the undamen-tals o proper ber management are to be ignored. In act, as this ber is

used in higher density applications, actors such as connector access and

cable routing paths become even more crucial. Reduced bend radius optical

ber is just one aspect o a complete strategy or ecient, uture-prooed

network management.


33/164

www.adc.com


Chapter 3Incorporating Passive CWDM Technology vs. Deploying

Additional Optical Fiber

The recent advancement in telecommunication applications or voice,

video and data places additional demands on ber optic networks.Adding additional ber to existing networks can be very costly to service

providers. In most cases, a ar betterand less costlyoption is ound in

coarse wavelength division multiplexing (CWDM) technology.

CWDM technology adds greater ber bandwidth while increasing the fex-

ibility, accessibility, adaptability, manageability and protection o the net-

work or applications up to 60 km.

What is CWDM?

CWDM can be viewed as a third generation o WDM technology. WDM

was developed as a ber exhaust solution and traditionally employed the

1310 nm and 1550 nm wavelength signals. In most WDM scenarios,

providers with a xed number o bers had run short o bandwidth due

to rapid growth and/or unoreseen demand. By multiplexing a signalon top o the existing 1310 nm wavelength, they could create additional

channels through a single ber to increase the networks capacity.

However, demand continued to increase dramatically with new inno-

vations and applications such as the internet, text messaging and other

high bandwidth requirements. This created the need or very ne channel

spacing to add even more wavelengths or channels to each ber.

Dense WDM (DWDM) was a major breakthrough as equipment provid-


34/164

www.adc.com

26 Central Ofce

ers pushed to oer new equipment, promising nearly unlimited bandwidth

potential. However, while DWDM was quickly adopted or long-haul and

transoceanic optical networking, its use in regional, metropolitan, and cam-

pus environments was, in most cases, cost prohibitive.

A more targeted and cost-eective solution ollowed with CWDM, a

more recent standard o channel spacing developed by the International

Telecommunication Union (ITU) organization in 2002. This standard calls

or a 20 nm channel spacing grid using wavelengths between 1270 nmand 1610 nm (see Figure 1). The cost o deploying CWDM architectures

today is signicantly lower than its DWDM predecessors.

Prior to ITU standardization, CWDM was airly generic and meant a

number o things. For instance, the act that the choice o channel spac-

ing and requency stability was such that erbium-doped ber ampliers

(EDFAs) could not be used was a common thread. One typical denition

or CWDM was two or more signals multiplexed onto a single ber, onein the 1550 nm band and the other in the 1310 nm bandbasically, the

original denition or early WDM.

1200 1300

O-band1260-1360

E-band1360-1460

Wavelength (nm)

Fiberattenuation(dB/km)

S-band1460-1530

C-band1530-1565

L-band1565-1625

1400 1500 1600

2

1.5

1

0.5

0ITU-T G.652 fiber

Waterpeak

1270 1290 1310 1330 1350 13701390

14101430 1450 1470 1490 1510 1530 1550 1570 1590

1610

Figure 1: CWDM wavelength grid as specifed by ITU-T G.694.2Todays standardized CWDM is better defned as a cost-eective solution

or building a metropolitan access network that promises all the keycharacteristics o a network architecture service providers dream

aboutoering transparency, scalability, and low cost.


35/164

www.adc.com


New developments

Even though the ITUs 20 nm channel spacing oers 20 wavelengths

or CWDM, the reality is that wavelengths below 1470 nm are con-

sidered unusable on older G.625 specication bers due to the in-

creased attenuation in the 1310-1470 nm bands. However, new bers

that conorm to the G.652.C and G.652.D standards, such as

Corning SMF-28e and Samsung Widepass, nearly eliminate the wa-

ter peak attenuation peak to allow or ull operation o all ITU CWDM

channels in metropolitan and regional networks.

This enables a CWDM system to operate eectively at the low end o

the ITU grid where attenuation was problematic or earlier bers. For

example, an Ethernet LX-4 physical layer uses a CWDM consisting

o our wavelengths near the 1310 nm wavelength, each carrying

a 3.125 Gbits/second data stream. Together, the our wavelengths can

carry 10 Gbits/second o aggregated data across a single ber.

