How do you calculate upstream & downstream bandwith per user on an ONT?
the ONT itself is going to be very high performance connected to fiber, so it itself is high performance, and often times can have multiple ports connecting to different devices. But, what is your general recommendations on planning for bandwidth in terms of the number of ONTs you’re going to need?
Most ONTs offer either four ports or eight ports, depending on how many devices you’re going to hook up. So if you’re going to hook up to a lot of wireless access points that consume a lot of bandwidth, that would be a consideration. One of the things you can do from the ONTs is dynamic bandwidth allocation. This allows you to go into each ONT and each port and allocate a certain amount of bandwidth guaranteed per port, and then you can also have the ability to burst, too. So it really depends on the different bandwidth requirements of the devices that are going to plug into the ONT.
Does an ONT need to be in an enclosure?
In a POL, the Optical Network Terminal (ONT) is can be wall mounted or ceiling mounted. The do not need to be an enclosure unless they are in a place where they might be accessed by the public — say, in a hotel or an MDU. In that case, they would likely be put in an in-wall enclosure.
Are there any special requirements for the optical fiber used in passive optical LAN?
Each ONT end device just requires one strand of single mode fiber. And the connectors typically used in passive optical LAN are SC/APC angle polish connectors. Note: some manufacturers now offer a way to leverage multimode fiber in applications where the fiber plant was already installed.
Are there any special requirements for the optical fiber used in passive optical LANs?
In a Passive Optical LAN each ONT end device requires one strand of single-mode fiber. The connectors typically used in passive optical LAN are SC/APC angle polish connectors on the end. If you have existing single mode in your backbone, then you’re halfway there.
How do I plug an MPO connector into my OTDR?
Many OTDRs have only one plug connector, which can make plugging in your MPO confusing. Luckily, there are different ways to connect your OTDR to the MPO connector. If you want to see the MPO connector, you’re going to need a launch fiber. That launch fiber is going to need to have the same MPO connector. If you’re doing 12 fiber, you need a 12 fiber MPO launch. And if you’re doing 24 fiber MPO, you’re going to need a 24 fiber MPO launch cable. The length of that cable is going to be a function of the distance of the fiber you’re testing, or the pulse widths that you’re using. If you’re using a short pulse width, a three or a five nanosecond, maybe a 10 nanosecond pulse width, you can get by with a pretty short launch fiber. Longer the better, but say at least 20 or 30 meters. Now on the OTDR end of that launch fiber, you could either have it be a fan out cable, or you can put a cassette there and move through the ports of the cassette. Also, there are manufacturers who make a switch. The switch has say an SC port and inside the switch it will move between 12 or 24.
Which MPO connector should I use?
The three most common MPO connector options are MPO-8, MPO-12 and MPO-24.
- MPO-8 is a legacy standard for the QSFPs, coming out of the transceivers running 40 Gigabit or 100 Gigabits. It is used for both multimode and singlemode transceivers and breakouts, but offers the lowest density option, because you’ll have to have more components for the MPO-8 as you move forward into higher speeds.
- The MPO-12 is the legacy embedded base and with the use of different modules and array fanouts can accommodate multiple configurations.
- MPO-24 is the newest option and is used on the trunk cables and modules. MPO-24 helps to future proof your network because it provides the highest panel density, allows you to use fewer components and may offer the lowest first installed cost. MPO 24 trunk implementations provide significant advantages for duplex and parallel implementations, providing for faster installation and better pathway efficiency.
What is a leaf-spine architecture and why is it popular?
In a leaf-spine architecture, a series of leaf switches form the access layer.These switches are fully meshed to a series of spine switches. Leaf-spine architectures are popular because they reduce the number of “hops” from server to server, which minimizes latency. One of the consequences of this approach is that it requires more physical connections to support that any-to-any connectivity. The physical layer becomes more important, because it needs to accommodate a simplex, duplex or parallel networking scheme. This means that sometimes the design, implementation, testing and validation can get fairly complex.
