Where can I get the transmit power & antenna gain of the Netgear WNDR3400 router?

I can't find the transmit power & antenna gain of the Netgear WNDR3400 home router.

Is there a secret place where this important spec is hidden?

I'm comparing routers and inherent power is one starting point.

Reply to
Bob Stevens
Loading thread data ...

If it isn't spec'd, I'd say go to the FCC product database. I'd look it up for you but I think Netgear makes crap so I have zero motivation to do the research. I've had two Netgear die. [Yes i check the wall warts. if only the failure were that simple.] I solved that problem by never buying their crap again. The FCC product database isn't for the timid. However, if you can dig up the FCC ID, then research is far easier.

So far this Buffalo wzr-hp-g450h is working well, but it isn't dual band. If you feel you need dual band, the equivalent Buffalo has its share of detractors on Newegg. That was one reason I avoided it. Also the dual band router don't seem to have removable antennas in general. With the single band buffalo, if I really wanted to do so, I could turn off the MIMO and run one port, then put on a BAA (big ass antenna). There are less options with dual band external antennas.

Incidentally I'm running DD-WRT on the Buffalo. It is completely factory supported. Now that I have DD-WRT, I don't think I could ever go back to proprietary software.

Out of curiosity, if Jeff is reading the thread, can you take a 3 port MIMO and put 3 120 degree sector antennas on it?

Reply to
miso

I would fire the person at Netgear who submitted that unit to the FCC. Seriously, the heat sink is visibly misaligned. You know with 100% certainty the inside of the unit will be photographed. Is it my imagination, or does that heat sink simply stay attached by the heat sink grease?

There is some mystery RTV on the some components. Perhaps the RTV near the inductor is to prevent a buzz, but the RTV on what I assume are ethernet filters is just weird. I'm surprised at the amount of through hole caps (other than electrolytics). But I don't have a router handy to compare it to. [I've tossed the busted stuff.]

On page 8, is the solder mask messed up, or is that heat sink grease that found its way to the bottom of the PCB?

OK, 2dbi on the antenna. Still it has a respectable output exceeding 20dBm.

Incidentally, this unit isn't dual band. It also isn't MIMO, but I don't obsess about that.

Needless to say, I'm not impressed by the build quality.

Interesting in that the PCB has much unpopulated real estate. Presumably it is used in more than one model.

Reply to
miso

Hell no. The whole idea behind MIMO is to have multiple paths (of slightly different length) between the access point and the client radio. If the paths are on 3 different antennas, pointed 3 different ways, you'll never get more than one stream. Same with beam forming. You can't beam form unless all the antenna patterns can be overlapped. Even worse, when the radio drops down to 802.11g speeds, it kicks in the usual diversity receive feature, which assigns one antenna as the "MAIN" and another as "AUX". It sits on the "MAIN" antenna and only switches to "AUX" if there are substantial errors that resemble the symptoms of frequency selective fading.

Reply to
Jeff Liebermann

Last weekend, I replaced a Netgear WNDR2000 v2 with a Linksys EN2500 at a house that was having range difficulties. I also brought along my "reference" WRT54Gv3 router with DD-WRT for testing. The WNDR2000 went about half way through a rather long (120ft) and narrow house. The EN2500 did somewhat better. The WRT54Gv3 was about the same. Not the best range test, but enough to convince me that the WNDR2000 has a problem.

The WNDR3400 is similar inside:

Dual band, two radios, and two antennas (diversity) per radio. Ignoring the FCC data, my experience with PIFA antennas are that they give a gain of -2dBi (note the minus sign) at best. Usually worse.

+17 to +20dBm is the typical power output. I think you'll find that almost all the wireless routers that have internal antennas have roughly the same specifications for antenna gain and tx power.

The problem is that you can always trade range, speed, and error rate. At a given error rate (usually 1 in 10^6), if I double the speed, I'll get much less range. If I want to go twice as far, the connection speed slows down.

It turns out that the limiting factors of speed and range is the noise and garbage produced by the CPU in the typical router. Also, such things like interference and reflections have a HUGE effect. CPU speeds and MAC layer implementation problems also show up as performance limitations. Only a few reviewers try to test routers under somewhat realistic conditions. I suggest you look at the reviews under:

Reply to
Jeff Liebermann

If there were a different station in each sector, would they get one stream each, total speed 3x1 stream, or would the aggregate throughput still be one stream? To put it another way, would it be one client per stream, as if there where three APs?

Reply to
alexd

MIMO spatial diversity works with a single MAC address at each end. The access point will have 2 or more streams with the same MAC address connected to a client radio with a single MAC address. There's no provision for handling multiple MAC addresses at the client end. If it's one MAC address, the MIMO chipset will combine and conglomerate the streams into what looks like a single 802.3 ethernet connection, encapsulated inside 802.11 packets. In order for that to work, the client radio has to be "seen" by all the radios and antennas. With a sector antenna arrangement, each client will only be (theoretically) heard in one sector, and therefore only one stream will be available.

On the other foot, beam forming will work with any number of client radios. Each one gets a peak in the pattern, with a corresponding null in the direction of interference sources. If it's 802.11g (non-MIMO), it will only get a 1x1 stream, but it will still benefit from the beam forming by improving the signal strength and SNR.

Reply to
Jeff Liebermann

To beam form, you need all the transmitters on at the same time. Is that really the case is MIMO?

Reply to
miso

Sorta. To do spatial diversity (multiple streams), you need to have all the transmitters on at the same time. It wouldn't make any sense to send the data sequentially on each stream, as the total thruput would be the same as a single stream. So, it gets sent simultaneously on multiple radios.

