Print

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

[burakkucat]

It's not usually connected at the XNTE, although ours is, and i'd assume all our neighbours too...Would this be wired up by Openreach, or the electrician, it's technically on our side of the wiring where it connects (To the right of the yellow plug)

@Daniel -- I would suspect the electrician has just followed the "Builders' Guide" document.

Now having read more about your substandard wiring, I would remove the very first socket and -- if it looks suspect -- the cable between it and the XNTE, then replace with a generic NTE5/A (manufactured by Pressac and available from Clarity or Solwise or Run It Direct, etc) and CW1308 specification cable. At your new internal NTE5/A, fit an NTE2000 -- the SSF, ADSL V1.0 integral Filter (again, Pressac manufacture, available via the same retailers, above). If at some time in the future you would like to have a working telephone extension socket, you would connect the builder's electrician provided wiring to IDCs 2 & 5 of your NTE2000.

Finally, I have replied to your PM. Not much help there. Sorry.

[danieltharris]

Hi guys,
Thanks again to everybody so far for all their help. Hopefully I can repay the favour in some way in the future.

I disconnected the Ring Wire at the External NTE, and at each of the 5 sockets in the house.

The sync speed has improved (3x the number it was before)...As I understand it, it may take 3 days of leaving the router on for us to start seeing an actual speed gain with downloads etc? Is this right? We are with Plus Net if that makes a difference.

Hopefully I haven't gotten too excited too soon, but it's gone from around 270-370 (Average), to around 1,200 (Which it is showing now).

So here's hoping we can atleast get 1 Meg throughput. Sounds silly but i'd be quite happy with that (For now), considering we have been on such a slow speed. With our high Line Attenuation I think 1 Meg would be close to the best we can expect on the current service.

I can't believe that 1 wire may make that much difference. I'm sure there are still improvements to be made to the wiring, such as those mentioned by numerous people including burakkuat and walter.

New stats are below:

Maximum Bandwidth (Up/Down) [kbps/kbps]:   588 / 1,248

Bandwidth (Up/Down) [kbps/kbps]:   443 / 1,272

Output Power (Up/Down) [dBm]:   12.5 / 15.5

Line Attenuation (Up/Down) [dB]:   42.5 / 67.5

SN Margin (Up/Down) [dB]:   11.5 / 6.0

Vendor ID (Local/Remote):   TMMB / IFTN

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):   0

Error Seconds (Local/Remote):   0 / 0

FEC Errors (Up/Down):   0 / 0

CRC Errors (Up/Down):   0 / 6

HEC Errors (Up/Down):   0 / 6

[razpag]

I tend to find around my patch on new-builds, that the 'sparkies' run a cable to a 'common-point' (an inaccessible junction box under the boards), and then tee-out in all directions from this point. As a result, there is no direct cable from the outside 'Master box' to any single socket ergo hefty drops in speed.

My advice (again), is to get us out, we can then test at the outside box, and test again on the internal sockets and make a judgement call about the internal wiring. Anything other than this is guesswork.

[razpag]

Great result (you posted as I was typing). The difference with that one single wire is, it imbalances the circuit and acts like an antennae for high-frequency 'noise' that combats your signal.

There could be lots more improvements to be made, ie- best cable to Cabinet, locating exactly which is the Master Socket, fitting an NTE2000.

See my post above for best results.

[danieltharris]

Great result (you posted as I was typing). The difference with that one single wire is, it imbalances the circuit and acts like an antennae for high-frequency 'noise' that combats your signal.

There could be lots more improvements to be made, ie- best cable to Cabinet, locating exactly which is the Master Socket, fitting an NTE2000.

See my post above for best results.

Thanks, definitely going to be taking all the advice on board and doing everything possible to get the best speed we can.

I *think* from what i've seen I see the route the wire takes, and thankfully seems like it does go direct from XNTE to the socket in the lounge first, but of course this is just an assumption at this point.

I've tried to explain below how I think it is wired, but it's not that easy to follow on a forum I know, especially with my amateur way of explaining it. Luckily we may not be in that situation where it goes to an inaccessible junction box, if it did then i'd assume that each socket would just have 1 "set" of wires coming into it. Looks like in our house they daisychain.

Which I reckon will make it easier to disconnect all but the first one.

XNTE => Lounge Socket (1 b/w in, 1 b/w out - 1 w/b in, 1 w/b out) => Bedroom right above (1 b/w in, 2 b/w out - 1 w/b in, 2 w/b out)

From here it seems the bedroom feeds one of it's "outward sets" of wires along to the hallway, which just has 1 of each cable and is the end of that branch

The other "outward set" of wires then goes to the smallest bedroom, which has 2 sockets. One of the sockets has 2 sets of wires (1 inward, 1 outward), the second socket in that bedroom just has 1 set of wires, and there that branch terminates.

I noticed the sync speed seems really solid now, before I would refresh and the sync speed would change every few seconds. Now every time I look at the DSL page it shows a solid number that doesn't seem to be moving. That's got to help too

I'll update here with any progress, just incase anybody else has problems too.

If this works i'll be getting in touch with all our neighbours just to let them know what I did, of course I don't want to take any responsibility for the advice, incase they end up doing something wrong 

[razpag]

I appreciate you trying to explain how it's wired Dan, but how can you tell ?? Are the wires not chased into the wall (ie- plastered over)

Ideally though, just the one socket working is deffo the way forward.

