"Veritasium" (YT) - "The Big Misconception About Electricity" ?

[Zero999]

In a lossless transmission line, energy won't be lost due to radiation.

I used to think so as well. However, the following paper:

J. E. Storer and R. King, "Radiation Resistance of a Two-Wire Line," in Proceedings of the IRE, vol. 39, no. 11, pp. 1408-1412, Nov. 1951, doi: 10.1109/JRPROC.1951.273603.

shows that a lossless transmission has a finite radiation resistance.

It depends on what is meant by lossless. If it has zero resistance, but is unshielded, then you're right, it will act as an antenna and radiate, but if it's perfectly screened as well, then it will not radiate.

Interestingly, this antenna will have several resonant modes, which will be excited, not only due to the length and separation of the wires, but the diameter of the conductor itself, which isn't stated.

[Sredni]

Nope, still not telling me anything useful, just stating that's a way to look at it.
What can looking at it that way DO FOR ME?

And what does thinking the power flows in the cable do for you? If all you want to see is voltage and current, you don't need that information either.
When you have to power a 1kW heater, do you look for a cable capable of sustaining 1kW of power? You just need to dissipate the power associated with the Poynting vector impinging in the cable.

Which makes me think of a way to verify if power comes from... outer space. Nope, surface charge will prevent that and make sure the Poynting vector, as soon as we cross the lateral surface of the resistor, will be directed inward in a uniform way (in hindsight it's logical if we consider that the E field inside is constant and directed along the axis).
Off the top of my head, knowing the Poynting vector field might help knowing which part of your heater will be hotter. If the battery-resistor circuit is not symmetric the power getting into the resistor (to be dissipated) will not follow a symmetric spatial distribution. In the improbable case you can have a perfectly homogeneous resistor, it should be possible to see certain parts of the heater glow more, while others will glow less. I wonder if its doable or if the process of redistribution in the material will make this effect drown in a sea of thermal uniformity.

Do you have any comment on how quantum field theory views this? or do you think it's bunk?

You don't need to go as far as QFT to muddy the waters. Even plain quantum mechanics can make things so complicated that you won't be able to have intuitive insights. For example: what makes the above heater glow? A classical physicist would say it's the collisions of electrons with the material's atomic lattice. A quantum physicist will object to that: what electrons? There's a collective wave function there, not a bunch of identifiable electrons. And if you have a perfectly spaced lattice that wave won't interact at all, so it can't be the lattice. Turns out it's imperfections in the lattice and the mechanism behind the transferal of energy is electron-phonon interaction. You need to throw in more than half Ashcroft Mermin to explain why your heater glow.

Back to QFT. My understanding is that more advanced theories can extend our knowledge to explain more of what we observe. So the question is: will QFT give different values for the electric and magnetic field in the space around the wires? I doubt it. These fields can be measured, so if QFT is that good of a theory they say (and it is) it will agree with experimental measures.

Do you think that the value of the electric and magnetic field in the middle of a circuit with battery and resistor will be different from what is predicted by classical ED? (I am not talking about vacuum fluctuation, but fields of the order of magnitude we can measure with 'ordinary' instruments).

[EEVblog]

You don't need to go as far as QFT to muddy the waters. Even plain quantum mechanics can make things so complicated that you won't be able to have intuitive insights.

Let me replace the text:

You don't need to go as far as QFT Poynting to muddy the waters. Even plain quantum mechanics Poynting Thereom can make things so complicated that you won't be able to have intuitive insights.

You are now saying the same thing that many people say about Poynting/Maxwell for DC and LF.
It's turtles all the way up.

[EEVblog]

Back to QFT. My understanding is that more advanced theories can extend our knowledge to explain more of what we observe. So the question is: will QFT give different values for the electric and magnetic field in the space around the wires? I doubt it. These fields can be measured, so if QFT is that good of a theory they say (and it is) it will agree with experimental measures.
Do you think that the value of the electric and magnetic field in the middle of a circuit with battery and resistor will be different from what is predicted by classical ED? (I am not talking about vacuum fluctuation, but fields of the order of magnitude we can measure with 'ordinary' instruments).

