Diagnosing ADSL problem with house wiring

I was trying to diagnose problems with a newly-activated broadband line for
a customer yesterday. Can I check that I've made the correct deductions
about where the problem lies?

BT Voyager 220 router (1 ethernet, 1 USB, no wireless)

Router in master test socket (extension wiring disconnected) works fine:

2.5-3 Mbps downstream
6 dB margin
63 dB attenuation


Router in other sockets (either the faceplate of the master socket or a
hard-wired extension from the back of the master socket) - sometimes works
fine but sometimes fails to sync. Remains in one state for a hour or so - it
consistently works or consistently fails. When it works I get

1 Mbps downstream
5.3 dB margin
63 dB attenuation


Given that attenuation is the same but the speed is much less (or the router
fails to sync at all), it looks as if it's the house wiring that's picking
up some noise that is causing the router to negotiate a lower speed in order
to achieve roughly the same noise margin.

I tested with the router in the front of the master socket and nothing in
any of the extension sockets, then gradually plugged in filtered phones, but
the statistics didn't change - so it looks like wiring rather than duff
equipment plugged into the wiring.

What sort of things could cause interference to ADSL such that it
consistently fails for an hour or so or else consistently works for an hour
or so. As far as we could establish, no electrical equipment was being used
during the dead time that wasn't also being used when the router could sync.

Who are the best people to investigate house wiring problems? Without the
right tools for re-crimping wires, I'm reluctant to start pulling out wires
from the back of the faceplate myself, especially as there is very little
spare wire - there's hardly room to swing a cat when I undo the faceplate.

Remove any unused but terminated cables. Replace the wiring , go from socket
to socket and ensure all daisy changed sockets have good connections


There is info on the DSL Zone site to help you with wiring


Cub

"Mortimer" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
>I was trying to diagnose problems with a newly-activated broadband line for
>a customer yesterday. Can I check that I've made the correct deductions
>about where the problem lies?
>
> BT Voyager 220 router (1 ethernet, 1 USB, no wireless)
>
> Router in master test socket (extension wiring disconnected) works fine:
>
> 2.5-3 Mbps downstream
> 6 dB margin
> 63 dB attenuation
>
>
> Router in other sockets (either the faceplate of the master socket or a
> hard-wired extension from the back of the master socket) - sometimes works
> fine but sometimes fails to sync. Remains in one state for a hour or so -
> it consistently works or consistently fails. When it works I get
>
> 1 Mbps downstream
> 5.3 dB margin
> 63 dB attenuation
>
>
> Given that attenuation is the same but the speed is much less (or the
> router fails to sync at all), it looks as if it's the house wiring that's
> picking up some noise that is causing the router to negotiate a lower
> speed in order to achieve roughly the same noise margin.
>
> I tested with the router in the front of the master socket and nothing in
> any of the extension sockets, then gradually plugged in filtered phones,
> but the statistics didn't change - so it looks like wiring rather than
> duff equipment plugged into the wiring.
>
> What sort of things could cause interference to ADSL such that it
> consistently fails for an hour or so or else consistently works for an
> hour or so. As far as we could establish, no electrical equipment was
> being used during the dead time that wasn't also being used when the
> router could sync.
>
> Who are the best people to investigate house wiring problems? Without the
> right tools for re-crimping wires, I'm reluctant to start pulling out
> wires from the back of the faceplate myself, especially as there is very
> little spare wire - there's hardly room to swing a cat when I undo the
> faceplate.
>

Mortimer wrote:

> I was trying to diagnose problems with a newly-activated broadband line for
> a customer yesterday. Can I check that I've made the correct deductions
> about where the problem lies?
>
> BT Voyager 220 router (1 ethernet, 1 USB, no wireless)
>
> Router in master test socket (extension wiring disconnected) works fine:
>
> 2.5-3 Mbps downstream
> 6 dB margin
> 63 dB attenuation

You're doing well there with 63dB of attenuation !


> Router in other sockets (either the faceplate of the master socket or a
> hard-wired extension from the back of the master socket) - sometimes works
> fine but sometimes fails to sync. Remains in one state for a hour or so - it
> consistently works or consistently fails. When it works I get
>
> 1 Mbps downstream
> 5.3 dB margin
> 63 dB attenuation
>
> Given that attenuation is the same but the speed is much less (or the router
> fails to sync at all), it looks as if it's the house wiring that's picking
> up some noise that is causing the router to negotiate a lower speed in order
> to achieve roughly the same noise margin.

It's almost certainly the imbalance caused by the ring circuit degrading the
ADSL signal actually. Been there done it myself.

Since the extension wiring isn't BT's, just remove the ring circuit. Most
(effectively all) phones don't need it any more anyway and a microfilter (which
you ARE using on the extensions aren't you ?) regenerates the ring signal
anyway.

Graham

On 10-Oct-2007, Eeyore <(E-Mail Removed)> wrote:

> You're doing well there with 63dB of attenuation !