Is there a difference between a PON power meter and a ‘regular’ power meter?
Yes, there are several differences. First, the PON power meter is a pass through device which measures the power of the downstream signal and the power of the upstream signal. A regular power meter is a single ended detector, so you can measure the downstream power, but then you can not validate that your ONT’s properly emitting because you cannot measure the upstream. That’s the first difference. Also, on your PON power meter you have preconfigured thresholds for the different wavelengths according to the different PON layers that you could have GPON, XGS PON, and GP PON2, all of those.
What’s the difference between PON, G-PON, and NG-PON?
PON is the general concept of Passive Optical Networking, which refers to the use of splitter typologies in point to multipoint technologies where one fiber is split out to multiple endpoints. GPON is the 2.5 Gbps Passive Optical Network, while NG-PON is in general, everything which is Next Generation, meaning 10 Gbps and beyond. In the standards they talk about Next Gen PON1, which is then the 10 Gbps variants, and Next Gen PON2, which are then the TWDM PON,
What is the number one cause of network failure?
The leading cause of network failures is dirty connections. Almost all the time, the dirt is completely invisible to the naked eye, but since the diameter of a fiber core is so small (50 microns for multimode fiber and only 8 microns for singlemode fiber) it doesn’t take much to block the light on the fiber endface.
When the dust cap is removed the ferrule end face can easily be contaminated by direct contact with skin oil, grease, salt, fingerprints, lint, uncured epoxy, grime or dust. Even just open air exposure may result in moisture and dust sticking to the end-face. When the dust caps are off your connectors are at risk of contamination.
This is why proper inspection and cleaning techniques are so important, especially with today’s tight loss budgets where too much loss from a dirty connection can have a direct impact on bit error rate, insertion loss and optical return loss which will impact system performance and reliability.
How do you test a POL network?
Since the POL is an enterprise adaptation of the GPON used in Fiber-to-the-Home applications, you can use the same tools and expertise that were developed for GPON.The recommended testing protocols for a passive optical LAN deployment are:
- Inspecting the connector
- Fiber characterization, with different tools like OLTS, OTDR, iOLM
- Validating service activation using a PON power meter
- Reporting tests and test results
In fiber optic networks, 80% of the problems are caused by dirty or damaged optical connectors, 10% of network problems are due to macrobends, which are responsible for signal loss and network quality deterioration, and the remaining 10% of problems can be attributed to other components – the ONT, the OLT and the splitters.
Does a Passive Optical LAN (POL) use multimode or single-mode fiber?
Passive optical LAN (POL) architectures primarily use single-mode fiber because of the medium’s high capacity and ability to support long link lengths — up to 20 km. In a POL architecture. In a POL, a passive single-mode fiber network runs down to the desktop, through an optical splitter and into a thin-client edge device or Optical Network Terminal (ONT). Recently single-mode to multimode optical splitters have been introduced that enable cost-effective re-use of existing multimode fiber cabling for Passive Optical LANs inside buildings and across a campus.
What is polarity? And why is it important?
Polarity defines direction of flow, such as the direction of a magnetic field or an electrical current. In fiber optics, it defines the direction that light signals travels through an optical fiber. To properly send data via light signals, a fiber optic link’s transmit signal (Tx) at one end of the cable must match the corresponding receiver (Rx) at the other end.
In duplex fiber applications, such as 10 Gig, data transmission is bidirectional over two fibers where each fiber connects the transmitter on one end and to the receiver on the other end. The role of polarity is to make sure that this connection is maintained.
Polarity in multi-fiber MPO type cables and connectors is more complicated. Industry standards call out three different polarity methods for MPOs—Method A, Method B and Method C. And each method uses different types of MPO cables.
Read more about the three options here.
What fiber is typically being used with the 200 400 Gig circuits?
