Using analogue phones on a structured wiring system

5 views
Skip to first unread message

Pandora

unread,
Jan 23, 2009, 7:02:46 PM1/23/09
to
I have a BT line which terminates at a BT master socket. I also have a
Cat 5e patch panel, which is connected to various RJ45 sockets around
the property.

I want to convert five of the RJ45 sockets to phone points so I bought
some PSTN line adapters from Comlynx:
http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36

On the patch panel, I intend to wire five spare ports in parallel back
to terminals 2 and 5 in the BT master socket. I can then patch across
to the port that has a phone on it.

My question is this:
The Comlynx adaptors have a capacitor/resistor/surge suppressor in
them. If I use five of them in parallel, won't it mess up the ring
tone?

I know that this is the reason that you can't wire BT master sockets
in parallel. The same would apply to the adapters I would have thought.

John Rumm

unread,
Jan 23, 2009, 9:11:54 PM1/23/09
to
Pandora wrote:

> I have a BT line which terminates at a BT master socket. I also have a
> Cat 5e patch panel, which is connected to various RJ45 sockets around
> the property.
>
> I want to convert five of the RJ45 sockets to phone points so I bought
> some PSTN line adapters from Comlynx:
> http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
>
> On the patch panel, I intend to wire five spare ports in parallel back
> to terminals 2 and 5 in the BT master socket. I can then patch across
> to the port that has a phone on it.
>
> My question is this:
> The Comlynx adaptors have a capacitor/resistor/surge suppressor in
> them. If I use five of them in parallel, won't it mess up the ring
> tone?

Probably not... the capacitor is only shunting the ring line from one of
the exchange lines. Extra capacitance will just *lower* the minimum
switching frequency that can "get through" the cap. Since I doubt you
will be using pulse dial phones, or ones with real bells for ringers, it
should not make any difference.

If you want a more flexible solution, you can pick up a cheap analogue
PABX from the far east via ebay. Then it certainly becomes a non issue.
(I have a 4/16 line/ext PABX waiting to go it when the required round
tuit is acquired and the structured wiring is done)

--
Cheers,

John.

/=================================================================\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\=================================================================/

meow...@care2.com

unread,
Jan 23, 2009, 9:23:54 PM1/23/09
to
John Rumm wrote:
> Pandora wrote:
>
> > I have a BT line which terminates at a BT master socket. I also have a
> > Cat 5e patch panel, which is connected to various RJ45 sockets around
> > the property.
> >
> > I want to convert five of the RJ45 sockets to phone points so I bought
> > some PSTN line adapters from Comlynx:
> > http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
> >
> > On the patch panel, I intend to wire five spare ports in parallel back
> > to terminals 2 and 5 in the BT master socket. I can then patch across
> > to the port that has a phone on it.
> >
> > My question is this:
> > The Comlynx adaptors have a capacitor/resistor/surge suppressor in
> > them. If I use five of them in parallel, won't it mess up the ring
> > tone?
>
> Probably not... the capacitor is only shunting the ring line from one of
> the exchange lines. Extra capacitance will just *lower* the minimum
> switching frequency that can "get through" the cap. Since I doubt you
> will be using pulse dial phones, or ones with real bells for ringers, it
> should not make any difference.

It depends on the value of those parts. The short story is you've got
the wrong items, you just want plug to socket adaptors, not stuff with
circuitry in it. Another option is to have a phone socket on the wall
next to the rj45 skt.


> If you want a more flexible solution, you can pick up a cheap analogue
> PABX from the far east via ebay. Then it certainly becomes a non issue.
> (I have a 4/16 line/ext PABX waiting to go it when the required round
> tuit is acquired and the structured wiring is done)

All you need do is connect the right sockets all in parallel, you dont
need a single electronic component anywhere.


