OpenWrt/LEDE Project

  • Status Closed
  • Percent Complete
  • Task Type Bug Report
  • Category Base system
  • Assigned To No-one
  • Operating System All
  • Severity Medium
  • Priority Very Low
  • Reported Version lede-17.01
  • Due in Version Undecided
  • Due Date Undecided
  • Private
Attached to Project: OpenWrt/LEDE Project
Opened by Rod Egan - 26.02.2017
Last edited by Adrian Schmutzler - 19.06.2020

FS#557 - Nexx wt3020 wifi rx sensitivity low with LEDE 17.01

The rx sensitivity reported by a Nexx wt3020 is ~ 20dB lower than a comparable tp-link device located sided by side with the Nexx wt3020.
There have been reports of a similar problem with OpenWRT versions.

There is a ticket open with openwrt with a similar reported issue as well as on the openwrt forum

lede trunk (and openwrt trunk) as of 24/2/17 also display similar poor rx sensitivity.


Closed by  Adrian Schmutzler
19.06.2020 11:27
Reason for closing:  Won't fix
Additional comments about closing:  

LEDE-17.01 has reached EOL (end of life). This bug has been closed automatically. If the problem persists with a newer version, please file an appropriate new bug report for that version.

Fabianhu commented on 13.03.2017 09:54

I tried hard by setting my repo to the commit mentioned by @psyborg55.
Unfortunately the same result like HEAD (080c6fa476d1320f83ad56e6aeaa292960f63dc1):

with LEDE (luci): I see only my own AP with 75% signal strength at < 1m distance
At 5m distance I have no stable connection.

With the original FW I see ~10 more from my neighbors.

does anyone have any directions for me?

Roger commented on 13.03.2017 11:56

The patch mentioned above was commited to the master branch, so the snapshot builds already have it.

You may want to give the snapshot compilation a try (if you are familiar with the command line, as they don't include the web GUI). You can find them at

Try flashing both 17.01.0 and the snapshot and check if there are any differences.

Rod Egan commented on 14.03.2017 04:15

See my comment in the original bug report. No difference between 17.01.0 and snapshot as of 24/2/17. The patch indicates it was committed 20/1/17 so it should have been in the 24//17 snapshot that I tested with.

Project Manager
Felix Fietkau commented on 18.04.2017 08:10

Please try the latest snapshot

Rod Egan commented on 18.04.2017 12:22

Tried with LEDE Reboot SNAPSHOT r3975-97d1c49
RX signal still poor
Also noted that with country code set to AU, TX power can supposedly be set to 36dBm!

root@LEDE:~# iw list
Wiphy phy0

      max # scan SSIDs: 4
      max scan IEs length: 2257 bytes
      max # sched scan SSIDs: 0
      max # match sets: 0
      max # scan plans: 1
      max scan plan interval: -1
      max scan plan iterations: 0
      Retry short long limit: 2
      Coverage class: 0 (up to 0m)
      Available Antennas: TX 0 RX 0
      Supported interface modes:
               * IBSS
               * managed
               * AP
               * AP/VLAN
               * monitor
               * mesh point
      Band 1:
              Capabilities: 0x2fe
                      SM Power Save disabled
                      RX Greenfield
                      RX HT20 SGI
                      RX HT40 SGI
                      TX STBC
                      RX STBC 2-streams
                      Max AMSDU length: 3839 bytes
                      No DSSS/CCK HT40
              Maximum RX AMPDU length 65535 bytes (exponent: 0x003)
              Minimum RX AMPDU time spacing: 2 usec (0x04)
              HT TX/RX MCS rate indexes supported: 0-15, 32
                      * 2412 MHz [1] (36.0 dBm)
                      * 2417 MHz [2] (36.0 dBm)
                      * 2422 MHz [3] (36.0 dBm)
                      * 2427 MHz [4] (36.0 dBm)
                      * 2432 MHz [5] (36.0 dBm)
                      * 2437 MHz [6] (36.0 dBm)
                      * 2442 MHz [7] (36.0 dBm)
                      * 2447 MHz [8] (36.0 dBm)
                      * 2452 MHz [9] (36.0 dBm)
                      * 2457 MHz [10] (36.0 dBm)
                      * 2462 MHz [11] (36.0 dBm)
                      * 2467 MHz [12] (36.0 dBm)
                      * 2472 MHz [13] (36.0 dBm)
                      * 2484 MHz [14] (disabled)
      valid interface combinations:
               * #{ managed, AP, mesh point } <= 8,
                 total <= 8, #channels <= 1
      HT Capability overrides:
               * MCS: ff ff ff ff ff ff ff ff ff ff
               * maximum A-MSDU length
               * supported channel width
               * short GI for 40 MHz
               * max A-MPDU length exponent
               * min MPDU start spacing
psyborg55 commented on 29.07.2017 11:05

