OpenWrt Forum Archive

Topic: How to set txpower above 17db on WNDR3700v2 with Backfire 10.03 RC5

The content of this topic has been archived between 31 Mar 2018 and 11 Apr 2018. There are no obvious gaps in this topic, but there may still be some posts missing at the end.

Hallo zusammen :-),

I saw a few posts on WNDR3700 with txpower above 17 like this one: https://forum.openwrt.org/viewtopic.php?pid=133386 ).
So, in principle it seems to be possible to run the WNDR3700 up to 30dbm

> I am able to notice differences ranging from 5-30dbm (27 on 2.4g).
> [...]
> iwconfig wlan1 txpower 30 - 5ghz to 30dbm

But how do I manage to get above 17db? For everything bejond 17db I get:

> root@OpenWrt:~> iwconfig wlan1 txpower 30
> Error for wireless request "Set Tx Power" (8B26) :
>     SET failed on device wlan1 ; Invalid argument.

Any hint would be highly appreciated :-)

Thanks,  hasselh


P.S.: Here my wireless config:
------------> cut <--------------

config 'wifi-device' 'radio1'
    option 'type' 'mac80211'
    option 'macaddr' 'c4:3d:c7:90:ab:1e'
    option 'hwmode' '11na'
    list 'ht_capab' 'SHORT-GI-40'
    list 'ht_capab' 'TX-STBC'
    list 'ht_capab' 'RX-STBC1'
    list 'ht_capab' 'DSSS_CCK-40'
    option 'txpower' '17'
    option 'htmode' 'HT40-'
    option 'channel' '40'
    option 'disabled' '0'
    option 'country' 'CA'

If you can set it to 1-17 but not above, I would guess it is the daemon preventing you from breaking the law ;-)

Check this:  http://en.wikipedia.org/wiki/DBm

nope, for my country it is significantly more:

http://www.bundesnetzagentur.de/cae/ser … 010pdf.pdf

which is:

- 20 dbm @2.4ghz
- 23 dbm @5ghz below channel 100
- 30 dbm @5ghz above channel 100

But this still does not answer my question what is required to get above 17dbm on my WNDR3700v2

Thanks, hasselh

Ich fasse es nicht. Jetzt habe ich das auch endlich gefunden:

http://www.bundesnetzagentur.de/cln_191 … _node.html

Aber nicht unter ISM gucken, sondern nach WLAN suchen. Ganz unten.

Maybe you should change "option 'country' 'CA' to DE?

(Last edited by scruffy on 14 Jun 2011, 00:28)

In the original post https://forum.openwrt.org/viewtopic.php?pid=133386, rjs_boy is running the 5GHz channel of his WNDR3700 with 10 20 25 and 30dbm and is able to see measurable differences

So is building your own image and setting CONFIG_ATH_USER_REGD=y the solution ?

@scuffy: option 'country' 'DE' also did not help. Openwrt is still limiting to 17dbm :-(

today I tried with Trunk (r27153)  with CONFIG_ATH_USER_REGD=y
And I still get the same result... :-(

root@OpenWrt:~> iwconfig wlan1 txpower 16
root@OpenWrt:~> iwconfig wlan1 txpower 17
root@OpenWrt:~> iwconfig wlan1 txpower 18
Error for wireless request "Set Tx Power" (8B26) :
    SET failed on device wlan1 ; Invalid argument.

Is there anybody who can help me how to get my WNDR3700v2 above 17dbm on a 5ghz channel ?

Thx. in advance,

hasselh

hasselh wrote:

Is there anybody who can help me how to get my WNDR3700v2 above 17dbm on a 5ghz channel ?

If you are in country US, you must be on the high channels (149, 153, 161, 165).

From the regdb:

654 country US (format: frequency range @ channel width, ( ant dbi, power dbm)): 
655         (2402 - 2472 @ 40), (3, 27) 
656         (5170 - 5250 @ 40), (3, 17) 
657         (5250 - 5330 @ 40), (3, 20), DFS 
658         (5490 - 5600 @ 40), (3, 20), DFS 
659         (5650 - 5710 @ 40), (3, 20), DFS 
660         (5735 - 5835 @ 40), (3, 30)

OK, meanwhile I am able to set txpower above 17dbm on a 5ghz channel.
The solution is at http://wiki.openwrt.org/toh/netgear/wnd … ory.issues

From dmesg:

cfg80211: Calling CRDA to update world regulatory domain
cfg80211: World regulatory domain updated:
cfg80211:     (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
cfg80211:     (2402000 KHz - 2494000 KHz @ 40000 KHz), (N/A, 3000 mBm)
cfg80211:     (4910000 KHz - 5835000 KHz @ 40000 KHz), (N/A, 3000 mBm)
[…]
ath: EEPROM regdomain: 0x0
ath: EEPROM indicates default country code should be used
ath: doing EEPROM country->regdmn map search
ath: country maps to regdmn code: 0x3a
ath: Country alpha2 being used: US
ath: Regpair used: 0x3a

iw phy1 info seems to support txpowers up to 30.0 dBm.
Setting iwconfig wlan1 txpower 27 now works without problems…  smile

