Digital Transmission and Twisted Wires

[galaxy_twirl]

Homework Statement

I am currently embarking on a project where I have to build a communications system which can take in an analog voice signal, translate it into digital signal, and send it across to the receiver where the original voice gets recovered and it can be played back using a speaker.

In the project, I have to use a twisted wire (I think it is just 2 straight wires twisted together) to link the transmitter and receiver ends of the system together. However, does the use of a twisted wire enhance the quality at which the data is transmitted? (i.e. through reducing interference or

Homework Equations

The Attempt at a Solution

We know that the benefits of having a twisted wire is such that electromagnetic interference is reduced by reducing crosstalk between the 2 pairs of wires. However, I am a bit confused here - does interference occur only when I transmit data via analog mode? This is because I am transmitting data digitally here via PCM, thus I was thinking, since digital systems are more resistant to noise, using a twisted wire for data should not pose a huge problem.

I also have to mention the possible problems when I use a twisted wire to link both the transmitter and receiver end together. Right now, I can only think of resistance affecting the quality of the data. May I know, if anyone has good online sources or ideas, on where I can find information to brainstorm about the possible problems I will run into when I use a twisted wire as my data link for the communication system?

Thank you.

 

[rude man]

Twisted-pair reduces the inductance of the wires. That's because the area between the wires is reduced as much as possible, and inductance rises with separation area: L = NΦ/i, N=1 here, and Φ = BA where A = area and B is found the usual way via Ampere's law.

Twisting also establishes a more or less repeatable characteristic impedance. For longer lengths, and with sharp edges typical of digital signals, you may want to terminate the line at the receiving or transmitting end with that impedance to avoid reflections and messed-up data.

 

[galaxy_twirl]

Hi rude man. Thank you for your answer. I will take your suggestion into consideration when I design my system. Just wondering, what do you mean by "terminating the line at the receiving or transmitting end with that impedance"? By impedance, you mean complex resistance, like in an AC circuit, right?

Actually, I am using the twisted wire as a temporary solution to the data link between my transmitter and receiver. I will replace it with a laser link later when I have ensured that my communication system is working fine. :)

Oh, may I also ask a few questions about using a laser link, just to be sure I have identified the possible problems of using laser as a data link?

1. Light from external environment causes noise and interference and cause attenuation losses at the receiving end, increasing errors there.
2. Poor focus of laser beam as I am using a cheap laser pointer.
3. Intensity of laser beam gets weaker over longer distances, plus interference from external light sources may cause more attentuation losses.
4. Difficulty in ensuring the laser beam strikes totally on the photodiode.

Are the problems I have listed above correct?

Thanks again to anyone who has helped me. :D

 

[NascentOxygen]

5. Fog, mist, rain, smog.

 

[galaxy_twirl]

Hi NascentOxygen. Ah I see. Thank you! :)

 

[rude man]

Just wondering, what do you mean by "terminating the line at the receiving or transmitting end with that impedance"? By impedance, you mean complex resistance, like in an AC circuit, right?

Your line has what is called a 'characteristic impedance" = Z₀. It's real for a lossless line and is the resistance your VOM would read if the line were infinitely long.

If you send a pulse with sharp edges along such a line, with zero ohm source and the receiving end has a resistor = Z₀ across its two wires, then your pulse will appear at the receiving end just like what it looked like at the sending end. Any other termination will result in a distorted received pulse due to reflections. If your source has finite resistance then you still get a distortionless pulse at the receiver, but attenuated. Basic rule of thumb is that the problem is appreciable if your return signal times are > the rise/fall times of your pulses.
Alternatively, one can show that if the receiving end is unterminated (infinite load) then, if your source resistance is Z₀ the pulse will again arrive undistorted. This is a broad subject and you'd need to spend quite some time studying it to understand.

About your laser ideas, what you say is all qualitatively correct but a lot depends on modulation mode, frequencies, etc. I suggest you play around with this empirically as this too is a broad discipline.

 

[galaxy_twirl]

Thank you rude man, for your replies.

I am studying this topic at the basic level, I guess, as I am a Year 1 student doing electrical engineering. I shall keep your responses in mind when I wire up my circuits. :)

Thanks again!