I am not talking about the measurements, they will be as they always have been. I'm talking about the the title of this thread "The Big Misconception About Electricity". Does the energy flow in the field around the wire or does it flow in the wire at DC? Poynting/classical field theory says outside, QFT appears to say inside.
I want to know what you and others who have been so (not incorrectly) dogged about anyone that dares think of this in any other way than Maxwell/Poynting think about this apparent conundrum.

[bdunham7]

And what does thinking the power flows in the cable do for you? If all you want to see is voltage and current, you don't need that information either.

As far as the work of an engineer regarding a DC circuit--specifying parts and calculating losses--the concept of 'power flow' probably doesn't factor in.  However, it helps to understand it intuitively when you are deciding what to worry about.  So if your cable has to run through a metal tube or take some odd shape, you need to know whether to consider 'the spaces in between the wires' and all that.  For an all-DC device--say battery powered thermal socks--the answer is you need not worry at all.  For a mains-powered toaster, you need to understand the magnitude of certain effects (primarily induction in surrounding objects) to know that you don't need to concern yourself with 'the spaces between' for the actual toaster and its cord, but you do have to construct your house wiring following certain rules, such as not having only one of a pair of current carrying conductors be within a metal conduit.  For a USB-3 circuit board, it's more 'spaces not traces', although that's not a definitively accurate statement either.

When you have to power a 1kW heater, do you look for a cable capable of sustaining 1kW of power? You just need to dissipate the power associated with the Poynting vector impinging in the cable.

Of course not and that is just another of your ridiculous straw men.  The cable needs to dissipate the power associated with the required current and the cable's resistance.  Somehow determining that by calculating Poynting vectors and an S-field would be the most ludicrously obtuse way that I can think of.  Here is a completely solved thermal sock circuit.  What I see is a circuit with its behavior completely defined regardless of how you configure the wires and regardless of what is in the 'spaces between', except perhaps a varying magnetic flux.  How would you apply Poynting to this in any helpful way?

[adx]

Wo - after a moment of clarity I was ready to call BS on the "energy doesn't travel in wires" claim a couple of pages back, but after getting year end tasks done I come back and find the thread has done it for me!

I wrote down what I could and never fleshed it out so may be a bit garbled now, but basically:

• power is one part force and one part movement
• it's a completely static magnetic field, nothing moves (it is an effect of moving charges)
• the (non) argument that "it's all fields" or "all energy anyway" only works to confirm the possibility of energy in the wires (the wires are energy)
• electric field is what causes the electrons in the wire to move, the magnetic field around the wire is a result of the electrons moving in the wire, nowhere else

(The idea I had in mind was more convincing than that.)

But then I realised that it's not up to me to prove where power flows. I only need ask for proof that power doesn't flow in the wires, because of the bold certainty with which the claim is made. It becomes a kind of theological argument. Especially as SandyCox said "Poynting’s theorem and charge flow lead to exactly the same answer". Is the expectation of a difference an illusion?

But this thread has confused me again. I had sort of settled on the opinion (some pages back) that DC power is a product of moving charge carriers and pressure difference (not field, but work function in a conservative field which boils down to a potential difference not gradient), inside and outside of the wires respectively. But Hontas Farmer's diagram of the electrons in the wire made it seem even simpler than that, and I'm now confused over the need for the pressure difference outside of the wires (something I had thought of before). Imagine a system where space outside the wires (or hydraulic pipes) doesn't exist, or at least couldn't hold anything. The pressure difference still drives down the pipes, through the resistor where it does its work, and returns via a parallel path. It divides along the path. But the concept of "power flow" is totally dependent on axial distance, and trying to locate it in the transverse direction requires setting up a field with gradient - that need not exist, but must if space exists. (Which is basically why I called the latter into doubt.)

