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

[vad]

The key message of Veritasium’s video is that the electro-magnetic energy does NOT travel in wires. But people, who were caught with their pants down by that video, are switching the topic, by highlighting video’s imperfections, ranging from messed up units of measure through omissions in the definition of the thought experiment.

[SandyCox]

Transmission lines don't account for radiation
[\quote]
You are absolutely correct. We also have to include a dipole in the model. I will see if I can figure out the details.

[Howardlong]

electro-magnetic energy does NOT travel in wires.

Who suggested otherwise?

highlighting video’s imperfections, ranging from messed up units of measure through omissions in the definition of the thought experiment.

I worked with an individual once whose MO was to omit key details, and only answer questions literally rather than considering context. It was a power game for him to keep himself indispensable. Technically, he was excellent, but we could never send him out to a customer. He didn't work well with others in his team. Information was siloed in his head. After a few months, we realised he wasn't an asset, he was a liability.

[SilverSolder]

[...] People are running around testing things like it's 1900.  [...]

Seeing is believing, though.  Nothing like getting the solder out to support or disprove a theory!

[adx]

Engineering is not the harbor of ignorance.

It really is. A working engineer might need to worry about annealing of copper in a dubious glass to copper seal one day, then coefficients of a biquad filter the next. So for that matter might a physicist. There is too much to do to maintain a clear working understanding of something that has never and probably never will come up in their career.

You see, in the past, engineers had insights in physics, that's why many of them contributed both to engineering and science.

Engineers number in the millions now. The type of insights you're talking about seem to come from once in a generation people, and from times when it was a lot more practical to make a notable fundamental contribution to science. Where are those opportunities these days?

Now engineers are seen as dogmatic people who cannot think outside of the box.

Engineering is mostly a commercial task-based function. Thinking outside the box can get you fired.

People you have to be careful not to trigger when you want to expose something that go against their preconceptions.

That's just belief at work. The dubiously existent backfire effect. Maxwellians have been equally triggered by comments which go against their worldview. It happens in medicine, and just about any situation where people are indoctrinated. It doesn't mean it's wrong to reject new facts, just that it is a process that has to be gone through when unacceptable contradictions arise.

More importantly, I don't think this "triggering" is as serious as you make out. Engineers are pointing out working realities that physicists might miss, and taking on the challenge of being "triggered" for a bit of fun and the opportunity to have a say. Neither Dave nor Mehdi actually disagreed with the core findings of Derek's video. I don't think any trained engineers on this forum have seriously taken exception to any core fact. I have trouble accepting an opinion that power 'flows' in a completely static magnetic field, but I am not arguing against the fact that Poynting's model works. Therefore people aren't arguing against fact or evidence. Nothing is being "exposed" beyond some raw nerves over things like education and the realities of an engineering life.

These engineers think they "own" circuit theory. They don't understand that circuit theory is just as physics as the Maxwell equations are. In fact circuit theory is just a special case of them.

No they obviously understand it and aren't arguing against reality. They are pointing out some impracticalities to going full-academic treatment. Yes, some people are rallying against the message (on YT comments etc), I don't think any are formally trained electrical engineers.

Some of these things lamentable, I'm not saying I like them or disagree with your sentiments, just pointing out facts.

No one has come up with a clean solution. Transmission line theorists, antenna masochists, electromagnetic solverists, or wire unreelists for that matter.

The solution was given by Derek. You have a Poynting vector pointing directly from the battery to the load. The electromagnetic field propagates at the speed of light, so energy will first arrive at the load at exactly 1 m/c seconds. That's simple and elegant. AND there's no other "approach" to the problem. Other "approaches" will give you the wrong answer.

Another approach is to say nothing propagates faster the speed of light, and through capacitance or magnetic induction, that's what happens, it can't be faster than 1m/c and the other answers aren't reasonable. That is simpler and eleganter.

Poynting is not needed, because the question is about current, not energy or power. Closing the switch is expected to cause a current burst, which creates a magnetic field, this travels outward and turns on the lamp at any current.

Maxwell is not needed, because a magnetic field will propagate without EM radiation to an infinitesimal degree over the distance, from an infinitesimally short dipole. One might guess to a correctness testably comparable to Maxwell's discovery, that the speed of light is involved.

