Sponsored Links

Out of the frying pan in to the fire??

It might be worth trying disconnecting the ring wire on your exisiting extension (if the BT engineer hasn't already done so)

see: http://yarwell.blogspot.com/2005/08/adsl-tweaking.html

And a telephone wiring diagram here: http://www.wppltd.demon.co.uk/WPP/Wiring/UK_telephone/uk_telephone.html

It worked for me, the SNR margin went up by about 3-4dB during the day and over 11dB in the evening.

When my modem was reading 23dB, the BT kit was reading 29dB. Either my modem introduces some noise or the measurements aren't that consistent.

Could also be the power supply if it has one.
 
Last edited:
For a reliable ADSL connection at a specific speed, it's not just the noise margin that's important, it's the downstream and upstream attenuation (particularly downstream). For 2M bit operation, the theoretical breakpoint for downstream attenuation is 43dB (the lower the number, the better the situation). Sometimes, however, if the noise margin figure is sufficiently high, you can still get reliable 2M bit operation at a worse attenuation figure than 43dB. For instance, I myself run happily at 47.5dB attenuation, as reported by my modem. Beyond 50dB attenuation, though, you're very unlikely to get reliable 2M bit operation. Incidentally, the figures you get from your modem will only be ballpark ones; they're not meant to be absolutely accurate.

You desperately need to update your modem's firmware and then run its WAN diagnostics, to get the downstream and upstream figures for attenuation and noise margin. Without knowing these, you'll be going nowhere. It might well transpire that your attenuation figure will be too high for reliable operation at 2M bits and that's why you've been getting dropouts. You might therefore have to revert to 1M bit operation instead. You probably already appreciate that the further you're situated from your local exchange, the higher will be the attenuation.

Do follow that yarwell link, as that person's a valuable contributor to the www.adslguide.org.uk website's Technical section. It sounds as if you need to disconnect the Ring wire on your extensions and to generally tidy up your extensions. Run them only as daisy-chained connections, eliminating spurs, as that's the only way you'll optimise the setup for speed. And ensure you use a plug-in filter only where it's truly needed, ie. where a phone is also plugged in.

From personal experience, a Belkin high-quality ADSL extension lead will help (due to the better twisting of the pairs in it) but don't bother with a so-called screened lead, as the degree of screening is actually minimal (it'll only be fully screened if the braid is connected to earth in the modem or the BT socket, which it won't be). EBuyer do those Belkins. Choose carefully, as some have RJ11s both ends and some have RJ11 one end, BT at other (avoid unnecessary plug-in filters).

Plug-in filters vary in their performance as well. Personally, I've found the cheap-and-cheerful ones from Solwise (www.solwise.co.uk), with built-in Ring reconstruction, to be perfectly adequate.
 
neale said:
To be more precise, the Solwise filters to which I've referred are:

http://www.solwise.co.uk/adsl_splitters.htm

They're the 'improved' version, but cost only £1.99 each, incl VAT.

I agree, I had BT branded filters, the ones with a wire attached, they both failed in a couple of months, I replaced them with the cheapo Solwise ones, never had a problem since, which just proves that expensive branded stuff aint always the best... ;)

Oh and to be honest, gold plated connections are rarely better than steel ones, as they are connecting with steel connectors in most cases....if both are gold plated, then there is some benefit.
 
neale said:
For a reliable ADSL connection at a specific speed, it's not just the noise margin that's important, it's the downstream and upstream attenuation (particularly downstream). For 2M bit operation, the theoretical breakpoint for downstream attenuation is 43dB (the lower the number, the better the situation). Sometimes, however, if the noise margin figure is sufficiently high, you can still get reliable 2M bit operation at a worse attenuation figure than 43dB. For instance, I myself run happily at 47.5dB attenuation, as reported by my modem. Beyond 50dB attenuation, though, you're very unlikely to get reliable 2M bit operation. Incidentally, the figures you get from your modem will only be ballpark ones; they're not meant to be absolutely accurate.