Most likely you are looking at either OM4 or OM5 multimode fibers. Of course single mode will always be an option, but on the multimode side you have OM5 fiber, which is optimized for the short wave division multiplexing (SWDM) so depending on the distances you need to support, those to fiber types will give you the most cost effective solution.
How do you inspect and clean an unpinned MPO connector?
An unpinned MPO is inspected in the same way as a pinned MPO, however from an inspection perspective there’s no contact made with the fiber. You’d use the same inspection probes that will find the fibers, inspect them, and do a pass fail analysis. For cleaning, you can use the cassette style cleaners. In fact in some ways they are easier to clean than a pinned connector
What is Wavelength Division Multiplexing (WDM) Technology?
Wavelength division multiplexing (WDM) allows multiple wavelengths, typically 2 or 4 wavelengths over a single fiber. The IEEE 802.3bs 200 Gb/s & 400 Gb/s Ethernet Task Force in 2016 added 200 Gb/s capability to support a cost and performance optimized migration path to 400 Gb/s that includes support for 200 Gb/s with at least 2 km of SMF (4l WDM duplex fiber) and at least 10 km of SMF (4 l WDM duplex fiber).
What changes were made to ANSI/TIA-942 when it was updated in 2017?
Some important changes to TIA-942-B Data Center Standard include:
- Added MPO-16, MPO-32 (ANSI/TIA-604-18) and MPO-24 (ANSI/TIA-604-5)
- Added Category 8
- Changed recommendation to category 6A or higher
- Added wideband laser-optimized 50/125 µm multimode (OM5)
- Added 75-ohm broadband coaxial cables and connectors (ANSI/TIA-568.C-4)
- Added recommendations for fiber in non-continuous pathways that could cause micro bends
- Recommends that cabinets be at 1200 mm (48”) deep and to consider cabinets wider than 600 mm (24”) wide
- Recommends considering pre-terminated cabling to reduce installation time and improve consistency and quality of terminations
- Recommends considering need for proper labeling, cable routing, cable management, and ability to insert and remove cords without disrupting existing or adjacent connections
- Maximum cable lengths for direct attach cabling in EDAs reduced from 10 m (33 ft.) to 7 m (23 ft.). (Direct attach cabling between rows is not recommended).
Other relevant changes include:
- Change in return loss for single-mode fiber from minimum 26dB to minimum of 35dB, making field terminations more difficult
- Polarity in Testing; used to be covered in ANSI/TIA -568.0-D and now is addressed in 568.3-D.
What types of connectors are used in Data Center applications?
In fiber optic systems for data centers, LC, SC and MPO optical fiber connectors are often used.
- SC (square connector) connectors have a push-pull coupling end face with a spring-loaded ceramic ferrule, and is ideal in data center applications.
- LC (Lucent connector) connectors – also push-pull connectors – came along after SC connectors, and feature a smaller ferrule (for this reason, it’s known as a “small form-factor connector”). Its smaller size makes it ideal for dense data center racks and panels.
- MPO connectors are used for ribbon cables with anywhere from eight to 24 fibers.
What type of fiber do you recommend for data center applications?
To support the high bandwidths required in data centers, most companies are installing at least OM4, laser optimized multimode fiber. Some companies are installing single-mode fiber, but that requires more expensive optics. A new option that is emerging is OM5, a wide bandwidth multimode fiber which allows short wavelength division multiplexing. This means the fiber can carry multiple wavelengths of light over the same fiber, increasing bandwidth significantly and yet still allowing the use of lower cost multimode optics.
What is wide band multimode fiber and how is it used?
WBMMF is a relatively new fiber medium specified in ANSI/TIA-492AAAE and given the designation of OM5 multimode fiber by ISO/IEC and TIA. This 50/125 µm multimode laser optimized fiber was originally developed to support Short Wavelength Division Multiplexing (SWDM) and supports 4 wavelengths of 25Gb/s transmission for an aggregated 100 Gb/s transmission on a duplex LC link, a popular and familiar interface in the data center. WBMMF supports four wavelengths between 850 nm and 953 nm, using multimode optics. WBMMF was designed for use in data centers.