As mentioned earlier, a major characteristic o the recent ITU CWDM

standard is that the signals are not spaced appropriately or amplication

by EDFAs. This limits the total CWDM optical span to somewhere near

60 km o reach or a 2.5 Gbits/second signal. However, this distance is

suitable or use in metropolitan applications. The relaxed optical requen-

cy stabilization requirements also allow the associated costs o CWDM

to approach those o non-WDM optical components.

Basic implementation

As stated earlier, CWDMs appeal is rmly rooted in meeting

the additional demands being placed on ber networks by a steady

stream o new, bandwidth-hungry applications. Adding more ber is

one solution, but there are many possible obstacles that will likely

make this solution cost prohibitive. Although every situation isdierent and brings unique considerations to the table, nearly any

ber deployment includes rights-o-way, trenching costs, additional equip-

ment, manpower, and considerable time.


36/164

www.adc.com

28 Central Ofce

Market studies have indicated accrued costs between $10,000 and $70,000

per mile to deploy new ber cable. The large disparity is due to dierent

situationsor example, it costs ar more to tear up a city street than to

simply trench ber in a rural setting. But the key issue is that network archi-

tects can incorporate a CWDM system or much less cost and still achieve

the bandwidth increases necessary to meet demand today and well into the

oreseeable uture.

Basically, a CWDM implementation involves placing passive devices, trans-mitters, and receivers at each end o the network segment. CWDM per-

orms two unctions. First, they lter the light to ensure only the desired

combination o wavelengths is used. The second unction involves multi-

plexing and demultiplexing the signal across a single ber link. In the multi-

plex operation, the multiple wavelength bands are combined onto a single

ber or transport. In the demultiplex operation, the multiple wavelength

bands are separated rom the single ber to multiple outputs. (See Figures

2 and 3)

ADCs passive network solution adds value by using the value-added

module (VAM) platorm to multiplex and demultiplex. These VAMs can

easily be incorporated into central oce (CO), multiple service operator

(MSO), and mobile switching center (MSC) environments or leveraging

the benets o CWDM. The MSC uses CWDM to multiplex the dierent

hosts on a wireless coverage system to multiple remotes using minimal

ber strands. Even a single ber can service our, six, or eight dierent re-mote units. From there, an antenna is attached to each device to enable

indoor wireless coverage.


37/164

www.adc.com


Figure 2: CWDMs in useFor example, MSOs can install a band system at the head-

end that will drop one wavelength to each node along a particular ring confgura-tion. This ring can be utilized as a single fber. Each CWDM device is packaged intothe VAM platormconnectorized and labeledor integration into the fber panelor cross-connect to save oor space and eliminate extra patch cords.

Designated, dedicated wavelengths

CWDM also oers the benet o individual wavelengths or allocat-

ing specic unctions and applications. Out-o-band testing capability isachieved by simply dedicating a separate wavelength or channel or nonin-

trusive testing and monitoring. In act, any number o dierent applications

can be applied to specic wavelengths. For example, a particular wave-

length might be allocated specically or running overhead or management

sotware systems.

This is a common practice in using CWDM or cable television net-

works, where dierent wavelengths are dedicated or downstream andupstream signals. It should be noted that the downstream and up-

stream wavelengths are usually widely separated. For instance, the


38/164

www.adc.com

30 Central Ofce

downstream signal might be at 1310 nm while the upstream signal is

at 1550 nm. Another recent development in CWDM is the creation o

small-orm-actor pluggable (SFP) transceivers that use standardized

CWDM wavelengths. These devices enable a nearly seamless upgrade

in even legacy systems that support SFP interaces, making the migration to

CWDM more cost eective than ever beore. A legacy system is easily con-

verted to allow wavelength multiplexed transport over one ber by simply

choosing specic transceiver wavelengths, combined with an inexpensive

passive optical multiplexing device.

Conclusion

ADC views the emergence o CWDM as the most cost-eective means o

moving ever-increasing amounts o inormation across metropolitan access

networks. For most providers, deploying new ber as a means o combat-

ing ber exhaust is not a viable option. There are too many high costs

involved with trenching the ber cable, and obtaining rights-o-way can be

an intensely complex issue.

CWDM simply makes sense, particularly with the technological

advancements in todays ber and transceiver options, including VAM sys-

tems. CWDM achieves the critical goals o transparency, scalability, and low

cost that providers seek in todays highly competitive industryan industry

where new applications and increasing demand dictate the pace or mod-ern telecommunication networks.