From personal experience, a Belkin high-quality ADSL extension lead will help (due to the better twisting of the pairs in it) but don't bother with a so-called screened lead, as the degree of screening is actually minimal (it'll only be fully screened if the braid is connected to earth in the modem or the BT socket, which it won't be). EBuyer do those Belkins. Choose carefully, as some have RJ11s both ends and some have RJ11 one end, BT at other (avoid unnecessary plug-in filters).

The BT guys fitted a filtered faceplate for me and I bought a new lead (I did get a screened one). The figures from my router now:

Operation Data Upstream Downstream
Noise Margin 30 dB 16 dB
Output Power 10.5 dBm 20 dBm
Attenuation 24.5 dB 41 dB

The Downstream noise margin figure varies between about 16 and 23 now. It used to vary between 10 & 12. I don't understand the significance of the upstream and downstream figures. Does the fact that the upstream figures are significantly better mean that most of the Downstream problems are caused by my distance from the exchange?
I'm going to play around a bit more. My router/modem is completely surrounded by other devices and power supplies and the modem cable winds through the whole lot, so there could be some improvements still to be made. Do you think my figures are good enough to attempt a 2M connection rather than my current 1MB?

Sorry for jumping on someone else's thread, but maybe my experiences maybe useful.
 
If you've been consistent in the way that you've presented your various upstream and downstream figures in your above posting, then I'd say you now have very decent figures and, provided the local DSLAM at the exchange is pushing out the required amount of power, you should be able to get 2M bit operation without any problem. Conceivably, you might still get an occasional CRC error or two (interrogate your router if you get a dropout) but that should be okay. That all presupposes that your modem is properly set up and that it has decent sensitivity; modems do vary. Your noise margins are now not too far different to mine, and your downstream attenuation is actually better than mine. In nearly a year of 2M bit usage, I've personally not had a single CRC error, as far as I can tell.

Upstream figures tend to be better because (a) that's the direction that's less busy, and (b) the upstream bitrate is much lower, so that crosstalk and noise from other lines in the underground/overhead telephone cable is less.

The significance of the noise margin is that it gives you some leeway either side of the attenuation figure. Thus, although 43dB is the nominal attenuation upper limit for 2M bits at present, whether you can operate successfully around that figure will depend on your noise margin. BT engineers need to take this into account when allotting a nominal speed.

Philby, take note!
 
Last edited:
Sponsored Links
BT's provs have a good team in now if they see a regrade that is over 43db att they will reject it, its so BT dont have to be trying to fine marginal faults on a line in the future, also the SNR is one of the most important things that will dictate if your line is stable or not anything under 10db on the SNR will cause drop out problems and these will be more often at night and when it gets colder.

I have found that 9 out of 10 routers/modems do display the correct figures for the line stats the best is a draytek 2600 which is no longer sold but IMO still the best helps drop the line noise by 5-7db with the UK2 firmware in some cases.

Simon,

By the looks of it your line stats are out from what i would expect and looks like you are just within the limits for a 2 meg line but the SNR will drop by about 5-6db after a regrade so you may see it get intermitant, remove any internal wiring you dont need from the phone line as that does cause alot of problems with the SNR margin.
 
LeeH said:
BT's provs have a good team in now if they see a regrade that is over 43db att they will reject it, its so BT dont have to be trying to fine marginal faults on a line in the future, also the SNR is one of the most important things that will dictate if your line is stable or not anything under 10db on the SNR will cause drop out problems and these will be more often at night and when it gets colder.

I have found that 9 out of 10 routers/modems do display the correct figures for the line stats the best is a draytek 2600 which is no longer sold but IMO still the best helps drop the line noise by 5-7db with the UK2 firmware in some cases.

Simon,

By the looks of it your line stats are out from what i would expect and looks like you are just within the limits for a 2 meg line but the SNR will drop by about 5-6db after a regrade so you may see it get intermitant, remove any internal wiring you dont need from the phone line as that does cause alot of problems with the SNR margin.


