why can the capacitor be charged by a battery (DC)

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Then how can current flow in a vacuum tube?
Yes, very good, but what if we replace the vacuum in the valve with a typical capacitor dialectric, maybe polystyrene or ceramic .....

Googling "does current flow through a capacitor ?" it's quite easy to find articles that say something along these lines

for a.c in a cap:
"A current will flow through the circuit, first in one direction, then in the other. However, no current actually flows through the capacitor. Electrons build up on the one plate and are drained off from the other plate in very rapid succession, giving the impression that the current flows through the insulator separating the plates.

Abbey I appreciate your technical knowledge is far greater than mine but can you please point me to a technical article to back up what you are saying.
Currently you're saying I'm wrong, I am keen to learn as much as the next person.
 
I think many conflate too closely the notion of "current flow" with that of "electron flow"? In my mind, Maxwell told us this 100+ years ago, by defining "current", as the net charge passing through an enclosed area per unit time. The "absence" of an electron is just as valid a unit of net charge passing "through" a point as is an electron.

Thus when charging a capacitor with DC, an electric field flux is established between the plates (which is the actual "work" being done by the voltage source), which exactly agrees with the net charge concept above.

One analogy I always liked that makes more sense intuitively: if we say current flow is analogous to water flow in a pipe, but what if you insert a flexible plastic barrier in the pipe? Think of a drum head stretched over and inserted in the middle of the pipe. When you begin to apply force to the water, the barrier will flex outwards to a fixed position based on how hard you push. However no water will flow! The drum head will push back equally as hard in order to maintain no net flow of water in the pipe. A bit of water on the other side of the drum head will be pushed out initially (so that the about of water entering the pipe is the same as seen leaving, while the pressure inside rises), however if you monitor water flow after the drum head after a steady state is reached, all you will be able to "feel" is a net pressure (with no actual flow of water).

If you want to transmit a signal down this pipe (and by signal, I mean some way you measure the amount of water that is flowing past), you must constantly vary the water pressure, which will cause the drum head to suck in and out, which will be felt as a modulating water pressure downstream of the drum head. This is your AC "current"/water flow.
 
Last edited:
...... Kirchhoff....
I knew something wasn't right.
I'm just wondering what the conclusion of this board is, is the current going through the capacitor or not?
My position hasn't changed a bit. Now, have I managed to convince anybody, I don't know...
Certainly not the OP, who has long since vanished.
 
for a.c in a cap:
"A current will flow through the circuit, first in one direction, then in the other. However, no current actually flows through the capacitor. Electrons build up on the one plate and are drained off from the other plate in very rapid succession, giving the impression that the current flows through the insulator separating the plates.
Then I must say the person who wrote this is an ignorant. "giving the impression", wow! super science.
Abbey I appreciate your technical knowledge is far greater than mine but can you please point me to a technical article to back up what you are saying.
Matador has given an analogy, I gave one with recirculating balls, and I mentioned Kirchhoff. What else can I say?
 

Attachments

  • Kirchhoff.jpg
    Kirchhoff.jpg
    76 KB · Views: 19
One analogy I always liked that makes more sense intuitively: if we say current flow is analogous to water flow in a pipe, but what if you insert a flexible plastic barrier in the pipe? Think of a drum head stretched over and inserted in the middle of the pipe. When you begin to apply force to the water, the barrier will flex outwards to a fixed position based on how hard you push. However no water will flow!
I think we are over analyzing the meaning of the word "through" in this context but that won't stop me! Does current really flow *through* a capcitor? Sounds like a good debate to have over some beers!

I like the water analogy but would argue that a pneumatic system would make for a slightly better analogy. While fluids have a pressure, they also have a lot of mass and inertia that comes with that. I think a gas like just dry air, is light weight and compressible and therefore might better represent the concept of voltage. I could see a classroom using such a system to make simple "circuits" constructed from common items. Capacitors could be, as you describe, a balloon type of material inserted into a section of pipe with an area widened so that the "drum" can easily deflect in each direction but limited to some maximum to account for the capacitors of different sizes. Resistors would be represented simply as holes of certain diameters perhaps using carburetor jets which are common, precise and cheap. A compressor can be used as a source of "power". Not sure how to implement magnetics or semiconductors though.
 
My position hasn't changed a bit. Now, have I managed to convince anybody, I don't know...

The symbol for the capacitor actually indicates that current does not flow between the capacitor plates (we are talking about an ideal capacitor here). The simplest example that proves this is a vacuum capacitor where there is no any material between the plates, and especially some that would transfer charge. Different insulators between the plates increase the capacity because they increase the possible electric field. But these insulators do not transfer the current defined with 1. Kirchhoff's law.
 
I think we are over analyzing the meaning of the word "through" in this context but that won't stop me! Does current really flow *through* a capcitor? Sounds like a good debate to have over some beers!
Indeed!

