In an article entitled "When good connections go bad" in the June 2007
issue of Australian PC World, the author writes:

"Unfiltered connected phones can disrupt your ADSL signal and, if you
lose your ADSL connection when the phone rings, there's a rogue,
microfilterless phone somewhere in the house."

The above claim makes absolutely no sense to me. Firstly, in a typical
Australian installation, the ADSL modem's connection to the phone line
is unfiltered. Hence the modem is designed to tolerate the 90VAC 20Hz
ring signal. Secondly, the ADSL filter is a low pass filter, ie it is
designed to pass voice frequencies but not ADSL frequencies. Hence an
unfiltered phone presents the same load as a filtered phone to any
voice frequency signal, including a ring tone.

Am I missing something, or is the author in error?

- Franc Zabkar
-- 
Please remove one 'i' from my address when replying by email.

Hi Franc,

> In an article entitled "When good connections go bad" in the June 2007
> issue of Australian PC World, the author writes:
> 
> "Unfiltered connected phones can disrupt your ADSL signal and, if you
> lose your ADSL connection when the phone rings, there's a rogue,
> microfilterless phone somewhere in the house."
> 
> The above claim makes absolutely no sense to me. Firstly, in a typical
> Australian installation, the ADSL modem's connection to the phone line
> is unfiltered. Hence the modem is designed to tolerate the 90VAC 20Hz
> ring signal. Secondly, the ADSL filter is a low pass filter, ie it is
> designed to pass voice frequencies but not ADSL frequencies. Hence an
> unfiltered phone presents the same load as a filtered phone to any
> voice frequency signal, including a ring tone.

Just a wild guess here, but some RF telephone devices, in addition to 
the normal Voice Frequency line line traffic, they have also been 
known to generate VHF signaling that can compete with ADSL signals on 
the copper, and so can compete with ADSL if they are using an 
unfiltered connection. So effectively the Filter works both ways, 
protecting the Voice device from VHF signalling and protecting the VHF
ADSL side from injected "noise" from the Phone.

Does this make sense?

Cheers.............pk.

-- 
Peter from Auckland.

Franc Zabkar <fzabkar@iinternode.on.net> wrote in part:
> In an article entitled "When good connections go bad" in the
> June 2007 issue of Australian PC World, the author writes:

> "Unfiltered connected phones can disrupt your ADSL signal and,
> if you lose your ADSL connection when the phone rings, there's
> a rogue, microfilterless phone somewhere in the house."

> The above claim makes absolutely no sense to me. Firstly,
> in a typical Australian installation, the ADSL modem's
> connection to the phone line is unfiltered. Hence the modem
> is designed to tolerate the 90VAC 20Hz ring signal. 

Undoubtedly.  This is also true in N.America


> Secondly, the ADSL filter is a low pass filter, ie it is designed
> to pass voice frequencies but not ADSL frequencies. Hence an
> unfiltered phone presents the same load as a filtered phone to
> any voice frequency signal, including a ring tone.

Some phone may have a tiny low ESR cap across the incoming
line.  That cap is too small to dampen LF voice, but will
soak up some HF DSL.  The DSL filter isolates the line.

-- Robert

fzabkar@iinternode.on.net (Franc Zabkar) wrote in
news:e3ht33pd3puqocbo6l9m2ql3hsjoa5fnfm@4ax.com: 

> The above claim makes absolutely no sense to me. Firstly, in a
> typical Australian installation, the ADSL modem's connection to the
> phone line is unfiltered. Hence the modem is designed to tolerate
> the 90VAC 20Hz ring signal.

Perhaps the modem's HP filter is internal, to avoid the possibility of
the end user forgetting to install it.

-- 
Bert Hyman | St. Paul, MN | bert@iphouse.com

On 07 May 2007 14:14:20 GMT, Bert Hyman <bert@iphouse.com> put finger
to keyboard and composed:

>fzabkar@iinternode.on.net (Franc Zabkar) wrote in
>news:e3ht33pd3puqocbo6l9m2ql3hsjoa5fnfm@4ax.com: 
>
>> The above claim makes absolutely no sense to me. Firstly, in a
>> typical Australian installation, the ADSL modem's connection to the
>> phone line is unfiltered. Hence the modem is designed to tolerate
>> the 90VAC 20Hz ring signal.
>
>Perhaps the modem's HP filter is internal,...

