Problem with operator's 5G NSA

waaaaski · Sun Oct 16, 2022 10:34 pm

Hello there,

I have two questions related to the 5G NSA. My current operator has mast with B1, B3, B8 and n41 and I'm connecting to this using 5G NSA.

1. I cannot connect using all bands - it could be done only (B3+B8+n41) or (B1+n41). The first option is done automaticlly by router the second option is possible by manual allowig on LTE side only B1 band - without forcing only B1 band, the router even doesn't see this band. It's my fault that I cannot aggregate all bands B1, B3, B8 and n41? Should I make some changes with pigtails on modem (I'm using Chateau 5G) or it could be the issue with operator mast? I was testing other operator (which had B20, B3, B8, n1) and here I didn't have any problem, router was picking all the bands in one time (in 5G NSA mode).

2. Today (I mean 16/10/2022 on evening) the router has problem with stable work with band n41:

the router even drops completely the n41 band (and switches from 5G NSA to LTE)
Here I also tested competition operator and the 5G band (n1) is working correctly. Not sure if the mast currently is facing the highest peak of users logging/using this mast - or I'm facing an other issue here? I collected some logs from modem.
Worth mention will be that I'm using this operator for couple of days. Normally the SINR for n41 I had 26-30dB (all the time) and this situation has never happened until today.

My setup: MikroTik Chateau 5G + QUPANEL 5G LTE HP MIMO 4X4, 5M (AP5G4).
The antenna is connected using ports A0, A1, A4, A6 on modem.

Thanks for any suggestions.

waaaaski · Mon Oct 17, 2022 2:01 pm

Now it started dropping the b1 band - no connection on 5G band.

attaching logs - there is a lot of:

 20:47:41 lte,packet,raw lte1 mbim: wdm >>> recv #0
 20:47:41 lte,packet,raw 07000080 74000000 00000000 01000000
 20:47:41 lte,packet,raw 00000000 a289cc33 bcbb8b4f b6b0133e
 20:47:41 lte,packet,raw c2aae6df 0b000000 48000000 63000000
 20:47:41 lte,packet,raw 63000000 05000000 02000000 ffffffff
 20:47:41 lte,packet,raw 1c000000 2c000000 02000000 55000000
 20:47:41 lte,packet,raw 00000000 ffffffff ffffffff 40000000
 20:47:41 lte,packet,raw 4f000000 4a000000 ffffffff ffffffff
 20:47:41 lte,packet,raw 20000000

normally I didn't saw such thing - what it is?

waaaaski · Mon Oct 17, 2022 9:53 pm

I put the same sim card to wife's mobile phone (which has modem 5G) and it's connecting to the LTE (one band B1 or B3) + 5G (I wasn't able to check which band it was) in 5G NSA mode.

On mobile phone it's working well - transfers 150-200+Mbits - where on router it has max 10-20+Mbits? - with dropping 5G bands.

Even weird that the primary bands gets sometimes very strange number on eafrcn (probably max int32?):

hecatae · Tue Oct 18, 2022 1:22 pm

Hi,

Which modem firmware are you using?
viewtopic.php?p=960260&hilit=chateau#p960286

The above may help, I'm currently testing a beta modem firmware for support

waaaaski · Tue Oct 18, 2022 1:54 pm

I tested on both versions:

RG502QEAAAR13A02M4G - it's dropping the 5G band
RG502QEAAAR11A06M4G - it was dropping the LTE band (in the same way like 13A02 is doing it for 5G and there is no 5G present)

From technical support of operator I heard that I never was connected to their 5G network - where router was notifying that I was (before the issue came out).
Not sure if this information is valuable - but I think that the operator expects, that router will login at 2600Mhz TDD (n38) but it's doing it on 2500Mhz (n41). I was trying to force login only on n38 but router wasn't able to do it at all (I mean - connect to the n38).

hecatae · Tue Oct 18, 2022 2:41 pm

Sounds similar to the issue I had on 3 UK, I could see the 5g n78 signal on the new r13a firmware but was unable to connect to it.
I would recommend raising a SUP ticket with support, noting that it's affecting connecting to n41.

waaaaski · Wed Oct 19, 2022 3:51 pm

I decoded the lte mbim reposonses (from the logs):

20:47:45 lte,debug lte1 mbim: >>> N service: connect, command: register state
 20:47:45 lte,packet lte1 mbim: status data: 
 20:47:45 lte,packet 00000000 03000000 01000000 60000000
 20:47:45 lte,packet 01000000 34000000 0a000000 40000000
 20:47:45 lte,packet 08000000 00000000 00000000 02000000
 20:47:45 lte,packet 7c000080 32003600 30003000 31000000
 20:47:45 lte,packet 50006c00 75007300

`4
@|26001Plus

 20:47:45 lte,packet,raw lte1 mbim: wdm >>> recv #0
 20:47:45 lte,packet,raw 07000080 64000000 00000000 01000000
 20:47:45 lte,packet,raw 00000000 3d01dcc5 fef54d05 0d3abef7
 20:47:45 lte,packet,raw 058e9aaf 04000000 38000000 01000000
 20:47:45 lte,packet,raw 01000000 28000000 10000000 00000000
 20:47:45 lte,packet,raw 00000000 00000000 00000000 00000000
 20:47:45 lte,packet,raw 00000000 69006e00 74006500 72006e00
 20:47:45 lte,packet,raw 65007400

d=ÜÅþõM
:¾÷¯8(internet

 20:47:45 lte,debug lte1 mbim: >>> N service: conn_ext, command: lte attach status
 20:47:45 lte,packet lte1 mbim: status data: 
 20:47:45 lte,packet 01000000 01000000 28000000 10000000
 20:47:45 lte,packet 00000000 00000000 00000000 00000000
 20:47:45 lte,packet 00000000 00000000 69006e00 74006500
 20:47:45 lte,packet 72006e00 65007400