Thank for all the advice. Here's the odd thing I've found now. When I connect my modem to the new filtered socket with my old cheap modem cable (bog-standard 2m cheapo) I get an average noise margin of 23dB. The lead isn't long enough to tidily run my my router without an extension, so I bought a 'better' longer one . WIth my new shielded, gold plated 4m cable I get an average value of about 17dB. Is the difference all down to the length of the cable? I might be tempted to re-arrange all my kit to move the router closer to the socket...
 
simonhancock said:
Thank for all the advice. Here's the odd thing I've found now. When I connect my modem to the new filtered socket with my old cheap modem cable (bog-standard 2m cheapo) I get an average noise margin of 23dB. The lead isn't long enough to tidily run my my router without an extension, so I bought a 'better' longer one . WIth my new shielded, gold plated 4m cable I get an average value of about 17dB. Is the difference all down to the length of the cable? I might be tempted to re-arrange all my kit to move the router closer to the socket...


Yep the cabling will be degrading the SNR and any one using extention cabling from the master socket i would strongly advice them to remove it, does the ATT go down aswell when you remove the new cable even if its only by 1-2db its enough to make BT think twice about refusing to regrade you, what you could do is put the old cabling in when you request the regrade and then go to the new cable after the regrade and see how it is before you start re-aranging things and getting different ethernet cables.

LeeH
 
As a retired electronics engineer with some considerable experience in cable technology, I disagree, LeeH. The length of cable he's using between the master socket and the modem will be a negligible percentage of the total between the modem and the exchange, so should have little effect on the total attenuation and noise margin (unless he's connecting more than just a modem at the end of it). Furthermore, he's now bypassing all the incrementally-wired slave sockets in the house so, if anything, the uniformity of the line connection will be better. There might, however, be a slight impedance mismatch between the line coming from the exchange and the new extension cable, which might be giving a few problems. Other than that, all I can think is that, by using the new extension cable, he might be unnecessarily carrying the Ring wire from the master socket all the way through to the modem. If so, the Ring wire needs removing from the connection inside the master socket. Then again, if the master socket he's got now has independent outlets for phone and ADSL (which is what he appears to be saying) , the ADSL should only come out on just two pins, so there shouldn't really be any issues at all. I think our friend should perhaps note the noise margin figures at different times of the day, to get a better idea of the average noise margin.

Also, I do wonder precisely which sort of 'improved' extension cable he's obtained. A very long cable that has an unearthed braiding might well act as an aerial and make matters worse. I would personally avoid braided twisted-pair extension cables, unless they're earthed correctly, and instead just stick with a UTP ADSL extension cable that has uniform, tightly-twisted pairs.
 
neale said:
As a retired electronics engineer with some considerable experience in cable technology, I disagree, LeeH. The length of cable he's using between the master socket and the modem will be a negligible percentage of the total between the modem and the exchange, so should have little effect on the total attenuation and noise margin (unless he's connecting more than just a modem at the end of it). Furthermore, he's now bypassing all the incrementally-wired slave sockets in the house so, if anything, the uniformity of the line connection will be better. There might, however, be a slight impedance mismatch between the line coming from the exchange and the new extension cable, which might be giving a few problems. Other than that, all I can think is that, by using the new extension cable, he might be unnecessarily carrying the Ring wire from the master socket all the way through to the modem. If so, the Ring wire needs removing from the connection inside the master socket. Then again, if the master socket he's got now has independent outlets for phone and ADSL (which is what he appears to be saying) , the ADSL should only come out on just two pins, so there shouldn't really be any issues at all. I think our friend should perhaps note the noise margin figures at different times of the day, to get a better idea of the average noise margin.

Also, I do wonder precisely which sort of 'improved' extension cable he's obtained. A very long cable that has an unearthed braiding might well act as an aerial and make matters worse. I would personally avoid braided twisted-pair extension cables, unless they're earthed correctly, and instead just stick with a UTP ADSL extension cable that has uniform, tightly-twisted pairs.

Here's the cable I'm using

http://catalog.belkin.com/IWCatProd..._Id=71505&Section.Section_Path=/Cables/Modem/

I now have a filtered faceplate on my master BT socket (with separate ADSL & phone outputs) with no other connections to that master. I only have my one modem attached to the new faceplate via the cable.
Apart from the length of the cables (the new shielded one is 4.6m, the old one is about 2m), the other difference is that the old cable has a little cylindrical device clamped to it (I never understood what these were!)
When I switch cables, the value for attenuation stays at 41dB, but the Noise Margin figure is consistently 16-17dB for the new cable and 23dB for the old one.
 