NT

John Rumm

unread,
Jan 23, 2009, 10:19:12 PM1/23/09
to
meow...@care2.com wrote:
> John Rumm wrote:
>> Pandora wrote:
>>
>>> I have a BT line which terminates at a BT master socket. I also have a
>>> Cat 5e patch panel, which is connected to various RJ45 sockets around
>>> the property.
>>>
>>> I want to convert five of the RJ45 sockets to phone points so I bought
>>> some PSTN line adapters from Comlynx:
>>> http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
>>>
>>> On the patch panel, I intend to wire five spare ports in parallel back
>>> to terminals 2 and 5 in the BT master socket. I can then patch across
>>> to the port that has a phone on it.
>>>
>>> My question is this:
>>> The Comlynx adaptors have a capacitor/resistor/surge suppressor in
>>> them. If I use five of them in parallel, won't it mess up the ring
>>> tone?
>> Probably not... the capacitor is only shunting the ring line from one of
>> the exchange lines. Extra capacitance will just *lower* the minimum
>> switching frequency that can "get through" the cap. Since I doubt you
>> will be using pulse dial phones, or ones with real bells for ringers, it
>> should not make any difference.
>
> It depends on the value of those parts. The short story is you've got

Not really - you could replace the cap with a wire and the phones would
still work[1]. The more caps you have the more the connection between
the exchange wire and the ring wire looks like a "wire"

[1] Many countries use two wire phone systems anyway. The only reason
for the inclusion of the ring wire was to prevent the bell on the phone
from "tinkling" when pulse dialing on another extension.

> the wrong items, you just want plug to socket adaptors, not stuff with

Yup, agreed. An ordinary secondary LAU is what is required.

> circuitry in it. Another option is to have a phone socket on the wall
> next to the rj45 skt.

Kind of defeats the point of structured cabling though ;-)

>> If you want a more flexible solution, you can pick up a cheap analogue
>> PABX from the far east via ebay. Then it certainly becomes a non issue.
>> (I have a 4/16 line/ext PABX waiting to go it when the required round
>> tuit is acquired and the structured wiring is done)
>
> All you need do is connect the right sockets all in parallel, you dont
> need a single electronic component anywhere.

agreed, unless you want to do extra stuff like being able to route calls
between rooms etc, or share multiple exchange lines effectively.

Gordon Henderson

unread,
Jan 24, 2009, 3:37:05 AM1/24/09
to
In article <3c2048b4-609d-4543...@g1g2000pra.googlegroups.com>,

Pandora <no--...@pandora.be> wrote:
>I have a BT line which terminates at a BT master socket. I also have a
>Cat 5e patch panel, which is connected to various RJ45 sockets around
>the property.
>
>I want to convert five of the RJ45 sockets to phone points so I bought
>some PSTN line adapters from Comlynx:
>http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
>
>On the patch panel, I intend to wire five spare ports in parallel back
>to terminals 2 and 5 in the BT master socket. I can then patch across
>to the port that has a phone on it.
>
>My question is this:
>The Comlynx adaptors have a capacitor/resistor/surge suppressor in
>them. If I use five of them in parallel, won't it mess up the ring
>tone?

It shouldn't as the capacitor will be on the output side of the socket. it
bridges to pin 3 on the BT side, and pin 3 won't be present on the
cat-5 side.

Looking at the data sheet, it suggests that part 42-212 is a
straight-through device with no active components. If you're paralleling
up 5 sockets, maybe thats the one to use... (And wire to pins 4 & 5 on
the cat-5 side)

The Natural Philosopher

unread,
Jan 24, 2009, 4:30:40 AM1/24/09
to

You can get slave CAT5 adaptors as well. As long as your extensions
carry the bell wire out of the master socket, all will be well.


However what I have done is to make telephone style wallplates on the
end of my cat5 as well as RJ45...

..but I also have a PABX, allowing me up to 8 phones.

Its a mixed digital analog panasonic. Second hand for about 100 notes.

Andrew Gabriel

unread,
Jan 24, 2009, 5:33:51 AM1/24/09
to
In article <3c2048b4-609d-4543...@g1g2000pra.googlegroups.com>,

Pandora <no--...@pandora.be> writes:
> I have a BT line which terminates at a BT master socket. I also have a
> Cat 5e patch panel, which is connected to various RJ45 sockets around
> the property.
>
> I want to convert five of the RJ45 sockets to phone points so I bought
> some PSTN line adapters from Comlynx:
> http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
>
> On the patch panel, I intend to wire five spare ports in parallel back
> to terminals 2 and 5 in the BT master socket. I can then patch across
> to the port that has a phone on it.
>
> My question is this:
> The Comlynx adaptors have a capacitor/resistor/surge suppressor in
> them. If I use five of them in parallel, won't it mess up the ring
> tone?

No, that's fine.
You don't need the resistor and probably not the surge suppressor
(unless your cat 5 lines are long and/or between buildings), but
they do no harm either.