39/164

www.adc.com


Chapter 4Adding New Video Services Warrants New Central

Oce Considerations

Although its air to say the distribution and access elements within the

outside plant (OSP) portion o the Fiber-to-the-Premises (FTTP) network de-mand the majority o attention during deployment, its still important not to

overlook implications to the central oce (CO). Any FTTP network requires

the same fexibility as the transport networkand it all begins in the CO.

The addition o video services to FTTP network presents challenges to the

CO requiring special consideration.

First, a review

Beore discussing the unique challenges o video, its important to briefy

review the overall implications that FTTP has on the CO architectureand

the importance o making inormed decisions in the early stages. The goal

o network planners is always to minimize capital expenses and long-term

operational expenses, while achieving the highest possible level o fexibility

in the network.

Architectural decisions involve connection strategies between optical line

terminal (OLT) equipment and OSP bers, fexibility in terms o test access

points, and WDM positioning. A key requirement or providing fexibility

evolves rom ensuring ull cross-connect capability. With all OLTs, as well

as OSP bers, connected at the optical distribution rame (ODF), easy ac-

cess and signicant long-term network fexibility is achieved, enabling easy

adds, moves, and changes to the ODF. Since the one constant in telecom-

munications has always been change, any assumption that the network

will remain static can result in signicant long-term capital expense and

fexibility issues.


40/164

www.adc.com

32 Central Ofce

The second critical architectural decision involves placement o the video

WDM within the CO environment. The video WDM combines the voice

and data signals with video signals onto a single bera key element

o FTTP deployment. Again, with expense and fexibility in mind, ADC

concludes that placing the video WDM in the cross-connect ODF lineup

is the best option.

This is done by using patch cords to connect the OLT equipment to the

inputs o the video WDM. A cross-connect patch cord connects the videoWDM common port to the designated OSP port, providing an immedi-

ate advantage o requiring just three connector pairs while still maintaining

maximum fexibility. With the video WDM located at the ODF and all OLT

patch cords routed directly to the ODF, even greater fexibility is provided

regarding how the OLTs are combined and congured. Any OLT is easily

combined with any other OLT, regardless o CO location.

Factoring in the video

The addition o video signals now presents new challenges to the con-

guration o the CO in order to maintain the same fexibility and price

points desired in deploying FTTP. The video overlay onto the FTTP net-

work adds additional ber cable management requirements. Also, in or-

der to split the video eed to multiple PONs, additional optical splitting is

necessary. Optical path protection switches are also incorporated wherethe video signal enters the service oce rom the video serving oce.

From the video OLT, video signals will pass through several erbium-doped

ber ampliers (EDFAs) used to ampliy and split the signal. Each EDFA

output will be urther split by additional optical splitters to maximize the

video output, allowing the most PONs to be served using the ewest

number o EDFAs. Each EDFA can have up to our outputs, each with its

own optical splitter, depending on signal strength.


41/164

www.adc.com


The use o optical splitters is critical, but there are several placement

options. For instance, the splitters could reside in either the OLT equip-

ment rame or the ber rame. Placing the optical splitter in the ber rame

enables even more fexibility. For instance, i a particular PON is located

a considerable distance away, a stronger video signal would be required

and the signal should not be split. By having the optical splitter in the

ber rame, a patch cord can be run rom the EDFA to the ber rame,

thus bypassing the optical splitter and allowing a stronger video signal

to go to that PON. This fexibility allows video signals o various powerlevels to reach PONs at various distances. These optical splitters would

reside in the ber rame in a chassis very close to the WDM chassis on the

1550 nm input side.

Assuming the oce providing the video service is not the same oce in

which the video signal originates, optical protection switching is also a

consideration. Through diverse path routing, both a primary and protect

video eed enters the optical protection switch in the video OLT equip-ment rame. The primary video eed throughputs to the video OLT, but

should that signal drop below a preset power threshold, the system

automatically switches to the redundant path (or protect) video eed.

The diverse path routing takes place at the transmission side where

a 1x2 splitter creates two diverse signals. This basically provides

SONET-like protection without all the electronics by using a splitter and

an optical switchmuch more cost eective.

Several important cable management considerations that apply in general

to the FTTP network architecture will apply to a very great extent when it

comes to video signals. Since video signals are usually high-power analog,

they require considerations or the use o angled polish connectors, con-

nector-cleaning techniques, and other cable management practices that

contribute to signal quality.


