Tomato - RT-N66u -Transmit Power

Discussion in 'Tomato Firmware' started by Daky, Aug 10, 2013.

  1. Daky

    Daky Network Guru Member

    Hello guys,

    What would be the highest \ safest Transmit Power on this router? (RT-N66U?)

  2. Daky

    Daky Network Guru Member

    Thanks for replying sir.

    I am from Canada, Advanced setting is set to Canada, and default power was set to 45, i increased it to 85.

    Since, the maximum value is 200mW, this should be safe?

  3. Marcel Tunks

    Marcel Tunks Networkin' Nut Member

    The number you enter is relative power and doesn't directly equate to mW.

    There are several threads on this topic as well as the "safe" maximum power. Zero is supposed to set it to the hardware default (which one would expect was sensibly chosen by the manufacturer after some testing.) Different driver versions also perform differently. At least one recent driver gave no change in power output at all with different numbers entered in the GUI.

    In general, keep the power to the minimum that allows reasonable connection. If the problem is devices not connecting or dropping connections, the problem is usually the client rather than the router.
  4. Planiwa

    Planiwa Network Guru Member

    May I ask:

    You say the default power was 45mW and you increased it to 85mW.

    1. How do you know the TX power? -- from the GUI or with the wl command?
    2. How did you increase it? -- GUI or wl command?
    3. Which Firmware are you using?

  5. Daky

    Daky Network Guru Member

  6. Planiwa

    Planiwa Network Guru Member

    FWIW, on Victek GUI, the Country was US, tx power setting was 0.

    wl txpwr returned: 31.75 dBm = 1496 mw.

    changed country to Canada, no change in wl output.

    Changed tx power to 50: 17.0 dBm = 50 mw.
  7. Mangix

    Mangix Networkin' Nut Member

    AFAIK, toastman did testing on this and found ~50 to be the optimal power setting. Higher power did not show an improvement.

    A way around this is to upgrade the antennas on your router(if possible). That way the extra power would result in signal instead of noise.
    HitheLightz and Marcel Tunks like this.
  8. Planiwa

    Planiwa Network Guru Member

    If the 3 Antennae are functional, rather than just a random wire in a random plastic enclosure.
    (Functional means expensive).

    Even then, increased TX power does not increase the RX sensitivity (au contraire), nor does it do anything to increase the client's TX power.

    I do not trust the wl command at all.

    Tomato say that 0 means Default.
    wl says that -1 means default.

    But this "default" is 1496 mW:

      wl txpwr1 -1
      wl txpwr1
    TxPower is 127 qdbm, 31.75 dbm, 1496 mW  Override is Off
    HitheLightz and Marcel Tunks like this.
  9. Victek

    Victek Network Guru Member

    The value showed in the gui or command it's not the real power delivered by the unit since the signal is boosted by the RF amplifier (SG devices) and then enhanced by the antenna gain, it's known as EIRP power, the real power.

    Then you can measure the signal strenght in one FFT Spectrum Analyzer to determine the real power. Measuring using your laptop,notebook is not giving any indication since the GAC control, driver version, filters may vary these values. At last you have to take in mind electrical ground noise also masking the real power, this is the reason why FFT Spectrum Analyzer shows more accurate values since you can filter the received frequency signal by mixing a master frequency, with this method (channel settings frequency/master frequency = clean frequency to measure).

    In RT-N66U, 2.4GHz, Channel 4, 40MHz width channel and Country=USA, driver I measured EIRP 27dB = 500mW. Same unit but 5GHz band I measured 25dB EIRP. Default values. Using countries adhered to CE regulations dismiss power to 100mW, better use FCC countries.

    These results are theoretical and done using Tomato RAF beta 'y', the location and neighbor radios sort big differences from user to user. Also take in consideration 2.4GHz interference over USB3.0 devices, a big issue discovered when RF power surpasses standard...
    Last edited: Aug 14, 2013
    HitheLightz and wlkn like this.
  10. Planiwa

    Planiwa Network Guru Member

    It would be useful to show wl command values along with the corresponding EIRP values.
  11. Daky

    Daky Network Guru Member

    The reason i asked for this is .. that i am having issues with some of my wi-fi devices, for example Samsung Galaxy S3, i am unable to get better rate then 54mbps, on both 2 ghz and 5 ghz, and router is in the same living room where phone is, also i am getting poor signals on ObiWifi .

    I tried many channels, but, no luck, i also tried to increase power, no luck.

