How can 2 signals go thru copper wire?

[rogersb]

talking to someone about direct tv systems, how they talk between HD boxes over the rg6 coax line one person said one part sends a signal and it travels down the center of the center coax conductor and another signal going the other way travels down the outer part of the conductor, the surface of the center wire.

but isn't the detail that the conductor is a transmission line that can be driven from different points?

like a steam pipe, you tap on it from the basement or 3rd floor and both hear it

 

[omega_minus]

I've never heard of sending signals down different parts of the wire. To put multiple signals on one line (as in a phone line in your neighborhood for example) a multiplexing technique is used. Either frequency division multiplexing, where each signal is at a different frequency which can be separated by a fast Fourier transform, or time division multiplexing where each signal "takes a turn" on the line.

 

[omega_minus]

I should add that an FFT isn't necessarily required for FDM, if the frequencies are far enough apart an RLC filter could be used on analog signals

 

[rogersb]

thanks. i was worried that some new theory had come out that I didnt know about. i thought that any wire is a transmission line that has unique RLC properties, and for any instant in time there is some discrete, fixed current flow and voltage. i thought that its like 2 people holding a long rope, you could modulate it from one side and a wave would propagate across, and the other side could modulate but the wave would combine

for current, i thought only one direction at a time, it could take turns like time multiplex but had not heard of different parts of the wire carrying a signal like that.

maybe in fiber optic cable it does, but that's not the same thing?

 

[Baluncore]

rogersb. The story you heard was partly true.

First consider a signal propagating in one direction along a coaxial transmission line. At any point on the line an RF current traveling on the outside of the inner conductor must be countered by an exactly equal and opposite current flowing on the inside of the outer conductor. The voltage at that point on the line is measured through the dielectric between those two surfaces. The characteristic impedance of the line in ohms is the ratio of voltage to the current.

A “linear” transmission line permits signals to propagate in both directions at the same time. You can put a different pulse in at each end, both pulses may pass on the line and arrive unmodified at the other end. There will be many variations in the instantaneous total current and voltage as they pass, but so long as the line is linear, the signals will remain separated by their direction of propagation.

When you look deep into another's eye, the light you see reflected from their eye is traveling the same path but in the opposite direction to the light reflected from yours. Like coaxial cable at RF, air is linear for visible light.

 

[CWatters]

i thought that its like 2 people holding a long rope, you could modulate it from one side and a wave would propagate across, and the other side could modulate but the wave would combine

Sure they would "combine" but with a suitable "polarising filter" at each end the wanted signal could be extracted and the unwanted ignored.

i thought that any wire is a transmission line that has unique RLC properties, and for any instant in time there is some discrete, fixed current flow and voltage.

That should be "at anyone position and at any instant in time there is some discrete, fixed current flow and voltage" but not relevant.

Two pulses/waves can be launched down a electrical wire (or a rope, a slinky or fibre optic cable) in opposite directions and they can pass each other and carry on as if nothing happened.

There are bound to be videos on youtube that show this.

 

[rogersb]

Thanks so much for the reply, I have questions inline:

rogersb. The story you heard was partly true.

First consider a signal propagating in one direction along a coaxial transmission line. At any point on the line an RF current traveling on the *** outside of the inner conductor must be countered by an exactly equal and opposite current flowing on the inside of the outer conductor***. The voltage at that point on the line is measured through the dielectric between those two surfaces. The characteristic impedance of the line in ohms is the ratio of voltage to the current.

*** at any point in the coax line? so if we pick some point x and measure the current on each of the 2 conductors, the center solid copper wire; and, the stranded/braided steel?

Are you saying that at any point x, x1, x2... the current on the inner solid wire must be equal and opposite to the current on the braided wire? But I thought these are 2 transmission line strips with different and unequal impedance, and the one thing we DONT get is balanced signal propagation? but we do get good shielding for the signal and this is the tradeoff?





A “linear” transmission line permits signals to propagate in both directions at the same time. You can put a different pulse in at each end, both pulses may pass on the line and arrive unmodified at the other end. There will be many variations in the instantaneous total current and voltage as they pass, but so long as the line is linear, the signals will remain separated by their direction of propagation.



GREAT! this is what I want to see. In all of the equipment doing data communication that I've seen, this method is not used so it may be possible in a lab setting but is not practical in real world equipment?

I want to set this up using equipment on a bench, any guidance is greatly appreciated!



When you look deep into another's eye, the light you see reflected from their eye is traveling the same path but in the opposite direction to the light reflected from yours. Like coaxial cable at RF, air is linear for visible light.

 

[Baluncore]

“outside of the inner conductor must be countered by an exactly equal and opposite current flowing on the inside of the outer conductor”

Yes, almost all RF current flows in the surface skin. It does not have time to penetrate before it reverses direction. This is the skin effect. The wave propagates close to the speed of light along the line but only at about 5 metres per second into the conductor, (=jogging speed).

But I thought these are 2 transmission line strips with different and unequal impedance, and the one thing we DONT get is balanced signal propagation? but we do get good shielding for the signal and this is the tradeoff?

