Author: Fiber-MART.COM

The need of quick provision of ports in data center environments is fulfilled by the use of multiport cables assemblies. This is very well achieved by a optical fiber cable strand, typically with 12 individual fibers and one MPO/MTP connector at the other end providing 6 parallel communication paths  and twice for 24 strand MPO cables. The quick provision is necessary in data centers between rack to rack  links. Using the connectors is a ‘plug and play’ solution with already tested patch optical budget properties.

MPO (Multi-fiber Push On) connectors are representing a standard for connecting technologies. In many cases, multi-fiber connector products are referred to as MTP connectors. The MPO connector is a multi-fiber connector that is defined by IEC-61754-7, “Fibre optic interconnecting devices and passive components – Fibre optic connector interfaces – Part 7: Type MPO connector family”; and TIA-604-5-D, ”Fiber Optic Connector Intermateability Standard, Type MPO”.

The term MTP (Multifiber Termination Push on ) is a registered trademark of US Conec. This is the term used by US Conec to describe their connector. The US Conec MTP product is fully compliant with the MPO standards. As such, the MTP connector is an MPO connector. The MTP connector is described by US Conec as, “a high performance MPO connector with multiple engineered product enhancements to improve optical and mechanical performance when compared to generic MPO connectors.”

Differences between MPO and MTP:

head shapes of the fibres differs: the MPO’s fibers are rectangular finish at head. The MTP fibres are round head terminated. For long use in terms of numbers of push in and pull up,  the MTP round head fibres maintain the lossless good coupling with female connectors.

You can not mate and have connectivity between the 12 strand MPO connector and 24 strand MPO connector.

12 Strand Applications

The new and improved next-generation MPO connector now delivers the optical, mechanical and environmental performance that service providers need to expedite the addition of fiber capacity and to support higher data-rate services. Among the numerous operational, financial and competitive benefits of using MPO connectors in the data center environment, are:

• optical insertion loss and return loss performance similar to single-fiber connectors

• maximum space savings for high-density fiber environments;

• reduced labor costs via fast, easy installation – because one 12/24 -fiber MPO connector replaces 12/ 24 single fiber connectors; and

• compliance with standards, i.e., IEC 61754-7; IEC 61755-3-31, IEC 61753-1

With so many service providers around the world now relying on the MPO connector to speed installation and deployment costs throughout their networks, it’s clear that the improved, next-generation MPO connector is ready for tomorrow’s high speed networks.

To achieve better optical performance, greater durability in the field and improved assembly quality, the design changes specifically include:

material changes to ensure reliable performance across a wide temperature range as specified in IEC 61755-3-31

extensively researching and refining the polishing process to achieve consistent low loss across all 12 fibers and replacing the flat ribbon cable assembly (for example, 0.178 inches x 0.08 inches) and its standard fiber with a round 3-mm cable utilizing single reduced bend radius fibers.

Practical implementations:

MPO/MTP styles cassettes have on MPO/MTP connector at one end, the fiber wires insides the cassettes rolled out and the individual  pair of fibre pairs at the other end.

End to end connectivity between cassettes is achieved with the use of trunk cables between cassettes.

The last connectivity pair is between cassettes individual pair of fibres terminations and switch or routers through a patch pair minding the Tx/Rx polarity.

The fight over 100G component market has been upgraded with the emerging of 100G QSFP28. This is an optical transceiver which can support 100G with the transmission mode of 4*25G. Usually, the move before was considered to be 10G→40G→100G. However, the new roadmap of 100G with QSFP28 is 10G→25G→100G or 10G→25G→50G→100G. One has a lot of questions for 100G migration. Why does 100G QSFP28 appear? Can 100GQSFP28 change our data center? The below article has answers to all your questions regarding QSFP28.

What QSFP28 offers?

Cost and power are considered the most important factors in data centers. A look back to the evolution of 100G modules in the past years has shown that things keep changing, from CFP to CFP2 and CFP4. All these changes are closely related to factors like cost and power.

High Port Density: The first generation of 100G transceiver was CFP and the drawback was large size. Then CFP2 and CFP4 were the next generations of 100G modules with a decreased size (quarter width of CFP) as compared to former.  QSFP28 with the same footprint and faceplate density as QSFP+ is even smaller than CFP4. Up to 36 QSFP28 can be installed on 1RU switch on the front panel, higher port density being a big advantage.

