Using 5GHz only

[tarassippo]

@tarassippo, that is not correct anymore. The lower channels have the same power as the higher ones for a few years now, but I don't think any router is allowed to transmit at 4W legally for any consumer router.

I'd be interested in reading such documents/sources... What I've learnt so far is that the all WiFi ecosystem is a kind of jungle - well, at least for me...

I have no clue if there are (consumer) routers capable of 4W on channels 149-161...

This is from an official document from FCC (April 2013):

 

[OzarkEdge]

Source: https://www.engeniustech.com/go-guide-channel-transmit-power-wi-fi-networks-2/

The transmit powers may or may not be current... but I suspect they are relatively accurate. The following argument can be adjusted for non-USA regulatory conditions.

The above article suggests why some channels are missing from some devices even though permitted by regulation. If I omit the 5.0 GHz channels not listed on my 86U for each bandwidth of 20, 40, and 80 MHz (not documented here)... so, only channels common to 20 and 40 and 80 MHz bandwidths... and further restrict those to only the primary channels for 80 MHz bandwidth... I'm left with channels 36, 52, 100, and 149. If I further restrict those to highest Tx power and no DFS interference, I'm left with channel 149 at 20/40/80 MHz bandwidth.

In my 86U experience, channels 149, 153, 157, and 161 have consistently logged an AiMesh wireless backhaul with 80 MHz bandwidth and the highest RSSI and Tx/Rx values (5G 3 bars at 77'). Since 80 MHz bandwidth spans all of channels 149, 153, 157, and 161; I suspect the selected/listed primary channel is more or less cosmetic... the primary channel is 149 at 80 MHz bandwidth (there are other interpretations of 'primary channel' but regardless... 80 MHz spans all four channels beginning at channel 149).

My current approach to WiFi channel configuration:

Disable Smart Connect and set different SSIDs and fixed channels for more reliable/stable WiFi (avoids band contention and Auto channel/DFS disruption, and allows segmenting clients across WLANs).

- set 2.4 WLAN to OE-24, N only, 20 MHz, ch 1,6,11 or 1,5,9,13 (use least congested), WPA2-Personal AES
- set 5.0 WLAN to OE-50, N/AC mixed, 20/40/80 MHz, ch 36,52,100,149 (149 has highest Tx power and no DFS), WPA2-Personal AES

Smart Connect might help clients connect to the preferred same SSID/band... reducing 2.4 GHz Tx power might help balance band contention... but disabling Smart Connect and using different SSIDs will help clients connect to the preferred different SSID/band AND allow you to maintain 2.4 GHz Tx power to extend coverage for select 2.4 GHz clients at distance AND allow you to segment clients across WLANs for better overall WiFi performance.

OE

 

[doczenith1]

This image show the Max EIRP at 4 W for both the lower UNII-1 (36-48) and the upper UNII-3 (149-161) channels.

 

[tarassippo]

@doczenith1 are you able to share the link ?

 

[OzarkEdge]

This image show the Max EIRP at 4 W for both the lower UNII-1 (36-48) and the upper UNII-3 (149-161) channels.

View attachment 24427

Trust but verify... I consistently get a significantly stronger wireless backhaul using the high channels on my 86Us.

OE

 

[ColinTaylor]

@tarassippo Your profile says you're in the UK so all these posts regarding power and channel allowances in the USA are meaningless to you.

Here is the current information for the UK: https://wifinigel.blogspot.com/2018/05/updated-white-paper-on-license-exempt.html

Note that this doesn't discuss 5GHz for things like outdoor point to point connections or SRDs.

Also note that even though the UK opened up usage of Band C recently the same is not true for the EU. Given that most consumer WiFi routers (like Asus) purchased in the UK are localised for the EU they will not allow usage of Band C.

So in summary the maximum permitted power output (assuming TPC) in the UK is 200mW for Band A, 1W for Band B and 200mW for Band C (if available). Just because the regulations allow 1W doesn't mean a particular device can produce that output.

 

[tarassippo]

@ColinTaylor I wouldn't use the word 'meaningless', nonetheless I'm learning lots of interesting stuff

Another good article: https://www.solwise.co.uk/downloads/files/intheuk5ghz.pdf

 

[ColinTaylor]

Another good article: https://www.solwise.co.uk/downloads/files/intheuk5ghz.pdf

Note that like a lot of documents this information is out of date because it doesn't cover the Band C changes introduced in 2017.

 

[doczenith1]

@doczenith1 are you able to share the link ?

Sorry but I do not have the link. I've had that graphic sitting on my desktop for a while now.

Regardless of what any of these images says the best thing to do is just try different channels and test your throughput to see which one works best in your environment.

 

[OzarkEdge]

I wouldn't use the word 'meaningless', nonetheless I'm learning lots of interesting stuff

Right. Adding the wrong numbers will yield the wrong answer, but you still learn how to add.

OE

 

[tarassippo]

Note that like a lot of documents this information is out of date because it doesn't cover the Band C changes introduced in 2017.

Ah my bad, and is this document up-to-date: https://www.ofcom.org.uk/__data/ass...wireless-telegraphy-exemption-regulations.pdf

Or do you have a link ?

 

[L&LD]

I agree with the testing for each environment/router and associated client devices.

In my experience I consistently see the lower 5GHz channels giving superior throughput and responsiveness for network use. Note not necessarily stronger signal rates, but higher and more consistent throughput in actual use.

 

[ColinTaylor]

Ah my bad, and is this document up-to-date: https://www.ofcom.org.uk/__data/ass...wireless-telegraphy-exemption-regulations.pdf

Or do you have a link ?

That's probably up to date. It has the same publication date as its companion document I indirectly linked you to earlier.

To be honest I would find it impossible to keep track of all the ongoing regulation changes, both in the UK and the EU. I just rely on WifiNigel to keep me informed on UK-related matters.

 

[MarkyPancake]

In my experience I consistently see the lower 5GHz channels giving superior throughput and responsiveness for network use. Note not necessarily stronger signal rates, but higher and more consistent throughput in actual use.

Same here. If I deviate too far from Channel 36, I see wireless performance issues across all devices, whereas it doesn't seem to matter how congested 36 is with neighbouring networks, it always performs the best in my environment.

 

[OzarkEdge]

In my experience I consistently see the lower 5GHz channels giving superior throughput and responsiveness for network use. Note not necessarily stronger signal rates, but higher and more consistent throughput in actual use.

I was inclined to accept that if a client connection RSSI and Tx/Rx values in the Wireless Log are maximized, then the 'throughput and responsiveness' would be optimum. The nearest house/WiFi here is about 400' away, so not much interference going on.

OE

 

[L&LD]

@OzarkEdge, that is exactly what I've found though. Higher Tx/Rx values and lower noise floor sometimes means nothing. That is exactly why I don't like using an app to set up the network.

 

[OzarkEdge]

@OzarkEdge, that is exactly what I've found though. Higher Tx/Rx values and lower noise floor sometimes means nothing. That is exactly why I don't like using an app to set up the network.

So how do I measure 'throughput and responsiveness'?

OE

 

[doczenith1]

So how do I measure 'throughput and responsiveness'?

OE

I believe iPerf would be the best tool to measure throughput. I ran my test between my wireless device(s) and a wired computer on my network. The options that worked best for me and provided the best throughput numbers was: iperf3 -c 192.168.1.11 -O 5 -i 5 -w 4M -t 60

On Android, Magic iPerf worked well more me. https://play.google.com/store/apps/details?id=com.nextdoordeveloper.miperf.miperf