CFOT Fiber Optics FOA CERTIFICATION.pptx

Editor's Notes

• #1: This is a basic fiber optic training program for FOA-Approved Schools to teach classes for the FOA CFOT certification. The program was developed by The Fiber Optic Association with inputs from many FOA instructors and technical advisors. It’s is based on their 30+ years of experience in the business, including training thousands of fiber optic technicians.
• #2: This is a complete training program covering the basics of fiber optics. You may edit it to include in training programs, including removing this slide which is for the instructor only. Usage of this PPT and other FOA provided training materials is covered by the license agreement. -The FOA licenses this program to instructors for their use teaching a CFOT course -It can be edited to use in the course -Copies with notes should be printed for all students -The material is subject to the license agreement -Intellectual property rights and copyright remain the sole property of The Fiber Optic Assn. Inc. -It may not be resold in any way, either in whole or part of another document or presentation. -Each slide is annotated. This tells the instructor and students what is important about each slide to help with presenting the materials, gives references and further study materials. -Printing Handouts For Students: Print all slides as "Notes" which provides each slide on a page with the Notes for each slide below the slide, explaining what is meant by the slide. Note: Slides 49, 51, 52, 53, 54, 146 and 202 have animated graphics. You must be in the “Slide Show” mode to see the animation.
• #3: FOA instructors: This section is intended to introduce the student to the learning resources available from FOA and provide a brief introduction to the history of communications and fiber optics. Review the jargon so the students begin to become familiar with the language of fiber optics.
• #4: The instructor will go over lab safety rules before each lab. The lab manual has several pages of rules for safety in fiber optic labs. Each student should be familiar with them and follow them carefully. Instructors must follow them too! See https://www.thefoa.org/tech/ref/safety/safe.html
• #5: What is The FOA? Fiber optics has become the predominant communications medium, not just for telephones and the Internet, but also for cable television, security systems and computer networks. Workers in all these fields are expected to understand fiber optics and, in many cases, be competent in its installation. Training in these areas has become extremely important and The Fiber Optic Association, the international professional association of fiber optics, was founded to promote professionalism in fiber optics through education, certification and standards. The FOA was founded in 1995 by a dozen prominent fiber optics trainers and industry personnel who felt an industry-wide independent certification program was important for the growth of the industry. Today, the FOA has more than 200 approved approximately training organizations, including technical high schools and colleges, professional training organizations, government agencies and manufacturers offering training programs. As of 2022 over 85,000 fiber techs have successfully completed requirements for the FOA CFOT Certified Fiber Optic Technician certification and 120,000 certifications in all specialties. For more information, see the FOA website https://www.foa.org
• #6: This presentation is based on 35+ years of experience in the fiber optic business and much of the material is now available on the web and in printed form. The FOA Reference Guides to Fiber Optics and the FOA Guide online are the reference text for the course and the certification exam. References to the proper chapters are given in the notes. The notes give an overview of what the slide means and provide hints to explaining the meaning of the slide. Translations into Spanish and French are available in printed form plus online in the FOA Guide where a version in Portuguese is also available. .
• #7: FOA websites and textbooks are based on the input of instructors and experienced fiber optic technicians with 25+ years of experience in the business. Much of the material is now in printed form in The FOA Reference Guide to Outside Plant Fiber Optics for CFOS/O certification, The FOA Reference Guide to Fiber Optics for CFOT certification and The FOA Reference Guide to Premises Cabling for CPCT Certification. For designing fiber optic networks, there is the The FOA Reference Guide to Fiber Optic Network Design and for testing The FOA Reference Guide to Fiber Optic Testing. The newest FOA book is the FOA Fiber To The Home Handbook covering all aspects of FTTH. The FOA has also created a complete online reference guide for fiber optics. The site is at http://www.thefoa.org/guide/. Included in the site are hundreds of technical pages, including a basic fiber optic section that includes quizzes on the materials and links to pages with more details. Either the textbooks or the website may be used by classes for reference.
• #8: The FOA has created a complete online reference guide for fiber optics aimed at the CFOT and other FOA certifications. The site is at https://www.foa.org/guide/. Included in the site are almost 1000 pages of technical information, including a basic section in 4 languages and the design section in 2 languages.
• #9: The FOA YouTube Channel “thefoainc” has about 50 short videos on fiber optics and premises cabling. The “Lecture Series on Fiber Optics” covers how fiber is used in communications, fiber optic components, network design, installation and testing. Dozens more videos cover “hands-on” topics like fiber optic cable preparation, splicing, termination and testing as well as how UTP cable used in premises cabling is terminated. For those wanting to know more about fiber optics or studying for FOA certifications, both basic and advanced, these are good study materials.
• #10: Fiber U at www.fiberu.org is a focal point for online learning about fiber optics. It's based on the giant FOA Online Reference Guide and hosts many free self-study programs and tutorials that can make learning about fiber much easier! After completing online courses students may take an exam for a Certificate of Completion and use that as part of the process of getting FOA certifications.
• #11: FOA participates in many US and international standards activities with ANSI, TIA, ISO/IEC and NECA plus specialized groups. In addition, we have added our own standards - Standards written BY contractors, designers, installers and users FOR contractors, designers, installers and users. These are based on TIA, ISO/IEC standards but summarized in 1 page and backed by tech info on the FOA website. FOA's Standards are concise standards created by FOA with the participation of experts in the field for the most common issues affecting fiber optic network owners, contractors, designers and installers. Each standard summarizes what the reader needs to know in just 1 page. Each of the FOA's Standards will reference other industry standards that are similar in scope and which are used as the basis of the FOA standard, allowing FOA standards to be substituted for them. These FOA standards can be used for reference in project paperwork when the user and contractor need to be certain they agree what is being specified for the project.
• #12: The FOA Installation Tech Bulletin, available free on FOA Tech Topics and the ANSI/NECA/FOA 301were written for installers to provide a single standard to cover all issues of installation, from safety through installation, termination/splicing and test. These standards define what is meant by installation in a “neat and workmanlike manner.” It covers both premises and outside plant applications. FOA certificants and students can get Free downloads – Links are on the FOA home page and FOA Guide.
• #13: Another good handout for the course is Lennie Lightwave’s Guide to Fiber Optics. Lennie has been around since the early 1990s as the beginner's source for technical information on fiber optics. You can see it on the web at www.LennieLightwave.com and a printable version in PDF format is available. Lennie is also available as a printed book from Amazon. Also see the “Virtual Hands-On” section of Lennie Lightwave’s Guide, FOA Tech Topics: http://www.thefoa.org/tech/index.html and videos on the FOA YouTube channel.
• #15: Let’s take a look at how our communications networks developed. Modern communications started with the telegraph – credited to the man who invented Morse Code but developed by many more inventors – Bell for the telephone and a guy you probably never heard of – Marconi – who started sending signals by radio waves to create the wireless transmission of information.
• #16: The telegraph expanded on land very rapidly, often following the railroads, But it took many decades before a transatlantic telegraph cable could be built. The first transatlantic telegraph cable was developed by Cyrus West Field’s American Telegraph Company and installed from Ireland to Newfoundland in 1958. It only lasted 3 weeks. It took to 1866 to get reliable connections. It took even longer to successfully install a transatlantic phone cable – the technology was much more complex. AT&T’s TAT-1 ran from Newfoundland to Scotland and offered only 36 voice channels.
• #17: Telstar 1 launched on top of a Thor-Delta rocket on July 10, 1962. It successfully relayed through space the first television pictures, telephone calls, fax images, and provided the first live transatlantic television feed. Satellite provided a major breakthrough in worldwide communications with a most annoying delay to transmit signals. Few satellites exist for voice communications today, mostly in remote areas of the world. But they do offer Internet connections where fiber has not yet penetrated.
• #18: The beginning of fiber optics is traceable to the 1960s and early 1970s. Dr. Kao developed the ideas which made optical fiber communications possible at Standard Telecommunications Labs, the ITT facility in the UK. Dr. Kao was awarded a Nobel Prize in 2009 for his work. Drs. Donald Keck, Robert Maurer & Peter Schultz developed a way to make fibers to Kao’s theories at Corning Glass Works AT&T Bell Labs developed the lasers and communications systems that made communications over fiber possible, but it was a group effort, not individuals.
• #19: In 1976, AT&T installed a test fiber optic system in the Chicago Loop’s coal tunnels in late 1976 and after testing, used it for commercial traffic early in 1977. At the same time, British Telecom built a link in Milton Keynes. Trivia – one of the techs that installed that link was an FOA instructor who retired in 2017. (Note the way the cable is being paid off the spool – over the side – which is all wrong! This method puts a twist in the cable for each loop taken off the cable. Cable spools should be placed on a spindle and the cable rolled off the spool to prevent twisting!)
• #20: AT&T developed and installed the first transoceanic fiber optic cable across the Atlantic in 1988, thus beginning the revolution in international communications. TAT-8 was more than 10 times faster than the fastest copper cables. It was still in service in 2002 when the bandwidth of new cables made it obsolete. Note the large drum of cable in the center of the ship. Submarine cables are built in very long lengths to reduce the need for splicing in the middle of the ocean.
• #21: Everybody knows you need fiber for communications!
• #22: The first commercial fiber optic installation was for telephone signals in Chicago, installed in 1976. The first long distance networks and computer links were operational in the early 1980s. By 1985, most of today’s basic technology was developed and being installed in the fiber optic networks that now handle virtually all long distance telecommunications and provide the backbone for most other communications and data networks.
• #23: The biggest advantage of optical fiber is the fact it can transport more information longer distances in less time than any other communications medium. In addition, it is unaffected by the interference of electromagnetic radiation which makes it possible to transmit information and data with less noise and less error. Fiber is lighter than copper wires which makes it popular for aircraft and automotive applications. These advantages open up the doors for many other advantages that make the use of optical fiber the most logical choice in data transmission.
• #24: Fiber has lots of technical terms that one needs to learn to understand the technology. We have provided an explanation of the jargon in the textbook: Chapter 2, FOA Reference Guide to Fiber Optics and Basics/Jargon on the FOA Online Reference Guide
• #25: Before we get started, let’s deal with some jargon – the language of fiber optics. You must understand the jargon to understand fiber optics. You can assume you need to know this jargon when you take a CFOT certification test. Fiber optics – the communications technology that sends signals over hair-thin strands of glass using pulses of light. Fiber - hair-thin strands of glass (or sometimes plastic) that carries the signals Cable plant – the fiber optic cable and hardware that transports the light signals from point to point. “Plant” is an old telco term for the cabling. Outside plant – outdoor cable plant – as opposed to the indoor – premises – cable plant.
• #26: There are two types of fiber - multimode and singlemode – defined by the size of the light-carrying core of the fiber. . Multimode fiber – large core fiber used for premises (indoor) networks. Singlemode fiber – small core fiber used for everything else – higher bandwidth and lower attenuation.
• #27: Here is some fiber optic jargon to remember: Wavelength is a measure of the "color" of light. Visible light is about 500-600 nm wavelength – trivia: the eye’s sensitivity matches the output of our sun which helps us to see in that kind of light. Beyond that we call it "infrared" light. At longer – infrared - wavelengths, fiber has lower attenuation, so most systems transmit in that region. Multimode fiber operates mostly at 850nm, just above the visible region. Singlemode fiber is optimized for transmission at 1310nm for shorter links and 1550 nm where the attenuation is lower for longer links. Safety note: Infrared light is invisible to your eye, so potentially harmful light can be present in a fiber, but you can't see it! Check with a power meter.
• #28: Fiber optic measurements are based on optical power. We measure optical power in dB. dB is a logarithmic expression that compresses the measurement to make it easier to measure big ranges of power. dB or decibels is a measure of power originally named for Alexander Graham Bell and used for measuring acoustic signal level. Later it was used for measuring the power of radio signals. dB is used because it allows measurements over large dynamic ranges and loss or gain to be calculated by simple addition and subtraction. The graph shows the relationship of dB to power – 10dB is a factor of 10 in power. Actual power is measured in dB relative to 1 milliwatt of power - dBm Fiber has signal loss – attenuation – caused by scattering and absorption of light in the fiber. Splices and connections have loss also caused by imperfect joints between fibers Loss is expressed in dB – a logarithmic term we’ll explain later – and the attenuation of a fiber is expressed in loss per length – dB/km
• #29: Fiber optics is metric – get used to it! Most fiber length is in kilometers or meters, fiber size is in microms or millionths of a meter. One km is about 5/8 mile or 3280 feet 1 meter is about 39.4 inches – a bit longer than a yard 1mm is 1/25 inch or 0.039 inch A human hair is ~50-100 microns in diameter.
• #30: FOA instructors: This section is intended to briefly introduce students to the many applications of fiber optics in communications and other fields where it has become an essential technology. Once you get to data links, slow down and cover the material carefully. From there on, the material is important to learn well because it shows up on the certification exam.
• #31: And the power of fiber optic communications connects the world. This is a NASA composite map of the world at night. The lights show where people live.
• #32: The red lines are all the submarine fiber optic cables that connect the places where people live and provide their communications links to each other. And the power of fiber optic communications connects the world.
• #33: This is a map of the Internet backbones in the US – all fiber. Note how it connects to the red submarine cables on the coasts. Fiber is how the world communicates.
• #34: These are but a few of the applications of fiber optics, as we concentrate on communications. Fiber optics are also used for lighting, signs, sensors and visual inspection (medicine and non-destructive testing).
• #35: Fiber is the most efficient, cost effective means of communications and is being used to transmit the explosive growth of communications.
• #36: Fiber optics has become widely used in telecommunications because of its enormous bandwidth and distance advantages over copper wires. The application for fiber in telephony is simply connecting switches over fiber optic links. Telecom systems carry more phone conversations over a single pair of fibers than could be carried over millions of copper pairs. Material costs, installation and splicing labor and reliability are all in fiber's favor - not to mention space considerations. In major cities today, insufficient space exists in current conduit to provide communications needs over copper wire. While fiber carries virtually all long distance communications and almost all local communications, the penetration of fiber to the home (FTTH) was hindered for a long time by a lack of cost effectiveness. The secret to making FTTH cost effective has been the development of the passive optical network (PON). Telecom systems operate at bit rates up to 100, 200 or 400 gigabits per second – with higher speeds coming quickly - and many links use WDM - wavelength division multiplexing - to put multiple channels of signals over one fiber.
• #37: Around 400 submarine fiber optic cables connect the world. Fiber optic cables have been installed underwater, beginning with TAT-8 in 1988, to provide worldwide telecom and Internet communications. Installing those cables is a very specialized process that requires special cable designs and custom cable-laying ships to pay out the cable over thousands of kilometer runs and place it on the ocean floor at great depths. Those applications, while interesting, are beyond the scope of this book. Top: Submarine cable map, actual undersea cable with fibers in the center surrounded by electrical conductors to power repeaters and strength members and covered by a heavy-duty jacket. Below: Cable being loaded onto ship spool, cable laying ship, deploying a cable – note the repeater in the lower right hand corner, cable landfall and the undersea cable in place.
• #38: FTTH has taken off because of the explosion of data use for the Internet and the reduction of cost of installing fiber to the home. Today hundreds of millions of subscribers worldwide are connected on FTTH. Over the last decade, the cost of connecting a subscriber directly on fiber had dropped by about 75-80%, making FTTH more cost effective. This is a combination of component cost reductions and new architectures like PONs – passive optical networks. Plus fiber is much cheaper to maintain than copper, especially aging copper networks typical of most areas. With fiber the new services like IPTV downloaded over the Internet are possible.
• #39: The passive optical network (PON) uses optical couplers, both wavelength division multiplexers and simpler splitter/combiners, to allow connection of many customers over only one fiber - like broadcasting TV or radio over air waves. Thus only a few fibers can support gigabit connections to many customers, typically up to 32 customers on one fiber from the CO to the local splitter. New PONs offer 10G connections and as many as 256 users per fiber from the CO (central office.) FTTH has caused the development of several new technologies to simplify installation and lower cost. Many FTTH networks use prefabricated cabling systems read to "plug and play" and bend-insensitive fiber that is rugged enough to simplify installation in tight spaces inside homes.
• #40: Wireless communications is also booming. The aim of cellular upgrades is to up bandwidth using small cells and 5G. Most wireless systems are not wireless - antennas are connected into the worldwide communications networks via buried or aerial fiber optic cables. Likewise, wireless LANs and metropolitan WiFi systems require cabling and fiber provides greater distances from hubs and switches and immunity to noise.
• #41: Most wireless towers are connected to the phone system using fiber backbones using standard singlemode optical fiber. Expanded 4G/LTE and new 5G service requires more antennas on the towers. Traditionally the antennas are connected on large coax cables to stations on the ground. Now fiber, usually multimode fiber for the short links, is being used for it’s lower bulk and weight, so only a single fiber optic cable and a power cable needs to run up the tower rather than the big bundles of coax shown on the towers on the upper left.
• #42: Most wireless towers are connected to the phone system using fiber backbones using standard singlemode optical fiber. Expanded 4G and LTE service requires more antennas on the towers. (5G will be predominately small cells covered in the slides following) Traditionally the antennas are connected on large coax cables to stations on the ground. Now fiber, usually multimode fiber for the short links, is being used for it’s lower bulk and weight, so only a single fiber optic cable and a power cable needs to run up the tower rather than the big bundles of coax shown on the towers on the upper left.
• #43: Small cells are low powered cell sites that cover a much smaller areas than the older high-power antennas. They are aimed at urban areas where big cell towers are less effective and generally disliked because of their size. Small cells are similar to another low power cellular system, DAS – distributed antenna systems – used indoors to provide cell coverage where normal signals do not penetrate or in areas where large numbers of users are gathered. Small cells are simple devices, integrated to require only a fiber and low power. But a city needs so many small cells that the fiber required is enormous – cities may have hundreds or thousands of small cells, requiring massive amounts of fiber. Small cells do not have the large racks of electronics at some cell towers. That equipment is placed in a central office and the antenna is directly connected on fiber. These systems are called fronthaul (as opposed to backhaul) or C-RAN for centralized radio access networks.