I'll have to do some more thinking.

[bdunham7]

To me it seems that when using wires the movement of electric charge alone could do work and transfer energy.

For an electric charge to do work, it has to be in an electric field--IOW it does it's work by moving from one potential to another.  So the arrival of an electron at the load alone doesn't accomplish anything, there also needs to be an electric field across the load in order for it to do work.  In a simple DC circuit, both the charges themselves and the electric field that propels them through the load are provided by the movement of charges through the conductor and the resultant charge distribution.  My Millikan tongs example separates the two functions.  Or does it?

[SandyCox]

I am not talking about the measurements, they will be as they always have been. I'm talking about the the title of this thread "The Big Misconception About Electricity". Does the energy flow in the field around the wire or does it flow in the wire at DC? Poynting/classical field theory says outside, QFT appears to say inside.
I want to know what you and others who have been so (not incorrectly) dogged about anyone that dares think of this in any other way than Maxwell/Poynting think about this apparent conundrum.

There’s nothing wrong with Maxwell’s equations. The problem is the misinterpretation of what the Poynting vector tells us. Here is what Haus and Melcher says in Section 11.3 of their book:

"we illustrate the danger of ascribing meaning to S evaluated at a point, rather than integrated over a closed surface."

Fig. 11.3.1 is a nice illustration of this misinterpretation.

I'm still writing a more detailed explanation of what they refer to. Veritasium is wrong about the steady-state transfer of energy.

Another example of this misinterpretation is the direction of the Poynting vector in the airgap of a transformer.

[bdunham7]

Fig. 11.3.1 is a nice illustration of this misinterpretation.

Can you post the illustration for those of us that don't have the book?

[rfeecs]

There’s nothing wrong with Maxwell’s equations. The problem is the misinterpretation of what the Poynting vector tells us. Here is what Haus and Melcher says in Section 11.3 of their book:

"we illustrate the danger of ascribing meaning to S evaluated at a point, rather than integrated over a closed surface."

They are using an alternative energy flux vector S, not the Poynting vector.

Starting with conservation of energy, looking at the energy in a volume of space, you get the change in that energy by integrating an energy flux over the surface of the volume.  The form you choose for that flux may be arbitrary.  If there are no sources or dissipation of energy in that volume, you may have zero flux on the surface or you may have equal flux going in as going out.

They chose an alternative definition of energy flux that only depends on current density.  It still satisfies energy conservation.  But it only works in certain cases.

Clearly it doesn't work for a propagating wave in free space.

The Poynting vector seems to work for all cases.  So why use two different definitions of energy flux?

[SiliconWizard]

And, IMHO, we still haven't defined exactly what "energy traveling in wires" means anyway.
If by that we mean the analogy of water flowing through a pipe, that is certainly bogus. Electrons do not "flow" from point A to point B like molecules themselves do.
And yeah, I agree with Dave on this point: ultimately it's quantum field theory.

So yeah. What does "in wire" means? What is in and what is out when it comes to energy? Isn't that just a question of probabilities?

[Uttamattamakin]

I got this one in my recommended list, which has only a few hundred views, so hence I link it here. The first half with formula's I skipped, but the 2nd half shows to me a valid point about electric fields in wires vs air gap when using quantum theory.

Very interesting, thanks for posting.
Nice explanation, sounds pretty solid to me. Quantum probability theory trumping Poynting?
I've sent this to Derek.

Thanks I am the one who made this video.  I would sum it up like what you said and add only this.  In a classical point of view of physics things happen or they do not happen, in a QFT point of view there are non zero and tiny probabilities of everything happening.  Quantum Electro Dynamics is the specific theory that applies to E and M.    As this video from PBS space time explains much better than me with my barely functional white boards.