42/164

www.adc.com

34 Central Ofce

Every network designer wants to get the most out o existing electronics.

In FTTP, that equates to getting the most PONs served and achieving the

highest network fexibility or the least amount o expense. But the con-

stantly-changing network still requires everyone to not only peer into the

uture, but to also design todays FTTP networks with the ability to adapt

to the uture.

Test access or the utureTesting the FTTP network is a serious challenge or service providers. Ad-

vanced ODF solutions are being adopted that enable remote test and

monitoring unctionality. With traditional ODF unctionality, perorming

tests or troubleshooting problems requires breaking into a patch and

basically taking the network out o service. But monitoring and testing ca-

pabilities can be incorporated into advanced ODF solutions that will enable

remote monitoring and trac identication, as well as reduce troubleshoot-

ing and ault isolation time. The net result is more eciency, reliability, and

cost savings.

By placing an optical NxN switch between the test equipment and the

access port on the bers, any ber can be tested with any test equip-

ment rom the network operations center (NOC). For example, i contact

is lost with several optical network terminals (ONTs), an optical time do-

main refectometer (OTDR) trace can be perormed over the particular berto isolate the ault. Perormance monitoring tests can also be

accomplished without having to dispatch a technician to the rame to man-

ually perorm testing.

Built-in diagnostics can identiy problems within the electronic equipment,

but to see whats happening within the ber requires specic test equip-

ment and non-intrusive access points. In any FTTP network, its a point-

to-point connection rom the OLT to the customer. I there is a ailure inthat network, the customer is out o servicethere is no redundant path

available. Thereore, the ability to restore the network quickly and easily is


43/164

www.adc.com


absolutely critical. The addition o this single switch provides technicians

with quick, easy, and reliable access to the networkall o which greatly

reduces network outage time and saves money.

Designing the CO to accommodate FTTP requires similar, i not more strin-

gent, cable management and architectural attributes as any transport net-

work. The video overlay makes even more demands on the CO in terms o

eciency, fexibility, and accessibility. Decisions made by service providers

today will signicantly impact the uture reliabilityand protabilityotheir FTTP network. But with careul planning, uture-proong the CO is a

good way to begin.


44/164

www.adc.com



45/164


Distribution


46/164

www.adc.com



47/164

www.adc.com


Chapter 5Its Happening in the Hub

The Fiber Distribution Hub (FDH) continues to play a vital role in supporting

rapid deployment and connection in Fiber-to-the-Premises (FTTP) networks.

Innovation in FDH design occurs at a rapid rate and next generation ea-tures appear in newer FDH enclosures. Key innovations include:

Miniaturized splitter modules with plug-in installation that allow

easy additions and upgrades

High-density termination elds with connectorized harnesses

allowing modular growth and fexible rearrangement

A wide range of sizes and mounting congurations that retaincrat-riendly ber management and maintenance eatures

Performance enhancements to optical connectors and splitters due

to the rigorous requirements o independent testing o all optical

components and enclosures

Time- and space-saving parking lots providing cross-connect function-

ality at interconnect loss and space levels

As a result, FDH products have been widely accepted in FTTP networks.

FTTP is now seeing large-scale deployment and FTTP deployment is de-

nitely still happening at the hub.


48/164

www.adc.com

40 Distribution

Network architectures

Fiber-to-the-Business

ONT

FDHCO/HE

OLT

Optical Distribution Network Fiber-to-the-Home

Fiber-to-theMulti-Dwelling Unit

Ater years o research and experimentation with access networks,

many network providers have settled on passive optical network (PON)

architectures as the direction or uture subscriber access. The PON ar-

chitecture has been adopted as a standard in ITU-T G.983.x that denesthe protocols, data rates, and operating wavelengths necessary to sup-

port network services. At the same time, the standards have established

power budgets and parameters or the ber optic plant to ensure reliable

transport all the way to the home. The technology o high-speed PON

equipment, combined with broadband ber oers the potential or con-

necting high bandwidth services directly to the home. The standards ensure

interoperability o equipment and thereore have driven down the cost o

deploying all optical networks. When adding in the cost savings associatedwith operating an all-passive optical plant, PON networks are attractive or

overbuild as well as new network construction.

