Читать книгу CompTIA Network+ Study Guide - Todd Lammle - Страница 26

A lot of us rely on wireless networking methods that work using technologies like radio frequency and infrared, but even wireless depends on a physical media backbone in place somewhere. And the majority of installed LANs today communicate via some kind of cabling, so let's take a look at the three types of popular cables used in modern networking designs:

● Coaxial

● Twisted-pair

● Fiber optic

Coaxial Cable

Coaxial cable, referred to as coax, contains a center conductor made of copper that's surrounded by a plastic jacket with a braided shield over it. A plastic such as polyvinyl chloride (PVC) or fluoroethylenepropylene (FEP, commonly known as Teflon) covers this metal shield. The Teflon-type covering is frequently referred to as a plenum-rated coating, and it's definitely expensive but often mandated by local or municipal fire code when cable is hidden in walls and ceilings. Plenum rating applies to all types of cabling and is an approved replacement for all other compositions of cable sheathing and insulation like PVC-based assemblies.

The difference between plenum and non-plenum cable comes down to how each is constructed and where you can use it. Many large multistory buildings are designed to circulate air through the spaces between the ceiling of one story and the floor of the next; this space between floors is referred to as the plenum. And it just happens to be a perfect spot to run all the cables that connect the legions of computers that live in the building. Unless there's a fire – if that happens, the non-plenum cable becomes a serious hazard because its insulation gives off poisonous smoke that gets circulated throughout the whole building. Plus, non-plenum cables can actually become “wicks” for the fire, helping it quickly spread from room to room and floor to floor – yikes!

Because it's a great goal to prevent towering infernos, the National Fire Protection Association (NFPA) demands that cables run within the plenum have been tested and guaranteed as safe. They must be fire retardant and create little or no smoke and poisonous gas when burned. This means you absolutely can't use a non-plenum-type cable in the plenum, but it doesn't mean you can't use it in other places where it's safe. And because it's a lot cheaper, you definitely want to use it where you can.

Thin Ethernet, also referred to as thinnet or 10Base2, is a thin coaxial cable. It is basically the same as thick coaxial cable except it's only about 5 mm, or 2/10″ diameter coaxial cable. Thin Ethernet coaxial cable is Radio Grade 58, or just RG-58. Figure 3.1 shows an example of thinnet. This connector resembles the coaxial connector used for cable TV, which is called an F connector.

Figure 3.1 A stripped-back thinnet cable

Oh, by the way, if you use thinnet cable, you've got to use BNC connectors to attach stations to the network, as shown in Figure 3.2, and you have to use 50 ohm terminating resistors at each end of the cable in order to achieve the proper performance.

You don't have to know much about most coax cable types in networks anymore, especially the thinnet and thicknet types of coaxial cable. Thicknet was known as RG-8, was about 1/2” in diameter, also requiring 50 ohm terminating resistors on each end of the cable. Nowadays, we use 75 ohm coax for cable TV; using coax in the Ethernet LAN world is pretty much a thing of the past, but we do use them for high-bandwidth runs in our data centers. RG-6, or CATV coax, is used in our broadband world.

Figure 3.2 Male and female BNC connectors

You can attach a BNC connector to the cable with a crimper that looks like a weird pair of pliers and has a die to crimp the connector. A simple squeeze crimps the connector to the cable. You can also use a screw-on connector, but I avoid doing that because it's not very reliable.

You can use a BNC coupler to connect two male connectors together or two female connectors together.

Table 3.1 lists some specifications for the different types of coaxial cable, but understand that we use only RG-59 and RG-6 in today's world.

An advantage of using coax cable is the braided shielding that provides resistance to electronic pollution like electromagnetic interference (EMI), radio frequency interference (RFI), and other types of stray electronic signals that can make their way onto a network cable and cause communication problems.

Table 3.1 Coaxial cable specifications

Twisted-Pair Cable

Twisted-pair cable consists of multiple individually insulated wires that are twisted together in pairs. Sometimes a metallic shield is placed around them, hence the name shielded twisted-pair (STP). Cable without outer shielding is called unshielded twisted-pair (UTP), and it's used in twisted-pair Ethernet (10BaseT, 100BaseTX, 1000BaseTX) networks.

Ethernet Cable Descriptions

Ethernet cable types are described using a code that follows this format: N<Signaling>X. The N refers to the signaling rate in megabits per second. <Signaling> stands for the signaling type – either baseband or broadband – and the X is a unique identifier for a specific Ethernet cabling scheme.

Here's a common example: 100BaseX. The 100 tells us that the transmission speed is 100Mb, or 100 megabits. The X value can mean several different things, for example, a T is short for twisted-pair. This is the standard for running 100-megabit Ethernet over two pairs (four wires) of Category 5, 5e, or 6 UTP.

