Hum from phone wires running next to mains?

It is quite difficult to induce hum into telephone wiring. Use twisted pair cabling rather than the flat ready-made extension cables.

Exactly! The phone company has millions of miles of cable running right below power lines and hundreds literally touching each other in the jacket of the cable. That little twist they put in the pairs is excellent in isolating them from crosstalk. Just don't use "straight through" door bell wire and you will be fine.

|>It is quite difficult to induce hum into telephone wiring. |>Use twisted pair cabling rather than the flat ready-made |>extension cables. | | Exactly! | The phone company has millions of miles of cable running right below | power lines and hundreds literally touching each other in the jacket | of the cable. That little twist they put in the pairs is excellent in | isolating them from crosstalk.

That twist is a great little means to ensure induced signals, whatever they may be, are induced in equal amount on both wires, so they do not contribute to the actual intended signal that is a differential between those two wires.

However, a risk exists when two different pairs are present next to each other and each pair is twisted at the same pitch. The signal carried by one can end up being induced differentially on the other. So don't twist those power lines, or if you do, twist them at a pitch with a ratio to the phone line twist that is not a whole number.

CAT5 cable is an example. It has 4 different pairs twisting along. Each of the pairs has a different twist pitch by design (unless you get some cheap cable not manufactured correctly).

I do not have any technical knowledge of this area.

I would like to ask about a cable which has two or more twisted pairs in it.

Is there is a greaterlikelihood of hum if I connect a "2 wire" phone extension by using one wire from a twisted pair and taking the second wire from a different twisted pair?

(As you suggest, I will not get the flat ready made extension cable which I guess is made from flexible multi-stranded wires.)

Is the sort of cable sold in the UK specifically for domestic telephone wall sockets (wuth single stranded wires) usually made up as "twisted pair" in the way you are recommending?

: : I do not have any technical knowledge of this area. : : : : I would like to ask about a cable which has two or more : : twisted pairs in it. : : : : Is there is a greaterlikelihood of hum if I connect a : : "2 wire" phone extension by using one wire from a : : twisted pair and taking the second wire from a : : different twisted pair?

Why would you want to do that..? The answer is very probably, so ensure that the pair of wires you use are twisted *together*..!

Ivor

Twist the power lines all you like. You *can't* physically twist them identically to that of a comm cable *and* get the two pairs to snuggle up to each other in a way that will create the problem described above.

Regardless, it isn't "smart" to run a comm cable in physical contact with power cabling. Even a couple of inches separation is sufficient to significantly reduce common mode coupling. And the fact that no hum is heard when it is first installed is *not* sufficient reason to accept such practice. The common mode voltage induced on the comm cable may not be a problem at any given time, but it means that in the future anything (such as kinks in the cable, dampness, damaged insulation, etc) that reduces the balance *will* cause excessive hum. The higher the common mode induced voltage, the less unbalance required to cause objectionable hum.

If you can avoid putting the two types of cable together, you *should*.

All multipair twisted-pair cable uses different twists for each pair. That is identically true for bundled pairs in telephone cable. Moreover, if there are multiple bundles the bundles are swirled within the jacket too.

That is referred to as a "split pair", and yes it will cause problems. It commonly happens with CAT5 cabling due to the different standards for pin assignments for a DS1 interface and for 10BaseT Ethernet. Typically a DS1 cable will work for Ethernet if the length is short, but if used for faster than 10baseT, it won't work at all, even for a 6 foot jumper cable.

On large telephone cables split pairs invariably have significant crosstalk (either hum or speech from other cables).

... snip ...

Yes. The idea of twisted pairs is that an interference appears on both lines, and thus tends to cancel itself. Separating the lines makes it easy for unequal induction.

yes

Much will depend on the power and light wiring method used and the quality of the cable used to carry your telephone circuits. What do you mean when you say "flex". I suspect you'll be amused to learn that in the USA that word is electricians short hand for flexible metallic conduit. I doubt that United Kingdom "flex" is anything like Flexible Metallic Conduit a photograph of which can be found at .

The best way to reduce the amount of noise in telephone lines is to use station cable that has the wire pairs continuously twisted around each other. In this way any electro magnetic fields that might otherwise induce an unwanted noise into the circuit is self canceling in the twisted pair of wires. Even with good quality station cable best practice is to maintain at least several inches of separation between the telephone cables and the electrical power and light wiring.

That practice is known in the North American communications industry as a split pair. It is usually the cause of a host of troubles of which induced noise is only the most common.

No I don't know why it is still used as most people in the UK seem to have adsl connections nowadays (not just phones)

Very interesting.

Please can you advise on how these twisted pairs compare with

1. shielded audio cable and
2. rf coax.

In case 1 both the wanted signal and the noise are in the audio frequency range.

