Print

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

[gab81]

hi guys,

After doing some posts here on the forum (and thanks to everyone's patience), it's clear that there is a major issue on my FTTC line, in specific it's noisy and the SNR is low for what it "should be".

I am posting this to see if anyone has an idea if there's a quick fix i could try to improve it or otherwise if you know where the problem might be.

The quickest recap i can give: i have 40/10 w/ sky, on install it was perfect, 39996 / 9996. after 4.5 days "something" happened (but i didnt have stats and graph back then), since then (55 days) i am synced @ 35000 down, up varied a bit and now stable 9996, ping also varied during the weeks and now stable at a great 5ms.

On install day engineer put his oscilloscope in, then couldn't find my line at the cabinet, asked me access to another cupboard but i didnt have the keys, then went back, put new batteries in the oscilloscope (it died eventually), back at the cabinet, found the line and up 39996 / 9996.

Let me do a quick digression. Few years ago in Italy i had an ADSL2+ line, one engineer connected me, it worked BUT not very high speed; i called the telco company, they sent another guy and the guy within 5 minutes found out i was connected on the wrong port at the building...changed it, connection speed improved, happy days.

Could it be a similar scenario?

Thanks
Gab

[Black Sheep]

No ........... if you were on any port other than the correct one, you would get synch, yes, but you wouldn't ever get a valid 'PPP Session'. In other words, you wouldn't ever get on the Web.

I left wondering about this 'Other cupboard' you mention ?? I'm assuming your line runs through it ?? Could there be a junction box in there with possible star-wiring connected, or maybe there's another socket in there that the VDSL runs through before it gets to what we think is the master socket ?

All scenario's I have come across, whereby the ill-informed either assume a socket to be the master, without checking the actual cable run physically, or they are told carte-blanche by the EU that it is the master socket ....... and subsequent investigations on one particular circuit found another 3 sockets that the EU didn't even know he had.

Is it possible to take photo's of your installation from where it enters the building through to the modem ??

[Bald_Eagle1]

After doing some posts here on the forum (and thanks to everyone's patience), it's clear that there is a major issue on my FTTC line, in specific it's noisy and the SNR is low for what it "should be".

What makes this even worse is that you are led to believe your line is only around 150m in length.

SNR & SNRM should really be very high as you are only using a 40/10 service.

I THINK we would see any evidence of bridged taps/star wiring in your Hlog graph.

[gab81]

No ........... if you were on any port other than the correct one, you would get synch, yes, but you wouldn't ever get a valid 'PPP Session'. In other words, you wouldn't ever get on the Web.

I left wondering about this 'Other cupboard' you mention ?? I'm assuming your line runs through it ?? Could there be a junction box in there with possible star-wiring connected, or maybe there's another socket in there that the VDSL runs through before it gets to what we think is the master socket ?

All scenario's I have come across, whereby the ill-informed either assume a socket to be the master, without checking the actual cable run physically, or they are told carte-blanche by the EU that it is the master socket ....... and subsequent investigations on one particular circuit found another 3 sockets that the EU didn't even know he had.

Is it possible to take photo's of your installation from where it enters the building through to the modem ??

hi

that wouldn't be much help as i cannot see the cabling at all, fibre cab is 150 meters away, then i have a classic green cab like 5 meters from street and in my building there is an electrical cupboard downstairs (for all flats here which i dont have access to), then im on first floor, i have just ONE telephone socket (is a small studio flat), so that's the main and the only one i have where my modem + router are connected, no extensions, everything is packed here actually.

in these flats the management installed their own network for tenants (but i don't use it bc is slow) and there are wall mounted Ethernet ports, in mine there are 2.... so MAAAAYBE the existing setup they did for the flats is somehow interfering with the telephone line itself, just my wild guess.

gab

[gab81]

After doing some posts here on the forum (and thanks to everyone's patience), it's clear that there is a major issue on my FTTC line, in specific it's noisy and the SNR is low for what it "should be".

What makes this even worse is that you are led to believe your line is only around 150m in length.

SNR & SNRM should really be very high as you are only using a 40/10 service.

I THINK we would see any evidence of bridged taps/star wiring in your Hlog graph.

only thing i could do right now is to verify the installation of the vdsl2 plate here that the BT engineer installed, in case he screwed some cabling there....any way i can test that?

[burakkucat]

only thing i could do right now is to verify the installation of the vdsl2 plate here that the BT engineer installed, in case he screwed some cabling there....any way i can test that?

Assuming that the SSFP (Service Specific Face Plate) was fitted to a pre-existing NTE5/A by the Engineer at service installation time, there is not really anything to go wrong. However, you can certainly check because it is your side of the demarcation point, the NTE5/A.

Just remove the two screws in the lower front faceplate and unplug it. Now you should be able to unplug the SSFP from the NTE5/A. From your earlier description, there should not be any wires attached to the IDCs (Insulation Displacing Connectors) on either the lower front face plate or the SSFP.

If you do discover any wires connected then the wiring is not as simple as you believe it to be. In that case, I would ask that you take a series of pictures of the wiring and make them available for viewing -- either attach them to a message in this thread or post them to some other public site.

[gab81]

only thing i could do right now is to verify the installation of the vdsl2 plate here that the BT engineer installed, in case he screwed some cabling there....any way i can test that?

Assuming that the SSFP (Service Specific Face Plate) was fitted to a pre-existing NTE5/A by the Engineer at service installation time, there is not really anything to go wrong. However, you can certainly check because it is your side of the demarcation point, the NTE5/A.

Just remove the two screws in the lower front faceplate and unplug it. Now you should be able to unplug the SSFP from the NTE5/A. From your earlier description, there should not be any wires attached to the IDCs (Insulation Displacing Connectors) on either the lower front face plate or the SSFP.

If you do discover any wires connected then the wiring is not as simple as you believe it to be. In that case, I would ask that you take a series of pictures of the wiring and make them available for viewing -- either attach them to a message in this thread or post them to some other public site.

Cheers for info, i will think about it if i feel adventurous :-)

on another note, i had a look at modem_stats.log and i noticed that SNR margin is going up:

24/07/2013 10:40 35000 9995 8.5 6.6 12.0 -6.6 46708 10953
11/08/2013 10:14 35000 9995 8.8 7.1 11.9 -7.2 47640 11477

values are growing sloooooooowly but seems steady and so is attainable speed. a good sign that things are improving perhaps? or is "normal" for fluctuations? it seems a steady improvement tho, goes up like 0.1 every 2-3 days.

[burakkucat]

Looking at the two sets of values, I would be inclined to regard them as showing normal fluctuation. The SNRM and attainable (sync) speed can vary, minute by minute. That is why a graphical plot, over a period of a few days, is so useful as it then becomes easy to see the overall trend of any change.

That said, small cumulative improvements all add up to something significant!