The initiative to build PON networks is oten reerred to as Fiber-to-the-

Premises (FTTP), to emphasize the vision o connecting ber rom the

central oce/headend (CO/HE) all the way to the premises. PON

architecture includes optical line terminal (OLT) equipment at the CO/HE

that bundles voice and data services. OLT equipment utilizes wavelength


49/164

www.adc.com


division multiplexing (WDM) technology to provide bidirectional voice and

data services (1310 nm/1490 nm) over a single ber. Additional WDM

components at the CO/HE allow integration o video services onto the

same ber at the 1550 nm wavelength.

OLT equipment ports are connected through optical splitters, allowing a

single port to serve multiple subscribers. The split ratio in PON networks

can vary, but typically networks are planned with 32- or 16-way splits. The

architecture may be congured by concatenating the splitters at a singlepoint. Most networks are planned with 1x32 splitters centrally located or

easy access or additions, service, and maintenance.

PON architecture includes optical network terminal (ONT) equipment at

the premises or resolution o voice, data, and video services. Standardiza-

tion o ONT equipment allows the same equipment to provide services or

Fiber-to-the-Home (FTTH), Fiber-to-the-Business (FTTB), and Fiber-to-

Multiple-Dwelling Units (MDU) applications. Combining these applicationsinto the FTTP network architecture provides economies o scale or con-

struction and service deployment.

The optical distribution network provides physical connection between

the CO/HE and the premises and includes various cabling segments

including eeder, distribution, and drop. These various segments are typi-

cally joined together by connectors and splices. The ber distribution hub

(FDH) is one o the key elements located between the eeder anddistribution segments and contains optical connectors and splitters to pro-

vide easy access and fexibility. The advantage o conguring the network

with connectors is to allow fexibility or service provisioning and or net-

work testing.


50/164

www.adc.com

42 Distribution

FDH network unctionFDH Pad and Pole

Central Oce/Headend

Underground Distribution

Aerial Distribution

The FDH is a key interace between eeder cables extending rom the cen-

tral oce to distribution bers routed to subscribers. The FDH serves an

analogous unction to serving area cabinets (SAC) used in copper-based

networks to interconnect the eeder and distribution segments o the net-

work. The hub becomes a primary point o fexibility in the network to con-

nect subscriber circuits. As service is required, technicians access the FDH

enclosure to route connections to complete subscriber circuits. The FDHalso serves as a central location or ber optic splitters. This is where the

PON network diers signicantly rom a copper network.

The optical splitters allow the PON OLT port to be shared among multiple

subscribers via the 1xn split, thus deraying the cost o the OLT. By locating

the splitters in the outside plant close to the serving area, the cost o eeder

ber is also signicantly reduced. For instance, when a 1x32 splitter is placed

in the FDH, one eeder ber may be routed into a neighborhood and provideservice connection to 32 subscribers. Another reason to locate splitters in the

FDH is that splitters can be deerred until they are needed to satisy service

requirements. The FDH can be accessed to add splitters as service demands

grow. Newer hub designs accept modular splitters that quickly plug into the

FDH to allow capacity to be expanded within a ew minutes.

Typically, the FDH is equipped with one stub cable that is spliced into a

eeder cable and another stub cable that is spliced to a distribution cable.Construction is usually completed using standard splicing techniques (usu-

ally mass splicing) with splices stored in standard splice closures. Some FDHs

are even equipped to handle the splicing inside the cabinet.


51/164

www.adc.com


Key FDH capabilities and innovations

The FDH enclosure provides a crucial crat interace in the outside plant

environment. Thereore each major unction o the hub supports easy crat

access or service and maintenance.

Fiber Management

Termination Splice Shel and Trays

Splitter Sheland Modules

Termination feld

The termination eld provides a location or terminating ber distribution

cable on optical connectors and adapters. The termination eld is sized to

support the number o subscribers located in the distribution serving area

downstream rom the FDH. FDH enclosures support a range o termination

eld sizes.

The termination eld provides easy access to both sides o the adapt-

er to acilitate cleaning and maintenance. ADC FDH enclosures eature

a unique swing rame design, a hinged chassis containing all the key

optical components including splitters, connectors, and splices. The de-

sign allows easy access to optical components rom the ront and rear or

cleaning and troubleshooting and is especially valuable in installations

where access is limited to the ront o the cabinet only, or example, in pole

mounted applications. Large cabinets deployed in ground mount applica-tions eature doors on the ront and rear to allow ull access to connectors

and splitters rom the ront and back.