Low Power Consumption: QSFP28 needs the lowest power consumption for transmission as compared with other 100G transceivers. It is less than 3.5 W whereas other transceivers consume around 6 W to 24 W.

Lower Cost: As seen above with higher port density and lower power consumption, the 100G QSFP28 be cost-efficient. Implemented with 4 lanes, this can increase the transmission capacity of every lane from 10G to 25G, which can effectively decrease the cost for each bit.

How could QSFP28 be a game changer for Data Center?

The 100G can be reached directly from 25G escaping 40G with QSFP28. We read how 100G uplink is converged by only four 25G links. In addition, the 25G network has the same cabling structure as a 10G network, but here capacity is much larger. The following table lists the related components and the suggested applications for QSFP28.

QSFP28 Series

Cabling

Applications

100G SR4 QSFP28

MMF,MTP/MPO

100G to 100G up to 100 m

100G LR4 QSFP28

SMF, LC Duplex

100G to 100G up to 10 km

100G QSFP28 to QSFP28 DAC

100G to 100G up to 5 m

100G QSFP28 to 4x25G SFP28 DAC

100G to 25G up to 5 m

100G QSFP28 AOC

100G to 100G up to 10 m

The challenge for long distance connectivity

QSFP28 is the smallest 100G transceiver. It’s a fraction of the size of the CFP. It is best for short distances. However, for longer distances, there have been some recent breakthroughs in transceivers with DWDM capabilities. PAM4 being the most significant of all still requires amplification for every short distance. For distances over 5 to 6 km, needs dispersion compensation. With this, it can handle traffic up to 80 km.

While the need for connecting 100G traffic is growing, no single small transceiver can solve the problem of connecting switches between data centers and other longer distance sites. That’s why organizations consider full-blown, DWDM platforms to handle their 100G data center connectivity. Here the output of the QSFP28 transceiver is taken to run through a complex web of transponders, amplifiers, signal conditioning, multiplexers, and network management.

Conclusion

There are many ways to transmit to the 100G network. QSFP28 modules are the suggested methods till today but no one can tell what the future will like. Both IEEE and MSA have published standards for 100G QSFP28. Customers can select the range according to their applications. QSFP28 full family is now available at CBO.

Fiber optic transceiver is popular nowadays. SFP and SFP+ transceivers are most widely among other transceivers. These are used for both telecommunication and data communication applications. The availability in various small sizes with various types. Allows users to select the appropriate one to provide the required optical coverage over single-mode fiber or multimode fiber. Keeping this in mind, manufacturers have designed various types of SFP and SFP+ modules to satisfy different demands. People are adept to use the compatible SFP/SFP+ transceivers instead of the original ones. In this passage, we will discuss this subject and will give you a basic understanding of compatible SFP and SFP+ modules.

SFP – Small Form Factor Pluggable Module

SFP transceiver is a hot-pluggable module used for both telecommunication and data communications applications serving high-speed demands. This can also be considered as an upgraded version of the GBIC module. SFP is most often used for Fast Ethernet or Gigabit Ethernet applications supporting speeds up to 4.25 Gbps.

SFP + – Small Form-Factor Pluggable Module

SFP+, as the name says, is an enhanced version of the SFP that supports data rates up to 16 Gbps. SFP+ supports 8 Gigabit/s Fibre Channel, 10 Gigabit/s Ethernet, and Optical Transport Network standard OTU2. This is a popular industry format supported by many network component vendors.

Why should we use compatible SFP or SFP+?

When one purchase a network component, one takes many aspects into considerations, such as price, functions, capability, etc. In the following section will list two main reasons why one choose compatible SFP and SFP+

Cost

As discussed in the above sections, there is no doubt that SFP and SFP+ are indispensable components in network communication. But when it comes to cost one has to take a wise decision as a cost for these transceiver modules are keeping rising for users. If we move towards brands like Cisco, Finisar, Dell or any other brand for transceivers, etc. we will find that these are too expensive for those who are deficient in capital. However, in the market, we have compatible SFP and SFP+ that are available at least half price of the original one. For example, a Cisco SFP-10G-SR compatible transceiver is available for only 16 dollars in the market today.

