Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[sevenlayermuddle]

Scanning this thread quickly one small point of detail that I'd question is 'attenuation should never change'.

I realise the problem seems to have been fixed by a new router but in terms of stitching up the loose ends, I can't help put my pedant's hat on and mention... any conductor's resistance is a function of temperature and it may be no coincidence that you saw slightly raised line attenuation over the past week or so.  The effect is measured on the absolute (kelvin) scale, in which 2 degree centigrade represents only less than 1 percent;  nevertheless the fluctuations over the past week or two have been more like 15 degrees, so I suspect it may well become measurable.

Although the effects of climatic conditions on line resistance are small, they do exist.   I seem to recall hearing that power distribution companies sometimes have to make allowance for it in HT wires.  My own router has been showing 57dB this past few days, compared with it's usual 55/56, but my connection is in the best of health in every other respect.  I may be wrong, but I've been attributing the slight change to the hot weather.

RE the faulty router (if we have confirmed it is faulty),  I'd bet a pound it a penny you'd find a bulging capacitor or two inside.  These components are also affected by temperature and, if on their last legs, the demise can be accelerated by higher temperatures.

Take care not to trip.  Reminds me I have my own tripping challenge  today, to work out why the mains RCD mysteriously tripped out last night . 

[kitz]

>>> I realise the problem seems to have been fixed by a new router but in terms of stitching up the loose ends,

No its not the router.
Perhaps I confused by saying yesterday afternoon that the line was running rock steady, but that was whilst the TG582n was still on.
Last week it ran rock steady for a few days still using the TG582n and the only reason I was putting a different router on was for the sake of elimination and rule out that it wasnt the TG582n/PSU.

Im now using a ST585, which I put on last night, and although its holding on to the line a bit better you can see there are still issues there.

Some time just before 4am Upstream attenuation changed  from 7dB to 9dB.
I'll attach the graph from Routerstats showing the dip in SNRM at just before 4am, and also periods this morning between 7am to 8.30am where you can see upstream SNRM is dropping down to zero. (You'll need to scroll image to the right)

These slight & brief bursts I also saw using the TG582n, and I think my previous posts may have said that aside from some overnight blips then it mostly ran rock steady for a few days... until it kicked off again this weekend.


>>>  you saw slightly raised line attenuation over the past week or so

Upstream attenuation has been fluctuating between 4.5dB and 10.5dB.   
Ive not seen anything like that before.. not on such a relatively short line, that has never before shown any changes.

>>> Take care not to trip.

lol..  I may be about to make matters worse.   

Yesterday I installed a wireless card in this PC to enable me to plot with routerstats with routerstats/dslstats whilst connected to the master socket.   However, the wireless signal from the ST585 isnt too good and my throughput speed is carp.   Thinking of temp trailing a long length of cat 5 cable out of here and over the banister, just so I can plot with RS (dslstats wont work on the ST585 ) whilst everything is in the NTE test socket.

[sevenlayermuddle]

No its not the router.

Sorry, my mistake.  I'd not been following the thread, read it in a hurry, and clearly misunderstood

All the same (clutching at straws to justify my contribution), I'd be open-minded to the possibility that whatever has changed(broken), the warmer weather may have been a factor in breaking it....  not that it helps a great deal, of course.

Thinking of temp trailing a long length of cat 5 cable out of here and over the banister, just so I can plot with RS (dslstats wont work on the ST585 ) whilst everything is in the NTE test socket.

Indeed, that seems to add a whole new dimension of tripping hazard.  Do take care!   

[kitz]

No worries, yes warmer weather may have something to do with slight changes in attenuation, and I welcome any suggestions or pointing out something I may have missed.

However not allowing for the fact that dB is a logrithmic figure... more than doubling of the attenuation is a heck of a lot.
Also as a very rough guide, upstream attenuation should be in the region of about 1/2 of the downstream.

I realise this post has got quite long now so I'll summarise

--------------

Symptoms

Previously rock steady short line with unwavering SNRM - it would sit quite happily at 3dB for months on end.

About 2 weeks ago started seeing changes in upstream attenuation and burst fluctuations in SNRM which causes the line to drop.
No obvious pattern to the times, although they mostly occur overnight, morning or weekends. 
I dont think Ive once seen it happen in the evenings which is when you'd most likely see changes in SNRm
Logging SNRM shows its burst type noise - length can vary between a few seconds... to a periods lasting a couple of hours.
Noise bursts will often, but not always, follow or accompany a change in upstream atten.
This post shows a log of how much the upstream attenuation can vary over a fairly short time span fluctuating between 4.5dB and 10.5dB

Tests performed:-

Change of adsl filter
Change of router
Test from master socket (direct into test socket)
QLT is ok
Using the phone doesnt trigger an event.