• #44: Large public gathering places like convention centers or sports stadiums require many antennas to provide sufficient bandwidth, so low powered antennas are installed – about one antenna per 100 users. A sports stadium typically has 1,000 or more DAS antennas connected on fiber optics.
• #45: The reason fiber is used in CATV networks is that the fiber pays for itself in enhanced reliability. The enormous bandwidth requirements of broadcast TV require frequent repeaters. The large number of repeaters used in a broadcast cable network are a big source of failure. And CATV systems' tree and branch architecture means and upstream failure causes failure for all downstream users. Reliability is a big issue, since viewers are a vocal lot if programming is interrupted! Applications in CATV were slow until the development of the AM analog systems. By simply converting the signal from electrical to optical, the advantages of fiber optics, especially reliability, became cost effective. Now CATV has adopted a network architecture that overbuilds the normal coax network with fiber optic links. The HFC network lets the CATV provider have a two-way connection to the subscriber that allows them to offer broadband Internet connections at a low cost. The fiber network will also allow easy conversion to digital TV when it’s ready.
• #46: But CATV operators needed something to combat the subscriber’s clamoring for fiber to the home, which lead to the development of RFOG, RF over Glass. RFOG is basically nothing more than a HFC/cable modem system built with less expensive components now available thanks to the volume pricing of components used in FTTH. It’s designed to operate over a standard telco PON (passive optical network) fiber architecture with short fiber lengths and including the losses of a FTTH PON splitter. There is one interesting side effect of this approach. Now telcos and CATV companies can deliver the same services over the same cable plant using totally different technologies. But that means that office or apartment building owners, developers or even whole towns that might be considering installing FTTH infrastructure themselves and leasing the fiber to a service provider can have a choice of service providers. One cable network can support either CATV or telco systems – or even someone else for that matter. That opens up a big market for private fiber optic systems.
• #47: Fiber is widely used in premises or structured cabling, supporting Ethernet to 100 Gb/s. Headed to 1 terabit/sec. Fiber is used for most backbones, some fiber to the desk and to connect WiFi access points, especially 802.11n. Some networks run a separate wireless network for visitors for security, keeping unauthorized users off the corporate network. A centralized fiber network allows using fiber without telecom rooms near the users, centralizing all the electronics in the computer room. Data centers are another big user of fiber, with connections at 10 Gb/s or higher where fiber is more reliable and consumes much less power.
• #48: Motorola, one of the largest suppliers of FTTH PON equipment is now offering systems similar to those used in large residential buildings for enterprise LANs in companies. They quote system costs that are much less expensive than installing a fiber optic backbone and copper cabling to the desktop. Because a POL brings to enterprise networks the management and control – plus encryption – of a PON FTTH network, it has become popular with organizations concerned over security and network management.
• #49: Data centers are the fastest growing application for computers used as servers. Connections are now at 10 to 100 Gb/s and new systems are becoming available at 200-800 Gb/s. Fiber links between these computers and storage devices or routers are quite common as fiber saves power, space and is much easier to install. Industry consortiums building hyperscale data centers led by Facebook, Amazon, Google, Apple and Microsoft have moved to singlemode (SM) fiber reduce the numbers of fibers and allow the continual upgrading they routinely do. Since these large data centers use massive numbers of links - 100,000+ is common – they have driven the price of electronics like the SM fiber transceivers down to make them affordable- often cheaper than multimode.
• #50: Fiber has found many other uses. Cellular systems are not wireless - most antennas are connected via buried fiber optic cables. Likewise, wireless LANs require cabling and fiber provides greater distances from hubs and switches and immunity to noise. Utilities have used fiber for managing their grids and communications throughout their networks for many years. Recent problems have had many upgrading their systems. Security systems use lots of fiber. CCTV cameras use fiber to extend their reach, for example in large airport terminals, outdoors in power plants or inside and outside big office buildings. Fibers can also be used as sensors, for example sensing intruders on fences or walking across buried fiber sensors. And, of course, fiber is very difficult to “tap,” making it popular for secure military and government networks.
• #51: The first military projects in the US were on ships, but over time fiber optics moved into tactical field networks for radar remote heads and battlefield communications, aircraft and helicopters and even the Internatinoal Space Station. Fiber’s immunity to EMI – electromagnetic intereference – no easy tapping or jamming – low weight and high bandwidth made it an ideal choice for these applications.
• #52: Many video links are available on fiber optics, from remote security cameras to broadcast TV cameras in studios or on location as in the auto race in Long Beach, CA shown. Audio links are used in concert halls, meeting rooms, or any large auditorium with powered speakers.
• #53: Woods Hole Oceanographic Institute started using optical fiber to connect their underwater robots called remote-piloted vehicles in the 1980s. The most spectacular result was the discovery of the Titanic by Dr. Robert Ballard who developed the technology with Jason, shown in the picture here looking into the window of a stateroom on the Titanic. Using fiber allowed the tether cables to be ten times longer than with copper and produce better signals! Now all RPVs use fiber tethers.
• #54: The electrical noise common to industrial environments makes it difficult to use copper data cables. But fiber is immune to electromagnetic interference and more flexible and withstands higher heat also. Industrial robots have fibers running along the arm. The machines are connected to a network, almost always on fiber.
• #55: Electrical utilities have been using fiber for many years for communications and to control their electrical distribution systems. Many use optical power ground wire (OPGW) that has fiber running inside of an electrical conductor. Fiber optic sensors are also used to monitor very high voltages and currents
• #56: Fiber optics have been used for decades in the energy business. In oil, fiber is used in exploration as super-sensitive sensors to gather geological data and monitor drilling. Fiber is widely used on drilling platforms and in refineries for communications, monitoring and control. Fiber is also used on pipelines to monitor flow and problems as well as communications.
• #57: Alternative energy production requires precise control and management to create electrical power compatible to the current grid. Wind and solar systems must be controlled to maximize outputs and control the processes. Solar using heat to generate steam, as well as those involving photovoltaic conversion, have reflectors that follow the sun, maximizing outputs. Windmills, of course, must fact into the wind and control the blades according to wind speeds. All this works on computer systems controlled by fiber. One solar facility in the Mojave has over 1200km (750 miles) of fiber!
• #58: Many networks are installed in cities. Some are owned by the cities to connect their offices, public services, emergency services, schools and libraries, etc. Some even lease fibers to high-tech companies. New techniques like microtrenching, shown here, is often used to prevent the disruption common to digging up streets to bury conduit or cables.
• #59: Most data links are used to connect two devices point-to-point and lack the protocols of a network. Most of these links offer fiber optics as an option - and some are only fiber. RS232 and RS422 are industrial links that have been around for many years. They have been available on fiber since fiber got started. Fibre Channel is a high speed link connecting computers to peripherals like disk drives and printers. HIPPI is a similar (and fairly obsolete) link. Fire Wire and Toslink used in consumer applications like digital audio. Most and Flexray are automotive networks, where fiber’s lighter weight and immunity to electrical noise make it a better choice than copper. Active optical links are being designed into the next generation of PCs to replace USB and FireWire.
• #60: Building management systems can use fiber in place of copper cable for longer distances and greater security. Industrial networks favor fiber for process control applications due to its distance capability and immunity to electrical noise. Fiber optic sensors are available for a number of applications, including measuring high voltages and currents as in power grids, dangerous chemicals and can operate in hazardous environments since they are intrinsically safe.
• #61: FOA Instructor: Important tech material starts here. Fiber optic transmission systems all consist of a transmitter which takes an electrical input and converts it to an optical output from a laser diode, VCSEL or LED. The light from the transmitter is coupled into the fiber with a connector and is transmitted through the fiber optic cable plant. The light is ultimately coupled to a receiver where a detector converts the light into an electrical signal which is then conditioned properly for use by the receiving equipment. Most links use two fibers transmitting in opposite directions for full duplex operation, but some, including the passive optical networks (PONs) used in FTTH transmit bidirectionally over 1 fiber at 2 different wavelengths. Just as with copper wire or radio transmission, the performance of the fiber optic data link can be determined by how well the reconverted electrical signal out of the receiver matches the input to the transmitter. The newest link technology uses coherent transmission. It can go to terabits/second or thousands of km of link length for submarine cables. It's very complicated so just take our word for it. The CFOT Lab Manual has an exercise on data links using media converters.
• #62: The communications world is electronic – processing and sending communications as electrical signals, but fiber optics uses optical signals. Electronic equipment may use plug-in transceivers that convert signals from electrical to optical for transmission over optical fibers or they may connect with copper cables to stand-alone transceivers called media converters that convert the electrical signals to optical signals. The Lab Manual has an exercise using media converters to demonstrate how fiber optic transmission links work.
• #63: Four types of sources are commonly used in fiber optics, LEDs, fabry-perot (FP) lasers, distributed feedback (DFB) lasers and vertical cavity surface-emitting lasers (VCSELs). All convert electrical signals into optical signals, but are otherwise quite different devices. All three are tiny semiconductor devices (chips). LEDs and VCSELs are fabricated on semiconductor wafers such that they emit light from the surface of the chip, while f-p lasers emit from the side of the chip from a laser cavity created in the middle of the chip. LEDs have much lower power outputs than lasers and their larger, diverging light output pattern makes them harder to couple into fibers, limiting them to use with multimode fibers. LEDs have much less bandwidth than lasers and are limited to systems operating up to about 250 MHz or around 200 Mb/s. VCSELs are faster and cheap sources and have made LEDs almost obsolete. Laser have higher power and greater bandwidth, making them ideal for long-distance high-speed links. Lasers have very high bandwidth capability, most being useful toaround 50 GHz or 50 Gb/s. LEDs have a very broad spectral output which causes them to suffer chromatic dispersion in fiber, while lasers have a narrow spectral output that suffers very little chromatic dispersion. DFB lasers, special versions of the F-P laser, which are used in long distance and DWDM systems, have the narrowest spectral width which minimizes chromatic dispersion on the longest links. DFB lasers are also highly linear (that is the light output directly follows the electrical input) so they can be used as sources in AM CATV systems.
• #64: The type of source used in a network depends on the speed of the network and the distance it must operate over. LEDs are useful to only several hundred megabits per second, so they are used in slower systems. LEDs are really obsolete in most networks today. Lasers can be used up to 10 gigabits per second or more, so they are the source of choice in high speed networks. Above 10Gb/s, wavelength division multiplexing is used for multiple channels to get 100, 200, 400Gb/s or more. Since the loss of the fiber goes down with longer wavelengths, long distance links tend to use the longest wavelength sources – ~1550nm. Above 100Gb/s, coherent links are now used for more distance and to overcome fiber bandwidth issues.
• #65: Receivers use semiconductor detectors (photodiodes or photodetectors) to convert optical signals to electrical signals. Silicon photodiodes are used for short wavelength links (650 for POF and 850 for glass MM fiber). Long wavelength systems usually use InGaAs (indium gallium arsenide) detectors as they have lower noise than germanium which allows for more sensitive receivers. Very high speed systems sometimes use avalanche photodiodes (APDs) that are biased at high voltage to create gain in the photodiode. These devices are more expensive and more complicated to use but offer significant gains in performance.
• #66: A fiber optic datalink works by transmitting from the transmitter to the receiver. The source in the transmitter couples power into the fiber which is then diminished by the attenuation of the optical fiber and losses from connectors and splices in the link. While we show a digital link, some fiber links transmit analog data. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #67: How Does WDM Work? It is easy to understand WDM. Consider the fact that you can see many different colors of light - red, green, yellow, blue, etc. all at once. The colors are transmitted through the air together, but they can be easily separated using a simple device like a prism, just like we separate the "white" light from the sun can be separated into a spectrum of colors with a prism.
• #68: The input end of a WDM system is really quite simple. It is a simple coupler that combines all the inputs into one output fiber. These have been available for many years, offering 2, 4, 8, 16, 32 or even 64 inputs. It is the demultiplexer that is the difficult component to make. The demultiplexer takes the input fiber and collimates the light into a narrow, parallel beam of light. It shines on a grating (a mirror like device that works like a prism, similar to the data side of a CD) which separates the light into the different wavelengths by sending them off at different angles. Optics capture each wavelength and focuses it into a fiber, creating separate outputs for each separate wavelength of light. Current systems offer from 4 to 64 channels of wavelengths. The higher numbers of wavelengths has lead to the name Dense Wavelength Division Multiplexing or DWDM. The technical requirement is only that the lasers be of very specific wavelengths and the wavelengths are very stable, and the DWDM demultiplexers capable of distinguishing each wavelength without crosstalk. To expand bandwidth, it's now common to add new wavelengths to current fibers rather than use new fibers.
• #69: Dense wavelength division multiplexing (DWDM) originally used optical signals multiplexed within the 1550 nm band compatible with erbium doped fiber amplifiers (EDFAs), which are effective for wavelengths between approximately 1525–1565 nm (C band), or 1570–1610 nm (L band). Dense wavelength division multiplexing (DWDM) channel plans vary, but a typical system would use 40 channels at 100 GHz spacing or 80 channels with 50 GHz spacing. Some technologies are capable of 12.5 GHz spacing (sometimes called ultra dense WDM). Such spacings are today only achieved by free-space optics technology. New amplification options (Raman amplification) enable the extension of the usable wavelengths to the L-band, more or less doubling these numbers. Generally DWDM is defined by laser frequency instead of wavelength, while CDWM is defined by wavelength. The DWDM spacing shown, ~0.8nm, corresponds to 100GHz frequency spacing. Systems may use 100, 50 or 25GHz spacing.
• #70: Much of the jargon associated with testing refers to the transmission “bands” so we offer this here as a reference tool. Much of the testing centers on the C band as it is used for so much of long distance transmission.
• #71: While the low loss of optical fiber allows signals to travel hundreds of kilometers, extremely long haul lines and submarine cables require regenerators or repeaters to amplify the signal periodically. In the beginning, repeaters basically consisted of a receiver followed by a transmitter. The incoming signal was converted from a light signal to an electrical signal by a receiver, cleaned up to remove as much noise as possible, then was retransmitted by another laser transmitter. These repeaters added noise to the signal, consumed much power and were complicated, which means they were a source of failure. They also had to be made for the specific bit rate of transmission and upgrading required replacing all the repeaters, a really difficult task in an undersea cable! Since the 1960s, researchers knew how to make fiber lasers. Proper doping of the fiber (introducing small amounts of active elements into the glass fiber) allowed it to be pumped with external light sources until stimulated emission occurred. While making fiber amplifiers was hypothesized early in the stages of fiber optic development, it was not until 1987 that working models were realized. Major contributors to the development included Bell Labs and NTT. The typical fiber amplifier works in the 1550 nm band and consists of a length of fiber doped with Erbium pumped with a laser at 980. The pump laser supplies the energy for the amplifier, while the incoming signal stimulates emission as the pulse passes through the doped fiber. The stimulated emission stimulates more emission, so there is a rapid, exponential growth of photons in the doped fiber. Gains of >40 dB (10,000X) are possible with power outputs >+20 dBm (100 mW). To date, the most efficient fiber amplifiers have been Erbium-Doped Fiber Amplifiers (EDFAs) operating in the 1550 nm range.
• #72: The ability of any fiber optic system to transmit data ultimately depends on the optical power at the receiver as shown above, which shows the data link bit error rate as a function of optical power at the receiver. (BER is the inverse of signal-to-noise ratio, e.g. high BER means poor signal to noise ratio.) Either too little or too much power will cause high bit error rates. Too much power, and the receiver amplifier saturates, too little and noise becomes a problem as it interferes with the signal. This receiver power depends on two basic factors: how much power is launched into the fiber by the transmitter and how much is lost by attenuation in the optical fiber cable plant that connects the transmitter and receiver. The optical power budget of the link is determined by two factors, the sensitivity of the receiver, which is determined in the bit error rate curve above and the output power of the transmitter into the fiber. The minimum power level that produces an acceptable bit error rate determines the sensitivity the receiver. The power from the transmitter coupled into the optical fiber determines the transmitted power. The difference between these two power levels determines the loss margin (loss budget or power budget) of the link.
• #73: After a fiber optic cable plant is installed, it may be used with a number of different types of fiber optic networks. Computer networks, telephone signals, video links, and even audio can be sent on the installed fibers. Each network type has a requirement for the performance of the fiber optic cable link. Most simply specify the maximum loss in the link that can be tolerated, a function of component specifications and installation quality. Others also specify the bandwidth performance of the fiber which is determined by the specifications of the fiber chosen. Every fiber optic link has a maximum loss of a cable plant over which it can work. That loss is determined by the output power of the transmitter coupled into the fiber and the sensitivity of the receiver, all expressed in dB.. The loss of the fiber optic cable it uses must be less than 1that maximum loss for proper operation. The drawings here illustrate the example in the textbook in Chapter 10. The transmitter couples a certain amount of power into the fiber in the cable plant. As the light is transmitted down the fiber, it is attenuated by the attenuation of the fiber and the loss in connectors and splices. In this link, the cable plant has 5 connections and a splice, plus the length of the fiber to cause loss.
• #74: While every link installed must meet some maximum loss to allow operation of the network intended to use it, different networks may have different link margins. Therefore we use a different approach. The loss of the link is considered acceptable if it is less than standard maximum values calculated from the characteristics of the link installation. What causes the losses in the fiber optic cable? First the fiber itself. The next loss factor is terminations. Splices are common in singlemode but rare in multimode networks Singlemode fiber is usually spliced with a fusion splicer which welds the two fibers together in an electric arc, with much lower losses. The final loss factor is stress in installation. Fiber optic cable pulled with too much tension may be damaged. Each time you make a bend with a fiber optic cable, you put some stress in the fiber which can cause loss. Even cable ties tightened on the cable can cause loss. Stress loss should be zero!
