Ubiquity AmpliFi HD

[Hermann]

Some experience with Ubiquiti Amplifi HD WiFi router could be helpful if someone wants to set up the same or similar network. Regardless of the excellent change in the audio stream, the network is unstable and cannot keep up with AVM in some discipline.
However, the first setup is exemplary simple. Figure 1 shows the configuration.
.

AmpliFi_Mesh2.jpg

.

There were no problems during the first three weeks until very short drops in volume drew attention. They kept occurring randomly and sporadically, so the first step was of course to check the WiFi signals between the routers. The routes were shown as perfect. Nevertheless, even after a restart, the interruptions remained, increased with subsequent crashes.

I noticed that the amplifiers were active at 2.4 GHz. Surprising, since AVM consistently outputs the first WiFi route in the 5GHz network between 590 and 650Mb/s and, on the other hand, only 4 devices are transmitting, while almost 20 appear in the 2.4GHz network. The first attempt to switch the middle router to 5GHz failed. After booting it was back to 2.4Ghz. Only when the last router was set to 5GHz could the middle one suddenly also use this frequency band.
In between, the amplifiers automatically switch to 2.4GHz. No reason can be seen. The Fritzbox logs show nothing unusual that could explain this. Unfortunately the Amplifi don't have any logs and I can't read the service data.
Another behavior is the last router logging into the first.
.

AmpliFi_Mesh1.jpg

.
If the last router is logged in to the first one, it stays there. That's not my understanding of mesh. It can only be changed by restarting (picture shows work in progress) with the hope that the middle one (about 8m away in the same room) will be recognized.

For a long time I thought the 5GHz network connected the routers, but I don't understand why both nearby routers switch to the 2.4GHz network at some point.

It is not possible to use a dedicated channel in the amplifiers in the 5GHz network, they are simply not selectable, but rather 'long range' and 'short range' with starting points. Whatever that means, it is not possible to choose a free channel. I once managed to set a different channel without auto, which simply didn't work. So back to auto and the router selects the most occupied channels in both networks. Stone Age!

I used the second AVM repeater in between to check the stability, but the network is stable and sets itself up automatically, and above all it regularly looks for less occupied channels and switches.

The Amplifi network is currently stable, but no one is allowed to ask why. I don't know the answer to that, but maybe it was the 30 cm change in location of the middle router. Who knows what to say when they show perfect signal levels consistently and in the event of an error.

My next attempt will be to take the first amplifier out of the network and feed the middle amplifier directly from the Fritzbox. If there is no qualitative difference to be heard in the music, the next step will be to remove the middle amplifier and use two AVM WiFi signal routes.

[matthias]

Hermann,
I don't know if it helps but I enabled "Band Steering" on my router so it searches automatically on both 2.4 and 5GHz for the best channel.

[Hermann]

Thanks Mathias,
Band steering is switched on. I guess by default. But maybe I'll turn it off to force the routers to 5GHz.

[tokenbrit]

There's nothing "perfect" about a bandwidth of 0.2 Mbps - even on 2.4 you should get over 50 if the AVMs get over 500 Mbps on 5GHz.

Which AmpliFi HD is the 'base' station, and how is it connected to your Internet? Do you have a separate modem, or is it a gateway with router capabilities? If it were me, I'd start with just one HD connecting to your Internet, and run a speed test - anything obviously low, like 0.2, would indicate a connection/config issue, and should be resolved before trying to get a reliable mesh network setup.

[Hermann]

Thanks for the tips tockenbrit.
The throughput is not the problem, the values serve to illustrate the behavior of the Amplifi routers.
.

Setup #1.png

The analysis continues. First a test. Downloaded some playlists from Spotify and turned off the monos. Spotify offline and the IPad mini logged into the middle amplifier (#2). The background is an attempt to determine the stability of the connection. And lo and behold, it ran smoothly for several days and nights, without any interruptions.
Then Spotify came back online and a short time later it started again, short drops. Checked the setting of the amplifier and was amazed again, a different channel was selected, instead of the usual 44, now channel 40, which is of course just as busy as the others. After leaving the apartment for a short time and returning, my wife says there were interruptions. Checked again, this time the 44 channel was displayed again.