Sponsored Links
simonhancock said:
When I switch cables, the value for attenuation stays at 41dB, but the Noise Margin figure is consistently 16-17dB for the new cable and 23dB for the old one.

The cabling that comes to your house has a bigger dia which doesnt have as much resistance (may be higher quaility copper aswell but i doubt it with BT) where as cabling in the house is alot thinner and have more resistance therefor over a distance the loss is exponential.

How ever i would of rather had been able to woosh the line for true readings but ADSL doesnt like it as soon as you put it onto house cabling.
 
Last edited:
A few months ago I had an intermittent disconection problem & I wanted to check it wasn't caused by my internal extensions.

So I tried plugging my router into the BT master socket with the extensions disconnected - using 2 old office telephone extensions a rj11 adapter and a long modem cable all plugged together (total length over 6 metres) to make it reach upstairs to my PC (I didn't fancy carting my monitor downstairs).

Although I did clean the plugs and checked the resistance with a meter, it was suggested to me that this would affect the noise & attenuation & that I could hook the pc up to a tv anyway so the router's cable would reach.

Anyway I got identical readings :shrug:

simonhancock said:
Here's the cable I'm using
the other difference is that the old cable has a little cylindrical device clamped to it (I never understood what these were!)

Possibly a ferrite bead to suppress radio interference ?

Could you use longer lan cabling and put the router near the master socket instead.
 
Last edited:
philby21 said:
My average connection time during 5pm and 10.30pm is 4 minutes
I am sure it is a line problem, have read of posts that street lights, or heating systems can cause problems. It will not be a Plusnet network problem, it was either is faulty equipment in the exchange, the line to your house, or your modem.

If you can get a friend to try his equipment from your BT master socket and you still have problems you can be sure its a BT problem which they will resolve free of charge.

For the record my stats are SNR Margin 34.3dB Down 31.0dB Up, line attenuation 15.1dB and 19.0dB, Errored Seconds 6 and 43, Loss of Signal 1 and 1, CRC Errors 0 and 4 in 24 days 2 hours. This is for a 1mbps line.

Conslusion it will not help changing ISP, best approch is to accept the offer and have BT out.
 
Top
Cheap BIG ISPs for 100Mbps+
Community Fibre UK ISP Logo
150Mbps
Gift: None
Virgin Media UK ISP Logo
Virgin Media £22.99
132Mbps
Gift: None
Vodafone UK ISP Logo
Vodafone £24.00 - 26.00
150Mbps
Gift: None
NOW UK ISP Logo
NOW £24.00
100Mbps
Gift: None
Plusnet UK ISP Logo
Plusnet £25.99
145Mbps
Gift: £50 Reward Card
Large Availability | View All
Cheapest ISPs for 100Mbps+
Gigaclear UK ISP Logo
Gigaclear £17.00
200Mbps
Gift: None
Community Fibre UK ISP Logo
150Mbps
Gift: None
Virgin Media UK ISP Logo
Virgin Media £22.99
132Mbps
Gift: None
Hey! Broadband UK ISP Logo
150Mbps
Gift: None
Youfibre UK ISP Logo
Youfibre £23.99
150Mbps
Gift: None
Large Availability | View All
Sponsored Links
The Top 15 Category Tags
  1. FTTP (6026)
  2. BT (3639)
  3. Politics (2721)
  4. Business (2439)
  5. Openreach (2405)
  6. Building Digital UK (2330)
  7. Mobile Broadband (2146)
  8. FTTC (2083)
  9. Statistics (1901)
  10. 4G (1816)
  11. Virgin Media (1764)
  12. Ofcom Regulation (1582)
  13. Fibre Optic (1467)
  14. Wireless Internet (1462)
  15. 5G (1407)
Sponsored

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved - Terms  ,  Privacy and Cookie Policy  ,  Links  ,  Website Rules