How We Test Wireless Products - Revison 8

[Razor512]

Not sure how accurate it is to measure power with a multimeter (cheap $20 one)

From when I tested the power consumption, I could not get the router to pull anything close to the max that the power adapter could output. e.g., you may see a router with a 2.5 amp 12V adapter but they will struggle to even pull 1 amp while benchmarking the router. while there is some overhead for powering external drives, since it is not good to really run a DC to DC regulator at its max rated output, it may be the case that some companies are adding a little extra headroom when it comes to the power supplies.

 

[sfx2000]

I think the power consumption is higher than you think...

Case in Point - P=IE = in other words - Watts = Amps * Voltage

Apple's Airport Extreme AC (AC17500 class) - consumes 1.5 Amps Max @ 110VAC -- So doing the math 165 Watts

The Airport Express is 0.2 Amps @ 110VAC -- Math here for a dual stream N600 class router... 22 watts consumed...

Case in point - given an average of $0.17 per KWH

Airport Express (0.2A/110VAC) -- $32.76 a year
Airport Extreme (1.5A/110VAC) -- $245.72 a year

These are real numbers quoted by the OEM (Apple) - using something like P3's Kill-A-Watt meter would provide even more relevant data for idle vs. full load testing that SNB already does as part of the reviews.

Most OEM's do not design for power efficiency, but perhaps they should as typically the devices are on 24/7

sfx

BTW - one can download an excel worksheet here that works the numbers

 

[sfx2000]

139.5 Million Access Points were sold in the consumer market, an additional 14M Enterprise AP's were projected to be shipped in 2013.

Improve efficiency by 25 percent - let's say draw the current down by 0.25 amps, and guess what, this is a very big number... something on the order of $5.8 billion dollars a year globally...

And that is just for WiFi access points...

 

[sfx2000]

139.5 Million Access Points were sold in the consumer market, an additional 14M Enterprise AP's were projected to be shipped in 2013.

Improve efficiency by 25 percent - let's say draw the current down by 0.25 amps, and guess what, this is a very big number... something on the order of $5.8 billion dollars a year globally...

And that is just for WiFi access points...

Correction - over $6.2B annually... when you factor in other SNB platforms such as NAS boxes, that number will jump up even further...

 

[RMerlin]

Not sure how accurate it is to measure power with a multimeter (cheap $20 one)

From when I tested the power consumption, I could not get the router to pull anything close to the max that the power adapter could output. e.g., you may see a router with a 2.5 amp 12V adapter but they will struggle to even pull 1 amp while benchmarking the router. while there is some overhead for powering external drives, since it is not good to really run a DC to DC regulator at its max rated output, it may be the case that some companies are adding a little extra headroom when it comes to the power supplies.

The Asus RT-AC87U AC adapter is rated for 1A draw (and 1.5A output @ 19V). That's enough to allow the USB3 port to provide up to 1700 mA to a plugged device, Up to 900 or 1000 mA to the USB2 port (I forgot the exact number quoted by Asus), and this is a router which has two CPUs to power (one of which is dual-core). So, I doubt that Apple's router would really need to draw that much power (even less that router with a 2.5A adapter).

Interesting thing about the AC87 adapter is Asus documents the current draw, the current output, and the hourly power consumption.

Input: 100-240V, 1A
Output: 19V @ 1.58A
Operating consumption:  33.48 Wh
Standby consumption:  0.12 Wh

(these last two lines seem to be in Spanish. Could be a requirement from one of the Latin countries?)

Measuring such low power draws isn't always easy unless you have precise equipment. Some UPS will give you an estimate on power draw, but they are more calibrated toward higher loads than in the 10-12W range.

