Characteristic Impedance

[PJ33]

Can someone give me some intuition of what characteristic impedance is.
When I google it, it's just give me the property, that in order not to lose any energy when an EM wave travels from one medium to another, you need to match the characteristic impedances of the mediums.
But what it actually means that air has impedance of 377Ω and copper has characteristic impedance of the magnitude 10^-2Ω.
I just need some better intuition.

Thank you in advance!

 

[Grossel]

If you have a look at youtube, searching for "impedance matching".

Here is a good one:

 

[crutschow]

A transmission line has a characteristic series inductance and parallel capacitance per unit length.
If you solve that for the lmpedance, you will get a resistive value.
From the maximum power transfer theorem, the line termination resistance must equal that value for the complete transfer of power.
If not then the leftover energy is reflected back to the source.
Make sense?

 

[KeepItSimpleStupid]

Beyond me, I guess: https://en.wikipedia.org/wiki/Impedance_of_free_space

 

[rjenkinsgb]

The best I can come up with is to imagine a wave traveling down a canal that has parallel sides.
There is little attenuation other than friction loss.

If the canal has a step change of width at some point, part of the wave reflects back at that point.
If it gets narrower, the wave reflects back from that part in the same polarity, eg a peak bounces back as a peak.
It it gets wider, a peak produces a trough reflection.


That's pretty much the same effect as with an electrical transmission line, wow a sudden change or discontinuity produces a reflection of part of the signal.

The best explanation I have come up with is it relates to capacitance per unit length; that's vaguely the "width" equivalent.

If you have a couple of full reels of coax and a fast scope, you can make eg. a 1KHz square wave osc and add a buffer capable of driving the cable impedance (like all sections of a cmos inverter in parallel) plus a matching resistor.

Put the scope on the same cable end and try different connections and loads at the end or between two pieces 7 the far end.
Trigger on the positive edge and you will see a half-amplitude edge as the osc output goes high, then that going higher or lower as reflections add to it, with anything but a matched load on the far end of the cable.

It's basically a time domain reflectometer (TDR) setup, that allows you to visualise cable or connector faults.

 

[nsaspook]

Can someone give me some intuition of what characteristic impedance is.
When I google it, it's just give me the property, that in order not to lose any energy when an EM wave travels from one medium to another, you need to match the characteristic impedances of the mediums.
But what it actually means that air has impedance of 377Ω and copper has characteristic impedance of the magnitude 10^-2Ω.
I just need some better intuition.

Thank you in advance!

For a EM wave (a real physical enity that obeys physical laws) traveling in a good conductor one characteristic impedance intuition would be resistance to transverse motion. EM propagation requires this, so if some component (like the traveling e-field) of the EM wave is effectively 'short circuited' until the source driving impedance low compared to the 'short' that is a indication of the media characteristic impedance to a EM wave.

 

[unclejed613]

if i have two parallel wires, each wire has a certain amount of inductance, and there's a capacitance between the two wires. the amount of inductance depends on the length of the wires, and the capacitance depends on the diameter of the wires AND their distance apart. the combination of the inductance and capacitance is the impedance. as the frequency goes up, the inductance increases in it's reactance and the capacitance decreases it's reactance, so with the combination of the two, the impedance remains constant because the reactances change, but their overall sum remains constant.

 

[Grossel]

if i have two parallel wires, each wire has a certain amount of inductance, and there's a capacitance between the two wires

This capacitance impact will strongly depend on whether it's some coax or if bare wire.

 

[unclejed613]

This capacitance impact will strongly depend on whether it's some coax or if bare wire.

i was trying to keep it as simple as possible... transmission line theory can be a deep subject... the point being for the capacitance, it's distance and surface area (which depends on the diameter and, maybe i should have stated it that way...). as a bit finer detail for the OP, for a given transmission line impedance, the proportion of the distance between the wires and the diameter (or capacitive surface area more accurately) stays pretty much the same. if you look at 300 ohm transmission line, the standard 300 ohm flatline has a spacing of about 0.3 inch and a wire diameter of about 0.025 inch (22Ga wire). 300 ohm ladder line for transmitting ("ladder line") using 18Ga wire is spaced at 0.5 inch.

 

[PJ33]

if i have two parallel wires, each wire has a certain amount of inductance, and there's a capacitance between the two wires. the amount of inductance depends on the length of the wires, and the capacitance depends on the diameter of the wires AND their distance apart. the combination of the inductance and capacitance is the impedance. as the frequency goes up, the inductance increases in it's reactance and the capacitance decreases it's reactance, so with the combination of the two, the impedance remains constant because the reactances change, but their overall sum remains constant.

Thank you, this was really helpful!