cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Highlighted
Recognised Expert
1,162 Views
Message 1 of 5

Infinity 2 Noise Margin and sync speed

Hi,

 

I guess my "problem" is nothing compared to some I read here, but thought I'd throw it in for discussion anyway.

 

In summary, from the Helpdesk page of my Home Hub 5A I see a maximum downstream data rate of approximately 54 Mbps, but it never seems to sync at anything above approximately 33 Mbps. The upstream is pretty much spot on.

 

Pertinent helpdesk page info as follows:

1. Product name: BT Home Hub
2. Serial number: +068543+NQ51009784
3. Firmware version: Software version 4.7.5.1.83.8.222.1.1 (Type A) Last updated 21/02/16
4. Board version: BT Hub 5A
5. DSL uptime: 2 days, 01:42:19
6. Data rate: 15605 / 33478
7. Maximum data rate: 15620 / 53095
8. Noise margin: 5.5 / 12.0
9. Line attenuation: 30.4 / 20.9
10. Signal attenuation: 30.3 / 20.3


So for some more detail.

 

My Infinity 2 service always gave me a data rate of approximately 33 Mbps downstream and 10-12 Mbps upstream. Aside from usual minor fluctuations, upstream and downstream noise margin were both always around the 6dB level.

 

Early January there was an issue where the upstream data rate changed to ❤️ Mbps and it stayed like that for nearly a week. Over the next few days I monitored it and did the usual throughput tests etc., and eventually called BT. After explaining the issue they did their remote diagnostics, had me power off/on the Home Hub etc., and then arranged for a boost engineer to come.

 

At this time I started logging the details from the "helpdesk" page of the HH with the RouterStatsHub program. A few hours after I reported the problem the service went back to normal (see RouterStatsHub log below), but I decided to let the engineer come anyway.

 

                          SNR (dB)  Sync          Max Data Rate
Date/Time                 Up   Down Up     Down   Up     Down
---------------------------------------------------------------------------
Sat 09 Jan 2016 21:59:03, 6.3, 6.6, 10098, 31150, 12254, 34111
Sat 09 Jan 2016 21:59:33, 6.2, 6.6, 10098, 31150, 12251, 34096
Sat 09 Jan 2016 22:00:03, 6.3, 6.6, 10098, 31150, 12251, 34115


When he turned up the following Friday the service was still running OK so I wasn't entirely sure what he could do. After discussion, and some confusion as he said he wasn't a boost engineer but was there for a BT Vision fault that had been fixed earlier that week, he said he'd change the socket. At the time I had a Type 4 faceplate with the Broadband Accelerator/I-plate which he changed to a Type 5 SSFP Infinity faceplate.

 

When I looked at the stats after we reconnected the HH I immediately saw some significant difference, specifically the downstream noise margin had gone to ~12dB and the maximum data rate had gone up to nearly 54 Mbps.

 

                          SNR (dB)  Sync          Max Data Rate
Date/Time                 Up   Down Up     Down   Up Down
---------------------------------------------------------------------------
Fri 15 Jan 2016 10:15:18, 6.0, 8.2, 11349, 33478, 11357, 38529
Fri 15 Jan 2016 10:15:48, 6.0, 8.3, 11349, 33478, 11424, 38569
Fri 15 Jan 2016 10:16:18, 6.0, 8.3, 11349, 33478, 11353, 38537

[snip where no log entries while faceplate replaced]

Fri 15 Jan 2016 10:27:18, 6.0, 12.1, 14530, 33478, 14530, 53928
Fri 15 Jan 2016 10:27:48, 6.3, 12.1, 14530, 33478, 14530, 53924
Fri 15 Jan 2016 10:28:21, 6.3, 12.1, 14530, 33478, 14530, 53884


I thought DLM would come into play so wasn't immediately worried that the line sync was still only 33 Mbps.

 

The service has been reasonably stable since the faceplate was changed. Looking at the WAN log on the HH I saw connections lost or reset on 20-Jan, 21-Jan, 23-Jan, 26-Jan, 04-Feb and 06-Feb, but then not again until 11-Mar. While there were six resets within 17 days (20-Jan to 06-Feb) the events on 23-Jan and 06-Feb were when the system was rebooted by BT. I also think the 11-Mar was also a BT triggered event as there's TR069 activity at the same time.

 

Despite the initial string of connection losses, I'd expect that if DLM were restricting things the stable period between 06-Feb and 11-Mar would have been sufficient for it to change things.

 

So I have hree questions

 

Q1:- Is an upstream noise margin of ~12dB a bad thing?

 

I'm thinking not based on the fact I've seen posts e.g., HELP ME: My BT Infinity 2 Connection Speeds are slowwwwwwwww where data rates of 19999/79999 Mbps are being achieved with noise margins of 15.5/10.6, and the explanations of SNR margin I've seen e.g., SNR margin, line attenuation, ADSL and a slow internet speed indicate high SNRM is a good thing. From the aforementioned:

 

"Some routers instead (or additionally) of the signal-to-noise ratio (SNR) figure are displaying the signal-to-noise margin (SNR Margin) which is the difference between the actual SNR and the SNR required to sync at a specific speed:

actual SNR = 44dB
SNR to sync at 8Mb = 35dB
SNR MARGIN = 44-35 = 9dB

 

A higher number is better and the numbers bellow [sic] 6dB may cause problems."

 


Q2:- Is the downstream data rate not running nearer to the maximum date rate of ~52 Mbps because the IP profile is limiting it?

 

I checked the IP profile over at Mouselike.org UK Broadband Tools and it's indicating the following:

 

Number: 01825******
The current Downstream BRAS rate is: 32.4 Mbps
The current Upstream BRAS rate is: 20 Mbps


Q3:- What should I do next?