With three devices it should be possible to identify an amplifi that may be susceptible to failure or is defective. So the amplifiers were swapped so that only two were actively connected at a time. Unfortunately also without success.

Next, the band steering was switched off, but without success, the interruptions remained.
.

Setup #2.png

Since I had to assume that the WiFi route between Amplifi #1 and #2 was unstable, I set up setup #2. The astonishing effect was the lack of influence on the musicality of the change. However, interruptions. Curiously, the band stearing was turned on again when I checked the settings.

Ok, but the sensitivity of connection #2 to #3 (now reduced to 6.5m) is extremely high. Every small movement of the Amplifi device #2 resulted in short interruptions. Also strange.

So the last consideration was to disable the automatic channel selection. And since it's on 153, no interruptions over 36 hours of Spotify online. This indicates a problem with the automatic channel switching.

If interruptions still occur, the next attempt is to have the Amplifi #3 controlled directly by the remaining AVM repeater, which should be called setup #2 with only one Amplifi and an AVM repeater. The last thing that comes to mind is to use the second AVM repeater again to connect the LAN output to the remaining Amplifi. That means swapping the Amplifi #2 for an AVM repeater.

[Sopper]

AVM is set to bridge mode? Or the Amplifi HD?

[Hermann]

There is a LAN bridge between the AVM router and the Amplifi #1. Setup #1 shows the Amplifi WiFi network, which is independent of the AVM WiFi network. The AVM repeaters inherently act as WiFi bridges with two Lan ports.

In this respect, the Amplifi network does not require the bridge mode.

Setup #2 is a mix of both networks.

[Sopper]

There is a LAN bridge between the AVM router and the Amplifi #1. Setup #1 shows the Amplifi WiFi network, which is independent of the AVM WiFi network. The AVM repeaters inherently act as WiFi bridges with two Lan ports.

In this respect, the Amplifi network does not require the bridge mode.

Setup #2 is a mix of both networks.

It’s my understanding that 2 different routers who both act as DHCP causes unstable internet.
One of them should be put in bridge mode.
In my case the supplied router from my ISP set to bridge and Amplifi HD as DHCP works best.

[Hermann]

Thanks Sopper, I don't have any experience with that. Both WiFi routers use different channels, both in the 2.4 GHz network and 5 GHz. In the last net mentioned, only 4 WiFi routers are active including my two networks and recently on different channels. I don't see any conditions for instability.

If I understand you correctly, the Amplifi takes over the DHCP function for the entire network. I wanted to avoid that. I'm now pretty sure the cause of the interruptions is due to the channel change, but that's not a well-founded finding. It should work uninterrupted for a few more weeks.

On the other hand, I don't want to change the setup right now, but I see the need to look into it further to understand what the bridge mode does in my setup.

[tokenbrit]

On the other hand, I don't want to change the setup right now, but I see the need to look into it further to understand what the bridge mode does in my setup.

For when you are looking into it further - https://help.amplifi.com/hc/en-us/artic ... ridge-Mode

Hope you get it sorted to your satisfaction. Happy holidays.

[Hermann]

Thanks again tokenbrit. The link is very helpful. The Fritzbox cannot be put into bridge mode. In order to use the Amplifi for this, I have to convert it again and use three Amplifis like in setup #1. Obviously this is the preferred setting.

[Azazello]

Finally managed to find a pair. Distinct improvement in sound quality over my previous Asus RT-AX92U. :)

[Azazello]

Let me clarify.

If you have Källa and another router: get Amplifi HD. Just do it.

[Hermann]

Short interim report. The above setup is stable so far. In fact, the stream seems to be less susceptible to interference with two Amplifi HD.

The test with just one Amplifi will have to wait.

[Hermann]

Over the past 14 days, the audio stream via two Amplifi HD has been stable. A fixed channel in the 5GHz network is set up in the first router. A stable signal is displayed between the routers. I am confident that there will be a solution.

[Kalle05]

Has enyone tried Amplifi HD with a ”better” power supply ?
I bought an iFi Audio iPower2 12volt and got an usb-c adapter so i could connect it to the router.
I got an overall improvement with the iFi, it is possible that it is a placebo effect but i hope it is a real improvement.
And many of you think right now that it should be a 5volt power supply,
the Amplifi needs at least 15 watts in ”power” and can operate with 5v - 9v and 12 volt power supplys.
Would be fun to hear if anyone has tried the same ?