52/164

www.adc.com

44 Distribution

Terminations in the eld are clearly marked to provide accurate identica-

tion o each subscriber termination. The termination eld provides organi-

zation and protection or ber jumper connections as they transition into

the ber management section o the enclosure.

Recent FDH innovations include high-density component packaging

resulting in signicant reduction o enclosure sizes. High-density termi-

nation elds with connectorized harnesses allow modular growth and

fexible arrangements.

High-density termination Early FDH termination requirements were

oten matched exactly to the requirements or subtending living units in

the immediate ber serving area. For instance, a 216 ber hub was speci-

ed to support a ber serving area o approximately 200 subscribers, pro-

viding a small (approximately ve to ten percent) portion o spare bers

routed into the serving neighborhoods. With more experience, planners

realized that additional ber capacity downstream could be required orunoreseen changes in the network or in services supplied. However, while

speciying increased numbers o spare bers, resulting in increased ber

termination requirements, users were reluctant to increase the overall

size o the enclosures. Thereore, ber termination elds had to handle the

increased capacity within already dened enclosure sizes. This involved in-

creasing termination density and also increasing the ber handling capac-

ity or a particular enclosure. For example, enclosures previously handling

216 bers were upgraded to terminate 288 bers. This increase in densityprovides the desired ber counts along with the spare growth capacity re-

quired or typical ber serving areas, while maintaining the overall size o

the enclosure.

Modular, scalable distribution In overbuild scenarios, the termination

eld on the distribution side is ully populated with connectors at the initial

installation, and the enclosure is provided with ully-terminated stub cables

sized or the enclosures direct termination needs. Network planners, how-

ever, considering newer greeneld developments, look or ways to deer

cost and match the FTTP build to the pace o the developments build.


53/164

www.adc.com


A new development, constructed in phases over a period o years, may

not initially require an FDH with a ully-populated termination eld. This

situation may be better served by gradually deploying terminations

as needed. To satisy this requirement, the FDH enclosure includes modular

blocks that allow terminations to be added as required. The modular termi-

nation block allows upgrades to the FDH to match the requirements o the

FTTP network deployment, thus deerring hardware costs.

Improved overall perormance Advances in planar splitter technologyhave dramatically decreased the amount o signal loss when a single ber

is split into several outputs. Innovation in component perormance has re-

sulted in lower loss connections, in both the termination elds and the split-

ters. Improved connector perormance or the widely used SC components,

allows connectorization to replace splicing on both eeder and distribution

bers while still meeting the overall loss limits within the FDH. Using con-

nectorization or input bers and distribution panels greatly reduces the

amount o time required to install and upgrade an FDH.

Splitter feld

Splitter modules are designed to snap-in to the splitter eld and can be

added as required by service demands. The splitter eld protects, organizes,

and routes both the input and output bers. The optical splitter modules

provide up to 32 connectorized pigtail outputs and one pigtail input.

Early generations o FDH were deployed ully loaded with splitter modules

that eatured storage ports, sometimes reerred to as parking lots, located

on the ront o the module to stage splitter output pigtails temporarily until

they were connected into service. The splitter module assembly included

modular parking adapters, each holding 16 or 32 connectors. As a split-

ter module was installed, the bers were ed into the ber management

trough and the parking adapters were snapped into place in the parkingarea. Individual connectors were then easily separated rom the parking

adapter and routed to the termination eld during service turn-up.


54/164

www.adc.com

46 Distribution

Recently, the parking lots have been relocated to a spot in the FDH away

rom the splitter modules. The parking adapters are removed rom the split-

ter module, allowing the splitter module to be reduced in size.

Today, most carriers take an incremental approach to adding splitter mod-

ulesdeploying FDH enclosures initially with just the splitter modules re-

quired to begin service connections. This reduces the number o parking

lots required or pigtail outputs. In essence, splitter outputs time share

parking lots; as the outputs o the initial splitter modules are placed intoservice, the parking lots associated with those outputs become available or

parking subsequent splitter module outputs This allows a signicant reduc-

tion in the size o the parking lot, and consequently, a reduction in the size

o the FDH.

