Can someone give me a simple explanation of impedance?

[OverDriven]

I apologize in advance for the noobie question, but I have looked everywhere and I cannot get my head around impedance.

I recently saw an analogy comparing a flashlight bulb powered by 4 AA batteries to a headlight bulb powered by a car battery. The flashlight bulb would represent a high impedance input and the smaller batteries a high impedance output. The car battery/bulb system is low impedance. If you mismatch the headlight bulb with the small batteries, it won't have enough power to light and the bulb will quickly drain the batteries. If you hook the small bulb to the car battery, it will light, but will not be using the car battery to it's fullest potential. I can understand this analogy, but when I read this paragraph last night:

A cathode follower is a means of coupling two sections of an amp together with minimal "insertion loss" or degradation. It has very low input capacitance and a very high input impedance so it will not load down the previous stage, and very low output impedance so very little signal is lost even when driving a very low input impedance. The cathode follower is an excellent buffer stage for driving a tone stack, effects loop, power valve or any circuit which would otherwise present a heavy load to a "normal" stage.

I am completely lost. If it has high input impedance, doesn't that mean it would severely IMPEDE the signal going into it? Can someone explain this impedance thing better to me in simple terms? Thanks in advance for any help!

[Enzo]

IMpedance is nothing more really than AC resistance. A resistor in a DC circuit has impedance, but it is just its resistance. In AC circuits capacitive and inductive effects can alter the response to a signal in different ways than simple resistance, so we add the capacitice and inductive reactance to the simple resistance, and the result is the impedance.

That is about as simple as I can make what impedance is. But impedance as a concept is used in many ways. Look at a battery and a light bulb. Current flows from the battery through the bulb and back to the other end of the battery. If we added a high resistance in series with the bulb, it would indeed resist - or impede if you like - the flow of current. But what if I connected that high resistance in parallel with the bulb? In other words wire the resistor right across the bulb socket's terminals. Assuming it is a large resistance like 1 meg - 1,000,000 ohms - it won't do much of anything to the bulb or its current. That 1 meg doesn't load down the battery. WHat if it was a 2 ohm resistor>? That WOULD load down the battery, and likely the bulb would get dim.

A high impedance input to something means the something won't load down the signal, it doesn't mean the impedance is in the way of the signal. It is like the resistor in parallel with the bulb unstead of in series with it. Make sense?

Think of this, you guitar pickups make a small voltage with very little current behind it. If you plug the guitar right into a speaker, the speaker's low impedance - 4 or 8 ohms or whatever - will totally load down the poor guitar signal. SOmewhere in the world of physics, the guitar signal is actually moving the speaker cone, but it is such a tiny amount you won't detect it. But with a high impedance input, the amplifier circuit won't load it down at all.

A low impedance output is more like you were thinking - a low impedance source (output) has littel resistance to impede the flow of current.

My description here is very crude and might not get you many points on an electronics exam, but I hope it makes it a little less fuzzy

[Steve Conner]

Indeed, the confusion is over the difference between series impedance and shunt impedance.

Series impedance is in the way of the signal, so you want to keep it low. An example would be the wires leading from your PA amp to your bass bins. Just try it with 1000' of bell wire, and then change it for 20' of #12, and see the difference.

Shunt impedance wastes signal by shorting it to ground, so you want to keep this one high. This is the impedance that is implied when the previous poster talks about "input" impedance.

[AkisTzortzis]

This is how I understand it (I am sure people will jump in to correct me)

Impedance is like resistance except it applies to AC circuits. AC behaves differently to DC, for example it jumps over a capacitor and gets slowed down by a coil. At high frequencies enough a "capacitor" is the distance between two tracks on the motherboard and a "coil" is a slightly bent track on the PCB. To measure impendance you cannot simply apply a DC voltage across your circuit and measure the voltage drop, you'd need to feed an AC signal and then measure the voltage drop across different frequencies, and then I am sure there will be a formula to put all that into a single value.

This "cathode follower" sounds like a "common collector" transistor equivalent. Remove all transistors and all tubes and think only of resistors. If we make a simple potentiometer, two resistors in series, and let's assume that each resistor is 10K. Let's also say the voltage across both is 12V. Since both resistors are the same the voltage drop across each is 6 Volts. If I plug my multimeter or other measuring device right at the junction between the two resistors I will measure 6 Volts. However if I plug a light bulb at the junction I am introcing a significant load, which means a very small resistance, and I have put that very small resistance in parallel with the bottom 10K resistor. Let's say the light bulb is 1K, effectively I am ignoring the 10K resistor because current prefers to flow through my light bulb. The circuit now has changed because we have put a 1K in parallel with the 10K, and since the 1K is so low (comparateively) the 10K can be ignored (or use the 1/R1+1/R2=1/Rtotal), without even looking at it I'd say the effective load is like 950Ohm. Which means I do not have 6 Volts at the junction anymore, I have a 10:1 ratio now, so I probably have more like 1 Volt !!! My 6 Volts has become 1 Volt. This is because by "loading" the circuit I have pulled significant current to alter the circuit's behaviour. My light bulb has a "low impendance" and as such it will "load" any circuit I will ever plug it on. In contrast my multimeter is "high impendance" and will not load the circuits I plug it on, and that is actually it's job, otherwise you'd not be able to measure anything. "Tapping" is also the same thing, plugging a seriously high impendance load onto a circuit so that you "steal the information" without the main circuit realising. In other words instead of the light bulb you attach a 100K resistor which is so high (compared to the 10K) that there is very little current flowing through it and the main circuit (the two 10K resistors) did not even realise.

It is therefore preferred to have input stages of high impendance so they do not load the previous stage (and so do not alter their characteristics) and have output stages of low impendance (lots of current flowing through them with no load attached) so that when you do attach a load the diverted current will not be a big deal (comparatively) and the voltages will not be affected.

However high impendance results in lots of noise too, so a balance needs to be maintained.

Finally it is neither voltage or current but energy. When you strum your guitar the mechanical energy of the strings gets converted to electrical energy at the pickups and then travels down the wire to the input stage of your amplifier. If the input stage is very very high impendance then you are borrowing a very small proportion of the original signal which you then proceed to amplify in all sorts of ways. That means your "original" is tiny and when you blow it up, when you zoom it you have introduced your own stuff. If your input stage is lower imepndance then you borrow more of the original signal and this serves better as the basis for amplifying it later. However if you borrow too much then you start altering the signal itself before you even process it. Apparently you want all the electrical energy of your original signal to be used to modulate the DC currents (and thus amplify) and not simply waste this energy on passive components that do nothing.

[BackwardsBoB]

Are you bating me?

Impedance is a fancy word for resistance. The impedance of a resistor is its value. Now, when you mix in capacitors and inductors and stuff, resistance becomes frequency dependant, and we start calling it impedance. We get used to it, and start talking about the impedance of resistors too.

Those of us who went to nerd school actually do impedance calculations using complex numbers that include the square root of -1 as a factor. We call them imaginary or complex numbers, and use them anyway.

Sometimes resistance can vary with voltage. A light bulb or tube heater is an example. Resistance is higher at higher DC voltage (once the thing warms up).

Sometimes impedances are discontinuous. A grid is an example. It's a very high impedance (because its a very high resistance) until the grid voltage gets above the cathode voltage, where current flows, and the impedance (resistance) drops. Diodes are another example.

Now let's imagine a three terminal device consisting of a bicyclist, two volt-meters, and a bicycle connected to a DC generator. The black leads of the two voltmeters and the black wire from the generator are connected together, and the red lead of one voltmeter (voltmeter #2), is connected to the generator output. This isn't a very practical circuit, since it is large, expensive, low-output, eats, defecates, sleeps and cheats. But it is a three-terminal circuit, with the red lead of voltmeter #1 and ground on the input, and ground and the generator output on the output.

The cyclist has been well trained to pedal so that the reading on voltmeter #2 (connected to the generator) matches the reading on voltmeter #1 (with the red lead in his hand). You can connect the - end of a battery to ground, and the cyclist will read the battery voltage with voltmeter #1, and start pedalling so that the reading on voltmeter #2 matches. You can put a low resistance on the output of the generator, and he'll pedal like mad, and you can put a high resistance on the output of the generator, and he'll pedal slow, making sure that the two voltmeters match.

The cyclist is effectively a cathode follower in pink spandex.

The input impedance (in this case, resistance), is the input impedance of voltmeter #1 (in his hand), and it better be really high, or it will load down circuits, make false measurements and cause manufacturer returns.

The output is the generator, and it puts out the same voltage as the input, regardless of load impedance (resistance). We can model it as an ideal voltage generator in series with a 0 resistance (impedance) resistor. The output impedance is 0, which is very low.

In this simile, the red lead of voltmeter 1 is the grid of the cathode follower, and the output of the generator is the cathode of the cathode follower. The cathode follower puts out a few more volts than the grid and doesn't need sustenance or clean-up, but it does need a fairly high voltage on its plate.
In both cases, if you put too high a voltage on the input, or put a way-too-low impedance on the output, things don't work out well.

Let us return to our cyclist circuit. It has a very high input impedance (resistance). You could hook up anthing, maybe even a static chage, and the output would match the voltage. If the input were low impedance (resistance), and you hooked up a 9V battery, the battery voltage would drop due to the divider with its own internal resistance. So we want high impedance. Doesn't have anything to do with impeding.

On the other hand, if the cyclist took a break, and we replaced the cyclist generator thing with a resistance from the input to the output, we'd want a very low impedance (resistance) to get as much of the battery voltage to the output as possible without dropping voltage and power in the resistor, delivering it to the load. But that's not like a cathode follower at all. This corresponds to Steve's series/shunt thing.

So cathode followers are as funky as guys in pink Spandex. Why don't we use them for all our tube circuits?

Ain't no gain. voltage in equals voltage out. A guitar puts out about 100mV, which would give you 0.8W into 8 Ohms if 8 Ohms weren't a ridiculously low impedance (resistance) to load a guitar with a 13K output impedance with. We need us some gain.

So instead of taking the output off the cathode, we take it from the plate and get something less than the voltage gain (mu) of the tube. Unfortunately, this circuit has a very high output impedance (the plate resitor value in parallel with the plate resistance of the tube), which will be 60K Ohms or something.

If we are very clever or copy neatly, we can follow this gain stage with a cathode follower, and get voltage gain and low(er) output impedance in one combined circuit. Maybe drive a reverb tank.

Tubes are fairly high impedance regardless of configuration, so talking about a cathode follower as a low impedance tube circuit is relative. Tubes are good at big voltage swings though. That's why we need output transformers. They convert large voltage swings at low current into low voltage swings at high current to drive low resistance 8 Ohm speakers to decent power levels.

Now you know that output transformers take high voltage, high impedance inputs and convert them to low voltage, low impedance outputs in order to drive low-impedance loads.

Good Night.

[tubeswell]

The cyclist is effectively a cathode follower in pink spandex.
Good Night.

Um Okay - I will keep a watch out for fruity cyclists following cathodes.

[Steve Conner]

LOL, that's surreal! Reminds me of Horowitz and Hill's "Transistor Man" (see fig 2.5 in link below)

http://books.google.co.uk/books?id=b...um=1&ct=result

So there's your first lesson in electronics. Transistors have a guy inside in a traffic cop-like uniform, and tubes have a tiny cyclist in pink spandex instead. That explains why they sound better.

[OverDriven]

Thank you...I think I am finally starting to really understand the concept. You guys are a great help even though I'm asking some pretty basic questions. That's what makes this place the best resource for amp building IMO. Thanks again...

[Enzo]

Is that mirror stuff on the tubes so the cyclist can look at himself?

[Chuck H]

Is that mirror stuff on the tubes so the cyclist can look at himself?

No. It's a rear view mirror.

(WARNING - Quickly advancing electrons may be closer than they appear)