If you wanna do just beam forming, you only need to have all the antennas operating at the same time, but with only one transmitter at a time. The antennas are phased to form peaks and nulls.

To the best of my limited knowledge, spatial diversity and beam forming are not done at the same time. I don't see any way to do both at the same time but maybe I missed something. I don't wanna read the IEEE specs. It turns my brain to mush.

Some clues:

Things get really messy with more than 2x2 MIMO systems:

Reply to
Jeff Liebermann

I'll read the papers when my brain is more alert. For the beam forming, it is far easier to play the phasing game during modulation (three different modulators) and just feed the antennas without any phasing circuits. [You can phase them simply with digital delay if narrow band, or Hilbert transformers if you want to get fancy. Or just generate I&Q and do phase with a vector sum.] I'm thinking back to my sonar days, but the idea is the same. Just a different frequency and medium. The ocean is a really lossy medium that needs to be equalized, so in a way RF is easier.

Having all the transmitters on at the same time does match the RF safety section of the FCC report.

Reply to
miso

I don't see a problem:

16 radios, 48 antennas, PCI-e(2.5Gbits/sec) interface, one box.

You can choose a 8-slot (XR-4000 - $2500) or 16-slot (XR-6000 - $4000) "UFO" chassis and for the radios you can choose 300Mbps ($250) or 450Mbps ($300) radios depending on your budget. So, a fully loaded 16 radio access point would be $8,800.

Linux inside:

Looks like Xirrus measures one xmitter and multiply by the number of active xmitters. Xirrus is grantee code SK6. There are 96 items listed.

I fished these comments out of the MPE (max permissible exposure) calculations: 5.15 GHz Band Four non-overlapping 20MHz channels available, or two 40MHz channels. Total EIRP in the band is restricted to 23.0dBm. output power per radio is reduced as soon as more than one radio is operational in the band to ensure the 23dBm eirp is not exceeded.

So, they have power control available. Nice.

As there can only be a maximum of 12 radios operational at any time (3 in the 2.4 GHz band, 5 in the 5.7GHz band and 4 in the 5.15 GHz band) the total eirp for the device is the sum of the highest eirps in each band: # Radios and Band EIRP (dBm) EIRP (mW) 3 radios in 2.4 GHz band 36.0 3981.072 5 radios in 5.7 GHz band 33.5 2238.721 4 radios in 5.15 GHz Band 23.0 199.5262 Total EIRP 38.1 6419.3 The rf exposure calculation using an EIRP of 38.1 dBm gives a minimum separation distance of 22.6cm (see below). The user?s manual requires a minimum separation of more than 25cm.

Yep. Each xmitter measured individually and added together.

Reply to
Jeff Liebermann

More on Xirrus.

Looks like they're doing WISP outdoor arrays:

Note the nearby sector antenna. Since the 3 coax cables go into the saucer, my guess(tm) is that the sector array runs the 2.4GHz radios, while the 5.1/5.7GHz stuff is inside the saucer.

Reply to
Jeff Liebermann

Friar Tech did a podcast on the mega-wifi box.

I haven't watched it or any of Friar Tech's podcasts yet, but they are on the list. I've seen him on the TWIT panel discussions and he seems to know what he is talking about.

Reply to
miso

The bottom of my WNDR3400v2 router says FCC ID: PY311100155. I called the FCC at 888-225-5322 to get the antenna gain & transmit power. The didn't know what I was asking & suggested I call the OET: 301-362-3000 FCC Office of Engineering & Technology

I left a message with the OET asking for the antenna gain & transmit power.

I went to that suggested web site:

formatting link
when I typed in the FCC ID, nothing came up.

What did I do wrong?

Reply to
Bob Stevens

Can you tell me HOW you found the document you are referencing?

I went to the FCC site:

formatting link
And then ran a search for FCC ID: PY311100155

And it came up blank.

Reply to
Bob Stevens

It works for me. Did you put the py3 in the first box and the 11100155 in the second?

Reply to
Pen

Duh. I realized now I have to type the first three characters in one line and the rest of the characters in the other.

Now I'm reading the documents - but trying to figure out what the antenna gain and transmit power are!

Reply to
J.G.

Thanks! That's what I did wrong!

Reply to
Bob Stevens

I called the FCC number 888-225-5322; but they gave me the Office of Engineering & Technology 301-362-3000

Then I went to

formatting link
And typed in my Netgear WNDR3400v2 FCC ID: PY311100155

The result is a report, completed on May 21, 2011 & submitted to the FCC by SPORTON International Inc, No. 52, Hwa Ya 1st Rd., Kwei-Shan Hsiang, containing a description of the antenna and chips and wiring.

This is the antenna gain that I could glean from that report:

5GHz Antenna 1 = 1.95dBi 5GHz Antenna 2 = 3.48dBi 5GHz Band 1: MCS0 (20MHz): 16.78 dBm, 5GHz MCS0 (40MHz): 16.57 dBm 802.11a: 24.48 MHz, 16.75 dBm

Never was the 5GHz conducted power greater than 16.75dBm (and it was as low as 14.16dBm).

Interestingly, the actual tested data rates never even got close to the claimed data rates on the box, but that isn't the topic of this thread.

I couldn't find the antenna gain for the 2.4GHz antenna. All I saw was the gain for the two 5GHz antennas 1 & 2: Antenna 1: Antenna 2:

So, what do you think the antenna gain is for 2.4GHz?

Reply to
Bob Stevens

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.