Compatibility

Now the other part of the story is many manufacturers restrict their devices to accept only original SFP modules as identified by their vendor ID. Each of them is unique and holds its own information in EEPROM. Third-party SFP manufacturers have introduced SFPs with “blank” programmable EEPROMs so that It can be reprogrammed to match any vendor ID.  For example a Finisar FTLF1318P2BTL is compatible with the original Finisar SFP transceiver, having all the same features as the original one does.

Conclusion

With the increase in demand for transceivers, compatible SFP and SFP+ with lower cost and high compatibility are becoming a perfect choice for most users. But one should keep in mind to have a firmware check for compatibility before installing, or the modules can’t work in good condition. At CBO we always strive to provide optical products and has earned a good reputation among our customers. You can find all kinds of transceivers here, such as 10G SFP+, 100Base SFP, XFP, and so on. We offer a variety of SFP and SFP+ transceivers which are fully compatible with major brands at a very lower price. We promise every product here is individually tested, walks through the testing challenges and 100% compatibility.

Fiber link performance has become a significant thing as data centers have started migrating to 40, 100, 200 and even 400G. A dead fiber link or a problematic module can bring a devastating impact in the form of system downtime – something not acceptable at any cost. Through this blog, we will discuss the basics around fiber link models and budget considerations.

Fiber Link Models

According to the IEE 802.3 standard fiber link model should be referred to as the “fiber link cabling model.” In this model specifications and characteristics of the various optical networking, elements are defined. In short, the EEE 802.3 standard deals with the characteristics of; connector performance, maximum reach, fiber cable performance, maximum acceptable connection loss, etc.

One or multiple numbers of optical fibers are utilized in the fiber optic cabling channels to support an optical link. Optical links provide an interconnection between transmitters at the MDI. When the components used in the construction of an optical link comply with specifications defined in the standard as mentioned earlier, then we can anticipate stable and optimized link performance at the physical Ethernet layer. Thus, our networks and data centers will work faster and better and with no or very less downtime.

The standard for fiber link performance covers these aspects:

Mechanical specifications of the optical interface aka MDI

Physical transmission media including Singlemode (OS1/OS2) and Multimode fiber (OM1, OM2, OM3, OM4, and OM5).

Physical transmission media including Singlemode (OS1/OS2) and Multimode fiber (OM1, OM2, OM3, OM4, and OM5).

Power budget (the maximum allowable spread between transmitter power and receiver power in OMA)

Power losses incurred by transmission over the fiber and through the transmitter

channel insertion losses (specified in dB and generally caused by fiber attenuation)

Maximum reaches in term of distance for various fiber types

Cable performance (cable skew)

Fiber connector performance (maximum insertion loss and total connector loss)

The relationship between Power Budgets and Fiber Link Models

As we know, channel insertion loss can be defined as a tolerable fiber link loss. Here, it is essential to understand that the power budget and channel insertion loss are two different things. Following is a general equation that can be used in the calculation of the power budget.

Power Budget = allocation for penalties + channel insertion loss + additional allowable insertion loss.

What is the Significance of the Power Budget?

Well, when we operate with having multiple points of connection in any fiber cabling system the knowledge of power budget and link model becomes critical. As we have discussed above, link performance becomes a critical parameter as data center undergoes upgradation and migrates to superior technology. Overloaded, broken or open fiber links are a no-no in today’s cutthroat environment.

How can this Information be linked to Data Center Performance?

Futureproof Your Network

Many data center engineers find themselves confused on whether they should buy and deploy new fiber cables or recycle installed cable. In such scenarios, making the right choice becomes very important as next-generation speed and performance impact enterprises. To futureproof your networking infrastructure, you have to take the right decisions in a timely fashion.

Minimize the Cost of Ownership

You have to understand that the cost of ownership includes a lot more than the costs of cabling and transceivers. The maintenance cost is also considered as an integral element of the ownership cost. You can have saved much money on the maintenance front by utilizing high-performance cable. These cables offer superior link performance and much flexible cabling provisions for cross-connects. Thus, acquiring quality cable alone can help you in cutting down on regular maintenance expenditures.

