A Pedantic Question

[nsaspook]

When is it of use to know/remember that individual electrons don't get from one plate to the other? Does this need to be in most people minds when using caps in electronics?
...
2. He is a professor of what exactly?

Is he a professor of practical electronics teaching basic electronics students?

It depends on what you are working on, if it's an led flasher then no but if it's a circuit that depends of very low currents, very high currents or high speed devices where the size of the components are close to the wavelengths of the signals moving in the circuit then yes. At this point the type of dielectric in a capacitor and even the material used in the plates becomes a important factor in good design. At this point understanding the physics of a capacitor enables you to make a intelligent choice of the type of device based on science instead of a rote selection from a book.

Knowing the basic physics of electronics enables you mentally visualize (it's very hard to express in words but imagine you see a circuit board as a layer of fog that condensed into shaped fields as components instead of seeing little electron balls whizzing around) electronic circuits in a different way that's not really useful when building a circuit board from a book but it's extremely useful when debugging or repairing a circuit board built by someone else where you have to analyze the functions to find a defect.
https://www.wiley.com/WileyCDA/WileyTitle/productCd-0471222909.html

...
I'll let the Dr. speak for himself: https://www.youtube.com/watch?v=sF-m3XZKvLI

 

[Ratchit]

shortbus,

Doesn't this explain how a cap "seems" to transfer AC?

No, because current "appears" to exist through a vacuum capacitor where no conduction or polarization exists. The details are complicated, but a simplified true explanation is that electrons pile up on one plate and delete on the opposite plate, thereby causing a current in the branch containing the cap for a transient period of time. No need to get wrapped around the axle explaining why it happens according to energy fields, or accept the lie that current exists through the capacitor. Simple explanations if desired, but always true is the best way.

Ratch

 

[shortbus=]

^This must be why "the Art of Electronics"(Horowitz and Hill) says that the math and physics of caps would take a whole book on themselves.

 

[MrAl]

^This must be why "the Art of Electronics"(Horowitz and Hill) says that the math and physics of caps would take a whole book on themselves.

Hi there,

Perhaps, but i think the basic understanding is not too difficult. A charge goes in one side and gets stored on that side, a different charge that was stored on the other side comes out that side. So the two sides have some independence from each other. The two sides are however connected by an external circuit so we see a potential difference across the two sides, and we see one charge going in and one coming out so it looks like just regular current flow to the external world.

 

[steveB]

Hi there,

Perhaps, but i think the basic understanding is not too difficult. A charge goes in one side and gets stored on that side, a different charge that was stored on the other side comes out that side. So the two sides have some independence from each other. The two sides are however connected by an external circuit so we see a potential difference across the two sides, and we see one charge going in and one coming out so it looks like just regular current flow to the external world.

MrAl,

I like that description. It is simple but it hints at a very interesting fact about capacitors. That is, we use them in a circuit mode operation. Circuit theory assumes no charges are built up in the device, which forces current in one side to be matched to that on the other side. If we deal with the more general case, as a physicist would do, we should immediately see that there is no reason why charge on one plate must be the negative of that on the other plate. If we explore this idea further, we would find that the value we call "capacitance" is really only completely desciptive in the special case of circuits. Two plates actually requires a 2x2 matrix of capacitance values. In circuit mode, the one value suffices, but in general it does not.

Students are often surprised to learn later (if they learn it at all) that a single conductor (consider the Van de Graaff generator) has capacitance. There, one capacitance value suffices. Students usually scratch their heads and can't make any sense of this. Well, that's because they were never exposed to the generalized theory of capacitance. For some reason, this subject is almost completely ignored. It should at least be discussed when a student gets to studying electromagnetic theory, but it seems to me it is ignored even there. The generalize capacitance theory is elegant and simple and would use-up only 2 pages of that "master volume" on capacitors, but it seems to be kept as a big secret, like a can of worms that no one wants to open.

I think what your well-chosen words show is that it is possible to keep things simple, but plant the seeds in a student's mind that will make them ask questions later. They will be using the simple ideas immediately, but the questions will work on their mind so that they are receptive for more details at a later time.