next time indicate in your issue report your device use MT7620N

Rod Egan commented on 29.07.2017 11:17

Yes Sir

psyborg55 commented on 24.08.2017 13:11

can you post RF calibration data from your device?

Rod Egan commented on 24.08.2017 13:21

Can point me to info about how to get the rf cal data? I've done a quick search on the lede forum but haven't found anything so far.

psyborg55 commented on 24.08.2017 13:55

run from ssh console: dd if=/dev/mtd2 of=/tmp/mtd2
scp mtd2 file to your PC and post here

Rod Egan commented on 24.08.2017 14:16

result of mtd2 download uploaded

   mtd2 (64 KiB)
psyborg55 commented on 24.08.2017 14:46

add chanregs 14, 17 and 18 to your rt2800lib.c file like this and rebuild image:

	if (rt2x00_has_cap_external_lna_bg(rt2x00dev)) {
		rt2800_bbp_write(rt2x00dev, 75, 0x68);
		rt2800_bbp_write(rt2x00dev, 76, 0x4C);
		rt2800_bbp_write(rt2x00dev, 79, 0x1C);
		rt2800_bbp_write(rt2x00dev, 80, 0x0C);
		rt2800_bbp_write(rt2x00dev, 82, 0xB6);
	} else {
		rt2800_rfcsr_write_chanreg(rt2x00dev, 14, 0x66);
		rt2800_rfcsr_write_chanreg(rt2x00dev, 17, 0x20);
		rt2800_rfcsr_write_chanreg(rt2x00dev, 18, 0x42);

try checking again if the RSSI got any better

Rod Egan commented on 25.08.2017 06:31

Built a dev system and made the suggested changes in ~/source/build_dir/target-mipsel_24kc_musl/linux-ramips_mt7620/compat-wireless-2017-01-31/drivers/net/wireless/ralink/rt2x00/rt2800lib.c around line 5287

flashed the wt3020 device and rx signal strength reported as -64/68 compared to -64/66 with 17.01.2 and snapshot r4741-7a8909411c near enough to no change.

As a check a physically adjacent device (a5v11) reports rx signal strength ~-44.

Both wt3020 and a5v11 have rx signal on wr842 (ap) as ~-40.

psyborg55 commented on 25.08.2017 12:44

this is the report i got about mt7620n devices. probably unrelated at all to these registers

Rod Egan commented on 25.08.2017 14:03

I won't be able to get back to this for a couple of weeks. I read a suggestion that the code by padavan doesn't have this issue. I'll see if I can get a copy and do some comparisons to see if there is anything obvious and/or try the code.

Rod Egan commented on 20.08.2018 07:41

Issue still present in 18.06.1
It isn't an issue with a specific device as I have a new device and the reported low rx sensitivity is still present.
Unfortunately I neglected to check with the oem firmware before loading 18.06.1 so it is unknown if oem firmware is any diferent.


Available keyboard shortcuts


Task Details

Task Editing