You have been mislead. A transmission line can be a wire over a ground plane, or a wire over the inside of a coaxial braid, they are the same. A transmission line is a ladder of capacitors distributed continuously between the continuous inductive conductors. The conductors do not have to be equal, it is the inductance sum that is important.

The wave on the line travels in the dielectric, the conductors are like mirrors that keep the wave in the dielectric and out of the conductors.

There is another separate transmission line between the outside of the coaxial braid and the environment, but that external line is not being driven with the same signal so is isolated from the internal line. The braid is a shield because the current in the braid is equal and opposite to the current in the core. The resulting magnetic fields cancel outside the line.

The internal signal will not leak through the braid because it must be accompanied by the equal and opposite current on the core, that current cannot flow through the dielectric so it keeps the braid current as close as possible, the inside of the braid.

In all of the equipment doing data communication that I've seen, this method is not used so it may be possible in a lab setting but is not practical in real world equipment?

It may not be used for the short signal runs you have encountered where there is no cost advantage. It takes more hardware to separate the TX and RX signals at the two ends so it is founsd where coaxial cable or optic fibre is used for undersea cables or telephone trunk lines.

To demonstrate the independent two way signals in a lab you will need a roll of coax with a hybrid coupler of some sort at each end. The line will need to be matched to the couplers to prevent reflections.

 

[rogersb]

Yes, almost all RF current flows in the surface skin. It does not have time to penetrate before it reverses direction. This is the skin effect. The wave propagates close to the speed of light along the line but only at about 5 metres per second into the conductor, (=jogging speed).

You have been mislead. A transmission line can be a wire over a ground plane, or a wire over the inside of a coaxial braid, they are the same. A transmission line is a ladder of capacitors distributed continuously between the continuous inductive conductors. The conductors do not have to be equal, it is the inductance sum that is important.




What I mean is that the coax is a set of 2 transmission lines. one is a center solid copper wire, the other is braided steel. I thought their characteristic impedances were not the same, hence unbalanced, meaning the signal does not pass thru the same. Are you saying this is not correct?




The wave on the line travels in the dielectric, the conductors are like mirrors that keep the wave in the dielectric and out of the conductors.


the wave does not travel in the conductor, it travels in the dielectric? could you please point to reference material on this? I've got access to lots of journals and textbooks and want to understand better.



There is another separate transmission line between the outside of the coaxial braid and the environment, but that external line is not being driven with the same signal so is isolated from the internal line. The braid is a shield because the current in the braid is equal and opposite to the current in the core. The resulting magnetic fields cancel outside the line.

The internal signal will not leak through the braid because it must be accompanied by the equal and opposite current on the core, that current cannot flow through the dielectric so it keeps the braid current as close as possible, the inside of the braid.


I thought it couldn't leak because the Electric Field won't pass thru the braid? if you had a 2 inner conductor coax and the shield as ground and the inner 2 conductors were the + and -, the signal still won't leak, right?



It may not be used for the short signal runs you have encountered where there is no cost advantage. It takes more hardware to separate the TX and RX signals at the two ends so it is founsd where coaxial cable or optic fibre is used for undersea cables or telephone trunk lines.

To demonstrate the independent two way signals in a lab you will need a roll of coax with a hybrid coupler of some sort at each end. The line will need to be matched to the couplers to prevent reflections.

I have a roll of coax of brand new RG6. couplers, i do have the old balun used for the old tv sets with the 2 screw connectors, a few of those sitting around, have a Tek 2465 i think its 100mhz or 400mhz

thanks

 

[Baluncore]

What I mean is that the coax is a set of 2 transmission lines. one is a center solid copper wire, the other is braided steel. I thought their characteristic impedances were not the same, hence unbalanced, meaning the signal does not pass thru the same. Are you saying this is not correct?

That is not correct. There must be a return circuit for any signal current that flows. The braid and the core are working together to make one internal transmission line. The equation to calculate the impedance of a coaxial cable uses the diameter of both the inner and outer conductors. It calculates only one impedance for the conductor pair.

the wave does not travel in the conductor, it travels in the dielectric? could you please point to reference material on this? I've got access to lots of journals and textbooks and want to understand better.

The energy wave on the line travels at a velocity determined by the dielectric constant of the insulation. It is not a function of conductor resistivity.
See; http://en.wikipedia.org/wiki/Transmission_line#Coaxial_cable and you will find reference to “TEM” mode. The electric and magnetic fields between the conductors must be in the dielectric as neither can penetrate a good conductor. Electro-magnetics will be a black art until it becomes a science. You need to read broadly on the subject as the concepts involved in transmission lines are not immediately obvious and it will take time for them to form in your mind.

I thought it couldn't leak because the Electric Field won't pass thru the braid? if you had a 2 inner conductor coax and the shield as ground and the inner 2 conductors were the + and -, the signal still won't leak, right?

A twisted pair inside a screen, usually called “twinax”. The transmission line is the balanced twisted pair. Their electric fields cancel and so do not reach the braid. In that case the braid is being used as a Faraday screen. It is a different case to coax cable where the inside of the braid provides the only return path for the centre conductor surface current.

Please clearly identify your questions where you embed them inside quotes.
You could italicise your questions, but it is better to use multiple quotes.