 

[atferrari]

A charge goes in one side and gets stored on that side, a different charge that was stored on the other side comes out that side. So the two sides have some independence from each other.

The two sides are however connected by an external circuit so we see a potential difference across the two sides, and we see one charge going in and one coming out so it looks like just regular current flow to the external world.

Wish I heard this (two-parts) explanation the first time I met the subject.

It seems that telling things as they are is not pedantic but simple AND useful. Gracias.

 

[MrAl]

Hello again atferrari and Steve,

atferrari:
De nada
It is pretty amazing isnt it? To think that one of the plates is getting more charge as time goes by and the other is actually giving up charge that it had as an originally neutral material. This kinda begs the question of whether or not we could ever run out of free electrons in the depletion plate. Turns out it looks like we can never run out, at least not without building a capacitor that is made specifically to see if we could actually run out of them, and even that may not be possible as i will get to next...

Steve:
Yes no reason why one plate could not have already had some charge or whatever. But if you could clarify what you are using the 2x2 matrix for that would be nice to hear too.
Also, as i was saying above, my attempts to come up with a physical cap that can be used to test this theory seems to turn up the fact that any cap we might find in the parts store would never run out of electrons in the depletion side simply because there are so many of them in even the thinnest of material. One of the constraints here is the voltage buildup between plates. As the cap charges and we loose more and more electrons from the depletion side before we get to the last electron the voltage builds up to such a high voltage that it would cause a spark to jump across the plates and that would end the experiment without ever being able to remove the very last electron.
Could it be that nature has a built in safety mechanism to NEVER allow the cap to run out of electrons?
If you care to take up this calculation it would be interesting to hear your results. I came with a quick estimate for a 1 inch square plate capacitor (2 plates each 1 square inch area on one side of each plate). To see one plate run out of electrons before the voltage sparkover seems to require a copper plate thickness of 8e-14 inches. That's extremely thin and even smaller than the diameter of one copper atom so it appears that it cant be built out of copper, and i have to wonder if there could be another material that could be used instead. Because it has to be so thin i dont see how just yet. But it would be interesting to hear your take on this too.
So for now it looks like we can never build a capacitor that can run out of charge on the depletion plate, but i'd be interested to hear more on other peoples thoughts on this too.

 

[ericgibbs]

hi Al,

Isn't a capacitor plate/conductor that has no free electrons an insulator?

E

 

[steveB]

MrAl,

A little later I'll post a couple of pages of text that discusses the generalized capacitance. I"ll attach it to this post with an edit later (EDIT: I've attached some pages from "Principles of Electrodynamics" by Melvin Schwartz. This section on The General Theory of Capacitance is one of many gems in this book). I don't want to go into great detail and sidetrack this thread, but it would be good for people to have a good reference to use in the future, should they ever need it.

On the charge running out question, I think any known conductor would explode before charge could run out. The negative side would have electrons flying off better than thermionic emission of any tube ever built and the positive side would have huge stresses that the material could not hold.

 

[Ratchit]

Eric,

Isn't a capacitor plate/conductor that has no free electrons an insulator?

Not exactly. The metallic plate of a cap has no free electrons when it is not energized. Each electron is bound and assigned to a metallic ion to make a electrically neutral material. A good conductor like a metal has its valance electrons very loosely bound, so they can be easily dislodged by a small voltage. A good insulator material holds its valance electrons tightly, which makes conductivity difficult. No need to worry about running out of electrons. They are exceeding great in number, and the back voltage caused by the charge separation will increase past any practical ability to implement further separation.

An N-type semiconductor does have a limited number of free electrons from doping, but that is another story.

Ratch

 

[ericgibbs]

hi Ratch,

I was being a little flippant with Al.

I am already well upto speed on the structure of the atom [ as we assume it to be at this time] and the doping techniques used in creating semiconductors, but I do appreciate you taking the trouble to post that info.

Eric

I would suggest you re-look at your understanding of 'free electron' theory in metals, attached a short pdf for your edification.