I'm not rejecting the Maxwell-Heaviside-Poynting idea though. If I understand it, and it works, why would I want to do that? If I don't understand it, and have to take it on faith that it works, then why would I reject it? Despise it through jealousy at being unable to comprehend? Then how come all these engineers got their degrees?

Edit: Oops, I deleted the main point of this section: In the original video, Derek's speed of light answer comes from an argument made from a simplified description (like my reasoning now above). Comparing this to a clean solution (in the sense of what a field solver would arrive at) is like the time I told a teacher I had done my homework and it was "in my head". Simple and elegant isn't clean and correct - it's one of the mistakes academics make.

[adx]

electro-magnetic energy does NOT travel in wires.

Who suggested otherwise?

bdunham7 was first I think, but: Me.

[Zero999]

I haven't read the whole thread, but Veritasium hasn't said anything incorrect, or controversial in his video. It's certainly easier to understand than the wind powered vehicle moving faster than the wind video.

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness. This is because the instant the circuit is made, the transmission line will have an impedance, equal to its characteristic impedance. Once the circuit has reached steady state, the battery will see the bulb's impedance.

[SandyCox]

I haven't read the whole thread, but Veritasium hasn't said anything incorrect, or controversial in his video. It's certainly easier to understand than the wind powered vehicle moving faster than the wind video.

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness. This is because the instant the circuit is made, the transmission line will have an impedance, equal to its characteristic impedance. Once the circuit has reached steady state, the battery will see the bulb's impedance.

This is exactly the conclusion I came to in the note I published on p 19. However, I since realised that there's more to the problem.  The transmission line model doesn’t account for electromagnetic radiation from the folded dipole.
Maybe the folded dipole looks like a short circuit at the moment the transient is initiated. In that case the light bulb will momentarily see the full battery voltage after 1m/c seconds.

[SiliconWizard]

I haven't read the whole thread, but Veritasium hasn't said anything incorrect, or controversial in his video. It's certainly easier to understand than the wind powered vehicle moving faster than the wind video.

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness. This is because the instant the circuit is made, the transmission line will have an impedance, equal to its characteristic impedance. Once the circuit has reached steady state, the battery will see the bulb's impedance.

Yeah. This is what AlphaPhoenix showed in his test, for those who have seen his video. It's not too bad, actually. He shows what happens when you open the line at one end, or both, as well.

[bsfeechannel]

I haven't read the whole thread,

We love people who have not read the whole thread.

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

He does that on purpose. He's challenging the idea that, because you have a 'transmission line", the transmission of energy will necessarily be confined to it.

This is not a problem of academic interest only. The other day we were discussing on this very forum how energy could possibly be transmitted inside a linear transformer, since the net magnetic field is just the magnetizing field, which stores energy from the primary in one half cycle, but returns it immediately in the next half cycle to the primary. Circuit theory led us to a dead end. The answer, you guessed it, is the Poynting vector.

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness.

You see? Circuit theory will lead you to the wrong conclusion. The bulb will not start to light up immediately. It'll take 1 m/c for the energy to arrive at the bulb.

So the "main criticism" you have is precisely the misconception he managed to debunk.

[Howardlong]

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

He does that on purpose.

I very very strongly doubt that.

He does bring up transmission lines very indirectly, discussing the undersea cables and the dispersion phenomenon, but does nothing to tie it into the discussion. Presented as such it's an interesting aside: he might as well have brought up the local weather forecast or the price of Emmental.

He's been fast and loose with accuracy, detail and omission. It's not a trait that engineers should espouse, and I doubt scientists would support it either.

A cynic and proponent alike would say he had 15 minutes to fill, and he accrued over 10m views: mission accomplished.

[EEVblog]

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

He does that on purpose.

I very very strongly doubt that.

I asked him this and I don't recall his answer directly (we talked for like 45min), if he sends me the recording I'll be able to answer this question.

[Zero999]

I haven't read the whole thread,

We love people who have not read the whole thread.

It's 15 pages long. Even though I'm still on Christmas break, I still don't have enough time on my hands to read every single post. At this point in the thread, it's completely unreasonable to expect someone making their first post to have read all of it.

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

He does that on purpose. He's challenging the idea that, because you have a 'transmission line", the transmission of energy will necessarily be confined to it.

This is not a problem of academic interest only. The other day we were discussing on this very forum how energy could possibly be transmitted inside a linear transformer, since the net magnetic field is just the magnetizing field, which stores energy from the primary in one half cycle, but returns it immediately in the next half cycle to the primary. Circuit theory led us to a dead end. The answer, you guessed it, is the Poynting vector.