BUT there is no visible difference in inSSIDer between 17 dB, 27 dB and 30 dB
Looks like OpenWRT or the WNDR3700 caps the output above 17dB:

http://img825.imageshack.us/img825/1672/inssider.gif

Did anybody see any visible difference in WNDR3700 output above 17dbm on a 5ghz channel ?
Any of my attempts with backfire 10.03.1-RC5 and backfire trunk have failed so far... sad

Thanks in advance, hasselh

(Last edited by hasselh on 24 Jul 2011, 21:44)

I replaced my regulatory.bin as well, and all channels are available to output to 30dBm according to:

iw reg get


However, when I do:

iwconfig wlan1 txpower 25

It doesn't get set higher than 24. Interestingly, I can set the 2.4GHz channel to 25dBm, but not more.

I don't get any errors or output. It seems to accept the command, and yet it doesn't change the power output past 24/25.

Is it working for you? What build are you using.

(Last edited by hechacker1 on 6 Aug 2011, 09:15)

hechacker1 wrote:

I replaced my regulatory.bin as well, and all channels are available to output to 30dBm according to:

iw reg get

That wont help you much when the limit of hardware is 20dBm per antenna chain at 5GHz which will give a max of 23dBm for both chains, and the hardware limit is 23 dBm per chain at 2.4GHz or a total of 26dBm for both chains.

JonnyM wrote:
hechacker1 wrote:

I replaced my regulatory.bin as well, and all channels are available to output to 30dBm according to:

iw reg get

That wont help you much when the limit of hardware is 20dBm per antenna chain at 5GHz which will give a max of 23dBm for both chains, and the hardware limit is 23 dBm per chain at 2.4GHz or a total of 26dBm for both chains.

Perhaps I'm missing something then? I've seen people set 27dBm on both radios according to various forums and posts.

It seems like a bug. Sometimes it accepts "txpower 25", but right now for example, I can't even go past 20dBm after I just lowered it to test.

EDIT:

http://www.dd-wrt.com/wiki/index.php/Netgear_WNDR3700

I guess this is what you are referring to? It would seem the particular version of the network cards in the 3700 can only do 27dBm on 2.4 with HT20, and 24dBm on 5GHz.

That still doesn't explain why sometimes when I set the txpower to 24, it doesn't take. I just tried again, and managed to get the 2.4GHz channel up to 26dBm.

(Last edited by hechacker1 on 6 Aug 2011, 20:09)

as JonnyM already said the power amplifier(s) cannot deliver more output power than what they have been designed to do even though the radio chip and driver software are providing and have been set to do more power.
When you look in the specifications of power amplifiers there is an issue with them starting to amplify less linearly when pushed to their limits, or in other words they start to introduce distortion. The sophisticated modulation schemes used in 802.11 (the way the radio signal is given its data content) require very linear amplification (the output being a precise multiple of the input). The effective data throughput may actually suffer as the linearity decay destroys more than the bit of extra power brings. For best performance it is wise to set the system a few dBs under the limit the power amplifier has been specified for (to account for calibration inaccuracies and spread between systems). Non linear amplification may also produce spurious signals.
There are only a few manufacturers of these amplifier ICs which require special production processes and you may have a look with a magnifying glass which actual PA chip has been designed in and check out its specs. Could not read the type numbers from the FCC photographs but they appear to be ICs AU6 and AU7 under the metal shield where antenna connectors AJ1 and AJ2 are located.

doddel wrote:

as JonnyM already said the power amplifier(s) cannot deliver more output power than what they have been designed to do even though the radio chip and driver software are providing and have been set to do more power.
When you look in the specifications of power amplifiers there is an issue with them starting to amplify less linearly when pushed to their limits, or in other words they start to introduce distortion. The sophisticated modulation schemes used in 802.11 (the way the radio signal is given its data content) require very linear amplification (the output being a precise multiple of the input). The effective data throughput may actually suffer as the linearity decay destroys more than the bit of extra power brings. For best performance it is wise to set the system a few dBs under the limit the power amplifier has been specified for (to account for calibration inaccuracies and spread between systems). Non linear amplification may also produce spurious signals.
There are only a few manufacturers of these amplifier ICs which require special production processes and you may have a look with a magnifying glass which actual PA chip has been designed in and check out its specs. Could not read the type numbers from the FCC photographs but they appear to be ICs AU6 and AU7 under the metal shield where antenna connectors AJ1 and AJ2 are located.

This may be all true, but I have seen real world signal strength increases by raising the radios to their limits.

With 17dBm (defaults), the range of the router barely reaches the 2nd story of the building, and doesn't get very far outside the building.

