Line Problems?

I�ve been having problems with my Talktalk Business ADSL2+ (LLU) connection, and would appreciate comments on the BQM trace here.

http://www.thinkbroadband.com/ping/share/d5bd865a06d...

Laptop just left on to log Routerstats data from 6pm to 10am, otherwise the rises in max latency coincide with internet activity.

Background:
Back in May I was having increasing problems with loss of connection and download speeds seeming to get slower with time. I also felt that at around 2Mb/s, it was not close to the 3.5 � 4 Mb/s that line speed estimators were coming up with. TTB support ran a line check and said there was �voltage on the line�, and after that was sorted I negotiated SNR down from 12 to 6dB.

(TTB advise that the line length is 3000m, which sounds as though the actual number has been rounded off, but is consistent with the point to point distance of 1645m to the exchange (WWPENZ)).

Since then I�ve been tracking performance with Routerstats, which has shown repeated re-syncs (17 yesterday), which explains the losses of connection. SNR also varies frequently, dropping to 3 dB or lower, but also increasing from time to time. (If some kind person can advise how to post pictures, I�ll put a typical day�s traces up � one for the FAQs?)

Again contacted TTB, who again said there was �voltage on the line�. When I chased them up, they said the line was now OK. They could arrange a callout, but charges would be made if they found a problem on my side.

Obviously I want to check everything out to eliminate any potential culprits before I commit to that. I�ve done all the usual checks � swapped filters, verified via the master test socket. Nothing untoward so far, and as usual, I can�t find any significant difference between the test socket and the extension I normally use; all connections via Ethernet, no wireless. The only other thing I can think of is to swap the router with one at another property we�ve got in the vicinity.

DSL Status - 10:50 Thursday 11th July 2013
DSL Status: Up
DSL Modulation Mode: ADSL2+
DSL Path Mode: INTERLEAVED
Downstream Rate: 2435 Kbps
Upstream Rate: 840 Kbps
Downstream Margin: 6 db
Upstream Margin: 6 db
Downstream Line Attenuation: 53
Upstream Line Attenuation: 35
Downstream Transmit Power: 0
Upstream Transmit Power: 0
PVC Connection
Encapsulation: RFC 2364 PPPoA
Multiplexing: VC
QoS: UBR
PCR Rate: 0
SCR Rate: 0
Autodetect: Disable
VPI: 0
VCI: 38
Enable: Yes
PVC Status: Applied

Setup: Router - Linksys WAG354G (v1), Dell Inspiron N5110, W7 Pro (64bit), Norton Internet Security, Buffalo LS-LG NAS, WD My Book Essential USB2

(If some kind person can advise how to post pictures, I�ll put a typical day�s traces up � one for the FAQs?)

You'll need to provide a link in your next post from an image hosting site > G**gle is your friend?

Thanks Alastair, but it'll have to wait until morning now...

In reply to a post by Mygri:

I�ve been having problems with my Talktalk Business ADSL2+ (LLU) connection, and would appreciate comments on the BQM trace here.

http://www.thinkbroadband.com/ping/share/d5bd865a06d...

Laptop just left on to log Routerstats data from 6pm to 10am, otherwise the rises in max latency coincide with internet activity.

Background:
Back in May I was having increasing problems with loss of connection and download speeds seeming to get slower with time. I also felt that at around 2Mb/s, it was not close to the 3.5 � 4 Mb/s that line speed estimators were coming up with. TTB support ran a line check and said there was �voltage on the line�, and after that was sorted I negotiated SNR down from 12 to 6dB.

(TTB advise that the line length is 3000m, which sounds as though the actual number has been rounded off, but is consistent with the point to point distance of 1645m to the exchange (WWPENZ)).

Since then I�ve been tracking performance with Routerstats, which has shown repeated re-syncs (17 yesterday), which explains the losses of connection. SNR also varies frequently, dropping to 3 dB or lower, but also increasing from time to time. (If some kind person can advise how to post pictures, I�ll put a typical day�s traces up � one for the FAQs?)

Again contacted TTB, who again said there was �voltage on the line�. When I chased them up, they said the line was now OK. They could arrange a callout, but charges would be made if they found a problem on my side.