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness.

You see? Circuit theory will lead you to the wrong conclusion. The bulb will not start to light up immediately. It'll take 1 m/c for the energy to arrive at the bulb.

So the "main criticism" you have is precisely the misconception he managed to debunk.

Well, of course, the electric field between the two plates of the capacitor, can only travel at the speed of light.

[Howardlong]

My main criticism is he fails to mention transmission lines, which are crucial to understanding this.

He does that on purpose.

I very very strongly doubt that.

I asked him this and I don't recall his answer directly (we talked for like 45min), if he sends me the recording I'll be able to answer this question.

For the avoidance of doubt, my opinion is that a transmission line model analysis was never even considered.

[SandyCox]

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness.

We know that energy from the battery will reach the bulb at approximately 1m/c seconds. So Veritasium’s question has been answered. The following is my attempt to gain some further insight.
The lumped-element equivalent-circuit of the transmission line can easily lead to misinterpretation:
1.   Do we start with a capacitor or an inductor when drawing the circuit? If we start with a capacitor, our intuition tells us that the transmission line looks like a short circuit at the moment the transient is initiated. If we start with an inductor, our intuition tells us that it will looks like an open circuit.
2.   In fact, our intuition is giving us the wrong answer. The lossless transmission line’s input impedance is purely resistive. It is neither capacitive nor inductive. During the transient the capacitors and inductors in the transmission line store energy. This happens in such a way that the transmission line looks like a resistor at its sending end.
3.   The transmission line looks like a short circuit in the steady state. The inductors are now short circuits while the capacitors are open circuits.
But there’s more to the story. We are looking at something like a folded dipole which also radiates electromagnetic energy during the transient, because of the unbalanced way in which it is driven. The transmission line model doesn’t account for this.
In [1] it is shown that a folded dipole may be analysed by considering its current to be composed of two distinct modes, namely, a transmission line mode and an antenna mode. A transient analysis of an ordinary dipole antenna can be found in [2]. The author later corrected the analysis in [3]. He shows that, wen driven by an ideal step voltage source, the initial input current to the antenna is infinite.
My conclusion is that we can gain some valuable insight from transmission line theory and from antenna theory, but that we cannot model the behaviour of this circuit exactly by theoretical means.
It is not necessary to take the full length of the wires into account in simulation studies. We are only interested in, let’s say, the first 50 ns during which the electromagnetic wave would only have travelled about 15m. Unfortunately, I do not have access to the necessary simulation software.

[1] G. Thiele, E. Ekelman and L. Henderson, "On the accuracy of the transmission line model of the folded dipole," in IEEE Transactions on Antennas and Propagation, vol. 28, no. 5, pp. 700-703, September 1980, doi: 10.1109/TAP.1980.1142400.
[2] Tai Tsun Wu , "Transient Response of a Dipole Antenna", J. Math. Phys. 2, 892-894 (1961) https://doi.org/10.1063/1.1724237
[3] Collin, R.E. and Zucker, F.J., “Antenna Theory: Part 1”, 1969

[bdunham7]

Unfortunately, I do not have access to the necessary simulation software.

You can download the student version of Ansys Electronics (includes HFSS) for free.  This is permissible even if you are not an enrolled student, as they say it is OK for 'self-learning', just nothing commercial.  I haven't had time to become proficient enough to make a working model, but I did start and it looks like it should work.

https://www.ansys.com/academic/students

[Zero999]

The most important thing to understand is, the wires are capacitively and inductively coupled. When the switched is closed, current will flow immediately, due to the capacitive coupling between the two wires. The bulb will immediately light, but not at full brightness.