The capacitors will each be feeding an individual phone -- to say
they are in parallel is rather misleading, and as they usually only
feed one appliance in this situation (and have to fit in a small space),
they're usually quite a bit lower value than the one in the master
socket that has to get a REN of 4 though it. Also, the one in the
master socket is not being used at all, because when you carry phone
over structured cabling, you do not carry the bell wire -- it's
recreated separately at each socket outlet by those capacitors.

> I know that this is the reason that you can't wire BT master sockets
> in parallel. The same would apply to the adapters I would have thought.

The bit that matters for parallel connecting, the bell wire (pin 3),
isn't being connected at all between your adaptors.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

Owain

unread,
Jan 23, 2009, 7:42:35 PM1/23/09
to
Pandora wrote:
> I have a BT line which terminates at a BT master socket. I also have a
> Cat 5e patch panel, which is connected to various RJ45 sockets around
> the property.
> I want to convert five of the RJ45 sockets to phone points so I bought
> some PSTN line adapters from Comlynx:
> http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
> On the patch panel, I intend to wire five spare ports in parallel back
> to terminals 2 and 5 in the BT master socket. I can then patch across
> to the port that has a phone on it.
> My question is this:
> The Comlynx adaptors have a capacitor/resistor/surge suppressor in
> them. If I use five of them in parallel, won't it mess up the ring
> tone?

It might do.

> I know that this is the reason that you can't wire BT master sockets
> in parallel. The same would apply to the adapters I would have thought.

Alternatives are
(a) remove components from adapters. Your phones /might/ work with
2-wire ringing.

(b) wire pin 3 in the BT master socket in parallel to the appropriate
wire in your patch panel and extend it thusly to the adapters

(c) install a small pbx with separate extensions

Owain

Pandora

unread,
Jan 24, 2009, 8:47:42 AM1/24/09
to
On 24 Jan, 10:33, and...@cucumber.demon.co.uk (Andrew Gabriel) wrote:
>
> <snip>

>
> The bit that matters for parallel connecting, the bell wire (pin 3),
> isn't being connected at all between your adaptors.
>

Thanks.

As Owain says, I suppose I could buy some "no components" adaptors and
connect the bell wire from the master socket pin 3 to the patch panel.
This way would be identical electrically to wiring secondary phone
sockets to the BT master.

However, since I already have the Comlynx PSTN master adaptors, I'll
try first with just connecting master socket pins 2 and 5 only.

Not sure which of the above two options is "industry standard" when
using structured wiring.

Pandora

unread,
Jan 24, 2009, 8:59:27 AM1/24/09
to
On 24 Jan, 02:11, John Rumm <see.my.signat...@nowhere.null> wrote:
>
> <snip>

>
> If you want a more flexible solution, you can pick up a cheap analogue
> PABX from the far east via ebay. Then it certainly becomes a non issue.
> (I have a 4/16 line/ext PABX waiting to go it when the required round
> tuit is acquired and the structured wiring is done)
>

Thanks, John.

I will reserve 8 ports on the patch panel for PABX phone extensions.

If I use a PABX, I have three issues to resolve.

What pins on the patch panel do I wire the four-core PABX outout
extension cables to?
What RJ45 adapter should I use if want to plug a PABX system phone?*
What RJ45 adapter should I use if I want to plug in an analogue PSTN
phone?**

(*42-214? on http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36)
(**42-212? on http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36)

Andrew Gabriel

unread,
Jan 24, 2009, 9:23:40 AM1/24/09
to
In article <136bd86d-fc05-41f8...@s9g2000prg.googlegroups.com>,

What you're doing in the "industry standard".

You never connect pin 3 through structured cabling.
There's no way to balance the current in twisted
pairs if you did so, making the circuit much more
liable to both generate and pickup interference
from other circuits in the structured cabling.
You'd need twisted triples, which don't exist in
structured cabling (or anywhere I can think of).

You only need all the pin 3's connected together
if you've got pulse dialing phones, and bells
in other phones on the same line which are liable
to 'tinkle' when pulse dialing is taking place
(and you care about stopping the bell tinkle).
Frankly, that never occurs in phone systems
installed on top of structured cabling.

The Natural Philosopher

unread,
Jan 24, 2009, 9:44:39 AM1/24/09
to

That is not true. A LOT of phones wont ring without the third wire
connected.

And what you said about CAT5 and balancing, is with respect, bollocks.

The third wire ONLY carries the ring current.