Obviously I want to check everything out to eliminate any potential culprits before I commit to that. I�ve done all the usual checks � swapped filters, verified via the master test socket. Nothing untoward so far, and as usual, I can�t find any significant difference between the test socket and the extension I normally use; all connections via Ethernet, no wireless. The only other thing I can think of is to swap the router with one at another property we�ve got in the vicinity.

DSL Status - 10:50 Thursday 11th July 2013
DSL Status: Up
DSL Modulation Mode: ADSL2+
DSL Path Mode: INTERLEAVED
Downstream Rate: 2435 Kbps
Upstream Rate: 840 Kbps
Downstream Margin: 6 db
Upstream Margin: 6 db
Downstream Line Attenuation: 53
Upstream Line Attenuation: 35
Downstream Transmit Power: 0
Upstream Transmit Power: 0
PVC Connection
Encapsulation: RFC 2364 PPPoA
Multiplexing: VC
QoS: UBR
PCR Rate: 0
SCR Rate: 0
Autodetect: Disable
VPI: 0
VCI: 38
Enable: Yes
PVC Status: Applied

Setup: Router - Linksys WAG354G (v1), Dell Inspiron N5110, W7 Pro (64bit), Norton Internet Security, Buffalo LS-LG NAS, WD My Book Essential USB2

Update: Here's the link to two Routerstats traces, one for each of the last two days:

https://picasaweb.google.com/mygriffiths/Broadband?a...

The 10th Jly has been the worst for re-synce to date, the 11th was the first time I've seen a change in downstream attenuation. It seems to me that this is a line problem, but I'd greatly appreciate your views and confirmation, or otherwise.

Have you done the usual tests of disconnecting all the phones and running the broadband from the master socket test socket.

A quick peek at the router stats show you are connecting at a slower than expected speed for the downstream attenuation, but the balance between upstream and downstream attenuation does look ok too.

In reply to a post by MrSaffron:

Have you done the usual tests of disconnecting all the phones and running the broadband from the master socket test socket.

I have done checks via the master test socket, but must admit that I did plug the phone into the splitter as well - so that's one thing at least to recheck!

In reply to a post by MrSaffron:

A quick peek at the router stats show you are connecting at a slower than expected speed for the downstream attenuation, but the balance between upstream and downstream attenuation does look ok too.

Hmm... - I'm trying to address two aspects, one the speed, which is why I requested the change of SNR. If all else fails I can just about live with 2M downloads, but it is the 2nd aspect, the re-syncs and the associated interruptions that really bug me. If I can find a cure those I'll be a lot happier... I've also got a suspicion that the two are probably not unrelated.

Can you confirm that when using the test socket that telephones don't work in any other sockets you have?

In reply to a post by MrSaffron:

Can you confirm that when using the test socket that telephones don't work in any other sockets you have?

I thought you might be on to something there, but I checked it out as soon as I got back, and it all appears to be wired up correctly - certainly no phone connection via the extension sockets when the master socket cover is off. I took a picture of the cover to show the configuration - https://picasaweb.google.com/lh/photo/IRhCP7RcFi7miR...

UPDATE: new router and results

I've now upgraded to a Netgear DGND 4000 N750 and typical results using RouterStats are as per this link:

https://picasaweb.google.com/mygriffiths/Broadband?a...

Line speed at present is -
Download 1999 KBits/s (249.875 KBytes/s)
Upload 726 KBits/s (90.75 KBytes/s)

The most obvious occurences are literally dozens of drop-outs per day which I didn't see with the old WAG354G, with the N750 nonetheless maintaining the link and resuming at the same sync rate. Also, the upstream SNR (Tx) is extremely variable, and I've no idea if this is normal. (I've got to say I'm impressed with the N750, it seems to be providing a remarkably stable connection on what to me looks like a pretty rubbish line.)

I think I've now covered all the checks, and to my mind this is all pointing to an inadequate line; I'd be grateful for any comments to support my conclusions, or otherwise.