Optimize Your Existing Networking Architecture

Understanding and analyzing your fiber link budget can assist you in optimizing your existing fiber link design. The fiber link budget enables you to access the channel insertion loss as well. As an example, consider shorter cable runs that can help you in creating more connection points. Whereas, you can achieve longer-distances by using low-loss fiber cable and low-loss connectors.

Fiber optic networks are networks where the transmission of data is done with the help of optical transceivers and optical cables. The optical transceivers transmit an optical light down an optical cable. As in the case with standard Ethernet copper networks, optical networks are also influenced by exterior stress and interior properties and as a consequence some power is loss. This optical power loss is called Attenuation.

Fiber optic cables consist of fiber optic glass core and cladding, buffer coating, Kevlar strength components and a protective exterior material called a jacket. Depending on the optical cable type. These components can vary in size and strength. Unlike the copper cables which use electricity to transmit data, fiber optic cables use pulses of optical light for the same function. Their core is made of an ultra-pure glass which is surrounded by a mirror like cladding. When the light hits the cable it travels down the core constantly bouncing of the cladding until it reaches the final destination. There are two types of optical cables, Multi-mode and Single-mode. From the outside they look almost the same, however their interior plays a huge role in the optical attenuation. Single-mode fibers are used for a long range, high speed connections because of their tighter core and cladding which improve the light transmission by limiting the light bouncing of the cladding. Multi-mode fibers have larger core thus the light will bounce more and more power will be lost until it reaches the destination.

However, the optical attenuation of optical fibers is not only the lost power due to the core of the cable. High optical attenuation can be caused by absorption, scattering and physical stress on the cable like bending. Signal attenuation is generally defined as the ratio of optical input power to the optical output power. As the names suggest, optical input power is the power injected in the optical cable by the optical transceiver, and optical output power is the power received by the transceiver at the other end of the cable. The unit of attenuation is described as dB/km.

Absorption is one of the biggest causes for optical attenuation. This is defined as the optical power lost due to the conversion of the optical power into another form. Absorption is typically caused by a residual water vapors. Generally absorption is defined by two factors:

Imperfection in the atomic structure of the fiber material

The extrinsic and intrinsic fiber-material properties which represent the presence of impurities in the fiber-material

The extrinsic absorption is caused by impurities like trace metals, iron and chromium, introduced into the fiber during the manufacturing process. These trace metals are causing a power loss during the process of conversion when they are transitioning from one energy level to another.

The intrinsic absorption is caused by the basic properties of the fiber material. If the optical fiber material is pure, with no impurities and imperfections, then all absorption would be intrinsic. For example in fiber optics silica glass is used due to its low intrinsic absorption at certain wavelengths ranging from 700nm to 1600nm.

Scattering losses are caused by the density fluctuations in the fiber itself. These are produced during the manufacturing process. Scattering occurs when the optical light hits various molecules in the cable and bounces around. Scattering is highly dependent on the wavelength of the optical light. There are two types of scattering loss in optical fibers:

Rayleigh scattering- this scattering occurs at commercial fibers that operate at 700-1600nm wavelengths. Rayleigh scattering occurs when the size of the density fluctuation is less than 1/10 of the operating wavelength.

Mie scattering- this scattering occurs when the size of the density fluctuation is bigger than 1/10 of the operating wavelength.

Bending the fiber cable also causes attenuation. The bending loss is classified in micro-bends and macro-bends:

Micro-bends are small microscopic bends in the fiber which most commonly occur when the fiber is cabled Macro-bends on the other hand are bends that have a large radius of curvature relative to the cable diameter.

Another type of optical power loss is the optical Dispersion. Optical Dispersion represents the spreading of the light signal over time. There are two types of optical dispersion:

Chromatic dispersion which is spreading of the light signal resulting from the different speeds of the light rays Modal dispersion which is spreading of the light signal resulting from the different propagation modes of the fiber

Modal dispersion is most commonly limiting the maximum bit rate and link length in Multi- mode fibers. The Chromatic dispersion is the main culprit for the attenuation in Single-mode fibers.

