Review: 4 powerline kits step in when Wi-Fi fails

20.08.2015
Are there places in your home or office where your Wi-Fi signal doesn't reach You're not alone. Because of Wi-Fi's limited range -- and old-school building construction techniques like thick plaster, brick or masonry walls -- even the best Wi-Fi networks have dead zones. If you have that problem, you have several options.

You can start drilling holes in your walls so you can run network cabling. You can try a Wi-Fi range extender (which might help, depending on the strength of your Wi-Fi and the efficiency of the extender). Or you can use a powerline kit, which routes data over your electric cables by piggybacking the data on top of the electrical current's 60-hertz wave and then extracting the data at the other end.

"The beauty of powerline networking is that it is inexpensive, has reasonable range and doesn't require loading any software," says Jeff Heynen, research director at Infonetics, a telecommunications research firm. "That makes it the easiest way to liven up a networking dead zone or extend a LAN."

Using a powerline kit is indeed about as easy as it gets. All you do is connect one of the adapters to your router via an Ethernet cable, and plug it into a nearby AC outlet. You then plug the other adapter into an outlet near your computer, connect it to your computer via an Ethernet cable and voila! Instant networking.

The first generation of powerline adapters could slow down your broadband considerably: They only allowed a single stream of data at once and topped out at a speed of 500Mbps. Current adapters, however, adhere to the HomePlug AV2 standard, which uses the MIMO (multiple-input, multiple-output) technology to move many streams of data at once without getting bogged down.

There are a few "gotchas" you should take into consideration, though, when it comes to powerline networking. To begin with, this class of networking equipment struggles to get to one-tenth of its theoretical maximum throughput of 1.2Gbps to 2Gbps, depending on the kit. In fact, the actual peak data flow is a little over 200Mbps of delivered data for some of the powerline kits and a lot less for others.

This should be more than enough for distributing a Web connection in a small office; in fact, it is in the same class with all but the newest 802.11ac Wi-Fi routers and access points. It will, however, feel like moving data through digital molasses when you're transferring big files.

There's an additional problem associated with powerline connections: The two powerline adapters must be plugged in within the maximum range of between 500 and 600 feet.

While that might sound like a lot (particularly in light of Wi-Fi's ability to reach only about one-third as far), we're not talking the distance between two points inside your home or office, but the length of the behind-the-wall electrical cables that will carry the data. As a result, it's a good rule of thumb to double the distance inside the building between the two powerline adapters to account for the behind-the-scenes wiring. In other words, if the two devices will be about 50 feet apart physically, you should count it as about 100 feet apart as far as your cabling is concerned.

And powerline adapters not only use an AC outlet, but can be large enough that they block one or more adjacent outlets. If that's a concern, some adapters contain their own power outlets (for a slightly higher cost) so that you don't get shortchanged. (None of the adapters devices reviewed here offers that, though.)

There are also a couple of things users should be aware of. First, while you can use an extension cord with a powerline adapter, it should have only a single port: a power strip can dissipate the data signal over several outlets.

In addition, make sure your electrical connection is grounded -- because MIMO adapters use the ground and will only provide a single stream of data without it. That means using a three-prong outlet; it is possible to use a three-to-two prong converter if you absolutely have to, but make sure you attach the converter's grounding wire to your outlet plate's center screw (which, in older homes, will usually be connected to the ground line).

Despite the shortcomings listed above, after trying out four of the latest powerline kits for a month, I came to think of them as the equivalent of networking extension cords -- really handy if you need one. These included: the D-Link PowerLine AV2 2000 Gigabit Starter Kit (model DHP-701AV), the Netgear Powerline 1200 (model PL 1200), the Trendnet Powerline 1200 AV2 Adapter Kit (model TPL-420E2K) and the ZyXel 1200 Mbps Powerline Gigabit Ethernet Network Adapter (model PLA5405).

I used them to supply my office with data, connect a printer that's out of the reach of a Wi-Fi access point and add network access to a storage room in order to better track inventory. And while they differ widely in range and throughput, all four can offer Internet access in problematic areas of a home or office.

In short, powerline Ethernet is one of those rare win-win situations, where you can get networking where it's needed without breaking the bank or raising a cloud of plaster dust.

Its adapters may not be the smallest or least expensive powerline devices available, but D-Link's PowerLine AV2 2000 Gigabit Starter Kit uses high-end components and excels at delivering data when the server and client are close to each other.

The D-Link kit comes with a pair of powerline units that each measure 1.3 x 2.9 x 3.9 in. They're the largest of the four powerline devices I looked at, but they don't obstruct the AC outlet above the one used by the unit. However, unlike the other three powerline adapters reviewed here, each D-Link adapter is so wide that it can block an outlet to the right or left of it.