Test not performed

AM radio
1) - I dont have a portable am radio
2) - Times are too sporadic to know when the problem is going to occur.

Other possible areas which could cause this type of fault

Line fault
Faulty line card at the MSAN
At first I thought it may be REIN/EMI when I started seeing the drops in SNRM... but unsure re the changes in attenuation


At this point Id normally suggest raising a fault with the ISP, but problem first occurred a couple of days after Id decided to move to FTTC
Migration and transfer to FTTC due to be completed this Friday.   I hardly think Be* are going to investigate after I asked for my MAC, but Im very aware that if its a line issue, then its not at all a good start for FTTC where its harder to monitor whats going on

[kitz]

Just noticed whilst I was typing out all the above, Ive had another noise burst lasting about 10mins. 

Admittedly the ST585v6 seems to be holding on to the line better when SNRm drops to 0dB.
Whilst the TG582n will drop the line when it gets to negative figures, I hardly think its a fault as such & I can rule out a hardware issue.

I'll keep the ST585 in the master socket for today, but tbh I dont see much point in keeping trialling wires all over waiting for an accident to happen... and it will be nice to get a decent wireless signal again from the TG582n for the ipad etc... and a proper wired LAN for the PC.

[kitz]

>>> Admittedly the ST585v6 seems to be holding on to the line better when SNRm drops to 0dB

hmmm  Im also beginning to suspect that routerstats is possibly throwing me a few false 0dB readings as the graphs shows quite a few zero dbs this afternoon but none of the surrounding spikeyness which I normally see around the time when my line starts going flaky.  Theres been about half a dozen or so dips this afternoon recorded in RS which Im beginning to suspect is false reports.

Unlike dslstats - afaik RS doesnt advise you when its not been able to get figures from your router.

[kitz]

and I know Im responding to myself but just this minute found out this little gem of information.

---------------

Someone else on the same Be MSAN as me is also seeing remarkably similar symptoms!
This is the response I just got after asking how his line was doing. - 
I hadnt even mentioned to him that I was seeing weird issues with my upstream attenuation

I am glad you asked actually. I don't know whats happened to the line here ... /snip/...

I don't have a clue how the upstream attenuation has changed by that much. Im thinking it's a line fault

Yep his Be connection is also seeing upstream atten changes and looking at his stats hes lost over half of his upstream speed and he is also beginning to wonder if he had a line fault 

Not 100% sure - but there is a strong chance we are both on the same line card, since we were both activated on about the same day as soon as the Be MSAN went live.

[burakkucat]

Ah, now that is interesting. Do you know if he has submitted a fault report?

[jabns]

I have not submitted a fault report yet but I will do tomorrow.

@Kitz I have installed DSLstats and will report back later with some stats. Ill leave it running all this week.

Edit: If you need any stats pulling or any screenshots posting just let me know.

Edit 2: Just reboot my router to try and get a decent sync and here are the results:

                                  Downstream    Upstream
Attenuation (dB):                28.0            17.1
Connection speed (kbps):    16905        1360
SNR margin (dB):                 3.3        3.8
Power (dBm):                     19.2        12.0

If you could post the results I sent you earlier kitz so we can compare, as I did not click save sent message so i can't see it.

[kitz]

Cheers jabns

I think what would be most telling is if you experience burst noise on the SNRm at the same time as me.

btw dslstats was written by our very own Eric. 
Check out its options and theres a logging facility for attenuation which he put in after I asked him nicely.  This may come in very useful to log your up atten too.

Since we dont live in the same estate then I suspect its now looking like it could be something on the Be MSAN.

[kitz]

Edit 2: Just reboot my router to try and get a decent sync and here are the results:

                                  Downstream    Upstream
Attenuation (dB):                28.0            17.1
Connection speed (kbps):    16905        1360
SNR margin (dB):                 3.3        3.8
Power (dBm):                     19.2        12.0


If you could post the results I sent you earlier kitz so we can compare, as I did not click save sent message so i can't see it.

ADSL Link                Downstream               Upstream
Connection Speed         14407 kbps               598 kbps
Line Attenuation           28.5 db                9.6 db
Noise Margin            7.3 db                 3.1 db

[kitz]

Attenuation changed again today, upstream atten now at 5.5 dB.  Still got the ST585 pm as Ive been out all day and not long been back home.