For the phenomena under discussion here Quantum Electro Dynamics would find that the path of highest probability is along the wire, where all the free electrons are and can interact with eachother at inter atomic distances VS the vast 1m void between them and the bulb.   This paper on Compton scattering gives a sense of the complications that we go through in exchange for this https://www.personal.soton.ac.uk/ab1u06/teaching/qft/qft1/christmas_problems/2014/xmas_problem_solution.pdf  I did a related problem Bahaba scattering, electron positron scattering in graduate school.  It was such a big problem that it was one of only two problems on a homework assignment, we had two weeks to do it.  So my informed estimate is  that the probability of conduction along the wire is 0.999999 with a 0.000001

It would make for a boring video.  What did not make for a boring video was this one. This video does the experiment that Veritasium showed but at a real scale.  What it finds is very interesting so I won't spoil it.



I'll be around from time to time.  My main focus of research is gravitational wave physics though.  Thanks for watching.

[bdunham7]

Isn't that just a question of probabilities?

It is all just a question of probabilities, but then you have to do the math.  It isn't just the electrons that are a wave function, molecules are as well.  It's just that the probability distribution for a molecule is more localized than a free electron in a conductor.  This doesn't mean that the concept of individual electrons flowing in a wire is any less valid than water molecules flowing in a pipe.  As for whether they flow through the wire or outside of it, that was the exact point made at the end of Farmer's video--both are theoretically possible, the former is astronomically more likely.

[SiliconWizard]

Isn't that just a question of probabilities?

It is all just a question of probabilities, but then you have to do the math.  It isn't just the electrons that are a wave function, molecules are as well.  It's just that the probability distribution for a molecule is more localized than a free electron in a conductor.  This doesn't mean that the concept of individual electrons flowing in a wire is any less valid than water molecules flowing in a pipe.

Oh really... Well, if you think QFT, or at least a part of it, can be applied to molecules, you must be following pretty recent advances in physics research: https://www.nature.com/articles/s41567-019-0663-9
because, before that, I don't think it was all that well established. Oh, and it's still partial, and the experiment, although interesting, certainly needs to be reproduced by others and more work is definitely needed on this. OK, that's for big molecules. But even for single atoms - I do think while we have indeed been able to assign them some wave functions, it's already not quite in the same territory AFAIK. Complex matter for sure. But, if that what you were thinking about, is molecular agitation really comparable to the movement of electrons, that's a tough one?

But, while giving some quantum properties to molecules so far, it still doesn't say that water molecules through a pipe would flow in any similar way electrons "flow".

As for whether they flow through the wire or outside of it, that was the exact point made at the end of Farmer's video--both are theoretically possible, the former is astronomically more likely.

Well yes! But we still haven't defined while "through the wire" exactly meant, unless I missed it. Can we find a definition?

[bsfeechannel]

Does the energy flow in the field around the wire or does it flow in the wire at DC? Poynting/classical field theory says outside, QFT appears to say inside.

Wait a minute, Dave. Are you faffing around the edges of physics to confirm your bias? What happened to the Ohm's law that we, engineers, "developed to give a more practical insight rather than what is actually happening at the physics level"?

[bsfeechannel]

Thanks I am the one who made this video.  I would sum it up like what you said and add only this.  In a classical point of view of physics things happen or they do not happen, in a QFT point of view there are non zero and tiny probabilities of everything happening.  Quantum Electro Dynamics is the specific theory that applies to E and M.

I like your video very much, especially that part where you say that, in QFT, there's no distinction between particles and fields. But I have a question. What if your wire, besides the interacting electrons you showed, also had protons?

As this video from PBS space time explains much better than me with my barely functional white boards.

Cool, so the energy-carrying particles are the photons, which are just oscillations in the EM field.

[bdunham7]

you must be following pretty recent advances in physics research...while giving some quantum properties to molecules so far, it still doesn't say that water molecules through a pipe would flow in any similar way electrons "flow".