Ideally use one ADSL filter to feed all the phones.
A premium quality one stabilised my line, plugged it into
the NTE faceplate, the ADSL modem into its RJ11, and
plugged the existing cheapo filter into the BT jack on
the quality filter, and the phones into that.
Effectively putting two phone filters in series.
Phones, extra ringer, CLID unit, and voice quality,
all ok, with stable BB speed.

Long high attenuation lines are problematical,
the weak received HF signals for your BB can suffer
crosstalk from neighbouring BB pairs, same
applies to the exchange end, your signal is weak
and competing against neighbouring BB pairs.
It's called NEXT (Near End Cross Talk) and
quality cables have a lower NEXT figure.
So line noise can vary considerably over time and
if sync is lost it may take time to re-establish
itself.
All you can do is use a modem that works
well with long high attenuation lines, then fit
the best filter(s) you can find, whilst
simplifying the extension phone wiring as
much as possible. Too much stuff on
the phone side is detremental. Unhooking
a solenoid trembler extension ringer helped a lot.

Eeyore wrote:
>
> Mortimer wrote:
>
>
>>I was trying to diagnose problems with a newly-activated broadband line for
>>a customer yesterday. Can I check that I've made the correct deductions
>>about where the problem lies?
>>
>>BT Voyager 220 router (1 ethernet, 1 USB, no wireless)
>>
>>Router in master test socket (extension wiring disconnected) works fine:
>>
>>2.5-3 Mbps downstream
>>6 dB margin
>>63 dB attenuation
>
>
> You're doing well there with 63dB of attenuation !
>
>
>
>>Router in other sockets (either the faceplate of the master socket or a
>>hard-wired extension from the back of the master socket) - sometimes works
>>fine but sometimes fails to sync. Remains in one state for a hour or so - it
>>consistently works or consistently fails. When it works I get
>>
>>1 Mbps downstream
>>5.3 dB margin
>>63 dB attenuation
>>
>>Given that attenuation is the same but the speed is much less (or the router
>>fails to sync at all), it looks as if it's the house wiring that's picking
>>up some noise that is causing the router to negotiate a lower speed in order
>>to achieve roughly the same noise margin.
>
>
> It's almost certainly the imbalance caused by the ring circuit degrading the
> ADSL signal actually. Been there done it myself.
>
> Since the extension wiring isn't BT's, just remove the ring circuit. Most
> (effectively all) phones don't need it any more anyway and a microfilter (which
> you ARE using on the extensions aren't you ?) regenerates the ring signal
> anyway.
>
> Graham
>
Note that some (cheaper) filters don't have the 'ring capacitor'
necessary if you remove the third wire.They are often a false economy
(he says from experience). Worth paying the £10 for proper ones from
Maplin or similar.

Is the house wiring proper paired cable (blue and blue/white pairs etc)
or is it old Blue/green/Brown/Orange stuff?
Or even worse alarm cable.

Neither of these are proper 'twisted pairs' and can severely upset things.

As others have posted, the filtered faceplate on the master socket cures
most ills. If you need to extend to a modem a bit further away, it is
often better to still install the faceplate and then purchase a long
RJ11/RJ11 cable for the ADSL data (As always however don't be conned
into PC W****'s highly expensive versions

Mike

Mike J wrote:

> Eeyore wrote:
> > Mortimer wrote:
> >
> >>I was trying to diagnose problems with a newly-activated broadband line for
> >>a customer yesterday. Can I check that I've made the correct deductions
> >>about where the problem lies?
> >>
> >>BT Voyager 220 router (1 ethernet, 1 USB, no wireless)
> >>
> >>Router in master test socket (extension wiring disconnected) works fine:
> >>
> >>2.5-3 Mbps downstream
> >>6 dB margin
> >>63 dB attenuation
> >
> > You're doing well there with 63dB of attenuation !
> >
> >
> >>Router in other sockets (either the faceplate of the master socket or a
> >>hard-wired extension from the back of the master socket) - sometimes works
> >>fine but sometimes fails to sync. Remains in one state for a hour or so - it
> >>consistently works or consistently fails. When it works I get
> >>
> >>1 Mbps downstream
> >>5.3 dB margin
> >>63 dB attenuation
> >>
> >>Given that attenuation is the same but the speed is much less (or the router
> >>fails to sync at all), it looks as if it's the house wiring that's picking
> >>up some noise that is causing the router to negotiate a lower speed in order
> >>to achieve roughly the same noise margin.
> >
> >
> > It's almost certainly the imbalance caused by the ring circuit degrading the
> > ADSL signal actually. Been there done it myself.
> >
> > Since the extension wiring isn't BT's, just remove the ring circuit. Most
> > (effectively all) phones don't need it any more anyway and a microfilter (which
> > you ARE using on the extensions aren't you ?) regenerates the ring signal
> > anyway.
>
>
> Note that some (cheaper) filters don't have the 'ring capacitor'
> necessary if you remove the third wire.