I think Maxwell would argue that yes, net charge passes through the capacitor (Maxwell called it a 'displacement current'). If you consider two infinite plates separated by a distance, then the only Gaussian surface that can be placed between without touching them is a plane. Since a charged capacitor has a net field between the positive and negative charges, and that field can be completely encompassed by the measurement plane, then such a field was established by the net charge passing "through" that surface. However unlike a resistor, the net charge per unit time quickly tapers off down to zero once the field is established, unless the applied voltage is changed.

It's explained here:

Past that, the debate becomes extremely metaphysical. :)
 
The symbol for the capacitor actually indicates that current does not flow between the capacitor plates (we are talking about an ideal capacitor here). The simplest example that proves this is a vacuum capacitor where there is no any material between the plates, and especially some that would transfer charge. Different insulators between the plates increase the capacity because they increase the possible electric field. But these insulators do not transfer the current defined with 1. Kirchhoff's law.
So by which miracle current passes from a generator to a load via a capacitor?
In the circuit I drew, do you agree the vectorial sum of the voltages along the loop is zero?
Unless the voltage across the capacitor is equal to the generator voltage, which is not confirmed by experience, there is a voltage across the resistor, so there must be current flowing, and this current has to flow through the capacitor, because there is no other possible way.

"there is no any material between the plates, and especially some that would transfer charge". It's a long time since scientists have established that energy can be transmitted without a physical medium, although some would try to revive the notion of ether.
Why would they need a physical support? If an object needed a physical support to travel, there would be no rockets, no satellites.
Electrons that accumulate on a plate are submitted to a force by the electrostatic field, so they migrate from one plate to the other, being replaced by others that come from the rest of the circuit.
Current is not a flow of electrons, it's a disturbance that is transmitted, just like sound moves at 360m/s, but the air molecules do not travel from the speaker to the listener.
I thought that, being a teach, you would know that.
 
So by which miracle current passes from a generator to a load via a capacitor?

A capacitor in an AC electrical circuit should be viewed as a charging and discharging element, just as a capacitor microphone capsule or electrostatic speaker works.

In the circuit I drew, do you agree the vectorial sum of the voltages along the loop is zero?
Sure. Second Kirchhoff's law.

Unless the voltage across the capacitor is equal to the generator voltage, which is not confirmed by experience, there is a voltage across the resistor, so there must be current flowing, and this current has to flow through the capacitor, because there is no other possible way."
Plain wrong.
You need to ask yourself why that current does not flow if the circuit/generator is DC after the capacitor is charged. Isn't there an electric field on the plates then, why doesn't the charge flow to the other side?
"there is no any material between the plates, and especially some that would transfer charge". It's a long time since scientists have established that energy can be transmitted without a physical medium, although some would try to revive the notion of ether.
Why would they need a physical support? If an object needed a physical support to travel, there would be no rockets, no satellites.
Charges (electrons) do not transform into the photons inside the capacitor.
Electrons that accumulate on a plate are submitted to a force by the electrostatic field, so they migrate from one plate to the other, being replaced by others that come from the rest of the circuit.
No way.
Current is not a flow of electrons,
Yes it is, but electrons move at few cm/sec, current slightly below c.
it's a disturbance that is transmitted, just like sound moves at 360m/s, but the air molecules do not travel from the speaker to the listener.
343m/s
I thought that, being a teach, you would know that.
Of course.
 
.......


Ripple current in selecting a reservoir capacitor for a power supply, or speaker coupling capacitor at the output of a single supply power amp.

Why a fully discharged capacitor draws huge current at t=0? (theoretically infinite, practically finite due to ESR).

The reason why, say a 10,000 uf 100V capacitor with a ripple current rating of 1A measures 15mm in diameter while the one with 10A will measure probably around 100mm in diameter.

Edit. I can see that Moamps has disected Abbey's later responses. But I firmly stand by that AC current flows through the capacitor. Think about ESR. If there is no current flow then there would be no heat dissipated across it.
 
Last edited:
Isn't it fascinating that something so fundamental to electronics is still debated. :)

You need to ask yourself why that current does not flow if the circuit/generator is DC after the capacitor is charged. Isn't there an electric field on the plates then, why doesn't the charge flow to the other side?

Because once charged, a DC voltage source cannot supply anything other than the electromotive force to hold the net charge constant. Once charge stops moving, then current must also be zero.
Charges (electrons) do not transform into the photons inside the capacitor.
But electrons aren't charges....they have charge. Nor do they need to change inside a capacitor, right?

Also, interpreting abbey's diagram that way isn't fair, because he specifically shows an AC voltage source. If it was DC, there would be some action at the beginning, but all currents would drop to zero once the cap's terminal voltage equalled that of the voltage source. Hence all current are equal, but zero (not the interesting case).

To be honest, I was perhaps 15 years into my career after obtaining my BSEE and BSCpE degrees before it began to gel what was really going on, and how these pieces were interconnected. Abbey's description of what's going on exactly matches my understanding of what Gauss, Faraday, and Maxwell were saying, but I would be open to hearing how others interpret it as well.
 