Blind-mate connections New miniaturized splitter modules eature

planar optical splitters and are 75 percent smaller, another contributing

actor in the reduction o the FDHs size. Additionally, innovation has im-proved the way splitter modules are installed into the enclosure. First gen-

eration modules were designed with the splitter module input extended as

a pigtail, which was spliced to eeder bers. As each subsequent splitter

was installed, it was spliced to eeder bers staged in splice trays. Splic-

ing consumes valuable time, and adds costs to service turn-up. Earlier

improvements included connectors on the eeder bers that allow quick

connection during splitter module installation, or a connector on the pig-

tailed input and a connector on the eeder bers mated at a connectorpanel in the enclosure. This approach provides a simple, much improved

method or quickly installing splitters. Connectorization o the eeder -

bers at the FDH also allows testing on the eeder rom the FDH i required.

However, connectorization o the eeder ber also raised a saety concern

regarding high power when analog video is transmitted over the path. To

address this concern, connectors can be angled or adapters with shutters

provided to prevent a technician rom accidentally looking into the high-

powered termination.


55/164

www.adc.com


Further innovations have resulted in a backplane connector system or in-

stalling splitter modules. In this conguration, eeder bers are terminated

with a standard connector pre-positioned on the backplane to receive a

plug-in splitter module with a mating connector. The backplane connector

is shuttered or saety so that a technician cannot accidentally look into an

unmated splitter module. As a splitter module is inserted into the backplane

receptacle, the module presses open the shutter to allow the splitter mod-

ule connector to mate with the backplane connector. This blind-mate ap-

proach using a common backplane technology improves eciency in utureexpansion activities.

Splice area

The FDH eatures a splice area to connect eeder bers or other cables

routed into the enclosure. One use or this area is the splicing o addition-

al splitter modules to eeder bers as the modules are added to the FDH

enclosure. An alternative to splicing the input is to include a connector at

this location.

Factory pretermination FDH enclosures typically include two pretermi-

nated stub cables. One stub cable is pre-connected to the optical splitter

module input so that it can be eld-spliced to the eeder cable. The other

stub cable is pre-connected to the termination eld, so that it can be eld-

spliced to the distribution cable. These cables attach to the enclosure usingstandard grip clamps and liquid-tight compression ttings seal the cables

at the enclosure entrance. Orientation o the enclosure stub cables varies,

depending on the FDHs mounting method.

Crat-riendly fber management

The FDH provides total ber management using a unique ront acing

cross-connect design. The ront ber management allows splitter module

outputs to be routed and staged within the enclosure or ecient connec-

tion into service at a later date.


56/164

www.adc.com

48 Distribution

Vertical channels using storage loops manage excess ber slack. The entire

cabinet can be interconnected without congestion. Connectorized pigtail

ends are stored on bulkhead adapters on the ront o the module so that

connector ends can be identied quickly and connected into service. Fiber

strain relie and radius control is provided through the enclosure.

Indoor confgurations

As FTTH moves into densely populated areas, the use o indoor berdistribution hubs becomes popular due to the number o units within

a particular building, as well as space restrictions outside the buildings.

Indoor FDHs provide all the same eatures as an outdoor FDH, but are

typically smaller and lighter. They do not need to meet the same harsh

environmental requirements as the outdoor FDHs. Fiber count capac-

ity ranges rom 72 bers to 432 bers, accommodating small to large

high-density buildings.

Below-grade confgurations

Another option or high-density areas, as well as areas that do not allow

above ground enclosures or zoning reasons, are below-grade ber distri-

bution hubs. These compact enclosures are stored in below-grade vaults

when not being accessed or service congurations.

Qualifcation

A complete FDH qualication program draws rom a wide array o exist-

ing standardized tests with existing procedures. In some cases, new test

procedures have been developed and rened to support the new con-

gurations and new technologies. The overall program is composed pri-

marily o testing regiments drawn rom Telcordia Generic Requirements.


57/164

www.adc.com


First and oremost, the qualication program involves testing optical con-

nectors to GR-326-CORE, Issue 3. All connectors utilized in the FDH en-

closure are subject to the complete outdoor service lie requirements

and to the ull spectrum o long-term reliability tests. In addition to testing

at 1310 nm and 1550 nm as required in GR-326, the test programs include

additional test wavelengths o 1490 nm and 1625 nm to assure users that

all operating wavelengths and all potential maintenance channels would

unction under the harshest conditions.