[danieltharris]

I appreciate you trying to explain how it's wired Dan, but how can you tell ?? Are the wires not chased into the wall (ie- plastered over)

Ideally though, just the one socket working is deffo the way forward.

They are all in the wall, only reason I came to this conclusion was that originally only the first socket was working, all the others needed wiring connecting properly, as I connected the bedroom one the one in the hall and 1 in the small bedroom started working (the one with 2 sets of wires).

The last socket in the small bedroom only worked after I reconnected a loose wire in the socket next to it.

So really it's based on assumption, but i'd like to try and find out for sure at some point.

Because after I re-attached a wire in the bedroom some others started working, I just assumed (hoped) it's all daisy chained.

I'm no pro that's for sure though

[jeffbb]

Hi
quote .As I understand it, it may take 3 days of leaving the router on for us to start seeing an actual speed gain with downloads etc? Is this right? We are with Plus Net if that makes a difference.

Well in a few days your IP profile will increase to 1000Kbps . The time it takes is variable see
IP profile    information

That should also improve your downloads 

Regards Jeff

[razpag]

I appreciate you trying to explain how it's wired Dan, but how can you tell ?? Are the wires not chased into the wall (ie- plastered over)

Ideally though, just the one socket working is deffo the way forward.

They are all in the wall, only reason I came to this conclusion was that originally only the first socket was working, all the others needed wiring connecting properly, as I connected the bedroom one the one in the hall and 1 in the small bedroom started working (the one with 2 sets of wires).

The last socket in the small bedroom only worked after I reconnected a loose wire in the socket next to it.

So really it's based on assumption, but i'd like to try and find out for sure at some point.

Because after I re-attached a wire in the bedroom some others started working, I just assumed (hoped) it's all daisy chained.

I'm no pro that's for sure though

Fair play Dan, it does indeed sound like the lounge is your first socket. What needs to be done now is to make this a proper Master Socket with an NTE5, and then have an NTE2000 SSFP fitted.

This goes gainst every rule laid down regarding xNTE's, as the wiring is wholly your concern, but it's the only way to acieve 'best results'. Maybe you can locally purchase a type of NTE5 and acquire your own SSFP and fit them yourself Just a thought pal.

[waltergmw]

Gentlefolk,

We've had a discussion a long time ago here about the lack of an internal test socket on a XNTE service unless an NTE5 type device is fitted.
Noting that Dan has a poor signal might it be a good idea to remove the NTE5 duplicated components in effect producing a "slave" NTE5 ?
I'm assuming the BT components or similar are fitted in the XNTE.

Kind regards,
Walter

[danieltharris]

Hi guys,

We are now getting a consistent 128 kb/s download speed, and able to stream non-hd videos on youtube without pausing and waiting 20 minutes for a 1 minute video to load

May sound silly but i'm very happy with the improvement. We were getting a very flaky 25-40 kb/s previously, and the DSL Sync speed seemed to change everytime I refreshed the page. Now it's looking solid and fixed.

I think this may be close to the best we can expect on this line and the current service, as our Line Attenuation is 67.5 at the moment, which according to the calculator means around 1 Meg - 1.5 Meg is about what we can expect.

I'll monitor it for about a week then look into implementing some of the other suggestions to possibly further increase the line.

Also investigating FTTC as an option, trying to find out which PCP our property connects to as the next street along seems to be connected one with FTTC as an option. We are in a new area, and perhaps our postcode not being in the databases makes companies think this isn't an option.

[roseway]

Sorry to be a pedant, but I guess you mean 128 kB/s (kilobytes per second). This is consistent with an IP profile of 1000 kb/s (allowing for small errors in measurement) and is a pretty good result on your long line. There may be a bit more to be squeezed out, but for the moment you're doing the right thing in just monitoring it for a bit.

[danieltharris]

Sorry to be a pedant, but I guess you mean 128 kB/s (kilobytes per second). This is consistent with an IP profile of 1000 kb/s (allowing for small errors in measurement) and is a pretty good result on your long line. There may be a bit more to be squeezed out, but for the moment you're doing the right thing in just monitoring it for a bit.

Yep, that's what I meant 

It's annoying being on such a long line, but I guess next time we buy a house we may just have to look beforehand at what's available.

I'm still gob smacked that removing that 'Ring Wire' actually had such a big effect on it.

[burakkucat]

Noting that Dan has a poor signal might it be a good idea to remove the NTE5 duplicated components in effect producing a "slave" NTE5 ?
I'm assuming the BT components or similar are fitted in the XNTE.

If I was in Dan's situation then, for testing purposes, I would make up two springy brass strips, connect them to a spare LJ2/3A and plug the strips into the network side of the yellow plug's receptacle in the XNTE. I would then connect a 2Wire 2700HGV to the LJ2/3A and leave it to synchronise. Once in sync with the DSLAM, I'd document the line characteristics -- thereby establishing base-line data for his pair.

@Walter,

Yes the XNTE does contain identical components to an NTE5/A -- capacitor, resistor and surge suppressor. So being a perfectionist, one would remove the duplicates from the NTE5/A.

[waltergmw]

@ BKK,

I suggest it's not just a matter of being a perfectionist but more of prudent engineering as, in this case, there is a distinct possibility that the A & B legs are swapped inside the house.
Even if this were not the case, I suggest adding unnecessary components to a line with an already poor signal is best avoided.

@ RP Can you please confirm that an XNTE does contain identical components to those in a NTE5/A ?

(Apologies to Paul in hijacking his thread for these comments.)

Kind regards,
Walter