Which is so low frequency as to be irrelevant to almost anything else
using the cabling.

The real reason why the standards are the way they are, is because
structured phone wiring implies PABX and PABX use a two wire multiple
master system. Or a 4 wire digital system.


I.e. by the time you now you want structured wiring, you are long past
the time when its one phone, three extensions and no PABX...

We've run Ethernet and phone down tme SAME cat 5 cables in
extremis..splitting the pairs so that different stuff runs on different
pairs. No interference.

Ron Lowe

unread,
Jan 24, 2009, 11:25:28 AM1/24/09
to
"Pandora" <no--...@pandora.be> wrote in message
news:3c2048b4-609d-4543...@g1g2000pra.googlegroups.com...


Read this document...

http://www.evonet.com/evonet/index.asp?Page=102&Download=cablechart.pdf
( note that the BT wiring shows pins 1-6 and also in reverse, 6-1.This is
due to a spectacular cock-up where the plug and socket numbering is reversed
from each other! I tend to always use the designation where 3 is the ring,
although you will also see it designated as 4. )

So now you know there are 3 different typed of LAUs ( tha adapters you
linked to ).

PSTN masters: contains ring cap, resistor, and surge suppressor.
PABX masters: contains ring cap.
Secondary: contains nothing, only wiring.

PSTN masters are intended to be used where the 2-wire raw incoming line is
presented over the structured wiring on pins 4+5. All the functions of a
normal BT master are provided in the LAU, including generating the ring
signal on pin 3 of the BT connector.

PABX masters are similar: they take a 2-wire signal from a PBX over pins 4+5
of the structured wiring, and generate the ring signal on pin 3. They just
ommit the surge arrestor, since the line is not coming in from overhead
plant.

Secondaries require you to have a master ( eg NTE5 ) elsewhere, and take a
3-wire feed in over the structured wiring: the line pair on 4+5 destined for
pins 2+5 of the BT connector, and the ring signal on 2 destined for BT pin
3.

In my installation, which is pretty much the same as yours, you should
really use secondaries. The raw incoming BT line goes to A+B on the BT
master, and then you wire from the extension outlets on the master pins 2,5
and 3 up to the patch panel. The mastering is provided by the NTE5, and
the ring signal is carried over pin 2 of the Cat6 to the LAUs. The line
pair is carried over pins 4+5 of the Cat6 to the LAUs.

In this way, you will have the same config as normal secondary extensions,
with no multiple masters to worry about.

--
Ron


Andrew Gabriel

unread,
Jan 24, 2009, 11:39:47 AM1/24/09
to
In article <123280828...@proxy00.news.clara.net>,
The Natural Philosopher <a@b.c> writes:

> Andrew Gabriel wrote:
>> What you're doing in the "industry standard".
>>
>> You never connect pin 3 through structured cabling.
>> There's no way to balance the current in twisted
>> pairs if you did so, making the circuit much more
>> liable to both generate and pickup interference
>> from other circuits in the structured cabling.
>> You'd need twisted triples, which don't exist in
>> structured cabling (or anywhere I can think of).
>>
>> You only need all the pin 3's connected together
>> if you've got pulse dialing phones, and bells
>> in other phones on the same line which are liable
>> to 'tinkle' when pulse dialing is taking place
>> (and you care about stopping the bell tinkle).
>> Frankly, that never occurs in phone systems
>> installed on top of structured cabling.
>>
>
> That is not true.

You've misread what I said.

> A LOT of phones wont ring without the third wire connected.

[I agree]

> And what you said about CAT5 and balancing, is with respect, bollocks.
>
> The third wire ONLY carries the ring current.

No, any current has to be carried by at least two wires,
and irrespective of carrying current, it's coupled through
the ring capacitor to one of the pair, and acts as an
antenna and/or capacitive coupler to other signals
(particularly to other similar unbalanced antennae wires).

> Which is so low frequency as to be irrelevant to almost anything else
> using the cabling.

The fundamental is low, but from PABX's is usually not a
sine wave (worse case square wave), so it's stuffed full
of harmonics. The interference is clearly audiable if you
stuff two 3-wire phone circuits down the same cable bundle,
and that stuffed up modems back when they were in common use.

> The real reason why the standards are the way they are, is because
> structured phone wiring implies PABX and PABX use a two wire multiple
> master system. Or a 4 wire digital system.