We know that energy from the battery will reach the bulb at approximately 1m/c seconds. So Veritasium’s question has been answered. The following is my attempt to gain some further insight.
The lumped-element equivalent-circuit of the transmission line can easily lead to misinterpretation:
1.   Do we start with a capacitor or an inductor when drawing the circuit? If we start with a capacitor, our intuition tells us that the transmission line looks like a short circuit at the moment the transient is initiated. If we start with an inductor, our intuition tells us that it will looks like an open circuit.
2.   In fact, our intuition is giving us the wrong answer. The lossless transmission line’s input impedance is purely resistive. It is neither capacitive nor inductive. During the transient the capacitors and inductors in the transmission line store energy. This happens in such a way that the transmission line looks like a resistor at its sending end.
3.   The transmission line looks like a short circuit in the steady state. The inductors are now short circuits while the capacitors are open circuits.
But there’s more to the story. We are looking at something like a folded dipole which also radiates electromagnetic energy during the transient, because of the unbalanced way in which it is driven. The transmission line model doesn’t account for this.
In [1] it is shown that a folded dipole may be analysed by considering its current to be composed of two distinct modes, namely, a transmission line mode and an antenna mode. A transient analysis of an ordinary dipole antenna can be found in [2]. The author later corrected the analysis in [3]. He shows that, wen driven by an ideal step voltage source, the initial input current to the antenna is infinite.
My conclusion is that we can gain some valuable insight from transmission line theory and from antenna theory, but that we cannot model the behaviour of this circuit exactly by theoretical means.
It is not necessary to take the full length of the wires into account in simulation studies. We are only interested in, let’s say, the first 50 ns during which the electromagnetic wave would only have travelled about 15m. Unfortunately, I do not have access to the necessary simulation software.

[1] G. Thiele, E. Ekelman and L. Henderson, "On the accuracy of the transmission line model of the folded dipole," in IEEE Transactions on Antennas and Propagation, vol. 28, no. 5, pp. 700-703, September 1980, doi: 10.1109/TAP.1980.1142400.
[2] Tai Tsun Wu , "Transient Response of a Dipole Antenna", J. Math. Phys. 2, 892-894 (1961) https://doi.org/10.1063/1.1724237
[3] Collin, R.E. and Zucker, F.J., “Antenna Theory: Part 1”, 1969

You're right of course, the magnetic fields due to the inductance can also only propagate at the speed of light. Yes, it's not accurate to think about it as a lumped model, but it does make it easier. It's true the transmission line initially looks like a resistor, until the steady state is reached. In a lossless transmission line, energy won't be lost due to radiation.

[rfeecs]

The Science Asylum has a new video about capacitors.  He sneaks in the Poynting vector and energy flow:



He corrects the mistake he made in his previous video where he said the Poynting vector points into the wires instead of along the wires.

His discussion of dielectrics is a little odd.  He didn't talk about the main reason to have a (high) dielectric material in a capacitor.

[SiliconWizard]

As I mentioned - and sorry if it had been already, the thread is now pretty long - as far as experimentation goes, AlphaPhoenix did the trick IMO Unfortunately, I can't see completely clearly on the video all the details of the setup, especially regarding the probing. That's the thing I would be particularly cautious about: making sure any voltage we see on the scope is not coupled to the probes themselves in any way. Other than that, what he gets is congruent with was is expected.

As to more theory, same remark, probably been mentioned already (so sorry in advance if the link has already been posted), but this lecture is rather good: https://www.feynmanlectures.caltech.edu/II_24.html

And yes, it has already been said multiple times in the thread, but this "switching" experiment involves both a transient analysis (with very large frequency components upon switching), and then a steady-state DC phase. So apart from the coupling between wires, so some energy being transmitted from the battery to the load in 1m/c s, the wire as a transmission line should also exhibit a particular behavior until we get into DC territory?

[bsfeechannel]

There is too much to do to maintain a clear working understanding of something that has never and probably never will come up in their career.
[...]
Engineers number in the millions now. The type of insights you're talking about seem to come from once in a generation people, and from times when it was a lot more practical to make a notable fundamental contribution to science. Where are those opportunities these days?
[...]
Engineering is mostly a commercial task-based function. Thinking outside the box can get you fired.

If engineers are so dumb these days as you say, why do they get into discussing things they not only do not understand, but, worse, also don't want to understand?

Oh! Of course! They're dumb.

That's just belief at work. The dubiously existent backfire effect. Maxwellians have been equally triggered by comments which go against their worldview.

Have you heard of an Einsteinian? If someone calls your attention to the fact that you might be making mistakes because you don't really understand the theory of relativity, you call this person an Einsteinian?

There's no such thing as a "Maxwellian". Maxwell's equations are the theory of everything classical electromagnetism. Everything that is classically electric/magnetic has to be checked against this theory and, if it fails, dismissed right away.

So Maxwell's equations are not a worldview, they are a theoretical tenet of our trade. That's why people get impatient when someone exhibits total ignorance of that fact and claims to be an electronics engineer at the same time. That's cringe worthy and embarrassing.