So why are the wires in this cable type twisted? Because when electromagnetic signals are conducted on copper wires in close proximity – like inside a cable – it causes interference called crosstalk. Twisting two wires together as a pair minimizes interference and even protects against interference from outside sources. This cable type is the most common today for the following reasons:

● It's cheaper than other types of cabling.

● It's easy to work with.

● It allows transmission rates that were impossible 10 years ago.

UTP cable is rated in these categories:

Category 1 Two twisted wire pairs (four wires). It's the oldest type and is only voice grade – it isn't rated for data communication. People refer to it as plain old telephone service (POTS). Before 1983, this was the standard cable used throughout the North American telephone system. POTS cable still exists in parts of the Public Switched Telephone Network (PSTN) and supports signals limited to the 1MHz frequency range.

Category is often shortened to Cat. Today, any cable installed should be a minimum of Cat 5e because some cable is now certified to carry bandwidth signals of 350MHz or beyond. This allows unshielded twisted-pair cables to exceed speeds of 1Gbps – fast enough to carry broadcast-quality video over a network.

Category 2 Four twisted wire pairs (eight wires). It handles up to 4Mbps, with a frequency limitation of 10MHz, and is now obsolete.

Category 3 Four twisted wire pairs (eight wires) with three twists per foot. This type can handle transmissions up to 16MHz. It was popular in the mid-1980s for up to 10Mbps Ethernet, but it's now limited to telecommunication equipment and, again, is obsolete for networks.

Category 4 Four twisted wire pairs (eight wires), rated for 20MHz; also obsolete.

Category 5 Four twisted wire pairs (eight wires), rated for 100MHz. But why use Cat 5 when you can use Cat 5e for the same price? I am not sure you can even buy plain Cat 5 anymore!

Category 5e (Enhanced)

Four twisted wire pairs (eight wires), rated for 100MHz but capable of handling the disturbance on each pair that's caused by transmitting on all four pairs at the same time – a feature that's needed for Gigabit Ethernet. Any category below 5e shouldn't be used in today's network environments.

Figure 3.3 shows a basic Cat 5e cable with the four wire pairs twisted to reduce crosstalk.

Category 6 Four twisted wire pairs (eight wires), rated for 250MHz. Cat 6 became a standard back in June 2002. You would usually use it as riser cable to connect floors together. If you're installing a new network in a new building, there's no reason to use anything but Category 6 UTP cabling as well as running fiber runs between floors.

Category 6a (Augmented) Basic Category 6 cable has a reduced maximum length when used for 10GBaseT; however, Category 6a cable, or Augmented Category 6, is characterized to 500MHz and has improved crosstalk characteristics, which allows 10GBaseT to be run for up to 100 meters. The most important point is a performance difference between Electronic Industries Alliance and Telecommunication Industry Association (EIA/TIA) component specifications for the NEXT (near-end crosstalk) transmission parameter. Running at a frequency of 500MHz, an ISO/IEC Cat 6a connector provides double the power (3db) of a Cat 6A connector that conforms with the EIA/TIA specification. Note that 3dB equals a 100 percent increase of a near-end crosstalk noise reduction. This is our future cable indeed!

Figure 3.3 Cat 5e UTP cable

Connecting UTP

BNC connectors won't fit very well on UTP cable, so you need to use a registered jack (RJ) connector, which you're familiar with because most telephones connect with them. The connector used with UTP cable is called RJ-11 for phones that use four wires; RJ-45 has four pairs (eight wires), as shown in Figure 3.4.

Figure 3.4 RJ-11 and RJ-45 connectors

Figure 3.5 shows the pin-outs used in a typical RJ-45 connector. Looking from the bottom of the connector, pin 1 would be on the left.

Figure 3.5 The pin-outs in an RJ-45 Connector, T568B standard

Most of the time, UTP uses RJ connectors, and you use a crimper to attach them to a cable, just as you would with BNC connectors. The only difference is that the die that holds the connector is a different shape. Higher-quality crimping tools have interchangeable dies for both types of cables. We don't use RJ-11 for local area networks (LANs), but we do use them for our home Digital Subscriber Line (DSL) connections.

RJ-11 uses two wire pairs, and RJ-45 uses four wire pairs.

There's one other type of copper connector called the RJ-48c, which looks exactly like an RJ-45 connector. This plug is very similar to the RJ-45 in that it has four wire pairs, but they are wired differently and used for different circumstances.

RJ-45 is mainly used in LANs with short distances (typically up to 100 meters), where the RJ-48c wiring type would be used with a T1 connection, which is a long-distance wide area network (WAN). In addition, to protect the signal in an RJ-48c, the wires are typically shielded, whereas the RJ-45 uses unshielded wiring.