 

At the moment any speed test I run tells me the service is fine, but I don't think it's as good as it could be, and hence looking for confirmation from those with far more knowledge of this than me. If I report this am I likely to get anywhere, and is there any advice to help the process?

 

Thanks and regards

Steve

0 Ratings
4 REPLIES 4
Highlighted
Distinguished Sage
Distinguished Sage
1,152 Views
Message 2 of 5

Re: Infinity 2 Noise Margin and sync speed

a  it's your downstream margin from your stats that is about 12db whereas upstream is about normal 6db

 

b  the ip profile is set by the conenction speed not tyhe other way round

 

c  from what you ahve posted all the previous resets/drops in conenction have resulted in a banded profile.  this takes time often weeks to release fully and all you can do is maintain a stable connection and DLM should start to reduce youe noise margin towards 6db with corresponding increases in data rate getting closer to attainable

 

this gives more info on DLM  http://community.plus.net/library/browsing/fttc-dlm-what-it-is-how-it-works/



If you like a post, or want to say thanks for a helpful answer, please click on the Ratings 'Thumbs up' on left hand side.
If someone answers your question correctly please let other members know by clicking on ’Mark as Accepted Solution’.
Highlighted
Recognised Expert
1,144 Views
Message 3 of 5

Re: Infinity 2 Noise Margin and sync speed

Hi imjolly,

 

Thanks for the prompt response, and noting the typo for Q1.

 

For Q1, typo aside, is a noise margin of 12dB a bad thing?

 

For Q3, as I pointed out the line was up and stable for five weeks from 06-Feb until 11-Mar. Is that not sufficient time for DLM to sort itself out?

 

The example in the link you provided talks of being placed on "barrier" 8 or 24. It then goes on to say that DLM would change the profile after 9 or 25 days, depending on which of those barriers the line is on. Either way, with a stable line of five weeks shouldn't DLM have sorted things out?

 

Thanks and regards

Steve

0 Ratings
Highlighted
Distinguished Sage
Distinguished Sage
1,127 Views
Message 4 of 5

Re: Infinity 2 Noise Margin and sync speed

the high noise margin of 12db is the reason your data down speed is not nearer your attainable speed.  so as DLM reduces your noise margin down to normal 6db your data speed will get nearer your attainable speed

 

your current conenction time is only 2 days - it is internet connection time not length of time hub has been powered up/system time



If you like a post, or want to say thanks for a helpful answer, please click on the Ratings 'Thumbs up' on left hand side.
If someone answers your question correctly please let other members know by clicking on ’Mark as Accepted Solution’.
0 Ratings
Highlighted
Recognised Expert
1,111 Views
Message 5 of 5

Re: Infinity 2 Noise Margin and sync speed

Hi,

 

Thanks again for the response.

 

I understand it's the length of time the line has been connected that's important. The current connection time is only two days as we had a power failure this week, but the period I'm questioning is between 06-Feb and 11-Mar.

 

From the WAN log of my HH we can see the line disconnected on 11-Mar and at that time the hub logged PPPoE down after 48359 minutes, or 33.5 days.

 

07:04:30, 11 Mar. (2901819.730000) WAN operating mode is VDSL
07:04:30, 11 Mar. (2901819.730000) Last WAN operating mode was VDSL
07:04:28, 11 Mar. (2901817.740000) PPP IPCP Receive Configuration ACK
07:04:28, 11 Mar. (2901817.730000) PPP IPCP Send Configuration Request
07:04:28, 11 Mar. (2901817.730000) PPP IPCP Receive Configuration NAK
07:04:28, 11 Mar. (2901817.730000) PPP IPCP Send Configuration ACK
07:04:28, 11 Mar. (2901817.730000) PPP IPCP Receive Configuration Request
07:04:28, 11 Mar. (2901817.720000) PPP IPCP Send Configuration Request
07:04:26, 11 Mar. (2901815.900000) PPPoE is up -​ Down Rate=33478000Kbps, Up Rate=14896000Kbps; SNR Margin Down=12.8dB, Up=6.2dB
07:04:26, 11 Mar. (2901815.890000) CHAP authentication successful
07:04:26, 11 Mar. (2901815.740000) CHAP Receive Challenge
07:04:26, 11 Mar. (2901815.740000) Starting CHAP authentication with peer
07:04:26, 11 Mar. (2901815.730000) PPP LCP Receive Configuration ACK
07:04:26, 11 Mar. (2901815.730000) PPP LCP Send Configuration Request
07:04:26, 11 Mar. (2901815.730000) PPP LCP Receive Configuration Reject
07:04:26, 11 Mar. (2901815.730000) PPP LCP Send Configuration ACK
07:04:26, 11 Mar. (2901815.730000) PPP LCP Receive Configuration Request
07:04:26, 11 Mar. (2901815.730000) PPP LCP Send Configuration Request
07:03:34, 11 Mar. (2901763.330000) PTM over DSL is up
07:02:53, 11 Mar. (2901722.400000) PTM over DSL is down after 48360 minutes uptime
07:02:53, 11 Mar. (2901722.400000) PPPoE is down after 48359 minutes uptime [Waiting for Underlying Connection (WAN Ethernet 2 -​ Down)]
07:02:47, 11 Mar. (2901716.770000) PPP LCP Send Termination Request [User request]


The question I have is, if the sync speed is not getting closer to the maximum data rate due to DLM, why was five weeks of stability not sufficient time for DLM to react and adjust things?

 

During this period the downstream noise margin was always around the 12dB level. The following crude graph shows the downstream noise margin as measured by RouterStatsHub.

snrm.png

 

As can be seen, the noise margin drops for a very short period on 08-Feb, after which it goes back up to approximately 12dB.

 

Are you saying that five weeks is not long enough for DLM to do it's job?

 

Regards

0 Ratings