Exactly.

>... to avoid the possibility of the end user forgetting to install it.

In Australia there is nothing to forget. AFAIK, we don't use central
splitters unless we have a back-to-base alarm system. Our telcos
supply self-installable ADSL kits which typically include one modem
and two phone filters.

- Franc Zabkar
-- 
Please remove one 'i' from my address when replying by email.

On 7 May 2007 23:18:06 +1300, "Peter" <SOMEONE@orcon.net.nz> put
finger to keyboard and composed:

>Hi Franc,
>
>> In an article entitled "When good connections go bad" in the June 2007
>> issue of Australian PC World, the author writes:
>> 
>> "Unfiltered connected phones can disrupt your ADSL signal and, if you
>> lose your ADSL connection when the phone rings, there's a rogue,
>> microfilterless phone somewhere in the house."
>> 
>> The above claim makes absolutely no sense to me. Firstly, in a typical
>> Australian installation, the ADSL modem's connection to the phone line
>> is unfiltered. Hence the modem is designed to tolerate the 90VAC 20Hz
>> ring signal. Secondly, the ADSL filter is a low pass filter, ie it is
>> designed to pass voice frequencies but not ADSL frequencies. Hence an
>> unfiltered phone presents the same load as a filtered phone to any
>> voice frequency signal, including a ring tone.
>
>Just a wild guess here, but some RF telephone devices, in addition to 
>the normal Voice Frequency line line traffic, they have also been 
>known to generate VHF signaling that can compete with ADSL signals on 
>the copper, and so can compete with ADSL if they are using an 
>unfiltered connection. So effectively the Filter works both ways, 
>protecting the Voice device from VHF signalling and protecting the VHF
>ADSL side from injected "noise" from the Phone.
>
>Does this make sense?
>
>Cheers.............pk.

Why would a phone which is designed to operate on a phone line with a
~4kHz bandwidth generate VHF?

- Franc Zabkar
-- 
Please remove one 'i' from my address when replying by email.

In article <pkdv33ttsk80vh8m7otb63ccbjb2vnkv19@4ax.com>, 
fzabkar@iinternode.on.net says...
> On 7 May 2007 23:18:06 +1300, "Peter" <SOMEONE@orcon.net.nz> put
> finger to keyboard and composed:
> 
> >Hi Franc,
> >
> >> In an article entitled "When good connections go bad" in the June 2007
> >> issue of Australian PC World, the author writes:
> >> 
> >> "Unfiltered connected phones can disrupt your ADSL signal and, if you
> >> lose your ADSL connection when the phone rings, there's a rogue,
> >> microfilterless phone somewhere in the house."
> >> 
> >> The above claim makes absolutely no sense to me. Firstly, in a typical
> >> Australian installation, the ADSL modem's connection to the phone line
> >> is unfiltered. Hence the modem is designed to tolerate the 90VAC 20Hz
> >> ring signal. Secondly, the ADSL filter is a low pass filter, ie it is
> >> designed to pass voice frequencies but not ADSL frequencies. Hence an
> >> unfiltered phone presents the same load as a filtered phone to any
> >> voice frequency signal, including a ring tone.
> >
> >Just a wild guess here, but some RF telephone devices, in addition to 
> >the normal Voice Frequency line line traffic, they have also been 
> >known to generate VHF signaling that can compete with ADSL signals on 
> >the copper, and so can compete with ADSL if they are using an 
> >unfiltered connection. So effectively the Filter works both ways, 
> >protecting the Voice device from VHF signalling and protecting the VHF
> >ADSL side from injected "noise" from the Phone.
> >
> >Does this make sense?
> >
> >Cheers.............pk.
> 
> Why would a phone which is designed to operate on a phone line with a
> ~4kHz bandwidth generate VHF?
> 

I think the theory proposed is it's a cordless phone, and generates RF
to talk to the handset, and the filtering isn't perfect so some of the
RF leaks into the phone wires (or alternatively, the phone wires pass
very close to the RF transmitter, and act as an antenna, picking up some
of the RF.)