Loads of PABX's (pretty well all at one point) use exactly
the BT system, because they use(d) exactly the same kit.
You still never carried the bell wire, as that caused very
audiable interference on other pairs. (You would only link
it locally between extentions on the same line.)

> We've run Ethernet and phone down tme SAME cat 5 cables in
> extremis..splitting the pairs so that different stuff runs on different
> pairs. No interference.

Ethernet, being properly balanced pairs, will neither transmit
nor receive from unbalanced antennae wires (and the frequencies
are so different it's unlikely there would be any problem
even if they did).

John Weston

unread,
Jan 24, 2009, 12:49:16 PM1/24/09
to
In article <123280828...@proxy00.news.clara.net>, a@b.c says...

>
> That is not true. A LOT of phones wont ring without the third wire
> connected.
>
> And what you said about CAT5 and balancing, is with respect, bollocks.
>
> The third wire ONLY carries the ring current.
>
> Which is so low frequency as to be irrelevant to almost anything else
> using the cabling.
>
> The real reason why the standards are the way they are, is because
> structured phone wiring implies PABX and PABX use a two wire multiple
> master system. Or a 4 wire digital system.
>
>
> I.e. by the time you now you want structured wiring, you are long past
> the time when its one phone, three extensions and no PABX...
>
> We've run Ethernet and phone down tme SAME cat 5 cables in
> extremis..splitting the pairs so that different stuff runs on different
> pairs. No interference.
>

Sorry, but that's not my experience. The unbalanced ring wire will pick
up all sorts of HF interference and couple it back, via the capacitor
onto one of the speech pairs. If this is done anywhere near a pair
carrying a DSL signal, then it will be seen as noise for the DSL signal.

If you are wiring just phones, then ring wire HF interference doesn't
matter. (But I can hear ring current crosstalk if pin-3 is extended over
structured cabling). IMO, if you are running phones over structured
wiring, then you are better using PBX mastering modules, or plug-in
converters with ring capacitors to generate the ringing signal on pin 3
at the remote end, so you maintin the twisted-pair integrity of your
structured wiring.

If you split the Ethernet pairs, then don't expect auto rate detect to
work. You'll have to nail everything to 10Mbps. I spent some time on
one job rewiring all the Ethernet cables that had been done by an
electrician using 1-2, 3-4, 5-6 & 7-8 for the pairs. It worked fine
when everything was 10Mbps, then they started introducing 100Mbps...

Using two Ethernets down one cat-5 cable, keeipn the standard two pairs
for one circuit and using the spare two pairs= for the other circuit
also works fine - until they go to Gigabit...

Do it right and do it once.

--
John W
To mail me replace the obvious with co.uk twice

Pandora

unread,
Jan 24, 2009, 6:01:40 PM1/24/09
to
Thanks to everyone for their comments.

It seems as though the method and the adaptors I am using are correct,
even though I wasn't sure why.

John Rumm

unread,
Jan 24, 2009, 7:46:04 PM1/24/09
to
Pandora wrote:
> On 24 Jan, 02:11, John Rumm <see.my.signat...@nowhere.null> wrote:
>> <snip>
>>
>> If you want a more flexible solution, you can pick up a cheap analogue
>> PABX from the far east via ebay. Then it certainly becomes a non issue.
>> (I have a 4/16 line/ext PABX waiting to go it when the required round
>> tuit is acquired and the structured wiring is done)
>>
>
> Thanks, John.
>
> I will reserve 8 ports on the patch panel for PABX phone extensions.
>
> If I use a PABX, I have three issues to resolve.
>
> What pins on the patch panel do I wire the four-core PABX outout
> extension cables to?

The extension cables on the PABX will be 2 wire typically on a analogue
PABX. Typically you would use a two wire RJ11 to RJ11 lead and plug one
end into the PABX RJ11 socket and the other into the RJ45 socket on the
patch panel. (RJ11s fit RJ45 sockets - they just can't connect to the
two outer pins)

> What RJ45 adapter should I use if want to plug a PABX system phone?*

You want a PABX master - i.e. a ring capacitor and nothing else. Most
analogue only PABX are controlled via an ordinary touch tone phone
rather than a system one.

meow...@care2.com

unread,
Jan 24, 2009, 8:09:25 PM1/24/09
to

but why fit caps when theres no reason for them. Its not as if the OP
is trying to save a penny on copper by running the extension in 2
core.