• #75: The drawings here illustrate the example in the textbook The graph below the link diagram shows the actual amount of light in the fiber along the length, directly corresponding to the link diagram above it. This diagram looks like an OTDR plot, since it is similar to what the OTDR measures. If you are not familiar with OTDRs, we will cover them in the testing sections. But look at the diagram closely. The power goes down as the light goes down the fiber, reduced by the attenuation of the fiber and the losses in connectors and splices. By convention, we include the loss of the connectors on the end of the cable plant, since when we test connectors, we do so by mating them to another reference connector. The power level starts at the transmitter output, coupled into the fiber, shown at the top of the X-axis of the graph. After the loss of the cable plant, it is reduced by the amount of the loss. In order for the link to work properly, the power at the receiver must be higher than the receiver sensitivity, shown at the bottom of the X-axis of the graph. The amount by which the receiver power exceeds the receiver sensitivity is the margin of the link.
• #76: The FOA Loss Budget Web App works with any browser. It estimates the optical loss of a fiber optic link. This will save time for the installer of a fiber optic link needing to know whether test results are reasonable and/or make a "pass/fail" determination. It can also help the designer of a link to determine if communications equipment will operate over this link. By choosing the type of link (singlemode or multimode) and specifying the length of the fiber and numbers of connections and splices, it will calculate the end to end loss of the link.
• #77: The FOA LossCalc Ap estimates the optical loss of a fiber optic link. This will save time for the installer of a fiber optic link needing to know whether test results are reasonable and/or make a "pass/fail" determination. It can also help the designer of a link to determine if communications equipment will operate over this link. By choosing the type of link (singlemode or multimode) and specifying the length of the fiber and numbers of connections and splices, it will calculate the end to end loss of the link. The app has default specifications for singlemode and multimode links or the user may create custom setups with specifications appropriate for any application.
• #78: This is a basic fiber optic training program for FOA-Approved Schools to teach classes for the FOA CFOT certification. The program was developed by The Fiber Optic Association with inputs from many FOA instructors. It’s is based on 30+ years of experience in the business, including starting one of the first fiber optic test equipment companies and training thousands of fiber optic installers.
• #79: Now let’s take a look at the components of a fiber optic system. We’ll examine each of these in detail and look at their installation. Most fiber optic components, including "Cables," "Termination" (Connectors and Splices) have their own PPTs that include information on the components and how they are installed.
• #80: Optical fiber is comprised of a light carrying core surrounded by a cladding which traps the light in the core by the principle of total internal reflection. Most optical fibers are made of glass, although some are made of plastic. The core and cladding are usually fused silica glass which is covered by a plastic coating called the buffer which protects the glass fiber from physical damage and moisture. When you prepare fiber for termination or splicing, you strip off the plastic buffer coating and leave the solid glass fiber. Most fibers are all glass, but glass core/plastic clad (PCS - plastic clad silica and HCS - hard clad silica) and all plastic (POF - plastic optical fiber) are made. Glass optical fibers are the most common type used in communication applications.
• #81: Total internal reflection is something swimmers know well – it’s the reflection you see when looking at the surface of the water from below. It’s an optical phenomenon caused by the difference between the characteristics of light and water. Water has an index of refraction much higher than air – 1.5 to 1 – that causes the reflection. The change in refractive index also causes light to bend at the surface between the two – causing the apparent bending in a stick stuck in the water. The index of refraction of a material is a property defined by the speed of light in the material compared to the speed of light in a vacuum. Glass has an index of refraction of about 1.4-1.5, meaning light travels 2/3 as fast in glass as in a vacuum. Whenever light crosses a boundary with materials of different index of refraction (like this white pole sticking into a pond), the speed of light changes and the light is refracted or bent. We can see this ourselves by looking at a stick stuck in the water as in the photo above. By making the core of the fiber of a material with a slightly higher refractive index, we can cause the light in the core to be totally reflected at the boundary of the cladding for all light that strikes at greater than a critical angle determined by the difference in the composition of the materials used in the core and cladding. Many students are curious how fiber is made. Good explanations are available on the FOA Guide and on YouTube from most fiber manufacturers. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #82: Optical fiber has two basic types, multimode and singlemode. Multimode fiber means that light can travel many different paths (called modes) through the core of the fiber, which enter and leave the fiber at various angles. The highest angle that light is accepted into the core of the fiber defines the numerical aperture (NA) .Two types of multimode fiber exist, distinguished by the index profile of their cores and how light travels in them. Step index multimode fiber has a core composed of one type of glass. Light traveling in the fiber travels in straight lines, reflecting off the core/cladding interface. The numerical aperture is determined by the differences in the indices of refraction of the core and cladding and can be calculated by Snell's law. Since each mode or angle of light travels a different path link, a pulse of light is dispersed while traveling through the fiber, limiting the bandwidth of step index fiber. In graded index multimode fiber, the core is composed of many different layers of glass, chosen with indices of refraction to produce an index profile approximating a parabola. Since the light travels faster in lower index of refraction glass, the light will travel faster as it approaches the outside of the core. Likewise, the light traveling closest to the core center will travel the slowest. A properly constructed index profile will compensate for the different path lengths of each mode, increasing the bandwidth capacity of the fiber by as much as 100 times that of step index fiber. Singlemode fiber just shrinks the core size to a dimension about 6 times the wavelength of the fiber, causing all the light to travel in only one mode. Thus modal dispersion disappears and the bandwidth of the fiber increases by at least another factor of 100 over graded index fiber.
• #83: Step index multimode fiber was the first fiber developed. The core and cladding were glass with an index of refraction chosen to accept light in a restricted cone limited by the total internal refraction of the light. The problem with step index multimode fiber is bandwidth – as you can see in the drawing, rays of light – we call them modes – traveling at larger angles travel longer paths in the core and arrive at the far end at different times. This causes dispersion – pulse widening – that limits the bandwidth of the fiber. Step index multimode fiber is limited to short links at a speeds of a few megabits per second – that’s slow in fiber world! Today, step index fiber is mostly plastic optical fiber used in slow links for consumer electronics, cars and for lighting. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #84: Remember the index of refraction of the glass determines the speed of light. If the core of the multimode fiber has a lower index of refraction toward the cladding instead of the same throughout, the light will speed up as it nears the cladding. By choosing the index of refraction profile carefully, it is possible to make the modes of light in the outside of the core speed up just enough to keep in synchronization with the slower modes in the center of the core. This technique works like this drawing shows. The core is formed of many layers of glass, forming a lens similar to the flat lenses used in lighthouses, called fresnel lenses, as shown. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #85: With the proper core design, graded index multimode fiber modes take approximately the same time to travel down the fiber, increasing the bandwidth by a factor of 100 or more. Multimode graded index fiber is used for premises cabling – local area networks (LANs) and building security and management systems. It is rarely used for outside networks. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #86: Singlemode fiber is like a step index multimode fiber except the core diameter is shrunk to a point where it can only support one mode of light. A singlemode fiber can provide much more bandwidth – thousands of times more than graded index multimode fiber. It also has significantly lower attenuation. Practically all outdoor networks – we call them ”outside plant” or outside cable plants – use singlemode fiber. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #87: Besides fibers made from pure glass, there are plastic optical fibers (POF) and hard-clad silica (HCS) or plastic-clad silica (PCS) fibers. POF is all plastic, mostly step index with a 1 mm outer diameter, although some graded-index POF of smaller size is becoming available. POF is used in consumer electronics, automobiles and industrial controls. HCS or PCS fibers use a step-index glass core and a plastic cladding. Both are used in industrial networks where flexibility and ruggedness are important. Unless specifically called out, we will concentrate on all-glass fibers in our presentation. Below are the three most widely used fiber types, all made from pure glass: Multimode fiber with core/cladding sizes of 50/125 and 62.5/125 microns. 50/125 is often referred to as “laser rated” fiber for it’s higher bandwidth capacity with laser sources (and two versions are available with different bandwidth ratings, OM2 with 500 MHz-km bandwidth for 850 nm VCSEL sources and OM3 with bandwidth of 2000 MHz-km at 850 nm.) 62.5/125 (OM1) is often called “FDDI fiber” since it was the standard for that network introduced in 1990. Do not mate dissimilar fibers – even the two multimode fibers 50/125 and 62.5/125 – as this gives high loss, as much as 4dB from the big fiber to the small fiber. From singlemode to multimode, it’s even worse – 17 to 20dB loss! Lab Illustrate: VFL with SM and MM fiber
• #88: Here is a cross-reference for the various names given to singlemode fiber types by standards groups around the world. IEC and ITU types are for the telcos, mainly outside plant, while TIA specs are for premises cabling specified in TIA-568 standards. The fiber outlined in yellow – the color of singlemode cable – is the most used fiber – G.652. A new class of fibers – G.657 – are less sensitive to bending – more on that in a minute. Don't try to memorize all the variations - they are in the textbook and on the FOA Reference Guide website, but you do need to be aware of the types when designing, installing and testing networks.
• #89: Multimode Fiber Types Here is a cross-reference for the various names given to multimode fiber types by standards groups around the world. Don't try to memorize them - they are in the textbook and on the FOA Reference Guide website, but you do need to be aware of the types when designing, installing and testing networks. Since the introduction of Gigabit Ethernet, we've seen premises cabling, designed to carry gigabit and 10 gigabit (and faster) traffic using 850 nm VCSEL (laser) transmitters, moving toward standardization on 50/125 laser-optimized fiber (now universally called OM3 fiber for it's international standards designation) often with LC connectors to match the manufacturers' standard for VCSEL transceivers. OM3/OM4 cabling even has it's own color, aqua, specified in TIA-598. OM1 fiber has had a long life - from the early 1980s to the early 2000s it was the primary MM fiber, but has been replaced by more modern, laser-optimized fibers for today’s gigabit+ networks. OM5 fiber was designed with a wider wavelength band – 850-950nm – to allow wavelength-division multiplexing with 4 VCSELs to achieve 40/100/higher gigabits/s on 2 fibers. It seems to not getting much support in the market. OM3 or OM4 fiber- is probably the most used fiber because it is the most cost effective for most applications.
• #90: Here is some fiber optic jargon to remember: Wavelength is a measure of the "color" of light. Visible light is about 500-600 nm wavelength. Beyond that we call it "infrared" light. At longer (infrared) wavelengths, fiber has lower attenuation, so most systems transmit in that region. Safety note: Infrared light is invisible to your eye, so potentially harmful light can be present in a fiber but you can't see it! Check fibers with a power meter.
• #91: The attenuation of the optical fiber is a result of the combination of two factors, absorption and scattering. The absorption is caused by the absorption of the light and conversion to heat by molecules in the glass. Primary absorbers are residual OH+ and dopants used to modify the refractive index of the glass. This absorption occurs at discrete wavelengths, determined by the elements absorbing the light. The OH+ absorption is predominant, and occurs most strongly around 1000 nm, 1400 nm and above1600 nm. The largest cause of attenuation is scattering. Scattering occurs when light collides with individual atoms in the glass and is anisotropic. Light that is scattered at angles outside the numerical aperture of the fiber will be absorbed into the cladding or transmitted back toward the source Scattering is also a function of wavelength, proportional to the inverse fourth power of the wavelength of the light. Thus if you double the wavelength of the light, you reduce the scattering losses by 24 or 16 times. Therefore , for long distance transmission, it is advantageous to use the longest practical wavelength for minimal attenuation and maximum distance between repeaters. Together, absorption and scattering produce the attenuation curve for a typical glass optical fiber shown.
• #94: More detailed fiber specifications can be found in the textbook or on the Reference website or from manufacturers websites or datasheets The specifications in parentheses are from TIA-568 which are more conservative than typical specs. The two specs for singlemode are for indoors (1 dB/km) and outdoor (0.5 dB/km). These are the wavelengths of light generally used for testing fibers for loss.
• #95: Manufacturers have been able to change their processes to make singlemode fiber with low-water absorption peaks, opening up a new transmission band (E-band) in singlemode fiber, which has proven useful for CWDM (coarse wavelength division multiplexing – a cheaper version of WDM used in shorter links like metro and data centers. CWDM uses inexpensive lasers with much wider wavelength spacing than DWDM for applications like 100G networks for data centers using 4 lasers.
• #96: Dispersion is the spreading of a pulse of light as it travels down an optical fiber. Dispersion limits the bandwidth of the fiber itself, but the bandwidth of the data link also depends on the bandwidth of the transmitter's source and receiver's detector. The causes of dispersion are different in multimode fiber and singlemode fiber. Some slides have animated graphics. You may need to be in the “Slide Show” mode to see the animation.
• #97: Fiber's information transmission capacity is limited by two separate components of dispersion: modal and chromatic dispersion. First modal dispersion: Step index multimode fiber has a core composed of only one type of glass. Light traveling in the fiber travels in straight lines, reflecting off the core/cladding interface. Since each mode or angle of light travels a different path link, a pulse of light is dispersed while traveling through the fiber, limiting the bandwidth of step index fiber. In graded index multimode fiber, the core is composed of many different layers of glass, chosen with indices of refraction to produce an index profile approximating a parabola. Since the light travels faster in lower index of refraction glass, the light will travel faster as it approaches the outside of the core. Likewise, the light traveling closest to the core center will travel the slowest. A properly constructed index profile will compensate for the different path lengths of each mode, increasing the bandwidth capacity of the fiber by as much as 100 times that of step index fiber. Singlemode fiber just shrinks the core size to a dimension about 6 times the wavelength of the fiber, causing all the light to travel in only one mode. Thus modal dispersion disappears and the bandwidth of the fiber increases by at least another factor of 100 over graded index fiber.
• #98: The second factor in fiber bandwidth is chromatic dispersion. Remember a prism spreads out the spectrum of incident light since the light travels at different speeds according to its color and is therefore refracted at different angles. The usual way of stating this is the index of refraction of the glass is wavelength dependent. Thus a carefully manufactured graded index multimode fiber can only be optimized for a single wavelength, usually near 1300 nm, and light of other colors will suffer from chromatic dispersion. Even light in the same mode will be dispersed if it is of different wavelengths. Chromatic dispersion is a bigger problem with LEDs, which have broad spectral outputs (their output light is comprised of many wavelengths of light), unlike lasers which concentrate most of their light in a narrow spectral range. Chromatic dispersion occurs with LEDs because much of the power is away from the zero dispersion wavelength of the fiber. High speed systems, based on broad output LEDs, suffer intense chromatic dispersion, about equal to the modal dispersion.
• #99: Another factor in fiber bandwidth is polarization mode dispersion. Polarization mode dispersion (PMD) is a bit more complex. Polarization is a phenomenon of light traveling in a medium as a wave with components at right angles. Some materials, like a glass optical fiber, have a different index of refraction for each of those components of the light wave, which is called birefringence. And a different index of refraction means light travels at a different speed, so in the simplest visualization, PMD in fiber looks like the drawing below, where each component of the polarized light travels at a different speed, causing dispersion. The magnitude of PMD in a fier is expressed as this difference, which is known as the differential group delay (DGD) and called Δτ (delta Tau). PMD is generally tested for fibers during manufacture or when being cabled. In the field, it is common to test PMD on newly installed fibers which are intended for operation at high speeds, generally above 2.5 Gb/s or when upgrading fibers installed some time in the past. Since PMD varies over time, a single test becomes an average and tests at a later time may be done for comparison.There are a number of commonly used test methods for PMD, some of which are limited to the manufacturing environment, while others can be used in the field. Essentially, all the test instruments have a source which can vary the polarization of the test signal and a measurement unit that can analyze polarization changes.
• #100: For almost fifteen years, one had only two choices if you were installing fiber. The de facto-standard multimode fiber had a core/cladding size of 62.5/125 microns and was rated for use with FDDI (Fiber Distributed Data Interface) or Fast Ethernet, both 100 Mb/s networks that used inexpensive LED sources as transmitters. Longer distances or higher speeds called for singlemode fiber with a small 8 micron core that required expensive laser sources. Fiber manufacturers had not put any real engineering effort into multimode fiber in fifteen years because 62.5/125 fiber met the industry’s needs. But with the advent of Gigabit Ethernet (GbE), calls for longer distances on multimode fiber sent them back to the labs. And what they came up with was a brand new twenty-year old fiber - 50/125! GbE uses a 850 nm laser for a source, so the fiber manufacturers revived a fiber that dated back to the “prehistoric era” of fiber - 1980 - when long distance telecom networks used newly-developed 850 nm lasers with a fiber that had a 50 micron core optimized for use with lasers. Now manufacturers have improved the performance of this 50/125 fiber even more, to allow use with >10 Gigabit Ethernet and Fibre Channel. Laser optimized 50/125 fiber is now available in 4 grades, with the top grade optimized for wavelength division multiplexing in the 850-950nm range).
• #101: Singlemode fiber have several versions depending on the application. OS1/OS2 are TIA specs for premises/campus applications. Low water peak fiber is used for CWDM – coarse wavelength division multiplexing. The other types are used for long links and DWDM-dense wavelength division multiplexing. ITU G.653    Covers single-mode dispersion-shifted optical fiber. Dispersion is minimized in the 1,550-nm wavelength range. At this range attenuation is also minimized, so longer distance cables are possible. ITU G.654: Covers single-mode fibre which has the zero-dispersion wavelength around 1300 m wavelength which is cut-off shifted and loss minimized at a wavelength around 1550 nm and which is optimized for use in the 1500-1600 nm region. ITU G.655    Covers single-mode NZ-DSF (nonzero dispersion-shifted) fiber) , which takes advantage of dispersion characteristics that suppress the growth of four-wave mixing, a problem with WDM (wavelength division multiplexing) systems. NZ-DSF supports high-power signals and longer distances, as well as closely spaced DWDM.