This theory seems reasonable and possible to me, but I'm not an expert.
(IANAEE ??)

> - Franc Zabkar
> 

-- 
John

On Mon, 07 May 2007 14:02:31 GMT, Robert Redelmeier
<redelm@ev1.net.invalid> put finger to keyboard and composed:

>Franc Zabkar <fzabkar@iinternode.on.net> wrote in part:
>> In an article entitled "When good connections go bad" in the
>> June 2007 issue of Australian PC World, the author writes:
>
>> "Unfiltered connected phones can disrupt your ADSL signal and,
>> if you lose your ADSL connection when the phone rings, there's
>> a rogue, microfilterless phone somewhere in the house."
>
>> The above claim makes absolutely no sense to me. Firstly,
>> in a typical Australian installation, the ADSL modem's
>> connection to the phone line is unfiltered. Hence the modem
>> is designed to tolerate the 90VAC 20Hz ring signal. 
>
>Undoubtedly.  This is also true in N.America
>
>
>> Secondly, the ADSL filter is a low pass filter, ie it is designed
>> to pass voice frequencies but not ADSL frequencies. Hence an
>> unfiltered phone presents the same load as a filtered phone to
>> any voice frequency signal, including a ring tone.
>
>Some phone may have a tiny low ESR cap across the incoming
>line.  That cap is too small to dampen LF voice, but will
>soak up some HF DSL.  The DSL filter isolates the line.
>
>-- Robert

I've only ever taken apart about a dozen phones (but lots of dialup
modems), but IIRC the only caps I've seen were ceramic RF noise
suppression caps of the order of tens or hundreds of pF, between each
of the tip and ring terminals and circuit ground. Are these the caps
you are referring to?

This is a photo of the innards of my ADSL modem:
http://www.users.on.net/~fzabkar/DSL-302G/Photos/LAN_USB_DAA.jpg

It bears some rough similarity to a dialup modem's DAA in that it also
has the abovementioned caps (C118/C120). RV1 is connected across the
line and appears to be a gas arrestor. It is normally open circuit.
Are you perhaps mistaking RV1 as your across-the-line cap?

- Franc Zabkar
-- 
Please remove one 'i' from my address when replying by email.

On Wed, 09 May 2007 01:09:26 GMT, John Santos
<john.santos@post.harvard.edu> put finger to keyboard and composed:

>In article <pkdv33ttsk80vh8m7otb63ccbjb2vnkv19@4ax.com>, 
>fzabkar@iinternode.on.net says...
>> On 7 May 2007 23:18:06 +1300, "Peter" <SOMEONE@orcon.net.nz> put
>> finger to keyboard and composed:
>> 
>> >Hi Franc,
>> >
>> >> In an article entitled "When good connections go bad" in the June 2007
>> >> issue of Australian PC World, the author writes:
>> >> 
>> >> "Unfiltered connected phones can disrupt your ADSL signal and, if you
>> >> lose your ADSL connection when the phone rings, there's a rogue,
>> >> microfilterless phone somewhere in the house."
>> >> 
>> >> The above claim makes absolutely no sense to me. Firstly, in a typical
>> >> Australian installation, the ADSL modem's connection to the phone line
>> >> is unfiltered. Hence the modem is designed to tolerate the 90VAC 20Hz
>> >> ring signal. Secondly, the ADSL filter is a low pass filter, ie it is
>> >> designed to pass voice frequencies but not ADSL frequencies. Hence an
>> >> unfiltered phone presents the same load as a filtered phone to any
>> >> voice frequency signal, including a ring tone.
>> >
>> >Just a wild guess here, but some RF telephone devices, in addition to 
>> >the normal Voice Frequency line line traffic, they have also been 
>> >known to generate VHF signaling that can compete with ADSL signals on 
>> >the copper, and so can compete with ADSL if they are using an 
>> >unfiltered connection. So effectively the Filter works both ways, 
>> >protecting the Voice device from VHF signalling and protecting the VHF
>> >ADSL side from injected "noise" from the Phone.
>> >
>> >Does this make sense?
>> >
>> >Cheers.............pk.
>> 
>> Why would a phone which is designed to operate on a phone line with a
>> ~4kHz bandwidth generate VHF?
>> 
>
>I think the theory proposed is it's a cordless phone, and generates RF
>to talk to the handset, and the filtering isn't perfect so some of the
>RF leaks into the phone wires (or alternatively, the phone wires pass
>very close to the RF transmitter, and act as an antenna, picking up some
>of the RF.)
>
>This theory seems reasonable and possible to me, but I'm not an expert.
>(IANAEE ??)