Like the others, each adapter is designed as a white plastic box with a three-prong grounded electrical plug. A trio of LEDs shows that the unit is powered, has linked with another powerline device and is connected to the host network. On the bottom of the unit, there's a recessed reset switch, an Ethernet port and a button for encrypting the data stream.

Inside is Broadcom's AV2 MIMO chipset, which is capable of a maximum data throughput of 2Gbps, 67% faster than the 1.2Gbps Qualcomm QCA7500 chip that the other three systems use. As its name implies, it can handle multiple lanes of data traffic at once.

In addition to the two adapters, the kit includes two short Ethernet cables. (You can connect up to 16 powerline adapters.) To connect the D-Link system, I just plugged one adapter in near my router and the other at the closest AC outlet to the client. Each adapter's lights started blinking, and in eight seconds I was connected.

Like the others reviewed here, the D-Link units can scramble the data using 128-bit AES encryption. To enable the encryption and pair the units, you press the button on the bottom of each adapter for a few seconds -- you have a couple of minutes to go from one adapter to the other.

The D-Link system's ability to dole out data blew the others away -- in fact, it was the only one of the four to break the 200Mbps barrier in testing. However, I found that it works best at close range, where it is least needed.

At 50 feet from the server, the D-Link registered a throughput of 204.2Mbps, which dropped off to 130.1Mbps at 250 feet and 30.9 Mbps at 450 feet.

The D-Link had a top range of 535 feet -- 35 feet farther than the Netgear system but not as far as the Trendnet kit's 595-foot range.

The units are far from the greenest of the four sets; I measured the power consumption for two adapters at 9 watts when on and 2 watts when asleep. It had an estimated annual power bill of $7.60 for the pair of devices if used for 10 hours a day during workdays, assuming a power cost of 12 cents per kilowatt hour (the national average). That's roughly four times what some of the others cost to use.

In my office, which is separated from my router by about 400 feet of electrical cabling, the device showed three green lights, indicating a strong connection. It was able to deliver a 45.6Mbps Web connection from my 50Mbps broadband line and delivered smooth HD and 4K videos; it also delivered fine smooth videoconferencing video with well-synched audio.

At the time of writing, the D-Link was on sale for $100. It is available only as a kit -- in other words, to add nodes, you'll need to buy them two at a time.

With the highest throughput at close range, the D-Link can help link different parts of a network, but its throughput can't match the others at longer distances, where it's really needed.

The long, narrow Netgear Powerline 1200 adapters are designed so that they won't block nearby AC outlets, and at $80, the kit is among the least expensive of its peers.

At 1.2 x 2.2 x 4.5 in., each Netgear adapter is slightly larger than the tiny ZyXel unit but much smaller than the D-Link. Its white plastic case has three LEDs on the front to show it's powered, has a network connection and has linked with another powerline device.

There are vent holes on the side, while underneath you'll find an Ethernet port as well as buttons for resetting the system and for encrypting the data flow.

Like the ZyXel and Trendnet systems, the Netgear kit is based on Qualcomm's QCA 7500 chipset; and, like them, it tops out at a theoretical maximum throughput of 1.2Gbps. It can work with several streams at once and link up as many as 16 individual powerline devices.

The Netgear kit includes a pair of adapters and Ethernet cables; as with the D-Link, you can't buy individual adapters. After I plugged the two devices into AC outlets, they started communicating in only seven seconds, making this the fastest of the four to get started.

If you want to encrypt the data, you just press the Encryption button on the bottom for a few seconds, first on the unit near the server and then on the one near the client. The devices automatically take care of the coding and decoding, as well as key management and distribution.

Everything worked well, and the Netgear proved to be a reliable way to move data around. However, its throughput was disappointing. At 50 feet, it moved only 151.2Mbps, 26% less than the D-Link's throughput. At 250 and 450 feet, the Netgear's flow dropped to 110.2Mbps and 20.9Mbps, respectively -- well behind the Trendnet's throughput. It had a top range of only 500 feet, the shortest of the group.

On the other hand, the two Netgear adapters consumed only 2.4 watts of electricity when being used and 0.4 watts when in idle mode, making it much less power-hungry than the D-Link. I estimated that it would cost $1.80 a year to use a pair of the devices if they were used for 10 hours every weekday and idle the rest of the time -- the lowest cost of the four kits reviewed here.