• #102: Most optical fiber manufacturers are now offering bend-insensitive fibers that can be bent tightly without much loss. This allows them to be used in close spaces like cable trays or run around the edge of a wall in a room.
• #103: These bend-insensitive fibers – both MM and SM – use an optical ring – like a guard ring – around the core made of lower index glass to reflect light escaping the core due to stress back into the core. It’s simple but very effective. SM is used for patchcords and premises use but is compatible to other SM fibers MM B-I fiber affects modal distribution, may have high loss connecting to legacy fiber
• #104: The standard multimode fiber for over 15 years had been 62.5/125 - the so-called FDDI grade fiber for it’s bandwidth to support FDDI but it’s days are over. There has been a changeover for high speed systems to 50/125 fiber for it's higher bandwidth performance with VCSELs (up to 10 Gb/s), but it cannot be mixed with 62.5/125! If you choose one fiber, all patchcords must be the same fiber! Fiber is cheap - as cheap as kite string! When specifying backbone cables, install lots of spare fibers including singlemode fibers in multimode cables - called a “hybrid” cable - for future high bandwidth applications. Sometimes it makes sense to put fibers into copper cables called a composite cable. This is becoming more common when using a combination cable (coax and UTP) for home networks that may include 2 fibers.
• #107: The choices for singlemode fibers is also many, but so are the applications. Short links (metro, premises) generally use G.652 – just plain SM fiber as has been used for decades. For longer links and DWDM, there are specialized fibers available with lower attenuation at 1550nm and dispersion characteristics that are more suitable for longer distances, higher speeds and DWDM – dense wavelength division multiplexing. Before buying cable, consult with the equipment and cable vendors for their recommendations for your application.
• #109: Because of the wide variety of conditions to which they are exposed, optical fibers have to be encased in several layers of protection. The first of these layers is the primary buffer coating, a thin protective coating made of ultraviolet curable acrylate ( a plastic), which is applied to the glass fiber as it is being manufactured. This thin coating provides moisture and mechanical protection. The next layer of protection is a buffer, that is typically extruded over this coating to further increase the strength of the single fibers. This buffer can be either a loose tube or a tight tube. The next layer is a strength member, usually an aramid fiber, that can be used for pulling the cable. Finally, the entire cable is covered by a jacket designed to withstand the environment into which the cable is going to be installed. Note: When you strip fiber for cleaving, splicing or termination, you only strip off the plastic buffer coating leaving the solid glass fiber.
• #110: There are 4 major types of cables. Tight buffer comes in simplex or zipcord versions. Distribution cable has multiple 900 µm coated fibers surrounded by aramid fiber (Kevlar) strength members and a jacket. Loose tube cable has one or more plastic tubes with 250 µm coated fibers inside the tubes, surrounded by strength members. The tubes are often filled with gels or powders to block water entry. Breakout cable is simply a number of simplex cables in a common jacket, designed for indoor use that requires a rugged cable. Tight buffer (a zipcord is shown), distribution and breakout cables are used indoors. Outdoors, loose tube cable is used to allow filling the cable with water-blocking materials to protect the fibers from moisture.
• #112: Zipcord is just two simplex cables attached by a thin web for convenience, since connections require two fibers transmitting in opposite directions Zipcord is primarily used for patchcords or short indoor runs It consists of two 900 micron tight buffered fibers, color coded, surrounded by aramid strength members and covered by a PVC jacket rated for flammability Zipcord can be used for patchcords on patch panels or connecting up equipment. It can be laid in cable trays or even pulled by the strength members in conduit.
• #114: Distribution cable is the most popular backbone cable, since it offers relatively high fiber density in a small cable that is easy to install Has many bundled 900 micron tight buffered fibers, color coded inside aramid strength members and covered by a PVC jacket rated for flammability Each individual tight buffered fiber can be directly terminated, but terminations are not well protected like in zipcord, simplex or breakout cables, so it should be terminated inside a patch panel or wall-mounted box.
• #116: Breakout cable consists of bundles of simplex cables inside a flame-retardant jacket for use indoors. It is a very rugged cable for harsh indoor environments but is bulky and heavy compared to distribution cable However you may directly terminate breakout cables for connections to equipment since individual fibers are protected inside individual jackets, making it very convenient to use.
• #117: Loose tube cable is used for most outdoor installations because it has higher strength for pulling long distances or aerial installs and can provide protection from moisture or water penetration and animal penetration. Fibers are loose in tubes for isolation from installation stress and allow the tubes and cable to be filled with water-block in gel or dry water-blocking powder. Loose tube cable contains one or more tubes which contain several individual fibers (usually up to 12). The fibers are only coated with the primary buffer coating of 250 micron diameter to keep the cable size small and are color-coded. Loose tube cables can be pulled with very high tension or suspended aerially without harming fibers. Loose tube cable is usually spliced and the splices stored in splice closures or the fibers are spliced to pigtails for termination. If directly terminated, the installer must use a breakout kit to terminate the fibers as they are too fragile to handle safely. Breakout kits use 1 mm tubing called furcation tubing to sleeve the fibers before termination.
• #118: Armored Cable Armored cable adds metal or dielectric armor over cable, typically to prevent rodent damage in direct burial installations Armored cable can be used in any application to prevent crushing, even indoors, for example in raised floors where there are many heavy cables already installed and potential of crushing the cable is high Armored cable is more difficult to prepare for splicing or termination, but ripcords are included to allow slitting the armor for relatively easy removal.
• #119: Ribbon cable allows maximum density of fibers - the smallest cable with the most fibers. Ribbons have 12-24 individual fibers held by plastic tape and ribbons are stacked in cable tubes or slotted cores. Twelve 12-fiber ribbons can fit in a 5 m (1/4”) square section and up to six of these ribbon assemblies are possible in a slotted-core cable design less than 25 mm (1”) diameter. Ribbon cables have been made with over 2000 fibers. This offers the maximum density of fibers but the cables are hard to work with. They require special tools to separate and strip the fibers and ribbon splicers to fusion splice them. Occasionally special mechanical splices using v-groove chips are used.
• #120: A new development in fiber optic cable is the microcable. Microcables use BI fiber with smaller diameter buffer coatings, typically ~200 microns instead of 250 microns so more fibers can be squeezed into a single buffer tube, typically 24 instead of 12. This 144 fiber cable is less than 10mm diameter – smaller than a pencil - and is designed to be “blown” into a 12mm microdcut. The microducts can be installed by microtrenching (covered later in this course) or by insertion into ducts that already have cable in them.
• #121: These small cables can be installed in a variety of microducts including installing ducts with subducts for future expansion. This is like adding multiple innerducts when regular trenching is used to install conduit or ducts in traditional underground installations.
• #122: New cable constructions using smaller fiber buffer coatings and flexible ribbons have led to a new category of extremely dense cables. These cables are heavier and much less flexible than standard cables and require special installation practices, bigger manholes (the bend radius is quite large!) and special splicing procedures. Installers are urged to get training from manufacturers on their cable types before attempting installation of these cables.
• #123: Aerial installations can use standard OSP cable lashed to a messenger for support. As shown here, sometimes several cables will be lashed to the same messenger and even splice cases can be hung from the messenger. Figure 8 style cable (not to be confused with the installation method of "figure 8-ing" cable when pulling) includes a strength member attached to the cable with a common jacket. ADSS cable is a special high tension cable with heavy jacket that can be installed without a messenger but requires special hardware that has soft inserts for mounting on poles and hangers at the cable ends. OPGW is a wire-armored cable where the wire can carry high voltage power. All these different types of cables have special hardware which is covered in the installation PPT.
• #124: OPGW stands for “optical power ground wire.” It is used as ground cable for high-voltage power lines, made possible by the fact that fiber is immune to electrical interference The fibers are inside plastic loose tubes which are then inside welded hermetic metal tube, covered by conductors that are also strength members. Finally, all is covered by a weather-resistant jacket. The fiber is suspended from the power line towers then cable ends are brought to the ground to be spliced. Splice closures are mounted on the power line towers.
• #125: Blown Fiber - An Alternative To Cable: Instead of installing cables with fibers, you can install “cable” with open tubes. After installation, you then blow special fiber into the tubes using high pressure compressed gas. Blown fiber allows easy installation of the cable and easy later installation of the fibers, but special equipment is required for the installation, limiting it to contractors who have made significant investment into equipment and training. Blown fiber requires special fibers designed to be installed this way. They are smaller and coated with a special coating that floats better in the tubes with less friction. It can be used with multimode or singlemode fibers or a combination of the two. Cable designs can have one or several tubes. More tubes means easy upgrades or separating fibers at junctions to go in different directions, a difficulty with normal cables where tubes are not easily split except in breakout cables. Blown fiber offers easy upgrades - blow out old fibers and blow in new ones. Blown fiber installations are more expensive initially but allows flexibility for future upgrades not possible with other cable designs.
• #126: Tight buffer cable is made in three versions: Single fiber, usually used for patchcords Zipcord used as a duplex patchcord for connecting transceivers or connecting devices together Breakout cable, which is simply a number of single fiber cables inside a jacket used for indoor cabling where each fiber is terminated and routed directly to a connection. Distribution cable is an indoor cable that has multiple fibers with a 900 micron buffer coating which can be terminated directly, but lacking the protection of the tight buffer cables above, must be protected by enclosure in a patch panel or box. Loose tube cable comes in many varieties for underground, aerial or even underwater applications. It can also be armored for protection against rodents for direct burial. Ribbon cable puts 12 or 24 fibers in ribbons and stacks the ribbons to make a small diameter very high fiber count cable. Applications are similar to loose tube designs.
• #127: Specifying the proper cable requires two major considerations: 1. How the cable will be installed. 2. What environment it will be facing after installation. These are simply guidelines to consider when looking for a cable for any particular installation. Different manufacturers have different cable designs for applications - and maybe different designs than other manufacturers. Therefore it is preferable to talk to several manufacturers when choosing a cable, especially in unusual situations.
• #128: Bending Radius Limits The normal recommendation for fiber optic cable bend radius is to ensure the minimum bend radius under tension during pulling is 20 times the diameter of the cable. When not under tension, the minimum recommended long term bend radius is 10 times the cable diameter. There is a Fiber U course on this.
• #129: Cable jackets for indoor cables may be color-coded in the scheme given above and covered in TIA-598. Colors are optional, however, so one may find cables in any color that was specified by the buyer. Outdoor cables are generally black due to the carbon black in the jacket to prevent UW from the sun damaging the cables.
• #130: Individual fibers or buffer tubes are color-coded in this 1-12 scheme. If there are more than 12 of any item, binder tape or tracer colors on a buffer tube, for example. In a loose-tube cable, 12 color-coded buffer tubes with 12 fibers each creates a 144 fiber cable. Ribbons follow the same color codes. Since these are only guidelines, alternatives may be seen. When splicing it is normal to splice like colored fibers.
• #131: Choice of cables Other factors to be considered when choosing a fiber optic cable are: 1. Current and future bandwidth requirements. 2. Acceptable attenuation rate. 3. Length of cable. 4. Cost of installation. 5. Mechanical requirements (ruggedness, flexibility, flame retardance, low smoke, cut- through resistance). 6. UL/NEC requirements. 7. Signal source (coupling efficiency, power output, receiver sensitivity). 8. Connectors and terminations. 9. Cable dimension requirements. 10. Physical environment (temperature, moisture, location). 11. Compatibility with any existing systems.
• #132: Breakout cable is larger and more expensive, but for short distances it offers more ruggedness and the ability to be terminated without the need for patch panels or termination boxes, saving that cost. For most backbone cables, distribution cables have a smaller size for the number of fibers, easing pulling of the cable, and are terminated in patch panels or boxes to protect the fibers. Remember that indoor cables must meet UL ratings!
• #133: Inspectors are not inspecting fiber for electrical safety (unless the cable is conductive), but are inspecting for conformance with fire codes. Outside the US, use appropriate local codes.
• #134: All outdoor cables are loose tube to allow inclusion of water-blocking compounds. Most are gels but some dry water-blocking cables are available (using materials developed for disposable diapers!) Outdoor direct buried installations will either be armored or installed in conduit to prevent rodent (or other critter) damage. Cables pulled through conduit must be chosen for the proper pulling tension, properly lubricated and pulled with some form of limiter (breakaway swivel or tension-controlled puller).
• #135: Hybrid or composite cables? These terms can be confusing as literature often mixes them up! These are the current TIA and ISO definitions. Adding to the confusion is the ISO has changed “hybrid” to mean fiber + conductors in some standards. Better to use the term hybrid for cables with different types of fibers or fibers and electrical conductors and then carefully define what the cable contains. These two types of cables are often confused in standards. The distinction between hybrid and composite cables has flipped several times in the history of fiber optics and differed among standards bodies. A hybrid cable originally meant a cable with two types of fibers, usually MM and SM, or a hybrid patchcord with, for example, a SC connector on one end and LC on the other end. A composite cable was defined per the US National Electrical Code: NEC Article 500.8(F) “Optical fiber cable contains conductors that are capable of carrying current (composite optical fiber cable)" More recent standards like the ones from IECA (Insulated Cable Engineers Association) use the term hybrid for cables with fiber and conductors. Two examples: Hybrid Cables and  FTTA cables
• #136: This is a basic fiber optic training program for FOA-Approved Schools to teach classes for the FOA CFOT certification. The program was developed by The Fiber Optic Association with inputs from many FOA instructors. It’s is based on 30+ years of experience in the business, including starting one of the first fiber optic test equipment companies and training thousands of fiber optic installers.
• #137: Fiber optic connectors and splices are used to join or couple two fibers together. Splices, however, are used to connect two fibers in a permanent joint. Connectors are also used to to connect fibers to transmitters or receivers, and, of course, connectors are designed to be demountable. While they share some common requirements, like low loss, high optical return loss and repeatability, connectors have the additional requirements of durability under repeated matings. Splices, meanwhile, are expected to last for many years through sometimes difficult environmental conditions, perhaps underground, underwater or suspended from aerial cables.
• #138: Fiber optic connectors and splices are used to join or couple two fibers together. Splices, however, are used to connect two fibers in a permanent joint. Connectors are also used to to connect fibers to transmitters or receivers, and, of course, connectors are designed to be demountable. While they share some common requirements, like low loss, high optical return loss and repeatability, connectors have the additional requirements of durability under repeated matings. Splices, meanwhile, are expected to last for many years through sometimes difficult environmental conditions, perhaps underground, underwater or suspended from aerial cables.
• #139: Most connectors work by simply aligning the two fiber ends as accurately as possible and securing them in a fashion that is least affected by environmental factors. The most common method is to have a cylindrical ferrule with a fiber-sized hole in the center, in which the fiber is secured with an adhesive. Note that fiber optic connectors are mainly “male” style with a protruding ferrule. Termination involves gluing the fiber into the ferrule and polishing the end of the ferrule. Other connector techniques like using lenses to create an expanded beam and alignment of bare fibers in a fixture like a mechanical splice have been tried and most have been abandoned for all but some very specialized applications. Connectors have used metal, glass, plastic and ceramic ferrules to align the fibers accurately, but ceramic seem to be the best choice. It is the most environmentally stable material, closely matching the expansion coefficient of glass fibers. It is easy to bond to glass fiber with epoxy glues, and its hardness is perfect for a quick polish of the fiber. As volume has increased, ceramic costs have become the lowest cost material for connector ferrules. The third part of a connection is an alignment sleeve, generally made as a split cylindrical sleeve. These sleeves can be made of molded glass-filled plastic, (metal (phosphor-bronze) or ceramic. The plastic ones are not recommended as they wear out quickly and get the connectors dirty. Metal ones are OK for MM but ceramic ones are recommended for all SM applications and testing.
• #140: Most connectors are physical contact (PC) connectors where the ferrules align the fibers and make contact. An alternative method of making connections is to expand the beam of light from the fiber using lenses. By creating a collimated beam (that is it stays the same size instead of expanding in a cone like the light exiting a fiber) that is several mm diameter, the alignment of the fibers becomes less critical and the larger beam is less sensitive to dirt. However, these are much more complicated optical assemblies and will typically have much higher loss and reflectance. Their higher cost makes them prohibitive for most applications, but they are still used in some MIL and aerospace connectors.
• #141: This MPO connector is called an array fiber optic connectors with all the fibers lined up in a single composite ferrule. The MPO connector was designed to provide very high density of fibers in patch panels and to connect to transceivers for parallel optics transmission (multiple channels of data over individual fibers) needed for very high speeds on multimode fiber. MPOs are also used in prefab cable plants (cables made in a factory and shipped to the job site for installation. There are multimode versions with flat ferrules and singlemode versions with angled ferrules to reduce reflectance. MPOs are not field terminated except by splicing onto cables. They can be difficult to test unless you have dedicated test sets. The pin/hole alignment method creates a polarized connector. In addition, it has two keying options. All told, there are about 32 variations of the connector meaning that users must know exactly which types of connectors can be mated.