 

[sinshiva]

according to a recent asus youtube video i saw on the ac87, the usb2 can supply something like 1500 mA while the USB3 port supplies 1750 mA

[edit/] just double checked that. https://www.youtube.com/watch?v=hxwy2uidCJo

 

[thiggins]

Meter: P3 international P4400 Kill a Watt

NETGEAR R8000
idle: 10W 18VA
5GHz IxChariot test: 15W 25VA

ASUS RT-AC87
idle: 11W 25VA
5GHz IxChariot test: 18W 39VA

 

[azazel1024]

Tim - I think so - esp at the present where the Tier 1's are throwing a lot of radio resources at WiFi and extended capabilities of the devices - and there doesn't seem to be much regards to overall power consumption...

And this is an indirect cost to the end-user - when you consider life-time expenses...

Saw a recent report/discussion on Cable Set-Top boxes, and was a bit surprised as to the vampire load they took when the TV was off...

sfx

Absolutely. They are getting better. My 2nd-ish generation Motorolla DVR for FIOS drew around 42W of power, TV on or off. It was so much heat it cooked itself after about 18 months in a poorly ventilated entertainment center. The replacement, a 3rd gen Motorolla DVR is much better. 20W TV on, 19W TV off. I'd still consider that considerable power for what it does, but it is enormously better.

With APs/Routers, a lot of users probably don't care, but I am sure at least a few do. I care, at least to an extent.

I've also seen widly different power consumptions. From around 2-3W for newer N150/N300 10/100 AP/Routers up to around 15W or so some of the higher end 11ac routers.

If nothing else it can make a big difference in locating as you might need ventilation/active cooling with some of the higher draw stuff if it isn't sitting in the open.

 

[azazel1024]

Meter: P3 international P4400 Kill a Watt

NETGEAR R8000
idle: 10W 18VA
5GHz IxChariot test: 15W 25VA

ASUS RT-AC87
idle: 11W 25VA
5GHz IxChariot test: 18W 39VA

Wow, I guess a good thing residences (well, at least in the US) don't pay for VA, but for watt consumption.

That is some downright horrible PF design. One thing I've never bothered to check on any of my stuff. I would have figured it would have been closer to 70-85% PF.

 

[sfx2000]

Wow, I guess a good thing residences (well, at least in the US) don't pay for VA, but for watt consumption.

That is some downright horrible PF design. One thing I've never bothered to check on any of my stuff. I would have figured it would have been closer to 70-85% PF.

Yep - and for many AP/Routers this is an easy thing to fix - worst case is less than $2USD to fix - challenge here is that most wall wart AC Adapters, they have a current cost of about $2.50 each, but when one looks at the Kit Bill of Material, this is an incremental cost...

I think SNB is in a great position to take a stand here - name and shame...

sfx

 

[sammi1923]

Spec question

In the new method, we need to change both path loss and multipath via programmable attenuators. We have some questions need your help about the test method.
1. Which multipath model ( B / C /D / E /F) do you use for the wireless product test?
2.What is the external loss value? Including antenna loss, air loss, and emulator loss. I know the test attenuations is from 0 to 63dB, but the external loss value should not be 0 in actually
3.What's brand and model of the emulator?

Thank you so much.

 

[thiggins]

1. Which multipath model ( B / C /D / E /F) do you use for the wireless product test?

B/C. Please read http://www.smallnetbuilder.com/wire...we-test-smallnetbuilders-wireless-testbed-v2-

2.What is the external loss value? Including antenna loss, air loss, and emulator loss. I know the test attenuations is from 0 to 63dB, but the external loss value should not be 0 in actually

Path loss for OTA coupling to the end of the cable in the lower chamber, including attenuators programmed to 0dB and MPE emulator is approximately 52dB for 2.4 GHz and 74 dB for 5 GHz.
3.What's brand and model of the emulator?[/quote]OctoScope OctoBox MPE. Please read http://www.smallnetbuilder.com/wire...we-test-smallnetbuilders-wireless-testbed-v2-

 

[sfx2000]

I've already touched on power efficiency - which is relevant as these are 24/7 enabled devices, and consume power all the time - idle vs. full load...

Anybody curious about IPV6 support - Native (e.g. your provider provides IPV6), and Tunnel Support - along with Dynamic DNS support, Firewalls, etc?

Just asking...