> [1] Many countries use two wire phone systems anyway. The only reason
> for the inclusion of the ring wire was to prevent the bell on the phone
> from "tinkling" when pulse dialing on another extension.
>
> > the wrong items, you just want plug to socket adaptors, not stuff with
>
> Yup, agreed. An ordinary secondary LAU is what is required.
>
> > circuitry in it. Another option is to have a phone socket on the wall
> > next to the rj45 skt.
>
> Kind of defeats the point of structured cabling though ;-)

well...

> >> If you want a more flexible solution, you can pick up a cheap analogue
> >> PABX from the far east via ebay. Then it certainly becomes a non issue.
> >> (I have a 4/16 line/ext PABX waiting to go it when the required round
> >> tuit is acquired and the structured wiring is done)
> >
> > All you need do is connect the right sockets all in parallel, you dont
> > need a single electronic component anywhere.
>
> agreed, unless you want to do extra stuff like being able to route calls
> between rooms etc, or share multiple exchange lines effectively.

connecting all the feeds in parallel at the wiring room doesnt prevent
installing a local exchange later, if its ever wanted one day.


NT

The Natural Philosopher

unread,
Jan 24, 2009, 9:38:26 PM1/24/09
to

That is not inconsistent with what I said.


> and irrespective of carrying current, it's coupled through
> the ring capacitor to one of the pair, and acts as an
> antenna and/or capacitive coupler to other signals
> (particularly to other similar unbalanced antennae wires).
>

So what? Its part of a twisted pair..ground that and its effectively
'shielded'. We aren't talking abiout running broadband on it or teh
other pairs. Just audio.


>> Which is so low frequency as to be irrelevant to almost anything else
>> using the cabling.
>
> The fundamental is low, but from PABX's

The OP specifically is NOT using a a PABX which is why he HAS to carry
the third wire.

So this isn't relevant.


is usually not a
> sine wave (worse case square wave), so it's stuffed full
> of harmonics. The interference is clearly audiable if you
> stuff two 3-wire phone circuits down the same cable bundle,
> and that stuffed up modems back when they were in common use.
>
>> The real reason why the standards are the way they are, is because
>> structured phone wiring implies PABX and PABX use a two wire multiple
>> master system. Or a 4 wire digital system.
>
> Loads of PABX's (pretty well all at one point) use exactly
> the BT system, because they use(d) exactly the same kit.
> You still never carried the bell wire, as that caused very
> audiable interference on other pairs. (You would only link
> it locally between extentions on the same line.)
>

HE ISN"T USING A PABX WHICH IS WHY HE NEEDS THE THIRD WIRE.,

>> We've run Ethernet and phone down tme SAME cat 5 cables in
>> extremis..splitting the pairs so that different stuff runs on different
>> pairs. No interference.
>
> Ethernet, being properly balanced pairs, will neither transmit
> nor receive from unbalanced antennae wires (and the frequencies
> are so different it's unlikely there would be any problem
> even if they did).

Exactly. Etherenet wont be an issue as its almost 100% immine from low
level interference.

As far as crosstalk between the phone exte=nsions goes, well had you
not realised that:

NO PABX IS INVOLVED

i.e. they are all carrying the SAME speech circuit.

I,e waht I said at the outset, you are talking standard PABX structured
able. He is NOT using it that way. If you run two wire from the BT
incoming you need 5 master sockets to regenerate the ring wire locally.
AN that is against BT rules for their lines: you are only allowed to run
one ring generator circuit AFAICR.

And the anti tinkle wire is a 4th wire surely?


>

The Natural Philosopher

unread,
Jan 24, 2009, 9:40:23 PM1/24/09
to
John Weston wrote:
> In article <123280828...@proxy00.news.clara.net>, a@b.c says...
>
>> That is not true. A LOT of phones wont ring without the third wire
>> connected.
>>
>> And what you said about CAT5 and balancing, is with respect, bollocks.
>>
>> The third wire ONLY carries the ring current.
>>
>> Which is so low frequency as to be irrelevant to almost anything else
>> using the cabling.
>>
>> The real reason why the standards are the way they are, is because
>> structured phone wiring implies PABX and PABX use a two wire multiple
>> master system. Or a 4 wire digital system.
>>
>>
>> I.e. by the time you now you want structured wiring, you are long past
>> the time when its one phone, three extensions and no PABX...
>>
>> We've run Ethernet and phone down tme SAME cat 5 cables in
>> extremis..splitting the pairs so that different stuff runs on different
>> pairs. No interference.
>>
> Sorry, but that's not my experience. The unbalanced ring wire will pick
> up all sorts of HF interference and couple it back, via the capacitor
> onto one of the speech pairs. If this is done anywhere near a pair
> carrying a DSL signal, then it will be seen as noise for the DSL signal.
>

ER thats why you put the filter on the BT master socket. To isolate that
trash from the DSL.