With 24dBm, it effectively reaches the 2nd story without being on the verge of losing connection all the time. It also allows me to travel further outside before I lose connection.

So, these radios seem to handle the extra power just fine.

What's disappointing is just how short the 5GHz range is. Even with 24dBm, it doesn't get much faster than 54Mbps on the 2nd story, no matter that it isn't very far (10 meters?).

My old WRT54-GL seemed to do much better with 251mW set. It would easily reach the furthest places on the property. Of course, the WNDR3700 can supposedly reach that, but I haven't tested it with a meter yet.

I may have to find the optimal location for the 3700. Frankly, the 4th gen Airport Extreme I have has better 5GHz range and throughput. I'm guessing the antenna configuration (and location) is responsible.

The 5th gen Airport supposedly does 400mW in most situations. So I don't thinks it's unreasonable to put out power that high (the FCC allowed it).

(Last edited by hechacker1 on 7 Aug 2011, 21:42)

backfire r28299

root@nodo_2:~# iw reg get
country US:
    (2402 - 2472 @ 40), (3, 27)
    (5170 - 5250 @ 40), (3, 17)
    (5250 - 5330 @ 40), (3, 20), DFS
    (5490 - 5600 @ 40), (3, 20), DFS
    (5650 - 5710 @ 40), (3, 20), DFS
    (5735 - 5835 @ 40), (3, 30)

so channel 44 (5220 MHz) allows 17dBm, then:

root@nodo_2:~# iw dev wlan0 set txpower fixed 1700

but no success:

root@nodo_2:~# iwconfig wlan0
wlan0     IEEE 802.11abgn  ESSID:"MeshRootNetwork" 
          Mode:Ad-Hoc  Frequency:5.22 GHz  Cell: 02:CA:FF:EE:BA:BE   
          Tx-Power=15 dBm   
          RTS thr:off   Fragment thr:off
          Encryption key:off
          Power Management:on

i have exactly the same problem with anything after r27268 on my 3700v2.
i can't get past 17 dbm on both radio 0 and radio 1 (2.4 and 5 ghz) regardless of the country used.

it just won't take the setting and reply with the same error as the other user (Error for wireless request "Set Tx Power" (8B26) )

anyone managed to fix this?

smile

Any solution in this case?
I have the same issue

don't the NA models have the EEPROM to prevent it from from being raised too high?

Some of you seem to be confused, Simply change the 5ghz channel to a channel lower in the list after applying the higher 30 db will be available.

and whats about the 2Ghz?

root@Gateway /root# iw reg get
country US:
        (2402 - 2472 @ 40), (3, 27)
        (5170 - 5250 @ 40), (3, 17)
        (5250 - 5330 @ 40), (3, 20), DFS
        (5490 - 5600 @ 40), (3, 20), DFS
        (5650 - 5710 @ 40), (3, 20), DFS
        (5735 - 5835 @ 40), (3, 30)
root@Gateway /root# iwconfig wlan0 txpower 27
Error for wireless request "Set Tx Power" (8B26) :
    SET failed on device wlan0 ; Invalid argument.
root@Gateway /root# iwconfig
wlan0     IEEE 802.11bgn  Mode:Master  Frequency:2.412 GHz  Tx-Power=10 dBm   
          RTS thr:off   Fragment thr:off
          Power Management:on

regulatory limit != eeprom limit

but I have an european model which ignores the eeprom, right?

netbus wrote:

and whats about the 2Ghz?

root@Gateway /root# iw reg get
country US:
        (2402 - 2472 @ 40), (3, 27)
        (5170 - 5250 @ 40), (3, 17)
        (5250 - 5330 @ 40), (3, 20), DFS
        (5490 - 5600 @ 40), (3, 20), DFS
        (5650 - 5710 @ 40), (3, 20), DFS
        (5735 - 5835 @ 40), (3, 30)
root@Gateway /root# iwconfig wlan0 txpower 27
Error for wireless request "Set Tx Power" (8B26) :
    SET failed on device wlan0 ; Invalid argument.
root@Gateway /root# iwconfig
wlan0     IEEE 802.11bgn  Mode:Master  Frequency:2.412 GHz  Tx-Power=10 dBm   
          RTS thr:off   Fragment thr:off
          Power Management:on

I have the same problem here. What is wrong with 20dB on 2.4 GHz?

Again, the power levels reported by iw reg and friends are *regulatory* limits. They're not related to hardware capabilities.
The FCC test reports for the WNDR3700 indicate around 17-23 dBm peak power output per antenna chain in OFDM (11g) mode. In 11n its even slightly less.

You can look it up yourself in test report 15C at page 66 and following; https://fjallfoss.fcc.gov/oetcf/eas/rep … 8300092%27

Could you explain a little?
Why can some people set more than 17dB while it doesn't work for me? Don't we have the same hardware?