i have been given a task to measure the impedance of a coaxial cable. i was doing some literature study and i came across two terms : transfer impedance and characterstic impedance. i am not able to understand the diference between these two terms for a cable.
Electronic – How is transfer impedance of a cable different from its characterstic impedance
cablescharacteristic-impedancecoaximpedance
Related Solutions
Simple Answer: No. It won't work. Not even close.
Long Answer:
There is a lot more to Ethernet, Cat-5 cables, and networking than just having wire that connects Point A to Point B.
Let's start with Ethernet. You have not mentioned what kind of Ethernet you are talking about. The commonly used 100Base-T Ethernet (a.k.a. 100 megabit Ethernet) uses 2 pairs of wire (4 conductors total). 1000Base-T (a.k.a. Gigabit Ethernet) uses 4 pairs of wires (8 conductors).
Cat-5 cable is made from 4 pairs of wire, with each pair twisted in a very specific way. Each pair is twisted at a different rate from the other pairs, to control the crosstalk between pairs. The insulation dielectric and thickness is carefully controlled, along with the twist rate, to control the impedance between the wires. Cat-5 (and Cat-5e, and Cat-6, etc.) has very specific specifications that must be met to be called Cat-5.
What you are calling "12 Core Coax" does not appear to be coax at all, using the traditional meaning of the word "coax". Rather, it appears to be normal 12-conductor cable with a wire braid shield. The link you provided has no specifications on things like twist rate, impedance, crosstalk, capacitance, bandwidth, etc.
Your first question, "Can I use this 12-conductor cable for Ethernet?", must be answered with a no. We have no way to compare Cat-5 with the cable you want to use. Your cable has no specs, while Cat-5 has very specific specs.
Your next question, "Can I fit two Ethernet links into 12 conductors?", is a definite maybe. I am, of course, ignoring the signal quality of that cable and only looking at the number of conductors. 100 Mbps Ethernet requires 4 conductors. 4+4=8. 8 <= 12. So yes, you can put 100 Mbps on 12 conductors. But Gigabit Ethernet requires 8 conductors. 8+8=16. 16>12. So there is no way to put two Gigabit links on a 12 conductor cable.
Next, you mention using some of the "cores as a common earth". Ethernet is transformer isolated at both ends of the cable and thus does not have a "common earth". If you want to get super technical about things, each pair of wires is in essence its own common earth. But don't read too much into that because you'll just get more confused.
I know what you're probably thinking at this point. You're thinking that you are only running slow-old-100-Mbps-Ethernet and all of this impedance controlled, twisted stuff can't be that important. And that you'll just chance it and run 100 Mbps Ethernet over the "12-core Coax" and hope for the best. I have talked to many people in this same position. People who ignored the advise of experts (not just me) and tried to run Ethernet over non-standard cables. All of these people have deeply regretted that choice-- in some cases they "lost" thousands of dollars in rewiring costs, and in other cases they lost their jobs. Don't do it!
But here is one solution that might work for you (and I have actually done this with success): Use standard Cat-5 or Cat-5e cable and run two 100 Mbps links over it. Two pairs for the first link, two more pairs for the second link. To make things better, Cat-5/5e cable is probably a lot cheaper than that 12-conductor stuff you found. Just make sure that when you select the wires for each pair of signals that you actually select pairs of wires! Just taking a Cat-5 cable and not paying attention to which wires are twisted together is guaranteed failure.
Imagine, for a minute that electricity travels quite slowly.
When you turn your light switch on what happens? Current starts to flow - it starts working its way down the wire and so does the voltage. The current that flows is determined by two things: -
- The voltage and
- An "impression" of what the load resistance might be.
Will the current be too small or will the current be too much? It is the cable (and its properties) that dictate the amount of current flowing.
Voltage and current are traveling to the "unknown" load and because V and I are flowing there is power flowing (P = VI). When the current and voltage reach the bulb, if the bulb's resistance doesn't match the V/I relationship not all the power is consumed.
This means the excess (or deficit) of power has to be reflected back up the wire to the switch. It's got nowhere else to go.
In the real world of data comms or radio, this causes "reflections" and these can add or subtract to the forward power traveling down the wire and, in the case of data, it can become misshaped leading to possible data corruptions. In the case of an RF carrier, there will be points along the wire where it appears unmeasurable.
The cable dictates how much current initially flows based (mainly) on its inductance, capacitance and resistance. The formula is this: -
Characteristic impedance = \$\sqrt{\dfrac{R+j\omega L}{G+j\omega C}}\$
R is resistance per metre, L is inductance per metre, C is parallel capacitance per metre and G is parallel conductance per metre. At high frequencies (>1MHz) the impedance starts to largely become: -
\$\sqrt{\dfrac{L}{C}}\$ and if you look at some coax specs you'll see that 50 ohms is the result of this calculation.
why can't I use a 75 ohm in place of a 50 ohm on the output of my transmitter?
Hopefully, by now you should be able to answer this.
Related Topic
- Electronic – Transmission line, cable impedance
- Electrical – How is “signal strength” characterized on a coaxial cable
- Electronic – How does this impedance matching circuit work? How to simulate it correctly in SPICE
- Electrical – Impedence Matching of Antenna and Coaxial Cable
- Electronic – How to estimate characteristic impedance for a random cable and tolerance of the termination resistor
- Electronic – Coax Cable Wave Impedance, Capacitive Load
Best Answer
Transfer impedance:
Transfer impedance is used to determine shield effectiveness at lower frequencies (< 1 GHz) against both ingress and egress of interfering signals.
This is an indication of how "easy" it is for external signals (outside the cable) to couple to the signal being transported by that cable. An ideal cable would block all signals from the outside.
Characteristic impedance:
The characteristic impedance or surge impedance (usually written Z0) of a uniform transmission line is the ratio of the amplitudes of voltage and current of a single wave propagating along the line; that is, a wave travelling in one direction in the absence of reflections in the other direction.
This description is more cryptic, characteristic impedance is a property of the design of the cable. This is not an impedance you can measure with a multimeter ! You need a network analyzer or cable analyzer to measure this directly.
When transporting a (high frequency) signal through a cable you do not want to distort that signal. For this you need a cable which can handle the frequency of the signal (not attenuate it too much). Such cables have a characteristic impedance which you must use to feed the signal into the cable but also to terminate the cable with at the other end.
This is similar to how a transmission line must be used. If the cable is 50 ohms it means that the impedance driving the cable must be 50 ohms and that the termination impedance at the output of the cable must also be 50 ohms. If you would use different values, the signal would reflect and distort itself.
You can also indirectly measure the characteristic impedance of an unknown cable by feeding a pulse into it while observing the voltage at the input of the cable. At the other end of the cable terminate the cable with different values like a short (zero ohm), open and 20, 40, 50 , 75 ohms etc. There will be a certain impedance where the pulse will not be reflected and that is the characteristic impedance of the cable.