The cheap ones I've seen do absolutely have the ring capacitor, but inany event
you'll be hard pressed to find aphone that hasn't already got its own !


> They are often a false economy
> (he says from experience). Worth paying the £10 for proper ones from
> Maplin or similar.

Nonsense.
http://www.maplin.co.uk/Module.aspx?...LTER&doy=10m10

IS a 'cheap' filter. Can be had for £2 or less on ebay.

http://cgi.ebay.co.uk/3x-BT-ADSL-Bro...QQcmdZViewItem

is basically the same thing.

Graham

"Mortimer" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
>I was trying to diagnose problems with a newly-activated broadband line for
>a customer yesterday. Can I check that I've made the correct deductions
>about where the problem lies?
>
> BT Voyager 220 router (1 ethernet, 1 USB, no wireless)
>
> Router in master test socket (extension wiring disconnected) works fine:
>
> 2.5-3 Mbps downstream
> 6 dB margin
> 63 dB attenuation
>
>
> Router in other sockets (either the faceplate of the master socket or a
> hard-wired extension from the back of the master socket) - sometimes works
> fine but sometimes fails to sync. Remains in one state for a hour or so -
> it consistently works or consistently fails. When it works I get
>
> 1 Mbps downstream
> 5.3 dB margin
> 63 dB attenuation
>
>
> Given that attenuation is the same but the speed is much less (or the
> router fails to sync at all), it looks as if it's the house wiring that's
> picking up some noise that is causing the router to negotiate a lower
> speed in order to achieve roughly the same noise margin.
>
> I tested with the router in the front of the master socket and nothing in
> any of the extension sockets, then gradually plugged in filtered phones,
> but the statistics didn't change - so it looks like wiring rather than
> duff equipment plugged into the wiring.
>
> What sort of things could cause interference to ADSL such that it
> consistently fails for an hour or so or else consistently works for an
> hour or so. As far as we could establish, no electrical equipment was
> being used during the dead time that wasn't also being used when the
> router could sync.
>
> Who are the best people to investigate house wiring problems? Without the
> right tools for re-crimping wires, I'm reluctant to start pulling out
> wires from the back of the faceplate myself, especially as there is very
> little spare wire - there's hardly room to swing a cat when I undo the
> faceplate.

My line has 60db attenuation. The only way I can get a stable ADSL
connection is to use a router connected to my master socket, and to connect
my household phone wiring via a microfilter. (One day I may get a fancy
filtered face plate.) This arrangement is not uncommon for people with high
loss lines.

Prior to this arrangement my line was most likely to fail at about 9pm
which is the time when most people are using ADSL.


--
Michael Chare

Eeyore wrote:

>
>
>>They are often a false economy
>>(he says from experience). Worth paying the £10 for proper ones from
>>Maplin or similar.
>
>
> Nonsense.
> http://www.maplin.co.uk/Module.aspx?...LTER&doy=10m10
>
> IS a 'cheap' filter. Can be had for £2 or less on ebay.
>
> http://cgi.ebay.co.uk/3x-BT-ADSL-Bro...QQcmdZViewItem
>
> is basically the same thing.
>
> Graham
>

The cheap ones from CPC definitely don't have the capacitor!!

Mike

Mike J wrote:

> Eeyore wrote:
> >
> >>They are often a false economy
> >>(he says from experience). Worth paying the £10 for proper ones from
> >>Maplin or similar.
> >
> > Nonsense.
> > http://www.maplin.co.uk/Module.aspx?...LTER&doy=10m10
> >
> > IS a 'cheap' filter. Can be had for £2 or less on ebay.
> >
> > http://cgi.ebay.co.uk/3x-BT-ADSL-Bro...QQcmdZViewItem
> >
> > is basically the same thing.
>
>
> The cheap ones from CPC definitely don't have the capacitor!!

That really is scraping the bottom of the barrel for cheapnes, but as I explained is unlikely to ever cause a problem.

Graham

Mike J wrote:
> Eeyore wrote:
>
>>
>>
>>> They are often a false economy
>>> (he says from experience). Worth paying the £10 for proper ones from
>>> Maplin or similar.
>>
>>
>> Nonsense.
>> http://www.maplin.co.uk/Module.aspx?...LTER&doy=10m10
>>
>> IS a 'cheap' filter. Can be had for £2 or less on ebay.
>>
>> http://cgi.ebay.co.uk/3x-BT-ADSL-Bro...QQcmdZViewItem
>>
>> is basically the same thing.
>>
>> Graham
>>
>
> The cheap ones from CPC definitely don't have the capacitor!!

This topic was discussed ad nauseum some months ago. Graham just can't seem
to grasp the fact the the more expensive designs provide a better solution.
As well as providing better filtration the design does not degrade the ADSL
side of things as the cheap models can sometimes do. On a high loss line it
can make the difference between ADSL working or not. On a better line there
is likely to be little difference in most, but not all, cases.


Peter Crosland

(E-Mail Removed)