Sounds reasonable to me.

FWIW, I used to have two phone lines, one for a dialup modem and one
for voice. I once tested my setup by disabling my modem's error
correction and connecting to my ISP via a comms app (HyperTerminal). I
then rang my voice line. During the ringing, the crosstalk between the
two wire pairs was serious enough to flood the terminal window with
dozens of spurious characters.

- Franc Zabkar
-- 
Please remove one 'i' from my address when replying by email.

Franc Zabkar <fzabkar@iinternode.on.net> wrote in part:
> has the abovementioned caps (C118/C120). RV1 is connected
> across the line and appears to be a gas arrestor. It is
> normally open circuit.  Are you perhaps mistaking RV1 as
> your across-the-line cap?

No.  Tip & Ring caps to ground are functionally equivalent
to caps in series across.  What they soak up is dependant
on their size and ESR.  An ADSL modem cannot tolerate much,
but a phone could.  And some would.

-- Robert

Franc Zabkar wrote:
> FWIW, I used to have two phone lines, one for a dialup modem and one
> for voice. I once tested my setup by disabling my modem's error
> correction and connecting to my ISP via a comms app (HyperTerminal). I
> then rang my voice line. During the ringing, the crosstalk between the
> two wire pairs was serious enough to flood the terminal window with
> dozens of spurious characters.

The telephone company uses twisted pairs with hundreds of pairs in a bundle 
in the aerial and buried plant without any problems at all with millivolt 
level voice signals running directly adjacent to 90 volt AC ringing.

Some possibilities to explain the cross-talk on your installation are:

1) You home has been wired with the old JK inside wire (known as "Jake" to 
telephone installers) which is untwisted pairs.
2) You are using a lot of 20 or 40 ft flat silver satin non-twisted station 
cords around your place.
3) Grunge in you wall outlets.
4) You might have low quality multi-line KSUless phones.

DTC <no_spam@move_along_folks.foob> wrote in part:
> Franc Zabkar wrote:
>> FWIW, I used to have two phone lines, one for a dialup modem and one
>> for voice. I once tested my setup by disabling my modem's error
>> correction and connecting to my ISP via a comms app (HyperTerminal). I
>> then rang my voice line. During the ringing, the crosstalk between the
>> two wire pairs was serious enough to flood the terminal window with
>> dozens of spurious characters.
> 
> The telephone company uses twisted pairs with hundreds of
> pairs in a bundle in the aerial and buried plant without
> any problems at all with millivolt level voice signals
> running directly adjacent to 90 volt AC ringing.

Fully agreed.  5-15 thousand feet of tight bundle run to the
typical phone.  The pairs are carefully twisted and allocated..
Crosstalk happens when pairs are split.


> Some possibilities to explain the cross-talk on your installation are:
> 
> 1) You home has been wired with the old JK inside wire (known
> as "Jake" to telephone installers) which is untwisted pairs.

Sure.  But quad isn't so horrible for crosstalk if the pairs
are allocated on the diagonal.


> 2) You are using a lot of 20 or 40 ft flat silver satin
> non-twisted station cords around your place.

Flat silver satin isn't great for crosstalk, and very much
worse if the pairs are not nested USOC.

> 3) Grunge in you wall outlets.

Unlikely.  But poor (hi resistance) connections might generate
noise with ringing current.

> 4) You might have low quality multi-line KSUless phones.

Quite possible.  Unplug to check.


-- Robert