In several days of testing over about 400 feet of electrical wiring, the Netgear adapter in my office was able to deliver only 24.4Mbps from my 50Mbps Web connection, the lowest throughput of the group. Part of the problem was that its Ethernet and powerline lights constantly blinked red, indicating that the device was nearly out of range. It did okay for basic HD videos, email and Web work, but faltered and halted when playing 4K videos. When I tested the device at about 350 feet, all videos played with no problem.

Priced at about $80, the Netgear Powerline 1200 ties the Trendnet system as the cheapest powerline kit, but it can't keep up with the others in terms of range and throughput.

Trendnet's Powerline 1200 AV2 Adapter Kit (TPL-420E2K) blows away the field with the best price, longest range and fastest throughput at long distances.

The Trendnet kit comes with a pair of powerline units that, at 1.5 x 2.3 x 3.1 in., are midrange in size. They don't obstruct adjacent outlets, but they do stick out the furthest from the outlet.

Each unit's white plastic case has a grounded power plug and three rectangular LEDs that are vertically arranged on the right side to show that the device is turned on, has an Ethernet link and has linked with another powerline device. On the bottom are the expected Ethernet port, a recessed reset button and a Sync button for getting the powerline devices to connect.

Like the powerline devices from ZyXel and Netgear, the Trendnet uses Qualcomm's QCA 7500 chipset and has a maximum theoretical throughput of 1.2Gbps. It can work with several streams at once and link up with as many as eight individual powerline adapters.

The Trendnet kit comes with a pair of Ethernet cables and a CD that includes the manual as well as a handy utility for finding the powerline devices on your network. You don't need to use it, but it's nice to have for troubleshooting or figuring out what's connected.

As with the others, to get the Trendnet working, you plug in one device near the router and the other where you want the client set up. Unlike the other kits in this roundup, however, the Trendnet can only connect using an encrypted stream of data, which means that the setup is a little more involved.

Once you've connected both units, you have to press the Sync button for 15 seconds on the device near the server to generate a new encryption key, and then press it again for about two seconds to distribute the key. You then press the Sync button on the other device for two seconds so that the pair will link using 128-bit AES encryption. During my tests, it took 30 seconds to complete the link-up versus less than 10 seconds for the others.

It's worth the wait, though -- the Trendnet had the longest range of the kits I tested: 595 feet, 60 feet farther than the ZyXel and D-Link, and 95 feet farther than the Netgear.

Although the Trendnet's throughput at 50 feet was a mediocre 179.5Mbps, at longer distances -- where the system is likely to be used -- it led the pack. At 250 feet it delivered 127.3Mbps, while at 450 feet, it pumped out 76.4Mbps, more than twice as fast as the D-Link, its closest competitor.

The Trendnet didn't use much power, either. The system consumed 3.3 watts when active and 0.4 watts when idle. This adds up to an annual cost of $2.20 for the pair of devices, assuming it's used for 10 hours every weekday and idle the rest of the time.

I set the Trendnet up in my office and used it for several days as my primary Internet connection. The device showed three green lights and passed through 47.9Mbps of data from my 50Mbps broadband line. It was great for viewing HD and 4K videos as well as general work (like email and Web browsing) and videoconferencing.

At a cost of about $80 through online retailers, the Trendnet kit is the bargain of the group; you can also buy individual adapters for about $43 each. It not only delivers the best range and throughput where it counts the most, but is economical to get and use.

Small and unobtrusive, ZyXel's 1200 Mbps Powerline Gigabit Ethernet Network Adapter (PLA5405) kit is able to connect with more units than the others reviewed here -- although most people won't need to add the number of devices it can support.

At 1.1 x 2.5 x 3.6 in., each white plastic ZyXel is one-third smaller than the D-Link adapters and leaves more than enough room to be able to use the AC outlets above, below and on either side.

It has a three-prong grounded power plug as well as three LEDs that show it's turned on, has found another powerline device and is connected to the network. There's an Ethernet port and a single button underneath that can encrypt the data flow with 128-bit AES coding or reset the system to its default settings.

The timing can be tricky, however, because the button performs different tasks depending on how long you hold it down. If you hold it down for up to three seconds, you'll encrypt the data stream; if you hold it for five to eight seconds, you'll wipe any network-related configuration data from it. Hold the button for 10 to 15 seconds and the unit is reset to its factory configuration. So you have to pay attention to your stopwatch.

The ZyXel adapter is powered by Qualcomm's QCA 7500 chipset, which hits a theoretical ceiling at 1.2Gbps. With the latest firmware, it can hit a theoretical maximum throughput of 1.3Gbps and can link up to 64 powerline devices at once. ZyXel engineers, however, recommend limiting a network to 16 units.

Like the others, the ZyXel kit comes with a pair of short Ethernet cables. It also has a CD that contains PDFs of the data sheet, quick start guide and manual.

