The maximum I can set my TX power in OpenWrt is 27 dBm, when the U.S. upper limit is actually 30 dBm for the frequency range I have chosen for my wireless network: UNII-3 (Upper).
I know it isn't a big difference in power, and apologies if this has already been answered; I am just looking for clarity since nobody seems to know why it can't be set above 27 dBm, seemingly under any circumstance.
30 dBm (1000 mW) for UNII-1 (5.15-5.25 GHz; Channels 36-48)
24 dBm (250 mW) for UNII-2 (5.25-5.725 GHz; Channels 52-144)
30 dBm (1000 mW) for UNII-3 (5.725-5.85 GHz; Channels 149-165)
Section 1.1.2 of the FCC test report lists RT3200 antenna gain, which is below 6 dBm. Reductions in limits for antenna gain above 6 dBm do not apply to the RT3200. I note the mac80211 antenna gain option specifies not absolute antenna gain, but reduction in gain from the regulatory limit, consistent with this limit definition.
Appendix C of the RT3200 FCC test report summarizes the power levels tested for the RT3200 FCC certification. Antenna gain ranges from 4.8 to 5.7 dBm; is listed in the first column; and is included in the power summation.
RT3200 UNII-1 is tested and FCC certified up to 27 dBm and UNII-3 up to 30 dBm using stock OEM firmware.
Either antenna gain is already included in OpenWrt txpower, or the OpenWrt allowance of 27 dBm for the RT3200 in UNII-3 is over the US limit of 30 dBm. I'm going to go with the former.
From this I conclude that OpenWrt is for reasons unknown neutering UNII-1 transmit power on the Belkin RT3200 / Linksys E8450 by limiting it to only 23 dBm (and UNII-3 to 27 dBm, but that is less an issue for me). It appears to me that up to 27 dBm should not be an issue for UNII-1. I've also noted more recent OpenWrt firmware has similarly limited the Reyee RG E5, which also uses the MT7622/7975/7915 hardware.
There is speculation here that Mediatek released a lot of production with the wrong tx power information written to the chips, and that OEM were informed about it and corrected it in the OEM firmware, but perhaps OpenWrt has not.
I'm assuming the FCC certification testing was done with stock firmware, so in this case, it would be Linksys that designed the RT3200 to go up to 27 dBm in UNII-1 and 30 dBm in UNII-3 and the FCC that approved it.
I'm actually arguing the opposite - OpenWrt is limiting transmit power on this device to less than is legally allowed and less then the txpower at which the device was designed, tested and certified to function.
The tested antenna gain for the RT3200 ranged from 4.8 to 5.7. Let's round up and call it 6 dBm.
Since OpenWrt allows setting txpower to 27 dBm for UNII-3, I am assuming that OpenWrt would not allow that if this exceeded the legal limit of 30 dBm (27+6=33). Therefore, I am concluding that antenna gain must already be included in the OpenWrt txpower setting.
As noted, this conclusion also appears consistent with how antenna gain is specified for mac80211, i.e., the antenna gain is entered as a difference from the regulatory gain limit (6 dBm in the case of FCC limits). If antenna gain exceeds 6 dBm, for example if the antenna gain were 8 dBm, then the FCC would require the maximum allowed txpower of 30 dBm be reduced to 28 dBm, to account for the antenna gain being 2 dBm higher than assumed in the regulatory limit.
I think that the MT7976 chip used for RF in the RT3200 5 GHz radio is rated for 23 dBm. That would total 30 dBm for the four MIMO chains, if the regulations work that way.
Desktop / indoor routers are seldom built for full legal power as the application does not warrant it.
Assuming the OpenWrt txpower setting already includes the antenna gain (6 dBm in the case of the RT3200, rounding up from 5.7), why would OpenWrt need to limit UNII-1 txpower to 23 dBm and UNII-3 txpower to 27 dBm, when the FCC report indicates the device was designed, tested and certified to 27 dBm on UNII-1 and 30 dBm on UNII-3, including antenna gain?
I guess Belkin / Linksys found a way around it in the OEM firmware then? Otherwise, how to explain the FCC test report results measuring 27 dBm on UNII-1 and 30 dBm on UNII-3?
The answer is right in your graphic LOL....you have the country code set to "US," yet you live in Canada. That's not a good idea if Canadian law requires lower txpower than US law. 30 dBm is allowed on 2.4 GHz in the US.
I suspect if you change your country setting from US to Canada, it will be properly limited to comply with Canadian law.
As to the 22 dBm max with country set to US, that may indeed be the most your hardware is capable of transmitting. Did you research the FCC test report or hardware specs for your device to determine its physical capabilities?
LOL We have the same limitations as the US, But just for the LOL's I did change it to Canada and to make sure it stuck I rebooted and here is the results
@eginnc Not sure if you seen this post it has a little bit of info and some revelations about the actual power output of 14~16db for a lot of devices that state in their advertised spec's of being ~30db
Sorry about the LOL - I shouldn't have. But, you did say your router was illegal in Canada when set to 22 dBm (158 mW), due to it exceeding your understanding of the Canadian limit to be 20 dBm (100mW)...
and the limit in the US is 30 dBm (1W) - up to an antenna gain of 6 dBm, with a one for one reduction above 6 dBm.
I would not have guessed your router has 16 dBm antennas to drop its US legal limit to 20 dBm, making the US limit the same as the Canadian 20 dBm limit, so I assumed the limits in Canada and US were different.
What puzzles me in the case of the RT3200 is that the FCC did in fact certify the RT3200 based on testing of the hardware at 27 dBm in UNII-1 and 30 dBm in UNII-3. I linked to the FCC test report and listed exactly where to look in it further up in this thread. Therefore, I do not think it's much of an assumption at all to say the OEM firmware and the OEM RT3200 hardware and design support this. Yet with OpenWrt firmware, UNII-1 is limited to 23 dBM and UNII-3 is limited to 27 dBm.
I've also used plenty of all-in-one devices sold in the US over the years that support 30 dBm in 2.4 GHZ and UNII-1 and 3 in 5 GHz. Perhaps that is not common with hardware targeted to the European market, but my experience has been it is quite common in the US (excepting low power - low cost options). I've even used targets that OpenWrt previously allowed setting txpower to 30 dBm, but within the last year or so, current OpenWrt versions now restrict txpower lower. I think something has changed in OpenWrt.