Optical splitters are ully tested to ensure trouble ree perormance over

the lie o the network. The splitters use planar technology and ollow a

qualication program aligned with service lie testing in GR-1209-CORE

and long-term reliability testing in GR-1221-CORE. Because o the nature

o testing very large devices (1x32 ports), special sampling techniques

were developed or optical measurement characteristics such as directiv-

ity. Splitter qualication is conducted at the ull operation spectrum o

our wavelengths including 1310, 1490, 1550 and 1625 nm. All testingis done in the ormat o the optical module that plugs into the FDH enclo-

sure, representing the exact conguration deployed in the eld. Tests or

the new enclosures include a ull range o environmental and mechanical

tests. Optical characterization is conducted at the same our wavelengths

as the connectors and splitters. Additionally, several o the tests such as

thermal cycling and seismic qualication are optically monitored during

the test at 1625 nm, which represents the worst-case scenario rom a

ber integrity perspective.

Independent testing o the qualication program demonstrated the FDHs

reliability, assuring a perormance level and longevity expected in an FTTP

network. Successul testing o all aspects o the enclosures, including per-

ormance o optical connectors and splitters, have given users the evidence

and condence to support wide scale deployment o FDH enclosures in the

distribution portion o FTTP networks.


58/164

www.adc.com

50 Distribution


59/164

www.adc.com


Chapter 6Extreme-Environment Perormance Considerations or FTTX

Splitter Modules

Optical splitter modules used in FTTX networks contain the splitters

that make passive optical networks possible. The module physically pro-tects the splitter and provides a means to connectorize the splitter inputs

and outputs.

Figure 1: Typical FTTX Splitter Module

Module housing(1xN splitter inside)

Bending Strain Relie

Input

Connectors

2 mm Furcation tube

A housing, constructed o plastic or metal, holds the splitter and provides a

means to up-jacket the splitter bers with 2mm urcation tube or connec-

torization. A certain number o outputs are connectorized. The input ber

may be connectorized, can be a pigtail, or can be attached to the module

by means o a backplane.


60/164

www.adc.com

52 Distribution

Industry standards

Telcordia GR-1209 and GR-1221 standards dene the operating

requirements or splitter modules in North America. GR-1209 denes ba-

sic optical perormance requirements such as insertion and return loss,

polarization-dependent loss (PDL), and uniormity. GR-1209 also de-

nes short-term environmental and mechanical requirements such as

input and output proo strength and side loading, and a temperature

and humidity prole. GR-1221 denes the splitter modules long-term

reliability requirements. GR-1221 requires splitters to go through2,000 hours o high temperature aging, low-temperature aging,

thermal cycling, and humidity aging. GR-1221 also subjects samples

to impact and vibration testing.

The operating extremes dened in GR-1209 and GR-1221 are -40C

to +85C and up to 95% relative humidity. GR-1209 and GR-1221 will

typically be called out by North American service providers deploying

passive optical networks. Some service providers may require their networkto unction at lower temperatures. In these cases, military specications

(MIL SPECs) requiring -55C minimum operating temperatures may be

called out.

These operating extremes present challenges when designing split-

ter modules. Beore large-scale North American deployment o FTTX in

2004, most modules containing splitters and connectors were used in

central oces. Splitter modules saw stable environments and were there-

ore not extensively tested. Testing to extreme conditions and deploy-

ment in outside plant environments orced service providers and equip-

ment manuacturers to re-evaluate the requirements o splitter modules.

GR-1209 and GR-1221 do not consider many characteristics that

are important or devices deployed in the OSP. For example,

GR-1209 and GR-1221 do not dene material properties such as

chemical resistance or installation considerations such as the handlingo urcation tubes at extreme temperatures.


61/164

www.adc.com


Furcation tubing

Furcation tubing is the material slipped over the splitter inputs and outputs.

The urcation tube protects the ber rom physical damage and makes con-

nectorization possible. The urcation tube is usually identical in construction

to a 2mm simplex jumper, but the .900mm tight buered ber is replaced

by a hollow tube. The hollow tube has a .900mm outside diameter and the

inside diameter is larger so that a ber can be inserted. Once the ber is

inserted into the inner tube, a connector can be terminated to the ends.

2 mm Outer Jacket

Inner .900 mm Tube

Aramid StrengthMembers

Splitter Input andOutput Fibers

Inserted Into This Space

Figure 2: Furcation Tube Construction

2mm simplex jumpers are typica

the book of next gen networks

Documents