I plugged both devices in and after 10 seconds of startup time, all three lights were glowing and I was connected and online. Despite a good start, hitting 188.4Mbps at 50 feet, its throughput had fallen to 55.8Mbps at 250 feet, barely half what the others were delivering. Its available bandwidth was just 12.3Mbps at 450 feet, one-sixth the level that the TrendNet system delivered.

Its 535-foot range matched that of the D-Link kit, but was short of the 595-foot range that the Trendnet system was capable of.

After I set the ZyXel up in my office, its HomePlug light glowed red, indicating that it was nearly out of range. Still, it sent through 35.9Mbps of throughput from my 50Mbps broadband connection. While it handled HD video and videoconferencing OK, it balked on 4K videos served up from my RAID storage device. The streams frequently stopped and faltered. (In other tests, though, the ZyXel delivered high-quality video at 350 and 400 feet, but at 450 -feet it was dropping frames.)

The device made up for its data-handling shortcomings with extremely low power consumption. While on, the two adapters used 3.2 watts, which dropped to 0.4 watts when idle. If it's used for 10 hours a day during the week and off the rest of the time, it should cost an estimated $2.20 to operate over a year, assuming that electricity costs the national average of 12 cents per kilowatt hour.

If you want single units, you can find them for about $80 (although at that price, you're probably better off getting the two-device kit, which you can also find for about $80).

The ZyXel kit doesn't cost much to operate and can support more adapters than its competition. However, not many users will need 64 (or even 16) adapters, and its throughput could be better.

Whether it's to plug a hole in a Wi-Fi network, extend the reach of a LAN or get a networked printer connected, range and throughput count for everything with powerline devices.

Although I really liked the size and economy of the ZyXel 1200, its mediocre range and subpar throughput were disappointing. While it does well with the sender and receiver close to each other, as soon as they are 200 feet apart, its performance drops off.

The same goes for Netgear's PL1200, which had the shortest range of the four. With estimated operating expenses of just $1.80 a year, however, it is the cheapest to use.

The only one of the four with a 2Gbps powerline chip, D-Link's Powerline AV2 2000 starts off very strong and moves over 200Mbps at short distances. However, at about 250 feet its throughput falls off quickly. It also uses a lot of power compared to the others, although at an estimated $7.60 a year in electricity costs, it won't be a major burden.

Of the four kits I looked at, one stood out head and shoulders above the rest: the Trendnet Powerline 1200. Despite its slow setup procedure, it dominated the competition when the two devices were 250 feet apart or more. In other words, it excelled where it would be most useful; it also had the farthest range of the group at 595 feet.

Inexpensive to get and use, the Trendnet can take networking to new places -- such as any nearby AC outlet.

To see how these powerline kits stack up, I tested each of them for speed and range using my office's Gigabit Ethernet network.

I started by plugging each powerline adapter into an AC outlet near my router (and the building's circuit breaker panel). Then I used the included Ethernet cable to connect the adapter to my router.

I used two test computers: a Surface 3 tablet (I used the Ethernet port on its docking station) and an Acer Revo One with a Core i5 processor, 8GB of RAM and 2.25GB of storage

To test the speed of the connection, I used the PassMark PerformanceTest networking suite of tasks. I used PassMark's variable block size that ranged from 1KB to 16KB in a 60-second test run, and also ran Ookla's SpeedTest.net to check on its Internet connection.

With the bandwidth baseline established, I turned to gauging the range of these kits. I repeated the tests in an adjacent room that has about 50 feet of hidden cabling behind the building's walls. I started adding long extension cords to the outlet to simulate behind-the-walls wiring. After three individual 100-foot cords, I added three 50-foot cords and then a 25-foot and a 10-foot cable. After each cord was connected, I repeated the benchmark tests and noted the total length of cabling.

When each system lost its network connection, I removed the last extension cord and added a smaller one and retested to get close to the actual range of the equipment. Its range was the last point at which it remained connected to the network.

I also measured how much power these devices consume while being used and while idle. This allowed me to calculate the estimated cost per year if the powerline equipment gets used for 10 hours a day during weekdays and is idle the rest. I used the national average of 12 cents per kilowatt hour for the price of power and multiplied the results by two for the pair of devices needed.

Finally, I set each adapter up in my office and used it as my primary Internet connection for a day or two. The office is on the opposite side of the building and up one floor from the router; it has about 400 feet of in-wall cabling (however, that is just an estimate). During that time, I used the powerline connection to watch HD and 4K videos from online sources and a local RAID server. Then I interacted with websites, sent and received emails and ran videoconferences.

(www.computerworld.com)

Brian Nadel

Zur Startseite