 

[killivolt]

As for me, Magic Smoke, Voodoo Electronics, EM Coupling, Capacitance, Electrostatic Charge, Double Slit, Current vs. Voltage control etc. It's what keeps me wanting to do "Hobby Electronics"

All of the Engineers talking about such things, is like sitting in the same room as Wizards and Warlocks speaking of their conjuring.

There is still magic in Electronics and will always be in my opinion, I don't see a loss in any conversations like these.

It really makes me think and want to seek more information about such subject matters even though it's well above me. Only to have a glimmer is enough for me.

Now, did I take the Blue Pill or the Red One. I can't remember all the conversations and debates like these that have opened my mind to "Electrons and Electron theory" it certainly will help many young seekers of such knowledge.

Regards to all the "Warlocks and Wizards"

With all do respect,
kv

 

[MrAl]

Hi again,


Eric:
Actually i was thinking along that line too if the free electrons could somehow be all removed, we'd be left with a huge resistance. That would stop the current flow.

Steve:
Yes there is also the constraint of the physical forces involved. I was looking at merely the voltage involved and found it amazing that for a 1 inch square capacitor with plate thickness of around 1/100 inch and plate separation of 1/100 inch would have to be charged to something like 100 trillion volts in order to remove all of the free electrons from the depletion side plate. That alone is quite amazing i think. That tells me that nature has a built in safeguard for making sure that capacitors always capacitate Seriously though this is amazing because if there was a lower limit some capacitors might stop working because they ran out of free electrons. But this also reminds us just how many electrons there are in materials like copper, and multiply that by 29.
But then i was looking at what it would take to build such a capacitor and found that if the plate is too thin then it has a current limit which means it might take too long to charge in order to run a real life experiment. I dont want to have to wait 200 years for a result.

Ratchit:
Yes i think as Eric indicated the free electrons are always there but we just dont use them. We call them free even if they are still held to the electron isnt that right? That's how we know how many we have to work with in the first place from the total of 29 or so. Just a small point here though.

killivolt:
Yes these things keep us interested dont they? Another interesting point is that in classical theory the air core inductor field reaches (at some future point in time) to the ends of the universe.

 

[Ratchit]

MrAL,

Yes i think as Eric indicated the free electrons are always there but we just dont use them. We call them free even if they are still held to the electron isnt that right? That's how we know how many we have to work with in the first place from the total of 29 or so. Just a small point here though.

I don't call them free because the valance electron is loosely bound to its copper atom. An atomic number of 29 means copper, which has 1 valance electron. There is no way you are going to strip off any electrons from the next completed 18 electron inner shell with a voltage or chemical reaction. Therefore you are stuck with only 1 electron for each copper atom.

As I said before, with N-type semiconductor, that is a different story.

Ratch

 

[ericgibbs]

MrAL,
I don't call them free because the valance electron is loosely bound to its copper atom. An atomic number of 29 means copper, which has 1 valance electron. There is no way you are going to strip off any electrons from the next completed 18 electron inner shell with a voltage or chemical reaction. Therefore you are stuck with only 1 electron for each copper atom.

As I said before, with N-type semiconductor, that is a different story.

Ratch

morning Ratch,

We know that the loosely bound valency outer band electrons are the one's that are considered to be the electrons that constitute the current 'flow' electrons.

Thats why classically they are referred as to 'free' electrons, as I stated in my previous post.

Why you have crept in with a comment about doped semiconductors when we are discussing capacitors, I can only guess.

Eric

 

[MrAl]

MrAL,



I don't call them free because the valance electron is loosely bound to its copper atom. An atomic number of 29 means copper, which has 1 valance electron. There is no way you are going to strip off any electrons from the next completed 18 electron inner shell with a voltage or chemical reaction. Therefore you are stuck with only 1 electron for each copper atom.

As I said before, with N-type semiconductor, that is a different story.