Category 5e Cabling Tips

If you want data rates faster than 10Mbps over UTP, ensure that all components are rated to deliver this and be really careful when handling all components. If you yank on Cat 5e cable, it will stretch the number of twists inside the jacket, rendering the Cat 5e label on the outside of the cable invalid. Also, be certain to connect and test all four pairs of wire. Although today's wiring usually uses only two pairs (four wires), the standard for Gigabit Ethernet over UTP requires that all four pairs (eight wires) be in good condition.

Also be aware that a true Cat 5e cabling system uses rated components from end to end, patch cables from workstation to wall panel, cable from wall panel to patch panel, and patch cables from patch panel to hub. So if any components are missing, or if the lengths don't match the Category 5e specification, you just don't have a Category 5e cabling installation. And certify that the entire installation is Category 5e compliant. I've got to warn you that doing this requires some pretty pricey test equipment to make the appropriate measurements!

Fiber-Optic Cable

Because fiber-optic cable transmits digital signals using light impulses rather than electricity, it's immune to EMI and RFI. Anyone who's seen a network's UTP cable run down an elevator shaft would definitely appreciate this fiber feature. Fiber cable allows light impulses to be carried on either a glass or a plastic core. Glass can carry the signal a greater distance, but plastic costs less. Whichever the type of core, it's surrounded by a glass or plastic cladding with a different refraction index that reflects the light back into the core. Around this is a layer of flexible plastic buffer that can be wrapped in an armor coating that's usually Kevlar, which is then sheathed in PVC or plenum.

The cable itself comes in either single-mode fiber (SMF) or multimode fiber (MMF); the difference between them is in the number of light rays (the number of signals) they can carry. Multimode fiber is most often used for shorter-distance applications and single-mode fiber for spanning longer distances.

Although fiber-optic cable may sound like the solution to many problems, it has its pros and cons just like the other cable types.

Here are the pros:

● It's completely immune to EMI and RFI.

● It can transmit up to 40 kilometers (about 25 miles).

And here are the cons:

● It's difficult to install.

● It's more expensive than twisted-pair.

● Troubleshooting equipment is more expensive than twisted-pair test equipment.

● It's harder to troubleshoot.

Single-Mode Fiber

Single-mode fiber-optic cable (SMF) is a very high-speed, long-distance media that consists of a single strand – sometimes two strands – of glass fiber that carries the signals. Light-emitting diodes (LEDs) and laser are the light sources used with SMF. The light source is transmitted from end to end and pulsed to create communication. This is the type of fiber cable employed to span really long distances because it can transmit data 50 times further than multimode fiber at a faster rate.

Clearly, because the transmission media is glass, the installation of SMF can be a bit tricky. Yes, there are outer layers protecting the glass core, but the cable still shouldn't be crimped or pinched around any tight corners.

Multimode Fiber

Multimode fiber-optic cable (MMF) also uses light to communicate a signal, but with it, the light is dispersed on numerous paths as it travels through the core and is reflected back. A special material called cladding is used to line the core and focus the light back onto it. MMF provides high bandwidth at high speeds over medium distances (up to about 3,000 feet), but beyond that it can be really inconsistent. This is why MMF is most often used within a smaller area of one building; SMF can be used between buildings.

MMF is available in glass or in a plastic version that makes installation a lot easier and increases the installation's flexibility.

APC vs UPC

The choice between angle-polished connectors (APCs) and ultra-polished connectors (UPCs) can make a pretty big difference on how your network will perform.

The ultra-polished connector looks like what you'd expect to find in a fiber-optic end. The cut is perfectly straight, as shown in Figure 3.6.

Figure 3.6 UPC typical fiber connector end

The angle-polished connector looks like Figure 3.7. Notice the perfectly cut angle, which seems odd, but there is a reason for this and it's a good one!

Figure 3.7 The superior APC fiber connector

Unlike the UPC, where the light is reflected back down to the core of the fiber cable, which causes a loss of db called a return loss because the angled connector causes the light to reflect back into the cladding – the thick sides of the glass instead of the core. But the APC doesn't cause nearly as much db loss when using this type of connector. Very cool design indeed!

Fiber-Optic Connectors

A whole bunch of different types of connectors are available to use with fiber-optic cables, but the two most popular are the straight tip (ST) and subscriber (or square) connector (SC). The ST fiber-optic connector (developed by AT&T) is one of the most widely used fiber-optic connectors; it uses a BNC attachment mechanism similar to thinnet's that makes connections and disconnections fairly frustration free. In fact, this is the feature that makes this connector so popular. Figure 3.8 shows an example of an ST connector. Notice the BNC attachment mechanism.

Figure 3.8 An example of an ST connector

The SC connector is another type of fiber-optic connector. As you can see in Figure 3.9, SC connectors are latched– a mechanism holds the connector in securely and prevents it from falling out.