Having this in mind we should always consider, test and calculate the possible attenuation of the fibers for deploying a stable network capable for future upgrades.

The Passive Optical Networks – PON are having a great expansion these days with the increased demand of the business and consumer of the Ethernet bandwidth. The expansion of the network it is possible if the total cost of ownership is lower than the revenues generated by customers.

In such PON systems, burst-mode optical transceivers are essential components. They are constructed by integrating transceiver-circuit, optical device, and module technologies.  The configuration of a typical optical transceiver is based on an optical sub-assembly (OSA). The transmitter and receiver modules are called TOSA and ROSA, respectively. A TOSA contains a semiconductor laser diode (LD), while a ROSA contains a photodiode (PD), optical lens, preamplifier, and passive electrical parts.

They are designed in small form-factor with some integrated optical sub-assemblies which can be suitable for high-density network. The major cost components of a transceiver module are the transmitter optical sub-assembly (TOSA), which converts an electrical and the receiver optical sub-assembly (ROSA). However, inside a BiDi (Bi-Directional) transceiver, there is a component with called “BOSA” (Bi Directional Optical Sub-Assemblies) which acts the role of TOSA and ROSA but with different principle.

In a PON system, one optical fiber is used for bidirectional transmission to reduce the network cost by using optical wavelength division multiplexing (WDM). For bidirectional transmission, a WDM filter is used: it lets transmitted optical signals pass through and reflects received optical signals. The optical bidirectional (BIDI) module is composed of a TOSA, a ROSA, and a WDM filter it lets transmitted optical signals pass through and reflects received optical signals.

TOSA Structure

The TOSA consists of a laser diode, optical interface, monitor photodiode, metal and/or plastic housing, and electrical interface. Depending upon the required functionality and application, other components may be present as well including filter elements and isolators. It is used to convert signal into an optical signal coupled into an optical fiber.

The transmitter mainly consists of an LD and its driver circuit with an automatic optical

power control (APC) circuit. Fabry-Pérot LDs (FP-LDs) and distributed feedback LDs (DFB-LDs) are widely used in optical transmission systems. FP-LDs are inexpensive and commonly used in ONUs. On the other hand, the LD of the OLT should provide a narrower wavelength for the optical signal than that of the ONU. The standardized allocation of optical wavelengths in a PON system is upstream around central frequency 1310nm and downstream data around central frequency 1490nm and video data downstream around 1540nm. To achieve this accuracy for the OLT transmitter, we used a DFB-LD, which can provide a narrow optical wavelength spectrum.

ROSA Structure

The ROSA consists of a photodiode, optical interface, metal and/or plastic housing, and electrical interface. Depending upon the required functionality and application, other components may be present as well including amplifiers. It is used to receive an optical signal from a fiber and convert it back into an electrical signal.

The receiver consists of a PD, which converts a received optical signal to an electrical current signal, and amplifiers. The amplifiers reshape input signals degraded by long-distance transmission. The amplifier circuit consists of a preamplifier and a post-amplifier. The preamplifier converts a current signal to a voltage signal and amplifies the converted signal. The post-amplifier equalizes the output signal of the preamplifier to an amplitude level suitable for input to the following digital circuit. The PD and preamplifier are assembled in a ROSA module because the preamplifier is very sensitive to mounting conditions. The ROSA module makes it easy to handle the optical module and better performance is obtained.

BOSA Structure

The BOSA consists of a TOSA, a ROSA and a WDM filter so that it can use bidirectional technology to support two wavelengths on each fiber. The most valuable advantage of BiDi transceivers is saving much cost on fibers.

Transceiver assembly

A small-form-factor pluggable (SFP) chassis, which can be attached to and detached from an electrical substrate. It contains a BIDI optical module with metal-can-type TOSA and ROSA, an LD driver IC, and a post-amplifier IC.

A burst-mode optical transceiver for gigabit-per-second-class PON systems. The use of various optical and electrical module techniques along with our developed receiver ICs enabled us to obtain high performance with an inexpensive PIN Photodiode. The transceiver is built on a small SFP chassis. It achieved a sensitivity of –29.7 dBm and an output optical power of more than +5 dBm. This optical transceiver will enable us to decrease the cost of gigabit-per-second-class PON systems.