• #142: Fiber optic connectors can have several different ferrule shapes or finishes, usually referred to as polishes. Air Gap: Early connectors, because they did not have keyed ferrules and could rotate in mating adapters, always had an air gap between the connectors to prevent them rotating and grinding scratches into the ends of the fibers. The air gap between the fibers causes a reflection when the light encounters the change in refractive index from the glass fiber to the air in the gap. This reflection was called back reflection or optical return loss, now called reflectance, and can be a problem in laser based systems. Flat PC (Physical Contact): Connectors use a number of polishing techniques to insure physical contact of the fiber ends to minimize reflectance. Beginning with the ST and FC which had keyed ferrules, the connectors were designed to contact tightly, what we now call physical contact (PC) connectors. Reducing the air gap reduced the loss and reflectance (very important to laser-based singlemode systems ), since light has a loss of about 5% (~0.25 dB) at each air gap and light is reflected back up the fiber. While air gap connectors usually had losses of 0.5 dB or more and reflectance of -20 dB, PC connectors had typical losses of 0.3 dB and a reflectance of -30 to -40 dB. PC: Soon thereafter, it was determined that making the connector ferrules convex would produce an even better connection. The convex ferrule guaranteed the fiber cores were in contact. Losses were under 0.3dB and reflectance -40 dB or better. APC: The final solution for singlemode systems extremely sensitive to reflections, like CATV or high bitrate telco links, was to angle the end of the ferrule 8 degrees to create what we call an APC or angled PC connector. Then any reflected light is at an angle that is absorbed in the cladding of the fiber. Reflectance is >~-50 to 60dB
• #143: You usually hear someone say “connector loss” when discussing the performance of a connector. But a single connector has no “loss,” because it is a component designed to make a joint between two fibers and loss only occurs when mating two connectors. That’s how we test connectors – we mate them to another connector, usually a high-quality reference connector (<0.5dB) in good condition ( and really clean.) It is not “connector loss” we measure, it is “connection loss,” the loss of a mated pair of connectors. If we are testing a unknown connector with a known good reference connector, it’s understandable that we would say the loss of that connector is what we measure, but it’s not technically correct. For example, if we test the same connector against another reference connector of lesser quality, it’s likely to show higher loss. Which result is correct? What about connecting to active devices like a transmitter source or receiver detector? Not all sources or detectors are the same, so there is no way to set test conditions for these devices. However, a connector that tests good when mated to another connector will always provide low loss connections to active devices.
• #144: Connection or splice loss is minimized when the two fiber cores are identical and perfectly aligned, the connectors or splices are properly finished and no dirt is present. Only the light that is coupled into the receiving fiber's core will propagate, so all the rest of the light becomes the connector or splice loss. End gaps cause two problems, insertion loss and reflectance. The emerging cone of light from the connector will spill over the core of the receiving fiber and be lost. In addition, the air gap between the fibers causes a reflection when the light encounters the change in refractive index from the glass fiber to the air in the gap. The end finish of the fiber must be properly polished to minimize loss. A rough surface will scatter light and dirt can also scatter and absorb light. Since the optical fiber is so small, typical airborne dirt can be a major source of loss. Whenever connectors are not terminated, they should be covered to protect the end of the ferrule from dirt. One should never touch the end of the ferrule, since the oils on one's skin causes the fiber to attract dirt. Before connection and testing, it is advisable to clean connectors with lint-free wipes moistened with isopropyl alcohol. Two sources of loss are directional; mismatches in numerical aperture (NA) and core diameter caused not by the connector but the fibers being joined. Differences in these two will create connections that have different losses depending on the direction of light propagation. Light from a fiber with a larger NA will be more sensitive to angularity and end gap, so transmission from a fiber of larger NA to one of smaller NA will be higher loss than the reverse. Likewise, light from a larger fiber will have high loss coupled to a fiber of smaller diameter, while one can couple a small diameter fiber to a large diameter fiber with minimal loss, since it is much less sensitive to end gap or lateral offset.
• #145: Reflectance is the term now generally used instead of optical return loss or back reflection. "Back reflection" is redundant – reflect means sending back to source. Light reflects at surfaces between materials of different indices of refraction. A glass to air interface yields about a 4% reflection. Reflectance in fiber optic cabling occurs at fiber optic joints, where connectors can have a small amount of air between dry surfaces causing reflections. Splices have lower reflectance due to the fusing of the fibers or using index matching fluid in mechanical splices. Domed (PC or physical contact) connectors have fiber end faces can minimize air to reduce reflectance .
• #146: What’s important in the performance of a fiber optic connector? Of course, the most important specification for a connector is loss - the less light loss the better. But we also want the connector to be repeatable - in two ways. If we terminate a lot of connectors, we need to be assured that most have about the same loss, so we can plan on that loss for calculating the likely loss of the cable. (We’ll look at power budgets later.) We also want it repeatable if we disconnect it and reconnect it many times, so we know the loss will not change when we reconnect it. Connectors must be designed to meet their specs over the environmental changes it will see. It’s no problem indoors, but outdoors, temperature and humidity can change, and think about connectors on an aircraft and the vibration they must endure! Reliability means maintaining low loss over its lifetime. Reflectacne is very important for Laser sources, as light reflected back can disturb the performance of the laser, plus reflected light can create optical “background noise” which confuses receivers. Ease of termination and cost probably need no further explanation.
• #147: Here are four generations of fiber optic connectors, showing how their size in particular has shrunk. On the bottom is the Deutsch 1000, one of the first commercial connectors. It held the fiber by vise-type action and connected fibers in a plastic lens with oil in it to assist the connection. Above it is a Biconic, AT&T’s first commercial connector and the first to work with SM fiber. The Biconic ferrule is a glass-filled thermoplastic. For The SC from NTT in Japan was one of the first to use ceramic ferrules and have very low loss, even with SM fiber. It’s still widely used today. The LC uses a very small ceramic ferrule to allow the connector to be so small, and it offers equal or perhaps even better performance than the SC.
• #148: ST (an AT&T Trademark) is probably still the most popular connector for multimode networks, like most buildings and campuses. It has a bayonet mount and a long cylindrical ferrule to hold the fiber. Most ferrules are ceramic, but some are metal or plastic. And because they are spring-loaded, you have to make sure they are seated properly. If you have high loss, reconnect them to see if it makes a difference. SC is a snap-in connector that is widely used in singlemode systems for it's excellent performance. It's a snap-in connector that latches with a simple push-pull motion. It is also available in a duplex configuration. LC is a new connector that uses a 1.25 mm ferrule, half the size of the SC. Otherwise, it's a standard ceramic ferrule connector, easily terminated with any adhesive. Good performance, highly favored for singlemode and practically the only connector used on transceivers for gigabit/s and above.   MPO is a 12 fiber connector for ribbon cable. It's main use is for prefab cable assemblies which can be installed as complete systems to avoid field terminations. It is also used on transceivers for parallel optics – multimode systems that use several channels over individual fibers to overcome the bandwidth limitations of multimode fiber.
• #149: Since so many connectors use the 2.5 mm ceramic ferrule, you can cross-mate them with adapters like these. You can get adapters for ST or SC to FDDI and ESCON also. There are some other hybrid types that claim to connect connectors with dissimilar ferrules, like SC to LC, but their alignment may not be adequate for low loss.
• #150: Since most fiber optic datalinks use two fibers transmitting in opposite directions, a duplex connector is an advantage, as it connects both fibers at once and maintains the polarity of the link (connecting transmitter to receiver.) The first connectors were designed as part of network standards, FDDI or Fiber Distributed Data Interface, and ESCON, the IBM peripheral network for mainframes, both introduced in the late 1980s. Both used ST/SC/FC compatible 2.5 mm ferrules in custom bodies. Later, the competition for "small from factor" connectors in the late 1990s spawned three new and very different connectors, the MT-RJ, Panduit Optijack and 3M Volition. The MT-RJ used a molded rectangular plastic ferrule with alignment by steel pins. The Optijack used regular ST/SC ferrules but closer together to fit in a body the same size as a RJ-45 copper connector. The Volition was a real deviation, using bare fibers aligned by V-grooves, a simple mechanical splice technique. None of these made any impact on the market. Another duplex solution was to join two regular connectors, SCs or LCs, into a duplex assembly. These are now the duplex solutions of choice. The latest type of duplex connector is the vertical ferrule designs like the MXC (shown) and the SN which offer very high density in patch panels, much better than the duplex LC.
• #151: With over 100 unique connector designs being offered in the history of fiber optics, lots of unusual designs have been tried. The Biconic was AT&T's first design, based on a molded plastic ferrule in a cone shape. It had to be polished carefully to an exact length measured by a gage provided in the termination kit. The Deutsch 1000 was a bare fiber connector with the fibers mating in a lens with index matching fluid. That connector was very large and heavy. The MT-RJ was very small, but problematic for field termination (mainly prepolished/splice type) and the alignment pins would wear out quickly. Testing was another problem. The Volition was a real deviation, using bare fibers aligned by V-grooves, a simple mechanical splice technique, but was not well-received.
• #152: There are always new connector designs being introduced – there have been more than 100 in the history of fiber optics. The vertical duplex design seems to be getting some interest for its higher density in patch panels.
• #153: Many versions of ruggedized fiber optic connectors have been built, including the MIL 38999 connector with fiber ferrules and the TFOCA, a hermaphrodic connector – it mates with another connector of the same type instead of having male/female or plug/jack variants.
• #154: Since fiber optics began, over 100 different styles of connectors have been offered commercially. Most have faded from use or never became popular, so only a few connector styles dominate today’s networks. Multimode installations generally use the LC or SC connector, with some legacy systems still using STs. Parallel optics used for high speed multimode networks and prefabricated cabling systems use MPOs, but polarity remains a complicated issue with so many fibers. LC has become the standard for transceivers at 1Gb/s and faster. Singlemode applications use mostly SC or LCs, but many installations are still using older designs.
• #155: The “normal” colors for connectors are as shown, but other colors are sometimes used, especially for multimode. The thing to remember is green means APC and it CANNOT be mated with other types of connectors. A new variant for MM OM5 fiber is lime green, but it may conflict with the APC green and may not be used.
• #156: Epoxy/Polish: Most connectors are the simple “epoxy/polish” type where the fiber is glued into the connector with epoxy and the ferrule end polished with special polishing film. These provide the most reliable connection, lowest losses and lowest costs, especially if you are doing a lot of connectors. The epoxy is cured in an oven in only a few minutes. Epoxy/polish termination is used for all factory-made patchcords and cable assemblies. Quick Setting Adhesives: These connectors use a quick setting adhesive with a curing agent to replace the epoxy. “Hot Melt”: This is a 3M trade name for a connector that already has the epoxy (actually a heat set glue) inside the connector. You insert the connector in a special oven. In a few minutes, the glue is melted, so you remove the connector, insert the fiber, let it cool and it is ready to polish. Crimp/Polish: Rather than glue the fiber in the connector, these connectors use a crimp on the fiber to hold it in. Expect to trade higher losses for the faster termination speed. Crimp/cleave connectors only cleave the fiber for termination - no polishing is required. Losses are higher as a result. Splice -on connectors (also called prepolished/splice connectors): Many manufacturers offer connectors that have a short stub fiber already epoxied into the ferrule and polished perfectly, so you just cleave a fiber and insert it like a splice or fusion splice it to a fiber. Fusion SOCs require a fusion splicing machine with adapters. Mechanical SOCs may require a special termination kit.
• #157: Let’s examine the process of terminating a fiber with a typical epoxy/polish connector. Since this is the most basic process of fiber termination, it's important for students to understand it, but it is probably a skill few will ever use in their work. Start by examining the parts of the connector you are terminating to see what parts are needed and how they are assembled.
• #158: Adhesive/polish connectors have these basic steps. The instructions that come with the connector parts will give directions for stripping the fiber, adhesive types to use, and polishing instructions. Most connectors use epoxies or other adhesives to hold the fiber in the connector ferrule and polish the end of the fiber to a smooth finish. Follow termination procedures carefully, as they have been developed to produce the lowest loss and most reliable terminations. Use only the specified adhesives, as the fiber to ferrule bond is critical for low loss and long term reliability! We've seen people use hardware store epoxies, Crazy Glue, you name it! And they regretted doing it. Only adhesives approved by manufacturers or other distributors of connectors should be used. If the adhesive fails, not unusual when connector ferrules were made of metal, the fiber will "piston" - sticking out or pulling back into the ferrule - causing high loss and potential damage to a mated connector. The polishing process involves three steps which only takes a minute: "air polishing" to grind down the protruding fiber, polishing on a soft pad with the fiber held perpendicular to the polishing surface with a polishing puck and a quick final fine polish.
• #159: At this point, inspect the polished end of the ferrule with the microscope to see that the epoxy is completely removed and that the tip is smooth and free of scratches. There are many inspection microscopes available with magnifications of 100X to 400X. Higher magnification may not be better, as it tends to make you more critical of scratches and imperfections. Lower magnifications work just fine.
• #160: Anaerobic adhesives are used instead of epoxy for a quick termination. There are several ways that are used to apply quick curing adhesives. 1. The method we recommend uses no accelerator. The adhesive we recommend (Loctite(R) 648) is an adhesive that will cure in 3-5 minutes without an accelerator, depending on the ambient temperature. If you are making more than one termination, you do not need an accelerator at all. This process will be described without using the accelerator and then the use of the accelerator will be described. 2. Inject the adhesive into the connector with a syringe then insert the fiber in the connector. Spray an accelerator on the tip of the ferrule to make the adhesive cure at the end quickly to allow immediate polishing. After spraying the tip, residue will be left on the connector ferrule that must be cleaned. Most accelerators are highly flammable, requiring care.     3. Inject the adhesive into the connector with a syringe, dip the fiber in an accelerator solution then insert the fiber in the connector. With this method, you must work fast and make sure the fiber is inserted rapidly or the adhesive will set before the fiber is fully inserted. Anaerobics do not leave the nice bead on the end of the ferrule, so you have to be more careful when cleaving the fiber and air polishing. Otherwise the polish process is the same.
• #161: 3 M Hot Melt connectors use a “hot melt” adhesive preloaded into the connector. The termination process involves heating up the connector until the adhesive becomes a liquid, then inserting the stripped and cleaned fiber. It is then set aside to allow the adhesive to cool and set before cleaving and polishing. The adhesive needs at least 1 minute in the oven to liquefy but after more than 10 minutes in the oven, it may not set when cooled, so the range of time in the oven is limited. To start terminations, turn the oven on and let warm up for at least 5 minutes. Insert a connector in the oven to let it warm up. While waiting for the connector to heat up, strip and clean the fiber. Remove the connector from the oven and insert the fiber. Let the connector cool to set the adhesive. Cleave and polish as normal. If you have a problem with the cleave/polish process, you can usually reuse the connector. If you pull the fiber back about 1 mm (1/16”) you can reheat it and push it forward without problems. Hint: Make sure you have the termination instructions for the exact connector you are using before you start! Different styles of connectors have slightly different termination processes. Caution: The Hot Melt oven operates at twice the temperature of the epoxy curing oven - 245 - 270 degrees C. or 473 - 518 degrees F. º C. It can cause burns if the metal parts are touched while hot. Be extremely careful with the oven! NOTE: Paper catches fire at 451 degrees F, so don't rest anything on the oven.
• #162: Most manufacturers now offer connectors that have a short stub fiber already epoxied into the ferrule and polished perfectly with a mechanical splice in the back of the connector, so you just cleave a fiber and insert it like a mechanical splice. It’s important to follow the manufacturer’s process exactly when terminating prepolished/splice connectors. The Corning UniCam process is: Insert connector in fixture Strip, clean and cleave fiber Insert fiber in connector Cam and crimp Remove from fixture Slide boot onto connector Photo courtesy Corning Cabling Systems
• #163: Fusion splice-on connectors (SOC) have a short fiber in the back that can be fusion spliced onto a fiber for termination. They offer significantly lower loss and higher reliability at at lower cost than prepolished splice connectors which include a mechanical splice. Most can be spliced in standard fusion splicers with a special connector holder. Details on the termination process are in the fusion splice section (next.)
• #164: Here is a possible alternative - installing a prefabricated or pre-terminated system. You design the network on CAD, give the information to a manufacturer and they provide a complete modular system. Some manufacturers terminate the cable in standard connectors (easier with small form factor connectors) and cover it with a pulling boot. The downside is they require extra care in design to get cable lengths correct and in installation to avoid damaging connectors and the higher loss for multifiber connectors used in many modular systems can cause problems with the total cable plant loss. Photos courtesy Nexans & Corning.
• #165: Many FTTH systems now use prefabricated cables for the drop to the house. Crews come into the neighborhood and install the drop closures on poles or in underground vaults and splice the fibers into the backbone fiber network that terminates in the central office or a local PON distribution hub. The tech doing the actual FTTH install merely plugs in the cables between the closure and the optical network terminal and spends the bulk of the time connecting the user to telephone, Internet and TV services.
• #166: Many FTTH systems now use prefabricated cables for the drop to the house. Crews come into the neighborhood and install the drop closures on poles or in underground vaults and splice the fibers into the backbone fiber network that terminates in the central office or a local PON distribution hub. The tech doing the actual FTTH install merely plugs in the cables between the closure and the optical network terminal and spends the bulk of the time connecting the user to telephone, Internet and TV services.
• #167: While connectors are demountable, splices are permanent connections. Splicing is only needed if the cable runs are too long for one straight pull or you need to mix a number of different types of cables (like bringing a 48 fiber cable in and splicing it to six 8 fiber cables - could you have used a breakout cable instead?) And of course, we use splices for restoration, after the number one problem of outside plant cables, a dig-up and cut of a buried cable, usually referred to as “backhoe fade” for obvious reasons! They may have different uses, but the basic specifications for splices are the same as for connectors. Splices may be fusion, where the fibers are welded together using a machine which produces a splice that looks like the one shown on the left in the lower photo, or a mechanical splice, several examples of which are shown in the lower photo.
• #168: There are two types of splices, fusion and mechanical. Fusion splicing is done by welding the two fibers together, usually with an electrical arc with an automated splicer which aligns the fibers exactly. It has the advantages of low loss, high strength, low reflectance (optical return loss) and long term reliability. Mechanical splices use an alignment fixture to mate the fibers and either a matching gel or epoxy to minimize back reflection. Some mechanical splices use bare fibers in an alignment bushing, while others closely resemble connector ferrules without all the mounting hardware. While fusion splicing normally uses active alignment to minimize splice loss, mechanical splicing relies on tight dimensional tolerances in the fibers to minimize loss.