[markiteight]

Has enyone tried Amplifi HD with a ”better” power supply ?
I bought an iFi Audio iPower2 12volt and got an usb-c adapter so i could connect it to the router.
I got an overall improvement with the iFi, it is possible that it is a placebo effect but i hope it is a real improvement.
And many of you think right now that it should be a 5volt power supply,
the Amplifi needs at least 15 watts in ”power” and can operate with 5v - 9v and 12 volt power supplys.
Would be fun to hear if anyone has tried the same ?

Would you be willing to post some clips in the Playground thread? I'm sure there are many of us who are curious to hear the stock vs. aftermarket power supplies.

[Kalle05]

Has enyone tried Amplifi HD with a ”better” power supply ?
I bought an iFi Audio iPower2 12volt and got an usb-c adapter so i could connect it to the router.
I got an overall improvement with the iFi, it is possible that it is a placebo effect but i hope it is a real improvement.
And many of you think right now that it should be a 5volt power supply,
the Amplifi needs at least 15 watts in ”power” and can operate with 5v - 9v and 12 volt power supplys.
Would be fun to hear if anyone has tried the same ?

Would you be willing to post some clips in the Playground thread? I'm sure there are many of us who are curious to hear the stock vs. aftermarket power supplies.

My music equipment is not up to your standards, if i post clips then it would sound bad to all of you regardless that i have had an improvement :-) I have a very small budget to spend on music gear and if i can get an improvement by adding a powersupply that costs about 70 euros i am "happy" with that, i am sure there are many out there who can in an "experimental mind mode" buy that powersupply and test it and post clips if there is others that would like to hear

[lejonklou]

Clips often sound bad. That’s not a problem as we only wish to hear the difference between the power supplies. So please go ahead and do it!

I have not tried any alternative power supplies for the Ubiquity Amplifi. It’s definitely possible there’s a better one out there.

The iFi supplies have however been tested with other products by several people I know, for instance to power the switch in between Amplifi and Källa. I think there were some posts here as well. The impressions have so far not been favourable when judged with the Tune Method.

I would personally rather test a range of ordinary power supplies than any brand that claim to have made a version for audio gear. Why? Because usually they don’t know how to judge musical performance. So when they build or tweak the power supply, they aim for things like ”it sounds bigger”, ”there’s more bass” or ”now there’s more detail”. And when we judge their result with the Tune Method, we find that the music is bloated, confused and boring.

[markiteight]

My music equipment is not up to your standards,

I have standards for other peoples' equipment? Huh...I didn't know that. Could you please explain, to me, what my standards are? Could you also please explain why that matters? This forum is a judgement free zone. We're here to collaborate and share, not point fingers and judge.

if i post clips then it would sound bad to all of you regardless that i have had an improvement

Yes, it would. And that's the point! I'll go a step further than Fredrik and say that ALL clips sound bad. They're supposed to! By removing the HiFi-ness from the equation it's much easier to focus on the relative differences in musical performance. That is what this forum is all about. Nobody here cares what your clips sound like, but we're all definitely interested in the musical merits of a potential upgrade.

I have a very small budget to spend on music gear and if i can get an improvement by adding a powersupply that costs about 70 euros i am "happy" with that, i am sure there are many out there who can in an "experimental mind mode" buy that powersupply and test it and post clips if there is others that would like to hear

Whether you or someone else posts the clips doesn't matter, people here will be equally interested in the comparison. Since you've already taken the plunge, why not share with the rest of the class? Besides, it sounds like you have some doubts as to whether the new power supply is indeed a genuine upgrade. What better way is there to help shed those doubts than soliciting the feedback of others who all share a common goal?

[Hermann]

Some time ago I tried to achieve an improvement with IFI power supplies. Repeaters and switches were equipped with these power supplies. I didn't understand why they were that hyped, because it wasn't an improvement. Another type was favored, Friwi medical plug-in power supplies. A total failure. All devices were returned.

However, there are a number of simple plug-in power supplies for various devices, so that the Netgear GS 308T is using an older GS 108T, which is much thicker than today's narrow plugs. I'm no longer looking for other power supplies etc., but I follow the discussions, of course, because that's exactly why this forum is so valuable.