Line dropped at about 12.40ish, lost some of my downstream speed, but as you can see its the upstream that takes the hammering.

line stats:-

Code: [Select]

Uptime: 0 days, 8:00:00

Modulation: G.992.5 Annex M US 56
Bandwidth (Up/Down) [kbps/kbps]: 2,407 / 19,979
Data Transferred (Sent/Received) [MB/MB]: 8.12 / 29.34
Output Power (Up/Down) [dBm]: 13.0 / 17.5
Line Attenuation (Up/Down) [dB]: 5.5 / 12.5
SN Margin (Up/Down) [dB]: 5.0 / 3.5
Vendor ID (Local/Remote): TMMB / µ
Loss of Framing (Local/Remote): 4 / 0
Loss of Signal (Local/Remote): 1 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): 0
Error Seconds (Local/Remote): 144 / 0
FEC Errors (Up/Down): 0 / 0
CRC Errors (Up/Down): 70,700 / 328
HEC Errors (Up/Down): 313,890 / 237
attached graphs from 12:40

[jabns]

I will post the last 48 hours of graphs in the morning. Im currently not a home but don't worry its been logging 

[kitz]

ok...   Ive not been home much either...
For the record these are my stats over the past few days. There are a few other tiny periods, but aside from these Ive been flatlining.

These are the main periods of instability:-

attenuation changes:-

15 Jul 2013 03:16:11   12.5   9.0
16 Jul 2013 12:40:06   12.5   5.5
17 Jul 2013 05:11:54   12.5   7.0



Periods of upstream SNRm fluctuations

18/07/2013   05:47 to 06:30   - Loss of about 4dB SNRm
17/07/2013   11:46 to 12:40   - Fluctuations of 2dB SNRm
17/07/2013   09:55 to 10:45    - Slight fluctuations of SNRm
17/07/2013   04:00 to 05:15   - Period of huge instability, downtime and lost of syncs


Stats just before I unplug the ST585 - Ive lost some sync speed both up and down, but as mentioned Ive not been around nor had chance to monitor things too closely.

Link Information

Uptime:   1 day, 7:56:49
Modulation:   G.992.5 Annex M US 56
Bandwidth (Up/Down) [kbps/kbps]:   2,385 / 19,392
Data Transferred (Sent/Received) [MB/MB]:   50.44 / 142.34
Output Power (Up/Down) [dBm]:   13.0 / 16.5
Line Attenuation (Up/Down) [dB]:   7.0 / 12.5
SN Margin (Up/Down) [dB]:   5.0 / 3.5
Vendor ID (Local/Remote):   TMMB / µ
Loss of Framing (Local/Remote):   8 / 0
Loss of Signal (Local/Remote):   3 / 0
Loss of Power (Local/Remote):   0 / 0
Loss of Link (Remote):   0
Error Seconds (Local/Remote):   380 / 0
FEC Errors (Up/Down):   0 / 0
CRC Errors (Up/Down):   72,717 / 121
HEC Errors (Up/Down):   339,498 / 82

[kitz]

The TG582n has been back on since yesterday afternoon.
Decentish sync speeds although the upstream line attenuation is back up at 10.3 dB

All seems relatively quiet last nite.  Downstream SNRm has complete flat-lined at 3dB.   Upstream is stable at around 3 - 3.5 dB
Possibly the last set of stats from me:-

DSL Type:   ITU-T G.992.5
Bandwidth (Up/Down) [kbps/kbps]:   2.507 / 20.871
Data Transferred (Sent/Received) [MB/MB]:   69,87 / 205,79
Output Power (Up/Down) [dBm]:   13,0 / 18,3
Line Attenuation (Up/Down) [dB]:   10,3 / 13,5
SN Margin (Up/Down) [dB]:   3,2 / 3,0
System Vendor ID (Local/Remote):   TMMB / ----
Chipset Vendor ID (Local/Remote):   BDCM / BDCM
Loss of Framing (Local/Remote):   0 / 0
Loss of Signal (Local/Remote):   0 / 0
Loss of Power (Local/Remote):   0 / 0
Loss of Link (Remote):   -
Error Seconds (Local/Remote):   27 / 0
FEC Errors (Up/Down):   0 / 0
CRC Errors (Up/Down):   290 / 28
HEC Errors (Up/Down):   71 / 68