• #169: Fusion splicers are expensive, highly automated machines that do most of the work. The operator uses a high quality clever to prepare the fibers and inserts them into the jaws of the splicer. The machine automatically aligns the ends, makes the splice and even gives an estimate of the loss. The operator then places the splice in a holder which also seals it and inserts it in a splice tray. While fusion splicers are expensive, each splice is cheap. So if you are doing lots of splices, fusion is more cost effective.
• #170: Fusion splicers are expensive machines, fully automated machines that do most of the work. The operator uses a high quality clever to prepare the fibers and inserts them into the jaws of the splicer. The machine automatically aligns the ends, makes the splice and even gives an estimate of the loss. The operator then places the splice in a holder which also seals it and inserts it in a splice tray. Fusion splicers can splice one fiber at a time or all 12 fibers in a ribbon. While fusion splicers are expensive, each splice is cheap. So if you are doing lots of splices, fusion splicing is more cost effective.
• #171: Ribbon Fiber Fusion Splicing In today’s telecom world, many of the splices are made with ribbon splicers, as one can splice 12 fibers in little more time than one can splice a single fiber. Plus the 12 fibers are placed in one protection sleeve and inserted into one slot on a splice tray, saving lots of space in splice closures. Ribbon splicing tools are more expensive but more automated - not just the splicing machine itself, but also strippers and cleavers. Lower cost, faster productivity, less space - it’s easy to see why ribbon splicing is so popular.
• #172: The fusion splicing process is as follows: Strip the plastic buffer coating from the glass fiber Clean the fiber Cleave the fiber Place fiber in splicer properly Repeat with second fiber Close cover, start automated program Splicer competes splice and estimates loss Remove fibers and place protector over splice Heat shrink splice protector to seal bare fibers and the splice Ribbon splicing works in a similar fashion but all 12 fibers are handled at once.
• #173: Splice-on connectors (SOC) have a short fiber in the back that can be fusion spliced onto a fiber for termination. They offer significantly lower loss and higher reliability at at lower cost than prepolished splice connectors which include a mechanical splice. Most can be spliced in standard fusion splicers with a special connector holder.
• #174: Mechanical splices have been offered in many varieties. All use some mechanical alignment fixture, v-groove, glass capillary, soft elastomeric or metal clamp v-groove, etc. and some means of securing the fibers in the splice. Mechanical splices are more common with multimode fiber but are used for singlemode restoration until fusion splicing can be done. The three examples of alignment shown here are the capillary as used on the Ultrasplice, a ribbon V-grove splice using glass or silicon V-grooves and a 3M Fiberlok which clamps the fiber in a stamped metal element. Mechanical splices themselves are more expensive per splice than fusion splicing which only has the cost of the splice protector, but the equipment necessary is relatively inexpensive. So if you are only making a few splices, mechanical may be the less expensive choice.
• #175: Mechanical splices, like this Ultrasplice, use a mechanical alignment fixture, a glass capillary in this case, and some means of securing the fibers in the splice, clamps secured by nuts in this case.
• #176: In order to get good fiber optic splices or terminations, especially when using the pre-polished connectors with internal splices, it is extremely important to cleave the fiber properly. Cleaving is the process by which an optical fiber is “cut” or precisely broken for termination or splicing. Just like cutting glass plate, fiber is cut by scoring or scratching the surface and applying stress so the glass breaks in a smooth manner along the stress lines created by the scratch. Properly done, the fiber will cleave with a clean surface perpendicular to the length of the fiber with no protruding glass on either end (called a lip) and no surface roughness (hackle or mist.) A cleaver is a tool that holds the fiber under low tension, scores the surface at the proper location, then applies greater tension until the fiber breaks. Good cleavers are automatic and produce consistent results, irrespective of the operator. The user need only clamp the fiber into the cleaver and operate its controls. Some cleavers, especially the handheld ones, are less automated, for example requiring operators to exert force manually for breaking the fiber, making them more dependent on operator technique and therefore less predictable.
• #177: Splices always require protection from the environment and are often outdoors. Protection for splices is provided by a splice closure which contains trays for individual splices. Completed splices are inserted in a splice tray which goes in a splice closure. Incoming cables are secured to the closure for mechanical strength and sealed. Loose tubes on the cable are secured to splice tray so the bare fibers are only exposed inside the tray. The closure is sealed to protect fibers and splices from moisture, water and anything else in the outside environment. Closures can be buried underground, supported on aerial cables or whatever means of cable installation is being spliced.
• #178: Splice closures come in all varieties, shapes and sizes, to accommodate just about any application. Requirements including fitting in the location, handling all the fibers and – especially – sealing properly to protect the fibers and splices from the outside environment.
• #179: This is a basic fiber optic training program for FOA-Approved Schools to teach classes for the FOA CFOT certification. The program was developed by The Fiber Optic Association with inputs from many FOA instructors. It’s is based on 30+ years of experience in the business, including starting one of the first fiber optic test equipment companies and training thousands of fiber optic installers.
• #180: Note: Testing is one of the most difficult topics in fiber optics but techs need to understand it in order to make tests and interpret results. Techs should understand dB and dBm (and the differences between them), how cables are tested using insertion loss tests, what an OTDR tests and what information is in the OTDR trace display and how to calculate loss budgets and use them for comparing to test results. These topics are included in the FOA KSAs and in certification tests. Testing fiber optic cables, connectors and splices is primarily done at the terminated phase on installed cable plants and patchcords. There are three ways of testing these cables: Continuity testing with a visible light source - a LED or incandescent bulb in a fiber tracer or a higher power visible laser in a visual fault locator. Microscope inspection allows finding damaged or dirty connectors. Optical power is used to measure transmitter output and receiver input power, plus differences before and after losses. Insertion loss simulated the way the cable will be used by a transmission system, using a source (LED or Laser at the same wavelength(s) as the system) and optical power meter, with two reference cables. OTDR testing uses a unique property of fiber - backscatter - to create a “snapshot” of the fiber and find faults. For more on testing, see the FOA textbook on testing.
• #181: Continuity testing is done with a visible light source - high power visible laser in a visual fault locator. The high powered laser in a visual fault locator (VFL) can trace fibers long distances (3 miles or 5 km) and even find breaks. In a break, the light lost can be seen through the jacket of simplex or zipcord cable and tight buffered fibers. Visual fault locators can also be used to optimize mechanical splices and prepolished/splice type connectors by adjusting the fibers to minimize the visible light lost.
• #182: You can visually inspect the polished end of a connector ferrule with a microscope to see that the ferrule is properly polished, there are no cracks in the fiber and that the tip is smooth and free of scratches. And of course, you can see dirt and any other contamination on the end of the ferrule that can affect light transmission through the connection. There are many inspection microscopes available with magnifications of 100X to 400X. Higher magnification may not be better, as it tends to make you more critical of scratches and imperfections. Lower magnification works just fine. A note on eye safety: A microscope focuses all the power in the core of the fiber into eye! Some microscopes have filters to remove potentially harmful infrared light but always test for power in the fiber optic cable with a power meter before inspection.
• #183: Video microscopes use small video cameras and microscope lenses to provide a display of the view on a small video screen or the display of a PC or tablet. They offer more flexibility in magnification and image manipulation. Many of these also offer automatic inspection to international standards for cleanliness and produce pass/fail results. Video microscopes also allow for storing an image of the connector, valuable for documenting the condition of the connector at installation and for reference in the future.
• #184: Most fiber optic measurements are based on optical power. Optical power is typically measured to check transmitter source power output at the transmitter or receiver power at its input. Measurements of optical power are also used to measure loss of fibers, cables and other components. Optical power is measured with the power meter attached to the system cable or, when testing source output, a reference test cable. Power can be measured in "milliwatts or microwatts" which is a linear scale of power or more commonly "dBm" which is dB referenced to 1 milliwatt.. Transmitter Power: The amount of light coupled into a fiber by a source is measured by attaching a patchcord to the source, either a known good system patchcord or a reference test cable. The cable used must have a connector that mates with the transmitter and a fiber size the same as the system cabling (50/125, 62.5/125 or SM) since the coupled power is highly dependent on the core size of the fiber. The meter connector adapter must be the same as the cable to allow connection. Connect the meter, set the range on dBm or watts as appropriate and be sure to set the wavelength to the wavelength of the source, as the meter’s calibration will be different due to the wavelength sensitivity of its detector!
• #185: Receiver power is measured by removing the cable connected to the receiver input and connecting it to the power meter. Set the meter range on dBm and be sure to set the wavelength to the wavelength of the source, as the meter’s calibration will be different due to the wavelength sensitivity of its detector! Measure the power and record the results. Loss is measured as difference in power before and after cause of loss, which is discussed in detail later. Standards for optical power measurement are in (US)- TIA FOTP-95 and International IEC 61280.
• #186: dB is a log expression of the ratio of two power levels. That’s why it’s so convenient to measure loss. If we take a fiber and induce loss by stressing it in a bend as we are doing here, we can measure the loss in the power meter. Note what happens when we we add loss – the display on the meter starts at -20.0 dBm and goes down to -22.4 dBm. The power goes down and the display in dBm becomes a bigger negative number. The difference between the power measured before the loss was induced in the cable, -20dBm, and the final value after the loss, -22.4dBm, is the loss, -2.4dB. We calculate this by simple arithmetic: -22.4dBm – (-20.0dBm) = -22.4 + 20.0 = -2.4dB. It becomes dB because the difference in power is relative power, not absolute power referenced to 1mW or 0dBm. While the number is negative if we have a loss, we usually say “the loss is 2.4dB” while the meter says “-2.4dB.” If we use the “dB” range on the power meter and zero out the power before we induce the loss, the meter will begin at “0.0dB” and go to “-2.4dB” since it is a relative measurement. In your lab, try this exercise yourself.
• #187: dB or decibels is a measure of power originally named for Alexander Graham Bell and used for measuring acoustic signal level. Later it was used for measuring the power of radio signals. dB is used because it allows measurements over large dynamic ranges and loss or gain to be calculated by simple addition and subtraction. dB is a measure of optical power on a log scale, simplifying measurements over a wide dynamic range. Fiber optics typically uses power levels from +20 to -40 dBm, a range of 1,000,000 to 1! But that translates to 60 dB, an easier number to deal with. Absolute power is measured in dBm or dB referenced to 1 mw. Positive dBm means the power is greater than 1 mw, while negative numbers mean the power level is less than 1 mw. A nice thing about dB is loss is easily measured by subtracting the reference level for “0” dB from the measured value of the loss. That is, if you measure -20 dBm from the end of the reference cable, then -22 dBm when testing cables, the cable loss is 2 dB. If you measure loss with a power meter and source, the loss will be a negative number. If you measure loss with an OLTS or OTDR, it will be a positive number which may cause confusion because OTDRs show gainers as negative numbers. It's a matter of the definition adopted by international standards in fiber optics. There is a Fiber U course on "dB."
• #188: Loss Measurements In dB Most measurements in fiber optics are of optical loss – relative power measurements as defined above – expressed in dB. If we have loss in a fiber optic system, the measured power is less than the reference power, so the measured value in dB is a negative number. If we were measuring gain of a fiber amplifier, the measurement would be displayed as a positive dB. When we set the reference value, the meter reads “0 dB” because the reference value we set and the value the meter is measuring is the same. Then when we measure loss, the power measured is less, so the meter will read “ – 3.0 dB” for example, if the tested power is half the reference value. Although meters measure a negative number for loss, convention has us saying the loss is a positive number, so we say the loss is 3.0 dB when the meter reads – 3.0 dB. Here’s Where dB Can Get More Confusing: Instruments that measure in dB can be either optical power meters, optical loss test sets (OLTS) or OTDRs. If you make an optical loss measurement with an OLTS, optical loss test set or OTDR, it will probably display loss as a positive number, not negative like a power meter. There are two reasons for this: 1) Most users think of loss as a positive number or a number without a polarity sign at all, and 2) Around 2000, international standards committees changed the definition of dB to make loss a positive number. It creates more confusion since a gainer, common in OTDR tests, shows a "- dB." Fortunately, they did not redefine dBm the same way, so power measurement of powers higher than 1 mW or 0 dBm are still positive numbers. Power measured in dBm – dB referenced to 1 milliwatt, a universal standard, is correct: higher power is "+" and lower power is "-". So +3dBm is 2 mw and -3 dBm is ½ mw.  
• #189: dB or decibels is a measure of power originally named for Alexander Graham Bell and used for measuring acoustic signal level. Later it was used for measuring the power of radio signals. dB is used because it allows measurements over large dynamic ranges and loss or gain to be calculated by simple addition and subtraction. dB is a measure of optical power on a log scale, simplifying measurements over a wide dynamic range. Fiber optics typically uses power levels from +20 to -40 dBm, a range of 1,000,000 to 1! But that translates to 60 dB, an easier number to deal with. Absolute power is measured in dBm or dB referenced to 1 mw. Positive dBm means the power is greater than 1 mw, while negative numbers mean the power level is less than 1 mw. A nice thing about dB is loss is easily measured by subtracting the reference level for “0” dB from the measured value of the loss. That is, if you measure -20 dBm from the end of the reference cable, then -22 dBm when testing cables, the cable loss is 2 dB. In the previous slide, we mentioned that OTDRs and OLTSs show loss as a positive number, but if you measure loss with a power meter and source, it will be a negative number. That's because the reference power out of the test source is larger than the measured power after the loss which means after the loss is a smaller (more negative) number in power. Fortunately, the international standards did not mess with the definition of power in dBm, so lower power is always a more negative number.
• #190: Insertion loss testing simulates the way the cable will be used by the systems operating over it. A source, similar to the system source is used for inserting light into the cable under test. A meter is used to measure the source output and the loss when the cable under test is added. A double-ended test like this measures the loss of the fiber and connectors on both ends, plus anything in the middle. The source should match the system source in type (LED or laser) and wavelength (850 or 1300 nm for LEDs and 850, 1310 or 1550 nm for lasers.) The power meter needs to be calibrated to NIST (standards US national standards labs) and be able to measure at appropriate wavelengths (850, 1310 or 1550 nm.) Reference cables provide the test conditions for the loss test. They mate to the connectors on each end of the cable under test to measure the loss of those connectors. The reference cables are critical to making good measurements. They must mate with the cable under test, so connectors must match or mating adapters be available, and the fiber must be the same type (MM or SM) and core diameter.
• #191: One makes the measurement by calibrating the output of the source and storing this measurement as “0 dB” loss reference. The attach the cable to test and receive reference cable and meter as shown, then measure the loss. The loss will include the connection from the launch cable to the cable under test, the loss of the cable under test (attenuation of the length of fiber in the cable plus all splices and connections) plus the loss of the connection to the receiver cable. (The connection to the power meter has no loss because the detector gathers all the light from the end of the fiber.) The loss will be a negative number. Most standards call for measuring the "0 dB" reference at the end of the launch cable.
• #192: There are really two methods of insertion loss testing So far we have talked about testing installed and terminated cable plants, where we want to test the connectors on each end and everything in between. So we use a meter and source with two reference cables - one on each end. This test is defined by a standard OFSTP-14 (OFSTP = optical fiber standard test procedure) for multimode and OFSTP-7 for singlemode. International standards are included in IEC 61280. Another test, FOTP-171, uses only a launch reference cable and the cable under test and is sometimes called a "single-ended" test. This method allows testing a single cable like a patchcord from each end separately to help find out if either connector is bad. Summary: Patchcords are generally tested single-ended and reversed to check connectors on each end separately. Cable plants are tested double-ended since that simulates how a system will use the cable plant.
• #193: FOTP-171 or single-ended test uses only a single launch reference cable to test the cable. This method allows testing a single cable from either end to find out if one connector is bad. It’s main use is testing patchcords to insure both connectors are good, but it can also be used to troubleshoot installed cables where one connector is suspected of being bad. The 0 dB loss reference is made by connecting the power meter to the output of the launch cable and measuring the power output. The cable under test is connected to the launch cable and the meter. The loss measured is only the loss of the mated connectors and any loss of the fiber in the cable, usually very small when testing patchcords this way. The fact that the connector on the launch cable and the cable under test are mated directly to the meter, with it’s large detector, means that the connection loss to the meter is calibrated out of the loss test, allowing testing of only the connector mated to the launch cable.
• #194: OFSTP-14 and OFSTP-7, double ended testing OFSTP-14/OFSTP-7 are used for testing installed and terminated cable plants, where we want to test the connectors on each end and everything in between. So we use a meter and source with two reference cables - one on each end. Different standards exist for multimode and singlemode fibers due to the requirements for modal power control in multimode fiber. The big issue with this test method is how one sets the 0 dB reference.
• #195: OFSTP-14/OFSTP-7 offers three options on how one sets the 0 dB reference. What is the reason for three different methods? It's determined by the compatibility of the power meter and source with the connectors on the cable plant, and whether the connectors are normal ferrule-type connectors that use mating adapters or plug and jack type connectors. The 1-cable method works when the connectors on the cable plant are compatible to the connectors on the test equipment. The 2-cable method works when the connectors on the cable plant are not compatible to the connectors on the test equipment, but can be mated with mating adapters. The 3-cable method works when the connectors on the cable plant are not compatible to the connectors on the test equipment and are “male/female” or “plug/jack” types that cannot be randomly mated. Read "5 Different Ways To Test" on the FOA Guide website for a complete explanation.