Background: decided to order the new router to rule that out as a possible source of problems. (Might seem like overkill, but there are other considerations behind the choice, which I won't burden you with.)

After some initial frustrations getting set up, the Routerstats traces looked similar to those with the (aged) Linksys, although I started off with it (N750) set to Auto rather than ADLS2+. The main difference was no change of sync when drop-outs occurred. I then changed the RouterStats sample rate to 5 sec, (previously 20 sec), to match the router, and a short while after, reset to ADSL2+. The increased sampling rates showed up all the drop-outs which I'd not previously observed, but then I'm pretty sure the Linksys only sample at around 30 sec ( I haven't had a chance to check so far).

And if you run the modem from the test socket, i.e. when the faceplate has been removed do things improve at all?

If the modem really is resyncing that many times in a day the standard ISP response is to change your profile to raise target noise margin and sacrifice some speed for stability.

There is no legal definition of an inadequate line for broadband, only that you are getting below what one would expect, and I suspect that your extension wiring may be a flat, rather than twisted core or if twisted it is not using a pair, thus picking up more noise than is ideal.

Running from the test socket, faceplate removed, makes no difference that I can discern. I even ran it with a flexible extension lead into the master socket, just a splitter but no telephone connected; this reduced the sync speed and increased the attenuation as one would expect, so although I thought that the fixed extension lines were flat, as you supposed, they are actually performing very well, so are probably twisted.

Not sure where this leaves me. You seem to suggest that noise is being picked up, which makes me wonder from where? While I can get the ISP to raise the target SNR, I should be getting a better speed - as you correctly point out. I guess the only option now is to bite the bullet and book an engineer(!) visit.

Edited by Mygri (Tue 13-Aug-13 11:53:13)

Noise that affects ADSL comes from many sources, some of which are:

AM radio masts
Plasma TV's
Misbehaving switched mode power supplies, e.g. mobile phone chargers
Florescent lights
i.e. anything that can cause noise that you would hear on an AM radio.

If the phone line is crystal clear when all ADSL kit is unplugged, then likely an engineer visit will produce little result, and you'll probably find the ISP just say the speed is reasonable.

Hmmm...

Within the house, I can't identify any obvious sources of interference. I have accessed the master socket, again, and plugged a splitter and just the phone into it. If I call the number from my mobile and wind the volume on the landline handset right up, I can just hear a low level white noise static. This is the best I can do in the circumstances as I can't get hold of a Talktalk Business quiet line number. (The BT number doesn't work.)

I note your comment about the ISP saying the speed is reasonable, although they did originally quote a speed of 5.7 MB, which is an awful lot better than I've ever had. That was back in October 2010.

So I'm just left to weigh up the likely success of an engineer(!) callout. Ho hum....

Btw, thank you for your prompt responses; despite my gloomy tone, you have helped me considerably.

In reply to a post by Mygri:

Within the house, I can't identify any obvious sources of interference. I have accessed the master socket, again, and plugged a splitter and just the phone into it.

Try it without the splitter, as that is filtering things out from the phone side. Also use the test socket behind the master.

Try it without the splitter, as that is filtering things out from the phone side. Also use the test socket behind the master.

Hello James, firstly, I have unwittingly mislead you and all those who have read and responded to this thread: when I said master socket, I should have said test socket, ie the one exposed with the faceplate off. My apologies for any confusion this has caused.

Secondly, thanks for your suggestion - I'm currently running from the test socket without the splitter, and I'll post the results a bit later on.

Cheers

Secondly, thanks for your suggestion - I'm currently running from the test socket without the splitter, and I'll post the results a bit later on.

Got some results, but no time to collate and post now - will update in the morning...

In reply to a post by Mygri:

Secondly, thanks for your suggestion - I'm currently running from the test socket without the splitter, and I'll post the results a bit later on.

The phone on its own in the test socket, and do the quiet line test.

You can't use the internet like this

The phone on its own in the test socket, and do the quiet line test.

Yessss.... I should have realised - Dohhhhh...

You can't use the internet like this

I can confirm this...

I've now done the phone test, using the test socket, and I can hear a distinct hum although it is at a low level. Any thoughts?