No, the concept of the localization of the wave function of any particle or object, quantified as its de Broglie wavelength, has been taught in introductory QM for decades and I had to demonstrate it experimentally (for electrons) in one of my undergraduate physics labs via the standard Davisson-Germer experiment.  Nothing new, except that the de Broglie wavelength for larger objects is so small and thus the distribution probability so localized that it is much more difficult to observe.

I did not say that the flow of water through a pipe was in any way 'similar' to the flow of electrons in a wire--nor did I say they were dissimilar-- I simply said that they were both valid concepts. 

Well yes! But we still haven't defined while "through the wire" exactly meant, unless I missed it. Can we find a definition?

Well, if you want a QM definition, I think it would simply be that the spatial probability distribution of the electrons in question at each point in time ( Ψ [x,y,z] (t) ) mostly falls somewhere within the dimensions of the wire.  IIRC, the de Broglie wavelength λ of an unbound electron within a copper conductor is less than a nanometer, λ for a water molecule would be in the femtometer range.  The 'odds' of a water molecule tunneling out of its pipe into free space is so low that we don't expect to ever observe it.  The electron may have a better chance, but it is still vastly more likely to be found within the wire.

[bdunham7]

The Poynting vector seems to work for all cases.  So why use two different definitions of energy flux?

Because a simple (and completely solved) problem like my thermal socks unnecessarily becomes a non-intuitive unworkable mess.  The real question for me is why you would analyze any problem from the perspective of 'energy flux'.  If there's a valid reason to do so, perhaps Poynting is the way to go.  There's a reason the Poynting's Theorem and the concept of the S-field shows up where it does.

[rfeecs]

The Poynting vector seems to work for all cases.  So why use two different definitions of energy flux?

Because a simple (and completely solved) problem like my thermal socks unnecessarily becomes a non-intuitive unworkable mess.  The real question for me is why you would analyze any problem from the perspective of 'energy flux'.  If there's a valid reason to do so, perhaps Poynting is the way to go.  There's a reason the Poynting's Theorem and the concept of the S-field shows up where it does.

I agree.  If you are analyzing an antenna you would probably use Poynting's Theorem (or your computer would).  But for DC, then P = VI works fine.  Use the right tool for the job.

Yet people can't agree that energy is outside the wire in the fields vs inside carried by electrons.

There is even talk of electrons in the battery "influencing" electrons in the lamp, maybe by the exchange of virtual particles.  Definitely not the right tool for the job.

[bdunham7]

Yet people can't agree that energy is outside the wire in the fields vs inside carried by electrons.

IMO, the question is wrong.  What is 'the energy'?  The energy of what?  A charge has potential energy if there is an electric field so that work can be done on it if it is allowed to move from a higher potential to a lower one.  The transfer of energy (flow) is in my view the delivery of such a charge to the load.  See my Millikan tongs example on the previous page if you haven't already.  Whether the energy is 'contained' in the charge or the field seems a questionable question since both are required.  You can have any field you want, but without a charge to work on there's no energy.  Likewise, the electron itself doesn't contain energy simply by its presence (well it does, but that is a different question) but it does by its motion, characteristics (mass and charge) and position within the conservative electric field.

So for a similar parallel example, suppose I have an object on a table.  We can say that it has potential energy relative to the floor due to the gravitational field.  If I push it off, that potential energy starts to be converted to kinetic energy.  We can develop a whole field of mechanics (LaGrangian) around this principle where objects are deemed to have potential energy plus kinetic energy.  Do we say that the energy actually flows or is contained in the gravitational field?  No, we describe the energy as being attributed to objects within the field, determined by their motion, characteristics (mass) and position within the field.

[SiliconWizard]

No, the concept of the localization of the wave function of any particle or object, quantified as its de Broglie wavelength, has been taught in introductory QM for decades and I had to demonstrate it experimentally (for electrons) in one of my undergraduate physics labs via the standard Davisson-Germer experiment.  Nothing new, except that the de Broglie wavelength for larger objects is so small and thus the distribution probability so localized that it is much more difficult to observe.