• #196: OFSTP-14/OFSTP-7 offers three options on how one sets the 0 dB reference with 3 options for reference cables. With one reference cable (the launch cable) This method sets the “0 dB reference” with the power meter measuring the output of the launch cable directly, so that no connector loss is included when setting the reference. Then when testing a cable with both launch and receive cables, the loss includes the loss of both connectors on the cable under test and the loss of all the components in between. With two cables (launch and receive cables) This method sets the “0 dB reference” with the launch cable mated to the receive cable, so that one mated connector loss is included when setting the reference. Then when testing a cable with both launch and receive cables, the loss includes the loss of connectors on the cable under test and the loss of all the components in between, less the loss of the mated connectors included in the reference.
• #197: OFSTP-14/OFSTP-7 offers three options on how one sets the 0 dB reference. With three cables (launch, receive and a “golden” reference cables) This method sets the “0 dB reference” with the launch cable and the receive cable, plus a “golden” reference cable mated to them, so that two mated connector losses and any fiber loss in the third cable are included when setting the reference. Then when testing a cable with both launch and receive cables, the loss includes the loss of connectors on the cable under test and the loss of all the components in between, less the loss of the mated connectors included in the reference. What is the reason for three different methods? It's determined by the compatibility of the power meter and source with the connectors on the cable plant, and whether the connectors are normal ferrule-type connectors that use mating adapters or plug and jack type connectors. Read "5 Different Ways To Test" on the FOA Guide website for a complete explanation.
• #198: Insertion loss measurement accuracy depends on a number of factors that need consideration by any fiber optic tech: First, it is important to be confident of the quality and condition of reference cables. The do not need to be special cables, just good quality patchcords with low loss connectors (certainly under 0.5 dB, preferably under 0.3 dB, tested single ended per FOTP-171 against each other.) One should consider modal distribution in multimode fiber. Most standards call for a source of calibrated output with a mandrel wrap on the launch cable. This subject is covered in detail on the FOA website on the page Modal Effects on Multimode Fiber Loss Measurements. The calibration of the source output when setting the 0 dB reference is very important. The method of setting the reference must be documented as it affects the loss measured and the value used will affect all measurements. It's a good idea to recheck the reference level occasionally to ensure the source has not changed. The cleanliness of reference cables is vitally important. Dirt on the connectors when setting the reference that's cleaned off later can cause loss measurements to be lower or even read as a gain. Setting the reference with clean cables that get dirty over time will cause a systematic increase in loss.
• #199: Insertion loss measurement accuracy depends on a number of factors that need consideration by any fiber optic tech. Remember fiber attenuation is different for different wavelength sources. The stability of the meter and source is very important since if either changes, the 0 dB reference will change, and all loss measurement will be in error. If you don't know if your equipment is stable, connect your source to a power meter with a reference cable and let it run for a while. And check it with a good battery and a discharged one to see if the battery level affects the power output. Always make certain that your batteries are good before beginning testing!
• #200: What determines a Pass/Fail when testing the loss of a cable plant? You calculate a link loss budget to estimate the loss of the cable plant if it is installed properly. The link loss budget = sum of the losses of all the components in a cable plant Fiber attenuation Splice loss Connector loss Splitters are also added in PON networks
• #201: After a fiber optic cable plant is installed, it may be used with a number of different types of fiber optic networks. Computer networks, telephone signals, video links, and even audio can be sent on the installed fibers. Each network type has a requirement for the performance of the fiber optic cable link. Most simply specify the maximum loss in the link that can be tolerated, a function of component specifications and installation quality. Others also specify the bandwidth performance of the fiber which is determined by the specifications of the fiber chosen. Every fiber optic link has a maximum loss of a cable plant over which it can work. That loss is determined by the output power of the transmitter coupled into the fiber and the sensitivity of the receiver, all expressed in dBm, and the difference between is the maximum loss in dB. The loss of the fiber optic cable (in dB) it uses must be less than that maximum loss for proper operation. While every link installed must meet some maximum loss to allow operation of the network intended to use it, different networks may have different link margins. Therefore we use a different approach. The loss of the link is considered acceptable if it is less than standard maximum values calculated from the characteristics of the link installation. When testing an installed fiber optic cable plant, we use a different approach. The loss of the link is considered acceptable if it is less than the loss values calculated from the characteristics of components used in the cable plant installation.
• #202: Every fiber optic link has a maximum loss of a cable plant over which it can work. That loss is determined by the output power of the transmitter coupled into the fiber and the sensitivity of the receiver, all expressed in dBm, and the difference between is the maximum loss in dB. The loss of the fiber optic cable (in dB) it uses must be less than that maximum loss for proper operation. When testing an installed fiber optic cable plant, we use a different approach. The loss of the link is considered acceptable if it is less than the loss values calculated from the characteristics of components used in the cable plant installation. What causes the losses in the fiber optic cable? First the fiber itself. The next loss factor is the loss of terminations and splices. Any losses from fiber optic splitters like those used in passive optical networks (PONs) are also included in calculating the loss budget. The final loss factor is stress in installation. Fiber optic cable pulled with too much tension may be damaged. Each time you make a bend with a fiber optic cable, you put some stress in the fiber which can cause loss. Even cable ties tightened on the cable can cause loss. Stress loss should be zero!
• #203: The graph below the link diagram shows the actual amount of light in the fiber along the length, directly corresponding to the link diagram above it. This diagram looks like an OTDR plot, since it is similar to what the OTDR measures. If you are not familiar with OTDRs, we will cover them in the testing sections. But look at the diagram closely. The power goes down as the light goes down the fiber, reduced by the attenuation of the fiber and the losses in connectors and splices. By convention, we include the loss of the connectors on the end of the cable plant, since when we test connectors, we do so by mating them to another reference connector. The power level starts at the transmitter output, coupled into the fiber, shown at the top of the X-axis of the graph. After the loss of the cable plant, it is reduced by the amount of the loss. In order for the link to work properly, the power at the receiver must be higher than the receiver sensitivity, shown at the bottom of the X-axis of the graph. The amount by which the receiver power exceeds the receiver sensitivity is the margin of the link. FOA has an online link loss budget calculator: https://foa.org/tech/ref/Loss_Budget/Loss_Budget.htm
• #204: Unlike sources and power meters which measure the loss of the fiber optic cable plant directly, testing just like the fiber is used for transmission of data, the OTDR works indirectly. It uses backscattered light of the fiber to imply loss (remember that scattering is the major cause of loss in the fiber.) The OTDR works like RADAR, sending a high-power laser light pulse down the fiber and looking for return signals from backscattered light in the fiber itself or reflected light from connector or splice interfaces. The OTDR test is an indirect test and will not generally correlate with insertion loss testing. However, it is useful to confirm splice losses and find faults in fibers, like breaks or severe stress losses such as from too tight bends. Many problems with fiber optic testing are caused by improper use of OTDRs, either on short cable plants where OTDR testing is not appropriate or by use by inadequately trained personnel. Modern, high-end OTDRs use artificial intelligence (AI) to analyze the OTDR data and provide test results. This is much better than the earlier autotesting option on OTDRs which often gave confusing results. However, operators still need to know how to use the instrument and how to set up measurements which shall be covered in the next few slides.
• #205: The OTDR works like RADAR, sending a high power laser light pulse down the fiber that is scattered in all directions including a small amount back toward the instrument itself. The OTDR receives the backscattered light and converts it into a display. It also receives reflected light from connector or splice reflectance. Only a small amount of light is scattered back toward the OTDR, but with sensitive receivers and signal averaging, it is possible to make measurements over relatively long distances. At any point in time, the light the OTDR sees is the light scattered from the pulse passing through a section of the fiber. The test pulse is attenuated by the fiber and connector or splice losses as it travels down the fiber, so the returned signal is lower as the pulse goes further. The attenuation is seen by the OTDR which processes the data. Since it is possible to calibrate the speed of the pulse as it passes down the fiber by knowing the time it takes and the speed of light in the fiber, the OTDR can measure time, calculate the pulse position in the fiber and correlate what it sees in backscattered light with an actual location in the fiber.
• #206: Thus the OTDR can create a display of the amount of backscattered light at any point in the fiber. Since the pulse is attenuated in the fiber as it passes along the fiber and suffers loss in connectors and splices, the amount of power in the test pulse decreases as it passes along the fiber in the cable plant under test. The portion of the light being backscattered will be reduced accordingly, producing a picture of the loss occurring in the fiber. Some calculations are necessary to convert this information into a display, since the process occurs twice, once going out from the OTDR and once on the return path from the scattering at the test pulse. The final OTDR display is dB on the Y-axis and distance on the X-axis.
• #207: The OTDR can create a display of the amount of backscattered light at any point in the fiber. Since the pulse is attenuated in the fiber as it passes along the fiber and suffers loss in connectors and splices, the amount of power in the test pulse decreases as it passes along the fiber in the cable plant under test. The portion of the light being backscattered will be reduced accordingly, producing a picture of the loss occurring in the fiber. Some calculations are necessary to convert this information into a display, since the process occurs twice, once going out from the OTDR and once on the return path from the scattering at the test pulse. The final OTDR display is dB on the Y-axis and distance on the X-axis. The OTDR trace shows a slope which is the attenuation coefficient of the fiber – dB loss over distance – and a drop where the fiber has loss as in a splice or connection.
• #208: There is a lot of information in an OTDR display and it is very important that a fiber optic tech understand what the OTDR can provide as test information. The slope of the fiber trace shows the attenuation coefficient of the fiber and is calibrated in dB/km (or sometimes kilofeet for the metric-challenged) by the OTDR. In order to measure fiber attenuation, you need a fairly long length of fiber with no distortions on either end from the OTDR resolution or overloading due to large reflections. If the fiber looks nonlinear at either end, especially near a reflective event like a connector, avoid that section when measuring loss. Connectors and splices are called "events" in OTDR jargon. Both should show a loss, but connectors and mechanical splices will also show a reflective peak so you can distinguish them from fusion splices. Also, the height of a reflective peak will indicate the amount of reflection at the event, unless it is so large that it saturates the OTDR receiver. Then peak will have a flat top and tail on the far end, indicating the receiver was overloaded. The width of the peak shows the distance resolution of the OTDR, or how close it can detect events. Understanding how to interpret OTDR traces requires lots of training and practice, and misinterpreting traces can be very expensive if good cables are rejected or bad ones accepted.
• #209: The slope of the fiber trace shows the attenuation coefficient of the fiber and is calibrated in dB/km (or sometimes kilofeet for the metric-challenged) by the OTDR. In order to measure fiber attenuation, you need a fairly long length of fiber with no distortions on either end from the OTDR resolution or overloading due to large reflections. If the fiber looks nonlinear at either end, especially near a reflective event like a connector, avoid that section when measuring loss. Connectors and splices are called "events" in OTDR jargon. Both should show a loss, but connectors and mechanical splices will also show a reflective peak so you can distinguish them from fusion splices. Also, the height of a reflective peak will indicate the amount of reflection at the event, unless it is so large that it saturates the OTDR receiver. Then peak will have a flat top and tail on the far end, indicating the receiver was overloaded. The width of the peak shows the distance resolution of the OTDR, or how close it can detect events. Understanding how to interpret OTDR traces requires lots of training and practice, and misinterpreting traces can be very expensive if good cables are rejected or bad ones accepted.
• #210: The large initial pulse of the OTDR causes recovery problems that limits the ability of the OTDR to see anything near it, even for high resolution instruments like the one shown here. That is caused by the high-powered test pulse reflecting off the OTDR connector and overloading the OTDR receiver. The recovery of the receiver causes the "dead zone" near the OTDR. To avoid problems caused by the dead zone, always use a launch cable of sufficient length when testing cables. Ghosts are causes by reflective events bouncing back and forth in a cable. Short cables with highly reflective connectors can show several ghosts. They can be detected by noting the reflective peaks are at multiples of the same distance. The limited distance resolution of the OTDR makes it very hard to use in a LAN or building environment where cables are usually only a few hundred meters long. The OTDR has a great deal of difficulty resolving features in the short cables of a LAN and is likely to show "ghosts" from reflections at connectors, more often than not simply confusing the user.
• #211: The large initial pulse of the OTDR causes recovery problems that limits the ability of the OTDR to see anything near it. That is caused by the high-powered test pulse reflecting off the OTDR connector and overloading the OTDR receiver. The recovery of the receiver causes the "dead zone" near the OTDR. To avoid problems caused by the dead zone, always use a launch cable of sufficient length when testing cables. Since the OTDR uses backscatter light to make measurements, it's dependent on the consistency of the backscatter. But scattering is the largest cause of loss in the fiber and various fibers have different losses - and different backscatter coefficients. When measuring connectors of splices, the loss measured will be higher going from a low loss fiber to a high loss one and lower in the other direction. The only way to overcome this error is to take data in both directions and average the measurement. Ghosts are causes by reflective events bouncing back and forth in a cable. Short cables with highly reflective connectors can show several ghosts. They can be detected by noting the reflective peaks are at multiples of the same distance. The limited distance resolution of the OTDR makes it very hard to use in a LAN or building environment where cables are usually only a few hundred meters long. The OTDR has a great deal of difficulty resolving features in the short cables of a LAN and is likely to show "ghosts" from reflections at connectors, more often than not simply confusing the user. The FOA Guide on OTDRs has much more detailed explanations of these issues.
• #212: Here is a real OTDR trace showing a fiber optic cable included in the FOA OTDR simulator files. There are 4 cables together here: a 250m launch cable, another 250m cable and two sections of approximately 2500m each. In this case there is no receive cable so you cannot measure the connector on the end of the cable, but you can easily see the end of the cable with a reflective connector and the trace showing OTDR noise after the end of the cable.
• #213: Here is an example of how to make measurements with an OTDR. The OTDR measures events chosen by placement of two or more markers on the display. In this case, the two markers are showing the attenuation coefficient measurement of a 2.5km section of the cable plant. In the data panel, you can see the position of marker L1 (0.526km) and L2 (3.024km) and the difference “d” (2.498km). You can also see the measured loss between the two markers (9.25dB) and the calculated attenuation coefficient of the fiber (3.70dB/km). Likewise, the two markers can be used to measure the loss of a connector or splice. More options with an OTDR include special methods to calculate loss using statistical methods that offer greater precision in some difficult test situations. OTDR Lab: With the FOA OTDR Simulator and the Parameter Traces files, compare the traces of the cable plant at different setup parameters: wavelength, pulse width, number of traces averaged, index of refraction, etc. Note the setup parameters that give the best traces to analyze. Then use the measurement methods to measure distance, fiber loss, connector and splice loss on the cable plant. With an OTDR, launch cables and simulated cable plant, show how an OTDR measures the cables. Compare manual setup to autotest.
• #214: Here is an example of a cable plant problem found by and OTDR. This photo was sent to FOA by one of our CFOTs. A cable had high loss in some fibers or no continuity – the OTDR showed the end of the fibers – at a location which did not have any splices or connections. The distance measured allow tracing the problem to where this cable transitioned from aerial to underground. The cable had been kinked and most of the fibers broken. The OTDR was the troubleshooting tool that found the problem.
• #215: Long distance high speed networks has special performance requirements for fiber. By long distance we mean more than 20-30km and by fast we mean >10Gb/s. Whether the fiber is newly installed or older fiber being tested for upgrades, additional testing is required to ensure the fiber performance meets the needs of today’s (and the future’s) high speed networks. Additional test include chromatic dispersion, polarization mode dispersion and spectral attenuation – and that’s in addition to connector inspection, insertion loss and OTDR tests.
• #216: Optical Reflectance (Return Loss) in Connectors If you have ever looked at a fiber optic connector on an OTDR trace, you are familiar with the characteristic spike that shows where a connection is. That spike is a result of the reflectance of the connector, or the amount of light that is reflected back up the fiber from the interface of the polished end surfaces of the connectors. It is called fresnel reflection and is caused by the light going through the change in index of refraction at the interface between the fiber (n=1.5) and air (n=1). In high-bit rate singlemode systems, reflectance can be a major source of transmission problems. The reflected light interferes with the laser diode chip, causing noise. Minimizing the light reflected back into the laser is necessary to get maximum performance out of high bit rate systems. In a short singlemode link, reflectance can also cause problems by reflecting back and forth many times in a link, creating “optical background noise.” Problems with transmission, e.g. high bit-error-rate, in short singlemode links can often be traced to highly reflective connectors. Since this is primarily a problem with singlemode systems, manufacturers have concentrated on reducing the reflectance of singlemode connectors The usual technique involves polishing the end of the ferrule to a convex surface (PC) or at a slight angle (APC) to prevent direct back reflections.
• #217: Here is what reflectance looks like in a fiber optic data link. If several reflective events are close, they can have multiple reflectance paths between them causing even more noise. If you have seen “ghosts” on OTDR traces, you have seen this happen. State-of-the-art connectors will have a reflectance of about -40 to -60 dB, or about one-ten thousandth to one millionth of the light being transmitted is reflected back towards the source. Reflectance can be measured two ways. One uses a setup similar to insertion loss testing but with an optical coupler while the other more common method uses an OTDR. The measurement of of reflectance is not a precise measurement because of the small amount of reflected light that must be measured. Uncertainties of up to +/-1to 3 dB are common.
• #218: Measuring reflectance per standard test procedure EIA FOTP-107 is straightforward, but requires a special test setup, shown in the accompanying diagram. This test setup can be used with a bare fiber output into which a connector pair is installed (analogous to a FOTP-34 connector insertion loss test) or with a connectorized output for testing patchcords (like FOTP-171). For this test, one needs a calibrated coupler which can be used to inject a source into the test cable or pigtail and measure the light reflected back up the fiber, along with a standard power meter and laser source. Due to the dynamic range required to measure return losses in the range of -25 to -60 dB, a high power laser source is necessary. The laser sends a signal down to the device under test and the power meter measures the amount of reflected light. The power meter also measures the output of the laser source to complete the measurement. This method is rarely used in the field but is used in factories making patchcords to test them for reflectance for quality control.