Figure 3.9 A sample SC connector

SC connectors work with both single-mode and multimode optical fibers and will last for around 1,000 matings. They're being used more now but still aren't nearly as popular as ST connectors for LAN connections.

Another type of connector I want to mention before moving on to the SFF connector is the FC connector, or field assembly connector, also called the ferrule connector, which isn't very popular. It's still used in telecommunications and measurement equipment with single-mode lasers, but the SC is a way more popular fiber end. The only reason I mention it here is because it is an exam objective; other than that you probably won't ever see it in production. These look identical to ST connectors.

You can also get a fiber coupler in order to connect an ST to an SC connector, for example, but you will lose a lot of your power (db) if you do so.

Small Form Factor Fiber-Optic Connectors

Another cool fiber-optic connector is the small form factor (SFF) connector, which allows more fiber-optic terminations in the same amount of space than its standard-sized counterparts. The two most popular versions are the mechanical transfer registered jack (MT-RJ or MTRJ), designed by AMP, and the Local Connector (LC), designed by Lucent.

Should I Use Copper or Fiber?

If your data runs are measured in miles, fiber optic is your cable of choice because copper just can't give you more than about 1,500 feet without electronics regenerating the signal. The standards limit UTP to a pathetic 328 feet.

Another good reason to opt for fiber is if you require high security because it doesn't create a readable magnetic field. Although fiber-optic technology was initially super expensive and nasty to work with, it's now commonly used for Gigabit or 10GB Internet backbones.

Ethernet running at 10Mbps over fiber-optic cable to the desktop is designated 10BaseFL; the 100Mbps version of this implementation is 100BaseFX. The L in the 10Mbps version stands for link. Other designations are B for backbone and P for passive.

The MT-RJ fiber-optic connector was the first small form factor fiber-optic connector to be widely used, and it's only one-third the size of the SC and ST connectors it most often replaces. It offers these benefits:

● Small size

● TX and RX strands in one connector

● Keyed for single polarity

● Pre-terminated ends that require no polishing or epoxy

● Easy to use

Figure 3.10 shows an example of an MT-RJ fiber-optic connector.

Figure 3.10 A sample MT-RJ fiber-optic connector

LC is a newer style of SFF fiber-optic connector that's pulling ahead of the MT-RJ. It's especially popular for use with Fibre-Channel adapters (FCs) and is a standard used for fast storage area networks and Gigabit Ethernet adapters. Figure 3.11 depicts an example of the LC connector.

Figure 3.11 A sample LC fiber-optic connector

It has similar advantages to MT-RJ and other SFF-type connectors but it's easier to terminate. It uses a ceramic insert just as standard-sized fiber-optic connectors do.

Media Converters

Sometimes, you'll need to convert from one media type to another. Maybe you need to go from one mode of fiber to another mode, or in an even more extreme case, you need to go from fiber to Ethernet. If you're faced with situations like these, you'll need to be familiar with some of the more common media converters:

Single-Mode Fiber to Ethernet These devices accept a fiber connector and an Ethernet connector and convert the signal from Ethernet and single-mode fiber (see Figure 3.12).

Multimode Fiber to Ethernet These devices accept a fiber connector and an Ethernet connector and convert the signal from Ethernet and multi-mode fiber (see Figure 3.13).

Fiber to Coaxial These devices accept a fiber connector and a coaxial connector and convert digital signals from optical to coax (see Figure 3.14).

Single-Mode to Multimode Fiber These devices accept a single-mode fiber connector and a multimode fiber connector and convert the signals between the two (see Figure 3.15).

Figure 3.12 Single-mode fiber to Ethernet

Figure 3.13 Multimode fiber to Ethernet

Figure 3.14 Fiber to coaxial

Figure 3.15 Single-mode to multimode fiber

Serial Cables

Except for multimode fiber, all the cable varieties I've talked about so far are considered serial cable types. In network communications, serial means that one bit after another is sent out onto the wire or fiber and interpreted by a network card or other type of interface on the other end.

Each 1 or 0 is read separately and then combined with others to form data. This is very different from parallel communication where bits are sent in groups and have to be read together to make sense of the message they represent. A good example of a parallel cable is an old printer cable – which has been replaced by USB, as I'll get to in a minute.

RS-232

Recommended Standard 232 (RS-232) was a cable standard commonly used for serial data signals connecting the DTE and the DCE, such as a computer's serial port to an external modem.

Figure 3.16 shows an example of one of the many types of RS-232 cables. These cables normally connect to a connector on the device called a DB-9.

Figure 3.16 An RS-232 cable end

Because laptops don't even come with these types of connectors anymore, they've pretty much been replaced by things like USB, Thunderbolt, and FireWire.