@Riley_S 

 

Thank you very much for your quick and informational response.  Much appreciated.

 

Your point about attenuation in the 5GHz range being greater than in the 2.4 is well taken.  However, when I use my old Archer C9 router, I have plenty of signal strength and download speeds using the 5GHz band, plus the ability to occasionally move my channel selection when my many neighbors start adding new internet services (which seems to happen more and more often).

 

If it is technically challenging to automate channel switching on the 5GHz band, I would certainly welcome at least the opportunity to manually choose a better channel.  I don't know how, but I have neighbors who are putting out 5GHz signals into my house that test stronger than my own, and are crowded around channel 40.  I can avoid them and getter better download speed by using the older, wifi 5 Archer.

 

Otherwise, the Deco XE75 is a great piece of kit.  It certainly makes my life easier on the 2.4GHz band!  I was getting chronic interruptions of service, despite manually jumping channels almost every day.  Now, the Deco does that automatically, and more effectively than I was doing.

If it could optimize the 5GHz that way, it would be wonderful.

 

I guess I'm not smart enough yet to understand why either automatic or manual adjustment cannot be made at 5GHz.  I was unable to follow that link you suggested.  I got a "Forbidden" rejection there.  I'll try again with a different browser later.

 

I'm trying to reduce my difficulties at 5GHz by running ethernet cable from the satellite to as many PCs and TVs as possible.  If Powerline worked better (across circuits), that would be my salvation.

 

Again, thanks for taking time to help me along.

  @Riley_S 

Thanks once again for your kind reply.

 

The link you sent is one I had already read, but I appreciate your providing it just the same.  That post does provide some useful perspective, however:

 

* It was posted well over two years ago, yet no real resolution has been developed.

 

* While I understand that the developers are trying to make the mesh system self managing (easy on the homeowner), I don't see how that goal conflicts with providing automatic channel adjustment in the 5GHz band.  In fact, it would seem a great improvement toward that goal.

 

* Despite the various discussions that have popped up concerning this issue over the past couple of years, I have still found nothing that explains WHY the same automation being used to optimize channels in the 2.4GHz band cannot also be applied to the 5GHz band, where we want to connect our speed-critical devices, and where there is a broader range of channels to choose from.  There may well be a good explanation, but I have been unable to find it.

 

If any sort (manual or automatic) of channel selection is inherently impossible for the 5GHz band, that would be a good thing to know.  It would help a (perhaps) small subgroup of potential mesh buyers to look elsewhere for a solution to the congestion that is now spreading beyond the 2.4GHz band and into the 5GHz band.

 

In line with your other kind advice, I have already been investigating replacing as much of my wifi network as much as possible with wire --- be it Powerline, pre-existing coax, or (over short distances) ethernet cable.  If I can accomplish enough along those lines, perhaps I can ditch the mesh idea altogether, and try to find a wifi 6 router with enough smarts to manage both the 2.4 and 5GHz bands with AI.

 

Once again, thank you very much for your insights and suggestions.  They definitely have been very helpful.

  @Iggy64 

Upon further review --------

 

After paying closer attention to the available settings in the Deco app, as well as diving more deeply into the 802.11 channel allocation in North America, I may be starting to understand why there is no automatic channel switching in the 5GHz band.

 

If we are to steer clear of the DFS channels, we are left with UNII-1 and UNII-3.  For reasons I don't completely appreciate yet, UNII-1 is generally preferable (more stable?) than UNII-3.  This puts the focus on UNII-1 -- roughly channels 36 -48.  

 

I note that the Deco's advanced settings allow us to choose either 80 or 160MHz (the latter recommended) as the channel width in the 5GHz band,  If I interpret the band structure correctly, even an 80MHz channel width will pretty much encompass the entirety of UNII-1.  Using 160MHz would require expanding into the DFS channels.

 

Back when I was still running on my TP-LINK Archer C9 router, I used 20 or 40MHz channel widths.  I therefore had a little bit of room to move within UNII-1.  Sometimes I even used UNII-3.  Now, with the Deco system, I have only the much wider channel widths to choose.  As a result, there is simply no room to move; I am using the entirety of UNII-1 whether I choose 80 or 160 band width. 

 

Perhaps TP-LINK figures we might as well go wide and take advantage of the bandwidth, since our neighbor's 5GHz signals usually attenuate fast enough to pose no serious interference.  My further testing and evaluating seems to indicate that this is indeed the case.

 

Where I remain confused is this:

 

If I choose the 80MHz channel width option and then open my Wifi Analyzer, it appropriately shows the expected quasi-parabolic signal curve arching from Channel 36 to 48, and numerically lists the channel width as 80.  That is, it blankets UNII-1.  I am getting all the bandwidth I can without trespassing into DFS.

 

However, if I choose the recommended 160MHz channel width, my Wfi Analyzer shows the same-sized curve from Channel 36-48, and lists the channel width as 80. 

 

At this point I therefore wonder:

 

- Are the 80MHz and 160MHz options really giving me the same result, such that DFS channels are being avoided, or

 

- Does the 160MHz option actually expand the channel width into the DFS range, but the Wifi Analyzer can't register that effect?

 

I'll have to do a bunch of testing to see if I can detect a consistent difference in system performance when I switch between the two channel-width options.

 

Perhaps someone can advise me where my thinking has gone off the rails, and help me further understand the big picture here.