More importantly, I don't think this "triggering" is as serious as you make out. Engineers are pointing out working realities that physicists might miss, and taking on the challenge of being "triggered" for a bit of fun and the opportunity to have a say. Neither Dave nor Mehdi actually disagreed with the core findings of Derek's video. I don't think any trained engineers on this forum have seriously taken exception to any core fact. I have trouble accepting an opinion that power 'flows' in a completely static magnetic field, but I am not arguing against the fact that Poynting's model works. Therefore people aren't arguing against fact or evidence. Nothing is being "exposed" beyond some raw nerves over things like education and the realities of an engineering life.
[...]
No they obviously understand it and aren't arguing against reality. They are pointing out some impracticalities to going full-academic treatment. Yes, some people are rallying against the message (on YT comments etc), I don't think any are formally trained electrical engineers.

It's a thought experiment. Thought experiments are designed to test the limits of a concept. You're not expected to really accomplish them. No one is dumb enough (not even the engineers you described) to put a cat inside a sealed box with a flask of poison that can be shattered as a consequence of some quantum effect to test the concept of quantum superposition. Schrödinger's cat is part of a list of thought experiments such as the Maxwell's daemon (yeah, the same Maxwell of our equations), and many others.



You can obviously discuss the feasibility of reproducing this thought experiment in practice, but to declare, as Mehdi did, that Derek is wrong will require that you be in good terms with the theory, which, as you say, most engineers aren't.

Maxwell is not needed

Avoiding Maxwell is not an option. Whether you are aware that what you doing is described by his theory or not. He's inescapable.

[EEVblog]

As I mentioned - and sorry if it had been already, the thread is now pretty long - as far as experimentation goes, AlphaPhoenix did the trick IMO Unfortunately, I can't see completely clearly on the video all the details of the setup, especially regarding the probing. That's the thing I would be particularly cautious about: making sure any voltage we see on the scope is not coupled to the probes themselves in any way. Other than that, what he gets is congruent with was is expected.

I mentioned this to Derek who didn't seem aware of the probing issue.
I suspect the video is not happening now, as he wanted to get it done before he went on holidays and he's just announced that he's now on holidays, so maybe the probing bit caught him off guard and he's rethinking it.

[SiliconWizard]

Yep. Other than this, as I mentioned above, I'm also interested in what happens along the wire during the transient phase, that is, how the waves are propagating along the wire while having high frequency components. The skin effect comes into play for a short amount of time. That is true for any kind of switched circuit, anyway, and here is going to happen in parallel with the other phenomenon, which is some of the energy coupled to the facing part of the wire loop.

[SandyCox]

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.

[SandyCox]

After doing a lot of reading and thinking, my current opinion is that Veritasium is wrong about electricity not flowing in wires.

At least at DC, Poynting’s theorem and charge flow lead to exactly the same answer. They are just two different ways of looking at the same problem. The one depends on the other and it’s impossible to decide which is the chicken and which is the egg.

Haus and Melcher shows something very interesting in Example 11.3.1 of their excellent book entitled "Electromagnetic Fields and Energy " . This example is about stationary electric and magnetic fields. They show that Poynting's theorem and charge flow gives exactly the same answer. We can choose which one we want to use. If we use both then we will incorrectly conclude that double the amount of energy is dissipated in the coaxial resistors. So, the two perspectives are just two sides of the same coin.
The situation changes when there is AC involved.

I will post a more detailed explanation in a few days.

[rfeecs]

A lot of people, myself included, have said we can model this with a transmission line model, and the transmission line just looks like a resistor equal to the characteristic impedance of the line during the initial turn on transient.

After reviewing Ben Watson's simulation video (https://youtu.be/aqBDFO1bEs8), it is clear that the transmission line model is wholly inadequate.  The transmission line model would result in equal and opposite current on the top and bottom wires.  In reality, the current is higher on the bottom wire.

No engineer should be satisfied with a model that doesn't calculate the correct current.

A next level model would be two coupled transmission lines for the top and bottom wires, shorted together at the ends.  This would at least allow for different currents on the top and bottom.

In reality, as has been pointed out, it is more like a long dipole antenna, or a long wire travelling wave antenna, coupling to another antenna.

This can only be practically modelled using an electromagnetic simulator as in Ben Watson's video.