> If you are wiring just phones, then ring wire HF interference doesn't
> matter. (But I can hear ring current crosstalk if pin-3 is extended over
> structured cabling). IMO, if you are running phones over structured
> wiring, then you are better using PBX mastering modules, or plug-in
> converters with ring capacitors to generate the ringing signal on pin 3
> at the remote end, so you maintin the twisted-pair integrity of your
> structured wiring.
>
> If you split the Ethernet pairs, then don't expect auto rate detect to
> work. You'll have to nail everything to 10Mbps. I spent some time on
> one job rewiring all the Ethernet cables that had been done by an
> electrician using 1-2, 3-4, 5-6 & 7-8 for the pairs. It worked fine
> when everything was 10Mbps, then they started introducing 100Mbps...
>
> Using two Ethernets down one cat-5 cable, keeipn the standard two pairs
> for one circuit and using the spare two pairs= for the other circuit
> also works fine - until they go to Gigabit...
>

Yup. But I didnt find a problem with 100Mbps stuff on split circuits.

Dave Liquorice

unread,
Jan 25, 2009, 5:41:07 AM1/25/09
to
On Sun, 25 Jan 2009 02:38:26 +0000, The Natural Philosopher wrote:

> And the anti tinkle wire is a 4th wire surely?

No, it's "earth" for earth recall.

The bell wire is the anti-tinkle/ring wire. Bells (sounders) should be
connected from it to the A wire. When a phone goes off hook the bell wire
is shorted to the A wire.

--
Cheers
Dave.

John Rumm

unread,
Jan 25, 2009, 11:52:59 AM1/25/09
to
meow...@care2.com wrote:

> but why fit caps when theres no reason for them. Its not as if the OP
> is trying to save a penny on copper by running the extension in 2
> core.

You don't really want to stick an unbalanced ring wire down your cat5
for reasons of noise immunity/generation. Hence its much better to run
structured wiring using 2 cores for phone.

>> agreed, unless you want to do extra stuff like being able to route calls
>> between rooms etc, or share multiple exchange lines effectively.
>
> connecting all the feeds in parallel at the wiring room doesnt prevent
> installing a local exchange later, if its ever wanted one day.

yup, true.

John Rumm

unread,
Jan 25, 2009, 11:57:07 AM1/25/09
to
The Natural Philosopher wrote:

> The OP specifically is NOT using a a PABX which is why he HAS to carry
> the third wire.
>
> So this isn't relevant.

You don't need to *run* the third wire - you run two wires from hub to
socket, and fit a ring cap at the socket - so the third wire exists only
for the length of the phones cable.

> And the anti tinkle wire is a 4th wire surely?

No, that was used for earthed (rather than timed break) "recall"
facilities. Generally not wired these days.

Pandora

unread,
Feb 8, 2009, 3:50:23 PM2/8/09
to
On 24 Jan, 00:02, Pandora <no--s...@pandora.be> wrote:
> I have a BT line which terminates at a BT master socket. I also have a
> Cat 5e patch panel, which is connected to various RJ45 sockets around
> the property.
>
> I want to convert five of the RJ45 sockets to phone points so I bought
> some PSTN line adapters from Comlynx:
> http://www.comlynx.co.uk/product_detail.php?pr_id=12&m_id=34&sb_id=36
>
> <snip>

Just coming back to report the outcome of my wiring experiment after
absorbing the various contributions on this thread.

I took a two-wire connection from BT master terminals 2 and 5 and
connected these to pins 5 and 4 of four patch panel sockets (wired in
parallel).

I used Cat5e patch leads to connect the four patch panel sockets to
the kitchen, dining room, bedroom and study. I inserted the Comlynx
PSTN adapters into the corresponding jacks in those rooms and plugged
the phones in.

It worked fine.

[Footnote: I have since bought a Panasonic TEA308 PABX and four
Panasonic hybrid phones. Once I've worked out the wiring on these,
I'll install them instead of the PSTN phones.]

Reply all
Reply to author
Forward
0 new messages