(internet

 20:47:45 lte,debug lte1 mbim: lte status: attached apn: internet iptype: 1
 20:47:45 lte,packet,raw lte1 mbim: wdm >>> recv #0
 20:47:45 lte,packet,raw 07000080 74000000 00000000 01000000
 20:47:45 lte,packet,raw 00000000 a289cc33 bcbb8b4f b6b0133e
 20:47:45 lte,packet,raw c2aae6df 0b000000 48000000 63000000
 20:47:45 lte,packet,raw 63000000 05000000 02000000 ffffffff
 20:47:45 lte,packet,raw 1c000000 2c000000 02000000 00000000
 20:47:45 lte,packet,raw 00000000 ffffffff ffffffff 40000000
 20:47:45 lte,packet,raw 51000000 49000000 ffffffff ffffffff
 20:47:45 lte,packet,raw 20000000

t¢Ì3¼»O¶°>ÂªæßHccÿÿÿÿ,ÿÿÿÿÿÿÿÿ@QIÿÿÿÿÿÿÿÿ

 20:47:45 lte,packet,raw lte1 mbim: wdm >>> recv #0
 20:47:45 lte,packet,raw 07000080 74000000 00000000 01000000
 20:47:45 lte,packet,raw 00000000 a289cc33 bcbb8b4f b6b0133e
 20:47:45 lte,packet,raw c2aae6df 0b000000 48000000 63000000
 20:47:45 lte,packet,raw 63000000 05000000 02000000 ffffffff
 20:47:45 lte,packet,raw 1c000000 2c000000 02000000 00000000
 20:47:45 lte,packet,raw 00000000 ffffffff ffffffff 40000000
 20:47:45 lte,packet,raw 51000000 49000000 ffffffff ffffffff
 20:47:45 lte,packet,raw 20000000

t¢Ì3¼»O¶°>ÂªæßHccÿÿÿÿ,ÿÿÿÿÿÿÿÿ@QIÿÿÿÿÿÿÿÿ <- the same thing like before

 20:47:45 lte,packet,raw lte1 mbim: wdm >>> recv #0
 20:47:45 lte,packet,raw 07000080 74000000 00000000 01000000
 20:47:45 lte,packet,raw 00000000 a289cc33 bcbb8b4f b6b0133e
 20:47:45 lte,packet,raw c2aae6df 09000000 48000000 00000000
 20:47:45 lte,packet,raw 03000000 01000000 60000000 01000000
 20:47:45 lte,packet,raw 34000000 0a000000 40000000 08000000
 20:47:45 lte,packet,raw 00000000 00000000 02000000 7c000080
 20:47:45 lte,packet,raw 32003600 30003000 31000000 50006c00
 20:47:45 lte,packet,raw 75007300

t¢Ì3¼»O¶°>Âªæß H`4
@|26001Plus

and the router is spamming with this informations - are those informations exposing some kind of problem with login using this APN?

gmsmstr · Wed Oct 19, 2022 6:58 pm

Went back to 7.5 no drops.. 7.6 drops.

waaaaski · Wed Oct 19, 2022 7:19 pm

Interesting thing - when I switch from (A0, A1, A4, A6) pigtails to (A0, A1, A5, A6) then this help me - now I have stable connection to the n41 (as CA):

Not sure if I get this correctly, but probably the router tried to connect to BTS using A0 and A1 for LTE B1, B3 + 5G n41? and this created some kind of race condition on modem?
and after when I splitted connection for n41 on port A5 and A6 could help? (I don't have any technical knowledge background with telecommunications, so I could be wrong)
On doc (from modem) I see that there is some bold information for n41 one pair on A0+A1 and second pair A5+A6 (so this suggests that we could link with n41 on A5+A6 - and that's why I changed pigtails):

Is there any way to check which bands are using antenna ports?

EDIT:
After restart of router it started again dropping connections (for n41) - but sometimes it is able to catch this for a long while and works stable.
On the end - still it's better than on previous setup of pigtails.

tlahitte · Fri Feb 10, 2023 7:36 pm

waaaaski have you found a solution by any chance ? I'm having exactly the same issue on band 78 using EE in UK. I can get internet running a 200mbs for hours and then it's dropping to LTE and not coming back. I have a chateau 5G ax with the antenna modification connected to a QUWireless 4x4 Mimo... I'm a bit desperate

Interesting thing - when I switch from (A0, A1, A4, A6) pigtails to (A0, A1, A5, A6) then this help me - now I have stable connection to the n41 (as CA):

Not sure if I get this correctly, but probably the router tried to connect to BTS using A0 and A1 for LTE B1, B3 + 5G n41? and this created some kind of race condition on modem?
and after when I splitted connection for n41 on port A5 and A6 could help? (I don't have any technical knowledge background with telecommunications, so I could be wrong)
On doc (from modem) I see that there is some bold information for n41 one pair on A0+A1 and second pair A5+A6 (so this suggests that we could link with n41 on A5+A6 - and that's why I changed pigtails):

Is there any way to check which bands are using antenna ports?

EDIT:
After restart of router it started again dropping connections (for n41) - but sometimes it is able to catch this for a long while and works stable.
On the end - still it's better than on previous setup of pigtails.

tlahitte · Mon Feb 13, 2023 1:09 am

I've been able to fix my issue by setting up my APN parameters manually. A simple Google search with the name of my provider followed by APN parameters gave me everything I needed. Connection is now stable, hope it could help others.

https://blog.linitx.com/mikrotik-lte-ho ... pn-issues/

Problem with operator's 5G NSA

Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA

Re: Problem with operator's 5G NSA