Ratch

Hi Ratch,

I believe you mean well here but your pedanticity is showing through here

The word "free" has many, many definitions of which i show two here now:

1. Not bound, confined, or detained by force. "The rising balloon was free to move with the wind".

2. Not permanently attached. "The electron was free to move in the metal".

For #1 it is like the balloon has to move and it might be called free because we notice it is moving.
But for #2 the electron does not have to actually move to be called free. The fact that it CAN move if agitated means we can call it 'free'. In other words, in the presence of an electrical field it is free to move. It is not as if it can not be called free because it is somehow stuck in the atom. It is loosely held, and compared to the other electrons it is free and they are not. So we also call the valence electrons 'free' in electronic work. In chemical work this might not be quite as common but in electrical work it certainly is and of course there is the free electron theory of metals as Eric pointed out nicely.

Make more sense to you now?

 

[Ratchit]

MrAl,

I believe you mean well here but your pedanticity is showing through here

I know, that is one of my problems.

2. Not permanently attached. "The electron was free to move in the metal".

So we also call the valence electrons 'free' in electronic work. In chemical work this might not be quite as common but in electrical work it certainly is and of course there is the free electron theory of metals as Eric pointed out nicely.

Make more sense to you now?

Don't you think that "available" would be a better description?

Eric,

Why you have crept in with a comment about doped semiconductors when we are discussing capacitors, I can only guess.

Because unlike metals, the electrons in N-type semiconductor are truly free in every sense of the word, to wander around the crystalline structure. I though it was a good example of what free electrons are in materials.

Ratch

 

[steveB]

But if you could clarify what you are using the 2x2 matrix for that would be nice to hear too.

Mr Al,

I updated my previous post to include some pages from a book "Principles of Electro Magnetics", by Schwartz. https://www.amazon.com/dp/0486654931

Perhaps you already have this, but if not, I recommend it highly. Many sections in this book are like priceless gems. Schwartz offers a unique perspective on many aspects of EM field theory. This book is well worth the modest price of < $12 price. I just noticed the Kindle version is available at $10, and so I'll buy that version too so that it is available when I'm traveling.

Could it be that nature has a built in safety mechanism to NEVER allow the cap to run out of electrons?
If you care to take up this calculation it would be interesting to hear your results. I came with a quick estimate for a 1 inch square plate capacitor (2 plates each 1 square inch area on one side of each plate). To see one plate run out of electrons before the voltage sparkover seems to require a copper plate thickness of 8e-14 inches. That's extremely thin and even smaller than the diameter of one copper atom so it appears that it cant be built out of copper, and i have to wonder if there could be another material that could be used instead. Because it has to be so thin i dont see how just yet. But it would be interesting to hear your take on this too.

In the past I've also done these types of calculations and they are fun to do and are very revealing. Actually, the above book does this kind of thing in many places, which is another reason you may appreciate it even more than most people. For example, there is another section where he mentions that if you take the electrons off one tenth of a cubic millimeter of material at the top of an Apollo rocket, and place them at the bottom of the launch pad, the electrostatic attractive force would be enough to keep that rocket on the ground at full thrust. Yet another way to show how nature won't let you build a basic cap that can run out of charge.

 

[tvtech]

I work at a magic smoke factory, we use only the best unicorn tears for our products.

Now THAT is funny

All the best

Regards,
tvtech

 

[MrAl]

Mr Al,

I updated my previous post to include some pages from a book "Principles of Electro Magnetics", by Schwartz. https://www.amazon.com/dp/0486654931

Perhaps you already have this, but if not, I recommend it highly. Many sections in this book are like priceless gems. Schwartz offers a unique perspective on many aspects of EM field theory. This book is well worth the modest price of < $12 price. I just noticed the Kindle version is available at $10, and so I'll buy that version too so that it is available when I'm traveling.



In the past I've also done these types of calculations and they are fun to do and are very revealing. Actually, the above book does this kind of thing in many places, which is another reason you may appreciate it even more than most people. For example, there is another section where he mentions that if you take the electrons off one tenth of a cubic millimeter of material at the top of an Apollo rocket, and place them at the bottom of the launch pad, the electrostatic attractive force would be enough to keep that rocket on the ground at full thrust. Yet another way to show how nature won't let you build a basic cap that can run out of charge.

Thanks Steve. I'll have to take a good look at the pages and maybe the book too, so thanks for the link to the book too.
And may the Schwartz be with you