Atheros AR9280 / AR7010 Initial low scaning signal range

Atheros AR9280 / AR7010 Initial low scaning signal range

[cybersnow_2001]

I'm using an Atheros AR9280 USB dongle (Panasonic UB94) under linux. I think a few years ago, it wasn't doing this behavior but now, I think it's starting to be a serious issue. When I plug the USB Wifi key, I can only see a few Access points but not all. Even near the main AP, the signal is seen as very low (1 bar out of 5). Sometimes, I cannot even see it. When trying to connect to it, it connects hardly. But after a few minutes, the signal becomes strong and all the neighborhood AP are shown. When the AP advertise the right TXPower, it seems to fix all the issues. The others AP appear and the signal become stronger on the actual connection. I see the signal jumping from 30% to 70% and the RX Rate jump too.

I suspect the ath9k_htc module is not enforcing a high enough TX Power at first glance or something else happening I don't understand with this chipset... The country code is set at 00 at first USB connection. But until I connect to an AP, the AR9280 stays at a low power state and on this country ID 00. Also, it's like the 5Ghz network scanning isn't activated. I don't know...

I've read in another forum that people experienced the same issue after a kernel update, maybe after 6.8. What do you think?

Re: Atheros AR9280 / AR7010 Initial low scaning signal range

[John Scott]

[-- Attachment #1: Type: text/plain, Size: 1185 bytes --]

On Wed, 2026-07-01 at 04:08 +0200, cybersnow_2001@tutamail.com wrote:
> I've read in another forum that people experienced the same issue after a kernel update, maybe after 6.8. What do you think?

I think I can partially reproduce the problem. I have a Sony UWA-BR100 on Debian Trixie (kernel 6.12.94) and a wireless scan only shows very few access points, never more than -65 dBm or so. The access point for my home, being the closest one to me, actually has the *worst* signal strength value reported as -80 dBm consistently. A scan for access points does show 5GHz ones, but it chose to connect to my AP on 2.4GHz (which may or may not be legitimate; iwd normally joins the 5GHz ones when I'm using dual-band NICs). I did not observe any circumstance in which the reported signal became better, though.

With a TP-Link TL-WN822N v2 (still AR7010, but the 2.4GHz-only AR9287 instead of dual-band AR9280), everything is excellent, so indeed it seems only the AR9280 devices are affected.
Could you share that forum link? I'm not sure I'll get around to it but my next steps would be to bisect the kernel using a virtual machine and USB pass-through. This is strange indeed...

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 411 bytes --]

Re: Atheros AR9280 / AR7010 Initial low scaning signal range

[cybersnow_2001]

Thanks for looking at it. Here is what I could find on the Web about the same issue. Someone said Kernel 6.4 and earlier worked. You might try with Debian Live DVD Buster, Bullseye and others distro if you want. I tried to use an older firmware of Atheros v1.3 instead of 1.4, same problem. I wonder if there are older firmware... But I'm not sure if this is related to the firmware. It might be more about the kernel 80211 modules... Or it might be an electronic issue coming from the chip itself. I need to do a cold/freeze and very hot test to see if it solve the issue...
https://discussion.fedoraproject.org/t/need-help-with-weak-wifi-signal-on-fedora-with-qualcomm-atheros-card/87854
https://www.spinics.net/lists/linux-wireless/msg90423.html
https://forum.mikrotik.com/t/ar9280-low-speed/62030
https://forums.freebsd.org/threads/atheros-ar9280-wifi-not-working.68357/
https://forum.opnsense.org/index.php?topic=28266.0
https://forum.mikrotik.com/t/compex-wle200nx-atheros-ar9280/166217/3
https://forum.netgate.com/topic/83364/atheros-ar9280-testing-settings-craziness-success-ymmv/2
https://forums.tomshardware.com/threads/qualcomm-atheros-ar9287-terrible-range.2687938/
https://forum.netgate.com/topic/53616/atheros-ar9280-ath0-stuck-beacon-resetting-bmiss-count-4-hangs-later/2
https://forums.linuxmint.com/viewtopic.php?t=299646



Jul 4, 2026, 17:25 by jscott@posteo.net:

> On Wed, 2026-07-01 at 04:08 +0200, cybersnow_2001@tutamail.com wrote:
>
>> I've read in another forum that people experienced the same issue after a kernel update, maybe after 6.8. What do you think?
>>
>
> I think I can partially reproduce the problem. I have a Sony UWA-BR100 on Debian Trixie (kernel 6.12.94) and a wireless scan only shows very few access points, never more than -65 dBm or so. The access point for my home, being the closest one to me, actually has the *worst* signal strength value reported as -80 dBm consistently. A scan for access points does show 5GHz ones, but it chose to connect to my AP on 2.4GHz (which may or may not be legitimate; iwd normally joins the 5GHz ones when I'm using dual-band NICs). I did not observe any circumstance in which the reported signal became better, though.
>
> With a TP-Link TL-WN822N v2 (still AR7010, but the 2.4GHz-only AR9287 instead of dual-band AR9280), everything is excellent, so indeed it seems only the AR9280 devices are affected.
> Could you share that forum link? I'm not sure I'll get around to it but my next steps would be to bisect the kernel using a virtual machine and USB pass-through. This is strange indeed...
>