• #219: OTDRs are more commonly used for measuring reflectance of connections in installed cable plants. Most OTDRs are programmed to measure reflectance from a connection or splice (usually only a mechanical splice will have reflectance.) Manu also measure return loss which includes the backscatter from the fiber, a more complicated topic than this course covers. The OTDR measures the peak of the reflectance above the level of the fiber, but it is a complicated process involving the baseline noise of the OTDR, backscatter level of the fiber and power in the reflected peak. (Refer to the FOA textbook on testing for more details.) Like all backscatter measurements, it has a fairly high measurement uncertainty, but has the advantage of showing where reflective events are located so they can be corrected if necessary.
• #220: Bandwidth and Dispersion: Some people have the impression that fiber has infinite bandwidth, but it’s not true. In fact, the distance fiber can carry network signals depends as much on bandwidth as loss - sometimes more. There are several factors that affect the bandwidth of singlemode fiber, but the two major ones for singlemode fiber are chromatic dispersion, of the fact that light of different colors travels at different speeds in glass (the definition of index of refraction) and polarization mode dispersion, caused by the varying speeds of planes of polarization. Equipment, expensive and complicated) are available to test these factors for long SM links, but it’s beyond the scope of this presentation. In multimode fiber, you have chromatic dispersion for the same reasons as in SM fiber, but you also have modal dispersion, caused by the different path lengths light follows in the larger core. While these factors are tested in the lab by fiber manufacturers, field testing is not done. However, bandwidth testers for MM fiber may become available in the near future due to the high bandwidth requirements of networks like 10 GbE.
• #221: One factor in fiber bandwidth is chromatic dispersion. Remember a prism spreads out the spectrum of incident light since the light travels at different speeds according to its color and is therefore refracted at different angles. The usual way of stating this is the index of refraction of the glass is wavelength dependent. Thus a carefully manufactured graded index multimode fiber can only be optimized for a single wavelength, usually near 1300 nm, and light of other colors will suffer from chromatic dispersion. Even light in the same mode will be dispersed if it is of different wavelengths. Chromatic dispersion is a bigger problem with LEDs, which have broad spectral outputs (their output light is comprised of many wavelengths of light), unlike lasers which concentrate most of their light in a narrow spectral range. Chromatic dispersion occurs with LEDs because much of the power is away from the zero dispersion wavelength of the fiber. High speed systems, based on broad output LEDs, suffer intense chromatic dispersion, about equal to the modal dispersion.
• #222: Another factor in fiber bandwidth is polarization mode dispersion. Polarization mode dispersion (PMD) is a bit more complex. Polarization is a phenomenon of light traveling in a medium as a wave with components at right angles. Some materials, like a glass optical fiber, have a different index of refraction for each of those components of the light wave, which is called birefringence. A different index of refraction means light travels at a different speed, so in the simplest visualization, PMD in fiber looks like the drawing below, where each component of the polarized light travels at a different speed, causing dispersion. The magnitude of PMD in a fiber is expressed as this difference, which is known as the differential group delay (DGD) and called Δτ(delta Tau). PMD can be affected by the fiber itself, especially the roundness of the core, or the stress on the fiber. This sensitivity to PMD means fiber may vary PMD under certain conditions like wind stressing aerial cable or vibrations affecting underground cables. PMD is tested for fibers during manufacture or when being cabled. In the field, it is common to test PMD on newly installed fibers which are intended for operation at high speeds, generally above 10 Gb/s or when upgrading fibers installed some time in the past to higher speeds.
• #223: DWDM or Dense Wavelength-Division Multiplexing creates a whole different situation for testing. With multiple wavelengths in one fiber, it may require a power meter with wavelength selectivity if testing a fiber carrying multiple signals. It will also probably require high power capability, since not only are multiple sources using the fiber, but they are generally amplified to high levels to allow very long distance links. All couplers and DWDM multiplexers have significant insertion loss which may need to be tested. A proper treatment of this subject is beyond the scope of this presentation. A FTTH presentation will cover some of the issues of DWDM and PON (passive optic network) testing.
• #224: With the development of low water peak fibers, the possibility of transmission from 1260 to 1675 nm has been considered. This results from careful manufacturing of the fiber to reduce the water in the fiber (in the form of OH- ions) that causes higher spectral attenuation at around 1244 and 1383 nm. Systems using coarse wavelength division multiplexing (CWDM) use lasers at 20 nm increments over this range. Since one may want to use available fibers of unknown spectral attenuation for CWDM systems which use lasers from 1260 to 1670 nm in 20 nm windows, it becomes necessary to test for spectral attenuation to verify the usability. At the water peaks, legacy fibers may have attenuation coefficients around 2 dB/km while low water peak fibers may be as low as 0.4 dB/km. Testing spectral attenuation is done several ways. One method uses broad spectral width sources like LEDs and a spectrum analyzer on the receiving end of the fiber. Some instruments used for CD testing can also measure spectral attenuation, doing both measurements at one time.
• #225: Specialized Test Equipment For Fiber Characterization The process of testing installed fiber optic cable plants for reflectance, CD, PMD and SA is called “fiber characterization.” It is often done on long fiber runs where high speed networks will be used. Fiber characterization requires specialized test equipment. The choice of test equipment is confused by the fact that there are quite a few ways to test each. However one patented method, available from one company (EXFO) requires only access to one end of the fiber and is the most widely accepted.
• #226: Fiber optic network design is another specialty in fiber optics and FOA has a certification just for design – and a textbook too. For techs in a CFOT course, this section is just to introduce them to the concepts of fiber optic network design.
• #227: What is fiber optic network design? Fiber optic network design refers to the specialized processes leading to a successful installation and operation of a fiber optic network. It includes determining the type of communication system(s) which will be carried over the network, the geographic layout (premises, campus, outside plant (OSP, etc.) and routing, the transmission equipment required and the fiber network over which it will operate. Next we have to consider requirements for permits, easements, permissions and inspections. When you know the communications equipment and routing, you can calculate a power budget to confirm the system will work once it is installed. Once we get to that stage, we can consider actual component selection, placement, installation practices, testing, troubleshooting and network equipment installation and startup. Finally, we have to consider documentation, maintenance and planning for restoration in event of an outage.
• #228: Writing Specifications For Cable Plants - It’s probably impossible to cover every possible issue in a design specification document, but here are some reminders to include: Specify network equipment or communication signals Specify route and installation methods, e.g. underground, aerial, cable pulled in conduit or blown into microducts, etc. Specify components: fiber and cable type, splices and connectors, cable plant hardware User may specify connector termination type if preferred, e.g. epoxy/polish, prepolished/splice, fusion-spliced pigtails for SM User may specify cable and hardware types, but should allow for alternate suggestions. Vendors may be able to offer alternatives that can save cost or enhance performance or reliability. User should have specification for max loss based on loss budget calculations and reflectance if important Other standard specs - the “boilerplate” put in every spec!
• #229: This section is an overview of fiber optic installation, both OSP (outside plant) and premises. In addition to this course, we suggest you take the free Fiber U online course on OSP construction to get introduced to OSP cable installation. The one thing to remember about any fiber optic installation is that every installation is unique and must be analyzed carefully before designing and installing it. Also, no class can prepare you for everything you will encounter in any installation, but the FOA goal is to have you understand cable prep, splicing, termination and testing so you can do the job properly.
• #230: “Fiber optics” is not all the same. “Outside plant” refers to fiber optics as used outdoors in telephone networks or CATV. “Premises” fiber optics is used in buildings and on campuses. Outside Plant: Telephone companies, CATV and the Internet all use lots of fiber optics, most of which is outside buildings. It hangs from poles, is buried underground, pulled through conduit or is even submerged underwater. Most of it goes relatively long distances, from a few thousand feet to hundreds of miles, over what we call “singlemode” fiber. Premises Cabling: By contrast, premises cabling involves cables installed in buildings for LANs or security systems. It involves short lengths, rarely longer than a few hundred to two thousand feet, of mostly “multimode” fiber. Both these applications are unique in the components they use, the installation methods and the testing procedures, but they share many of the basic principles we learn in this course. Whether the installation is outside plant or premises is very important to the installer of fiber optic cabling systems.
• #231: Outside plant installations are almost all singlemode fiber, and cables often have very high fiber counts. Cable designs are optimized for resisting moisture and rodent damage. Installation requires special pullers or plows, and even trailers to carry giant spools of cable. Long distances mean cables are spliced together, since cables are not made longer than about 4 km (2.5 miles), and most splices are by fusion splicing. Connectors (SC, ST or FC styles) on factory made pigtails are spliced onto the end of the cable. After installation, every fiber and every splice is tested with an OTDR. The installer usually has a temperature controlled van or trailer for splicing and/or a bucket truck. Investments in fusion splicers and OTDRs can add up to over $100,000 alone. Outside plant installations require more hardware (and more investment in the tools and test equipment.) Pullers, splicers, OTDRs and even splicing vans are the tools of the trade for OSP contractors.
• #232: Cable Installation - OSP Buried Know the installation method - direct buried, conduit or innerduct, aerial, etc. and have an experienced crew leading the install. This is not a time for inexperienced people learning on the job. Bring along some new personnel as helpers so they can learn from the experienced ones. Know limits in tension and bend radius for the cable you are installing. Attach pulling eyes properly to the strength members and jacket. Despool cable by rolling it off the spool, not pulling off the sides of the spool, as that will put a twist in the cable and perhaps kink it. Figure 8 the cable on the ground for midspan pulls to avoid putting a twist in the cable. The biggest cause of fiber optic network failure (and equally for other buried utilities) is “backhoe fade!” Never dig until you know what is underground where you plan to dig! A new nationwide service is available: dial 811 to get information on buried utilities before you dig! See http://www.commongroundalliance.com/ for more information.
• #233: Cable Pulling - OSP Use powered capstans for applying tension on long pulls through conduit or innerduct to control tension properly. Use automated tension control equipment available with the capstans. Plan for lubrication on conduit pulls. See the American Polywater video (www.polywater.com)
• #234: Remember the slides in the cable section on bend radius? Bending Radius Limits: The normal recommendation for fiber optic cable bend radius is to ensure the minimum bend radius under tension during pulling is 20 times the diameter of the cable. When not under tension, the minimum recommended long term bend radius is 10 times the cable diameter. There is a Fiber U minicourse on this topic also. It is very important to manage the bend radius of a fiber optic cable, especially in underground cable installation. Pulling cables can cause permanent – and very expensive – damage!
• #235: Cable Installation – Microtrenching Microtrenching is becoming popular in metropolitan areas because it disrupts traffic less and leaves less of a mess on roadways. The biggest cause of fiber optic network failure (and equally for other buried utilities) is “backhoe fade!” Never dig until you know what is underground where you plan to dig! A new nationwide service is available: dial 811 to get information on buried utilities before you dig! See http://www.commongroundalliance.com/ for more information.
• #236: Microtrenching is often used to install microducts and the fiber is installed by blowing the cable in. It’s called “blown fiber” but the actual process uses compressed air to float the cable in the duct to greatly reduce friction and then push the cable through the duct with the help of the air flow. This technique has been gaining popularity because it is fast and less expensive than other installation methods.
• #237: Aerial cable can be self-supporting (ADSS), supported by a messenger or, as used by many electrical utilities, inside the optical power ground wire. Cable manufacturers should be consulted for specific design and installation instructions for their cable designs.
• #238: Many installations involve splitting the fibers in a backbone cable or dropping a small fiber count cable from a large backbone cable. A Fiber U MiniCourse will explain how a high fiber count backbone or distribution cable can drop just a few fibers at a location without splicing all the fibers in the cable, saving time and cost. 
• #239: Submarine/Underwater Cable Transoceanic links require giant ships and long cables stored in special reels on the ship. Most times a complete link is prepared and installed "hot" - transmitting data" so the condition can be monitored during installation. If anything fails, it can be more easily pulled up and repaired at that time than after the whole cable is installed. Crossings of lakes and rivers are simpler, but it's a good idea to trench for the cable to bury it and reduce the chances of it being snagged. Most underwater installs will require special permits. And safety is more complicated!
• #240: The latest application for fiber is direct connection to the home, providing virtually unlimited bandwidth, limited only by the electronics delivering services. FTTH involves OSP installation to the curb, drops into the subscriber premises and then requires a home that can deliver the services to telephones, TVs and computers. Construction to the curb is conventional – aerial, underground or direct buried - while the drop to the home may be made in many different ways – trenching, directional boring, or aerial – depending on the requirements of the neighborhood. The FTTH drop is often made with prefabricated cable that allows "plug and play" installation.
• #241: Premises cabling is mostly multimode in short lengths, rarely longer than a few hundred feet, with 2 to 144 fibers per cable typically. Some users install hybrid cable with both multimode and singlemode fibers. New passive optical LANs us only singlemode cables. Splicing is practically unknown in premises applications except for splice-on connectors. Most connectors are SC or LC style. Testing is done with a source and meter, but every installer has a VFL (visual fault locator) to check fiber continuity and connection. Unlike the outside plant technician, the premises cabling installer (who is often also installing the electrical power cable and Cat 5 for LANs too!) probably has a much smaller investment in tools and test equipment. Premises applications usually mean lots of cables - both copper and fiber - terminated in telecom rooms.
• #242: Premises installers generally install cables in cable trays and need only a termination kit for attaching connectors and a simple test kit for their installations. Working in crowded telecom closets or communications rooms is the norm. Working conditions can involve working in various types of environments from old telecom and IT equipment like this to brand new buildings like in the next few slides.
• #243: Premises applications usually mean lots of cables - both copper and fiber - run inside the building in conduit, cable trays, under floors or proper hangers and terminated in telecom rooms.
• #244: Premises Installation Cable may be suspended on J-hooks, placed in cable trays or pulled in conduit or fire-rated innerduct All cable must meet fire codes and all installation practices must meet local building and fire codes. Mixed with copper cables, fiber should be run on top or suspended below cable trays to prevent crushing the fiber cables.
• #245: Premises Installation - Codes All cable must meet fire codes - look for NEC ratings and testing on the cable jacket All penetrations of fire-rated walls or floors must be firestopped.
• #246: Pre-Installation:
No installation should begin until there is a complete design, all equipment and components have been chosen, the cable routing is determined and any permits or coordination with other groups is ready. Cable documentation should be started before installation, so the installation is properly documented and ready for labeling and recording test data. Documentation will facilitate installation, allow planning for upgrades and provide data needed for restoration. Components must be ordered and delivered to the job site before installation can begin. Relevant personnel who will be affected by the install, for example those located in the installation area or who may lose communications services, must be notified. If the installation takes more than one day, arrange security to guard the equipment and components left on the construction site.
• #247: During The Installation Inspect all installation workmanship during the installation itself so any problems can be identified and solved before they become major issues. Daily supervisors and installers should review processes, progress on the job and test data. All affected personnel should receive immediate notification of problems and solutions, shortages, etc. Be careful when installing cables to avoid stress, hazards that may snag cables and kink them or installing cables where heavier cables may be placed on top of them. Bundling cables for neatness is fine, but be careful using cable ties. Tightening them can put harmful stress on the fibers (or pairs in UTP copper cables), so hand tighten them and cut off the excess length. Even better, use soft "hook and loop" ties that can be reopened to move cables.
• #248: Every fiber optic network, especially an outside plant installation, is susceptible to outages, either by damage to the cable plant or problems with the communications equipment. Network operators must be prepared for outages and have a plan for restoration. Typical problems are cable digups called "backhoe fade" and aerial cables damaged by "target practice" There is a Fiber U online course on Restoration.
• #249: The instructor will go over lab safety rules before each lab. The lab manual has several pages of rules for safety in fiber optic labs. Each student should be familiar with them and follow them carefully. Instructors must follow them too! https://www.thefoa.org/tech/ref/safety/safe.html Once you start working in the field, these rules still need to be followed plus the standard workplace safety rules also are important – you are essentially in a construction project so personal protective equipment and workplace rules are very important for your safety!
• #250: This is a basic fiber optic training program for FOA-Approved Schools to teach classes for the FOA CFOT certification. The program was developed by The Fiber Optic Association with inputs from many FOA instructors. It’s is based on 30+ years of experience in the business, including starting one of the first fiber optic test equipment companies and training thousands of fiber optic installers.
• #251: The FOA CFOT certifies those who demonstrate knowledge, skills and abilities appropriate to tasks involving fiber optics. The FOA develops appropriate reference and training curriculum materials to use for teaching or studying fiber optic technology covered in the exam. It is the job of the instructor to verify that the student has shown the ability and skills to perform typical fiber optic tasks.
• #252: FRG, FOTM, FOA Online Fiber Optic Reference Guide, Understanding Fiber Optics, The Basics: Plus CFOT Study Guide, linked on Contents Page Don’t forget the FOA YouTube Lectures also!
• #253: We've reviewed many tests and these are the questions most missed. Take a minute and review these topics. All are covered in the slides above. Start with the jargon slides.
• #254: We've reviewed many tests and these are the questions most missed. Take a minute and review these topics. Yes ¼ of all those taking the CFOT test think you can strip the cladding from the core. EVERY fiber drawing we found on the Internet search with one exception (which we will show in a second) showed the same thing - the core of the fiber separate -sticking out of the cladding and the cladding sticking out of the primary buffer coating. Those drawings are not all from websites that you might expect some technical inaccuracies, several were from fiber or other fiber optic component manufacturers and one was from a company specializing in highly technical fiber research equipment. The only drawing we found that does not show the core separate from the cladding was - really! - on the FOA Guide page on optical fiber. The fiber is one solid piece of glass, pulled at high temperature from a large glass preform. The core and cladding are simply different types of glass that form the structure that allows fiber to transmit light with low loss.
• #255: End