How does WDM technology differ from parallel optics?
Whereas Wavelength Division Multiplexing (WDM) utilizes a single fiber, parallel optics uses multiple fibers and lanes at various transmission rates. At the 25G transmission rate and at 100 meters, for example, IEEE’s 400 Gb/s migration path requires 16 lanes (16 MMFs to transmit and 16 to receive) for a total of 32 multimode fibers. This parallel configuration launched new 16-fiber and dual-row 32-fiber MPO connectors for which TIA recently published a standard. At 500 meters for SMF, four lanes / 8 fibers total are required by the IEEE for 400 Gb/s. Currently, the IEEE P802.3cd is defining standards for 50 Gb/s, as well as next generation 100 Gb/s and 200 Gb/s Ethernet. Published release is expected in Q3 of 2018.
Will WDM/SWDM ultimately replace parallel fiber optic solutions?
The consensus among FOTC members is that both solutions will coexist and depend upon specific data center network goals and applications. Employing parallel fiber solutions are currently necessary to transition between speeds, so that breakout capability is essential.
What’s the difference between Tier 1 and Tier 2 Testing?
Tier 1 testing looks at loss, length and polarity. While Tier 1 fiber optic tests can identify problems in terms of pass or fail, they cannot determine the root cause or location of the problem. Tier 2 fiber optic testing is used to pinpoint root-cause locations and the amount of loss and optical return loss (ORL) from each problem contributor and is performed selectively in addition to Tier 1 testing under specific conditions and situations. Tier 2 fiber testing provides a deeper level of link visibility unlike any other fiber infrastructure tests. The optical time-domain reflectometer (OTDR) is used to perform Tier 2 fiber optic testing.
Should I install singlemode or multimode fiber in my network?
While some people choose to install singlemode fiber because of it’s high bandwidth, multimode fiber continues to be a popular choice for enterprise applications. Newer grades of multimode fiber, such as OM4 laser optimized fiber and OM5, wideband multimode fiber, have the bandwidth to support most applications over the distances required, plus the cost for the optics remains lower than the cost of singlemode optics.
In TIA standards what is the difference between a TSB and an Addendum?
A TSB is a Telecommunications Services Bulletin. It is purely informational: we might look at a subject, collect information and share it with the industry. There is no normative information, – no “shall” statements. TSBs are often stepping stones in the standards process as we start to explore topics that might become part of a standard. In fact, a TSB can be referenced by a standard. An addendum is an official addition to a publish standard. Anything in the addendum has the same enforcement as a full, published standard. The contents can be normative or informative depending on the content and how the engineering committee positions the material as an addendum
Can the same fiber optic transceivers that are used with OM3 fiber, like SFP + 10Gbps pluggable modules, be used with OM4 fiber or are there new transceiver types that need to be used?
Yes, you can use the same fiber-optic transceivers for both OM3 and OM4 fibers because the two fiber types are basically the same except that OM4 fiber has higher bandwidth. The IEEE 10Gbps Ethernet standard states that 300m OM3 and 400m OM4 link lengths are supported with 10GBASE-S compliant transceivers.
Are Bend Insensitive Multimode Fiber approved by standards?
Yes. ANSI/TIA/EIA-568B.3 sets performance specifications, minimum bend radius standards and maximum pulling tensions for 50/125-micron and 62.5/125-micron fiber optic cables. For inside plant cable, the fiber cable bend radius is 10 times the cable’s outside diameter under no pull load, and 15 times the cable’s outside diameter when subject to tensile load performance measurements and qualification processes. If you are interested learning more about the status of BIMMF or participating in the discussion you can access the schedule for TR-42.12 here and download recent meeting reports.