In case 2 the electricity supply noise contains harmonics of similar frequency to the wanted rf signal.

Can't comment on the hum ... but it looks like others have.

One contribution I would make is that you are aware that your phone service will support 4 REN and that each phone is normally 1 REN, meaning that you can have a maximum of 4 phones. My parents had more of this and whilst from their perspective it seemed to work (they could call out), it stopped people from calling in because their phones stopped ringing.

So be aware that if you are adding four or five additional phone sockets that you won't be able to use all of them (at the same time).

In alt.engineering.electrical George wrote: | On Tue, 04 Mar 2008 21:22:30 +0000, phil-news-nospam wrote: | |> In alt.engineering.electrical snipped-for-privacy@aol.com wrote: |> | On Tue, 4 Mar 2008 19:13:07 -0000, "Graham." wrote: |> |> |>It is quite difficult to induce hum into telephone wiring. Use twisted |> |>pair cabling rather than the flat ready-made extension cables. |> | |> | Exactly! |> | The phone company has millions of miles of cable running right below |> | power lines and hundreds literally touching each other in the jacket of |> | the cable. That little twist they put in the pairs is excellent in |> | isolating them from crosstalk. |> |> That twist is a great little means to ensure induced signals, whatever |> they may be, are induced in equal amount on both wires, so they do not |> contribute to the actual intended signal that is a differential between |> those two wires. |> |> However, a risk exists when two different pairs are present next to each |> other and each pair is twisted at the same pitch. The signal carried by |> one can end up being induced differentially on the other. So don't twist |> those power lines, or if you do, twist them at a pitch with a ratio to the |> phone line twist that is not a whole number. |> |> CAT5 cable is an example. It has 4 different pairs twisting along. Each |> of the pairs has a different twist pitch by design (unless you get some |> cheap cable not manufactured correctly). | | Very interesting. | | Please can you advise on how these twisted pairs compare with | | 1. shielded audio cable | and | 2. rf coax. | | In case 1 both the wanted signal and the noise are in the audio frequency | range. | | In case 2 the electricity supply noise contains harmonics of similar | frequency to the wanted rf signal.

I don't have specific data on the quality of noise immunity. I'd bet that kind of research has been done. It most certainly would vary by quality of construction of the cables in question.

RF coax comes in various levels of quality based on a stated shielding percentage. I've seen lows of 60% all the way up to 100%. The latter could be a foil, or a solid metal encapsulation (quite a variety of different coax types with this).

I've seen cables, including CAT5, with both twisting _and_ shielding around the whole cable assembly. I don't know how much the effectiveness works together. I have not had a case where I would consider using it.

|> QUESTION: I would like to know I this will increase the level of hum. |>

|> ISTR UK phones have a transformer and some other components to |> neutralise hum but would that be good enough to prevent hum from a messy |> setup like mine? Some details are below. |>

|> ------------------------- |>

|> In my situation the phone extension wires and the mains wires will run |> close to one other. |>

|> There will be about four or five additional extension phone sockets. | | Can't comment on the hum ... but it looks like others have. | | One contribution I would make is that you are aware that your phone service | will support 4 REN and that each phone is normally 1 REN, meaning that you | can have a maximum of 4 phones. My parents had more of this and whilst from | their perspective it seemed to work (they could call out), it stopped people | from calling in because their phones stopped ringing.

I used to see phones rated in terms of their "ringer equivalence" here in the USA. These numbers were, for some phones, as low as 0.2. I do not recall ever seeing one about 0.9. That would suggest to me that you could readily have more than 4 phones on such a phone circuit. I never had any reason to actually do a scientific test of this.

| So be aware that if you are adding four or five additional phone sockets | that you won't be able to use all of them (at the same time).

Or check your phone specs for an REN or ringer equivalence number.

If you want to put DSL on your phone line, I also suggest a splitter at the entrance of the phone line and a separate NON-branching higher grade (e.g. twisted pair) wire for the run from the DSL side of the splitter to the intended connection.

There is obviously a significant difference in the phone systems in the two countries. Our phones have the bells in parallel and if thee are too many the wrong impedance is presented to the exchange, and no ringing voltage will get sent. I have never seen a UK approved phone with a REN less than

You can buy a REN booster ( a mains powered device) which allows many more phones.

It depends on how old the phones, or these days with modern electrically powered phones, cordless/etc, the REN, number may be very low per phone. IMO, no good reason not to get a cordless phone these days as some are dirt cheap, except it's nice to have at least one non-electric in case the power goes out.

I just looked at my phoneset, a wireless extension by Uniden. the base is rated at 0.08 REN. We have a regular phone and 2 of these wireless base stations with 2 sets each, so everyone has a set handy and the neighbors can listen in too.