Uh, yeah. Precisely. The wave function applied to "any object" of any scale is a cute theory, but it just failed for anything other than very small particles. Electrons, sure nothing new about that! The paper I quoted above may reshuffle the cards somewhat, although I'm not hugely holding my breath at this point.

I did not say that the flow of water through a pipe was in any way 'similar' to the flow of electrons in a wire--nor did I say they were dissimilar-- I simply said that they were both valid concepts. 

Both valid concepts, but for describing different things.

Well yes! But we still haven't defined while "through the wire" exactly meant, unless I missed it. Can we find a definition?

Well, if you want a QM definition, I think it would simply be that the spatial probability distribution of the electrons in question at each point in time ( Ψ [x,y,z] (t) ) mostly falls somewhere within the dimensions of the wire.

Yeah, that's a start. It's already beyond what was bluntly put, without much details, in the original video. (Of course this is still somewhat simplified IMO: the dimensions of the wire, for any practical wire, are a more complex concept, and more difficult to model than it appears.)

But after reading all those various approaches, I'm still under the impression that the question is both unanswered and ill-defined.

[rfeecs]

Whether the energy is 'contained' in the charge or the field seems a questionable question since both are required.  You can have any field you want, but without a charge to work on there's no energy.

So a Megajoule laser pulse doesn't contain any energy?

Charges gave up energy to create the pulse.  Charges will eventually absorb energy when the pulse hits them.  But how did the energy go from one location to the other?

Apparently not in the fields, because without a charge there's no energy?

OK, the energy is in photons.  That's another model.

Perhaps you are saying that static fields do not contain energy.

It seems like a chicken and the egg thing.  All electromagnetic fields are created by charges.  The energy was put into the system when the charges were moved.  You can't have one without the other.

[SiliconWizard]

It seems like a chicken and the egg thing.  All electromagnetic fields are created by charges.  The energy was put into the system when the charges were moved.  You can't have one without the other.

Now I think we're moving forward... =)

[bdunham7]

Uh, yeah. Precisely. The wave function applied to "any object" of any scale is a cute theory, but it just failed for anything other than very small particles.

QM, QED and QFT are non-intuitive and hard to grasp, but they certainly aren't a cute theory that fails.  More than one cynic struggling to understand them has noted that many of simpler mathematically predicted non-intuitive results are either too small or too infrequent to be observed, except of course for those that are observed--tunnel diodes, diffraction and while we're at it, things like incandescent light bulbs.  Most of the actual physical phenomena that make the universe work are things classical physics, including EM theory, fails to explain properly.  The fact that some of those QM results are difficult to observe directly doesn't invalidate or marginalize the theory in any way.  In any case, last I heard someone had managed to demonstrate diffraction in and calculate the de Broglie wavelength of fullerenes, a.k.a. buckyballs which are a spherical C₆₀ molecule.

But after reading all those various approaches, I'm still under the impression that the question is both unanswered and ill-defined.

If you are referring to whether the energy flows in the wire, OK, I'll agree that there is room for discussion.  But if it is whether the electrons flow in the wire, I'd regard that as not really worth debating.

[bdunham7]

So a Megajoule laser pulse doesn't contain any energy?

Charges gave up energy to create the pulse.  Charges will eventually absorb energy when the pulse hits them.  But how did the energy go from one location to the other?

Apparently not in the fields, because without a charge there's no energy?

Context and previous replies (which are context, I suppose) are important.  I specifically addressed this a page ago and you are misconstruing what I've said.

https://www.eevblog.com/forum/chat/veritasium-(yt)-the-big-misconception-about-electricity/msg3910100/#msg3910100

Perhaps you are saying that static fields do not contain energy.

Perhaps you could read my post that I've linked and the one after it and then see whether you agree, disagree or don't understand my position on that.  Can you have a static E-field without charges?