    I dont know what to do\try any more ...

    Any suggestion would be very much appreciated
  12. Planiwa

    Planiwa Network Guru Member

    I very much doubt that this has anything at all to do with transmit power.


    See what data rates you can get on devices that you know to be fast enough not be be a processing bottleneck.

    As an overall approach, I tend to observe, quantify, measure, diagnose *first*, in order to understand the problem. *Then* I think about creating a solution.

    Another approach is to try this and that in hopes of stumbling on something that works.

    Sometimes the second approach is useful too. :)
    Last edited: Aug 15, 2013
  13. Marcel Tunks

    Marcel Tunks Networkin' Nut Member

    Do any of your devices connect higher than 54mbps? Sometimes the issue is security type (must use WPA2), advanced settings (WMM must be on) or other settings on the router. Trying 20MHz channel width on 1, 6, 11, or 13 in 2.4GHz and trying both upper and lower bands in 5GHz is also recommended.

    Again, though, transmission power on the router is generally less of an issue in residential WiFi than the transmission power of the clients. The default setting is likely fine. In fact, it may be better to turn the transmission power down below the default value if the clients are close to the router.

    You likely know this as well, but how many bars you get in signal strength and the RSSI on InSSIDer don't equate to throughput or stable signals. You can have very high RSSI and full signal strength on the device with terrible performance.

    Last, but not least, small changes in orientation and placement of obstacles make a huge difference in wireless performance. It's not the number of obstacles but the type of obstacle and the distance through which the signal must travel. Perpendicular to a wall is better than an angle, for example. Tilting the router 45-90 degrees, moving it to higher or lower location, placing metal objects behind the router to reflect RF (metal strainer) are quick and easy ways to experiment with this issue.
    Last edited: Aug 15, 2013
  14. Daky

    Daky Network Guru Member

    On 5Ghz, the most i saw was 54 mbps for some reason.

    I just returned the power to 45, as this was default.

    My S3 phone is varying from 6mbps to 54 mbps and its sitting like 2 meters from the router.

    I did try many different channels (upper \ lower ) etc, and the same output.

    Please advise

    Attached Files:

  15. Marcel Tunks

    Marcel Tunks Networkin' Nut Member

    If the numbers on that screenshot are to be believed, then the problem is not inadequate transmission power. In fact, lowering the transmission power would be reasonable (and possibly helpful.)

    Any chance you could post screenshots of your Basic and Advanced Wireless pages?
  16. Daky

    Daky Network Guru Member

    Sure, thank you for help!

    Attached Files:

  17. Marcel Tunks

    Marcel Tunks Networkin' Nut Member

    Disable afterburner on both radios, turn transmit power down (maybe around 20)

    Next step would be to change 2.4GHz channel width to 20MHz, switch to n only on both radios, forget the network on your S3, update S3 firmware if possible, then reconnect.

    S3 has a reputation for problems.
    wlkn likes this.
  18. Daky

    Daky Network Guru Member

    I will give it a try, thanks.

    I have custom ROM on S3, and i never experienced this problem before.

    What Aferburner usually do, if you don't mind telling me.

  19. Marcel Tunks

    Marcel Tunks Networkin' Nut Member

    Afterburner/speedboost is an 802.11g technology to try to increase throughput. Doesn't work very well. Doesn't work with 802.11n.
    wlkn likes this.
  20. Daky

    Daky Network Guru Member

    MAN, you ROCK !!!!!!!!!!!!!!!!!

    My S3 is connected to 150 mbps now !!!!!!!!!!!!!!!!!!!!!!!!!

    THANK YOU !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
  21. Marcel Tunks

    Marcel Tunks Networkin' Nut Member

    No problem.

    I suggest checking throughput with large file downloads over wifi on 20MHz channel width, then try 40MHz again and compare.

    If the S3 goes back to 54mbps on 40MHz channel width, then it may be a compatibility issue between the drivers.
  22. Daky

    Daky Network Guru Member

    Will do more tests.

    Thank you very, very much.

    You just made me very happy, as this was going to my nerve since couple days ago.

    User ERROR (mine) ;\
  23. occamsrazor

    occamsrazor Network Guru Member

    I remember this a bit, but after much searching can't find that Toastman post and would like to read it. Anyone know where it is?
    So is there any conclusion to all this - i.e. optimum transmit power?
  24. Victek

    Victek Network Guru Member

    Toastman settings was for other routers, drivers and wireless bands, it's not applicable for this model.
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice