capacitors and audio signals

As i was sitting there modding away last night i realized that i don't get a fundamental concept of caps.

If a capacitor blocks any current after it is charged up, how does any audio signal get to the other side of the circuit??

simple question but every resource explains the charging of caps but not how they work in practice.

thanks

A capacitor as an ideal circuit element behaves according to this equation:

instantaneous current (in amperes) = capacitance (in farads) times instantaneous rate-of-change of voltage (in volts).

The current is the current through the capacitor and the voltage is the volatge across the capacitor.

So for example if I could generate a voltage ramp, i.e. say 1 volt increase per second, and applied this to my 1 Farad cap (a really big cap btw) I would be pulling 1 ampere exactly and continuously from my source of the voltage ramp.

But sooner or later I would run out of available volts! So at that point I flip my ramp polarity switch and start to ramp DOWN now at the same ramp magnitude but the opposite polarity. Abruptly my current goes from 1A to -1A, and my voltage is now going down at 1V per second.

That's one way you get current to keep flowing in a capacitor. I will leave the sinusoidal voltage case to another poster ;-).

hey, what caps do is block DC but AC will "pass" through them. however in reality the frequency cutoff is also determined by the value of the cap.

do some searches on here as there have been many posts about this before.

:thumb:

EDIT: I see Bcarso has again beaten me to the punch with a much more elegant explaination than my layman's terminology..

thanks for both explanations.

It seems that the ac current isn't blocked (only dc) and the frequency of attenuation is dependent on the value of the cap.

thanks

for all reasonable purposes, yes that's a roger.

but again, that's a simplified way of putting it, to go into depth would likely make more confusion so we'll leave it at that for now.

:thumb:

Some points about wording

current does not go though a cap(except leakage and DA) electrons move via other parts of the circuit to the plates.

Current does not flow(like water etc) If I understand corrrectly electrons get kocked out of the outer cloud one goes in one goes out at normal temps(not talking cooper pairs).

I think the simple flow idea can hurt a persons understanding

it is all probability QED

This is how I imagine a how a capacitor works.. correct me if I'm wrong.

When voltage on both sides of the capacitor stays the same nothing happens. even when the voltage on one side is higher than on the other side. cause a capacitor doesn't conduct electricity.

When you increase the voltage on one pin of a capacitor it will charge and "push out" electrons on the other side. And when you decrease the voltage on one side it will "suck up" electrons on the other side.

So a varying voltage on one side will cause a varying voltage on the other side. But as soon as you stop varying the voltage (dc) everything will return to normal. Bigger caps will be better equiped to slower varying voltages so they will form less of an obstacle for lower frequencies.

yeah the current/water analogy gets screwy when you go beyond elementary electronics but it is similar as the electrons jump orbits from atom to atom until the potential of the system reaches equilibrium just like water would flow if it were pressurized. the pressure would be voltage and the amount of flow would be current. if it were turning a wheel you would need two things, pressure(voltage) and flow(current) which would be doing a form of work equaling a measurement of power(wattage). the analogy is not to explain the way electrons move in the system but rather just equating the net movement to something simpler in idea and tangible to something that is unseen and untouchable(unless you like getting shocked..)

as for the caps:

It's not so much that the lower frequencies are "slower" as it is the lower the frequency the more energy it contains. the larger the value of the cap, the more charge it can hold and the lower the frequency it can sustain. If you look at the opposite, since lower frequencies are higher energy, a smaller value cap cannot support it and the response through the cap suffers.


again sorry for the layman's terms. I really don't care for geekspeak when trying to convey ideas for everyone..
:guinness:

> how does any audio signal get to the other side of the circuit??

The voltage across a cap changes "slowly".

How slow depends on the resistance in the circuit.

For audio, we usually pick a combination of R and C so that it takes most of one second for the cap voltage to change a lot.

So during a 20 cycle per second wave, the voltage changes very little.

Say you have an amplifier output that sits at 100V DC, and an input that has to sit at zero V.

Put a cap in between, with proportional resistors on each side.

When you turn-on, the "100VDC" side naturally starts from zero (dead) and rises to 100V DC as the tubes warm up, around 10 seconds.

The "zero V" side tends to rise, but the resistor on the zero V side "slowly" drains off the capacitor.

The capacitor ends up (after many seconds) with 100V across it: 100V on one side, zero V on the other side.

Now bump the "100V" side to 101V. Instantly the "zero V" side rises to 1V. And then it drains away, "slowly". But audio waves are always changing, back and forth, they don't stand still. If the capacitor-resistor combination is a good size for audio, you get a 1V jump that starts to drop to 0.9V but barely gets there in the time of a 20Hz wave. So all audio 20Hz and up is passed at 90%-99.9% of the original level, 1Hz (which we don't want) passes at about 10%, and DC is nearly zero. (It would be zero if we could wait for infinite time.)

> current does not go though a cap... electrons move

Capture an electron in a bottle and show it to me.

"Current" DOES "flow" in a capacitor. Electrons (if they really exist) don't cross the insulator, but so what? Trucks don't cross oceans, but you put a part on a truck in Japan, a week later you take that part off a truck in NJ, who cares about the truck?

> current does not go though a cap... electrons move

Capture an electron in a bottle and show it to me.

"Current" DOES "flow" in a capacitor."

PRR, I think what they meant was that the belief was that current "flows" through the cap, as if you were to tag one single electron and put it in one side of the cap that it would pop out the other side eventually...

wow, im shocked at such an awesome reply to such a basic question...thought it might get ignored. thanks..

i tend to try to think of circuits as audio paths instead of charges and loads. thats why it seemed like the audio path was being cut if the 'music' couldn't get through the cap to the other side.

in-----ll------out

> if you were to tag one single electron

My point is: this experiment can't be done. Electrons are probably real, or at least useful concepts. But you can't tag one. And there are so MANY electrons in any conductor that, even if they had a "real" aspect, any single electron is unlikely to "flow".

What actually happens in a cap is: there are so many electrons that a few more or less on each plate hardly matters. But it matters a little, because all those electrons push on all other electrons in the universe, especially the ones on a nearby conductor. If those electrons have a place to go, one of them will wander away.

By that accounting, there is a maximum amount of charge that a cap can hold: the number of electrons in its conducting plates. But even in atom-bomb ignition caps and other hard-working condensers, the total number of off-balance electrons is a Teentsy fraction of the total electrons.

In practice: the electrons in the insulator also lean one way or the other. If they lean too far, they break loose, they flow, they rip apart the molecules of the insulator, and we have a burnt-out capacitor.

There is a great piece by the late Dick Feynman about what really goes on when a cap is charged. He talks about it in his fashion as if he were considering it for the first time, and gets deeper, and deeper, and...

I used to think it was somewhere in the classic three-volume Lectures, but alas I have those in storage (@#$%^&&). Actually it may be somewhere else--maybe I do a leetle googling, no??

Another great book to help one unlearn everything is Carver Mead's Collective Electrodynamics.

I could be way off here but isn't a capacitor like a thunder cloud striking the earth? The earth is one side of the capacitor and the cloud is the other. Right?






I am new to the game,so don't bash me too hard if this isnt right.
Patina