A capacitor in an AC electrical circuit should be viewed as a charging and discharging element,
And when charging/discharging, no current flows? Q=I.t. If there is a charge, there needs to be some current over some time. This current does not stop at the electrode. How come the current measured on one side is exactly replicated on the other side.
Just like the pressure that is applied at one side of a hose is replicated at the other end (notwithstanding losses, that are dynamic, not static)

If this is not passing through, I don't know what your definition of through is.
What wording would you recommend for a substitute to "passing through"?

just as a capacitor microphone capsule or electrostatic speaker works.
Not a very good example.
A) You use a particular case to examplify a general case.
B) Wrong example: a condenser mic is presumed operating with constant charge
Sure. Second Kirchhoff's law.
OK. then why do you disrespect his first law?
Plain wrong.
You are entitled to think I'm wrong...
You need to ask yourself why that current does not flow if the circuit/generator is DC after the capacitor is charged. Isn't there an electric field on the plates then, why doesn't the charge flow to the other side?
Corect. I need to re-think my claim.
Yes it is, but electrons move at few cm/sec, current slightly below c.
That's exactly what I was referring to earlier. At the atomic level, the electrons that "move" at one point of a circuit are not those that circulate at another point. But on an electrical POV, the current is the same. Just like the flow of water at one end of a hose is the same at the other end, but the water molecules are not the same.
Typo; 340m/s is a good enough approximation. geography dependant.
 
And when charging/discharging, no current flows?
It flows, but not through insulators between plates. It just charges and discharges plates.
Q=I.t. If there is a charge, there needs to be some current over some time. This current does not stop at the electrode.
Yes, it stops. If not, a capacitor is faulty.
How come the current measured on one side is exactly replicated on the other side.
Electric field is the key. As magnetic field in transformers.

If this is not passing through, I don't know what your definition of through is.
What wording would you recommend for a substitute to "passing through"?
The current we measure on capacitors leads do not go through capacitor from one to another plate.

Not a very good example.
A) You use a particular case to examplify a general case.
B) Wrong example: a condenser mic is presumed operating with constant charge

After you, a condenser microphone will be impossible as concept because the AC voltage generated with displacing the membrane will be dumped with the current passing through/between plates.

OK. then why do you disrespect his first law?
Not at all. It is preserved.
Corect. I need to re-think my claim.
You should.

Agreed. We are wasting a lot of electrons and BW for a matter of semantics.
No. We are discussing here the fundamental principles. That isn't semantic.
 
. Think about ESR. If there is no current flow then there would be no heat dissipated across it.
The heat is mostly generated by the current flowing thru metal plates. The plates are in practice made of a thin material wound into rollers or such to obtain a large effective surface of the plate. This material is heated by charging and discharging the capacitor.
 
The current we measure on capacitors leads do not go through capacitor from one to another plate.
Current is not individual electrons. In an electrical circuit, if you put a cassock on a particular electron, you may never see it at the finish line.
If I follow your drift, current does not pass either in a copper wire, because copper is constituted of 99.999xx% vacuum.
And the concept of "capacitor impedance" is wrong, because impedance and current are related.

Would you say that a xfmr does not pass power "through" because the electrons that generate the magnetic domain rearrangement are not the same that the secondary shakes?

After you, a condenser microphone will be impossible as concept because the AC voltage generated with displacing the membrane will be dumped with the current passing through/between plates.
There is current only when there is a variation of voltage across a capacitor. In a condenser mic, this current is extremely small, because the RC product is high enough, but is real. Since it's AC, the average value is zero, so what is "dumped" during one phase is reinjected at the following phase.
 
Current is not individual electrons. In an electrical circuit, if you put a cassock on a particular electron, you may never see it at the finish line.
You will never see it on the finish line if it is a capacitor as an obstacle. There is an insulator that prevents it from doing so.
If I follow your drift, current does not pass either in a copper wire, because copper is constituted of 99.999xx% vacuum.
Haven't you ever heard of the term "free electrons" in metal? Are you trying to twist what I said?
And the concept of "capacitor impedance" is wrong, because impedance and current are related.
No, that concept is completely valid. The current referred to here is the charging and discharging current.
Would you say that a xfmr does not pass power "through" because the electrons that generate the magnetic domain rearrangement are not the same that the secondary shakes?
No, you obviously misunderstood something I said, which is not impossible because English is not my native language.
There is current only when there is a variation of voltage across a capacitor. In a condenser mic, this current is extremely small, because the RC product is high enough, but is real. Since it's AC, the average value is zero, so what is "dumped" during one phase is reinjected at the following phase.
Does this current flow through capacitor capsule, from one to other side? If not, why not in your model?

Utterly sterile discussion.
Like the sex of angels.

Okay, we can conclude that you think the current is passing through the capacitor, I don't think it's passing. Well let everyone if they’re interested do a little research and they’ll find out what’s right.
Let's move on.
 
Back
Top