### Author Topic: EEVBlog 1439 - Analyzing Veritasium's electricity video  (Read 8348 times)

0 Members and 1 Guest are viewing this topic.

#### Sredni

• Frequent Contributor
• Posts: 491
• Country:
##### EEVBlog 1439 - Analyzing Veritasium's electricity video
« on: November 24, 2021, 03:15:57 pm »
Dave, you should correct your video, the errors are too big.

At minute 3 something ops 9:30 you say that at DC the Poynting vector would be pointing towards the battery. This is wrong.
At minute 41 something you show a picture of a resistive wire from Feynman that shows the Poynting vector going inward, into the wire. Well, that's a resistor, power is getting in to be dissipated.

Even at DC the Poynting vector is going out from the battery, then goes nearly parallel to the copper wires, and then finally points inside the resistor. You misread Feynman (who, by the way, for fig 27-5 explicitly talks about 'resistive wire' which has a resistance and a voltage drop).

Simple experiment:
9V battery shorted by copper wire: the wire gets hot. Poynting vector pointing out of the battery and into the wire. (the wire is the dissipative element in this context).
9V battery with a 100 ohm resistor connected via copper wires: the resistor gets hot, the wire doesn't. Poynting vector directed out of the battery, nearly parallel to the wires, and then plunging into the resistor. As correctly shown by Veritasium (for DC)

And no, skin effect has nothing to do with that.

EEVblog 1439 - Analysing Veritasium's Electricity Video

Dave analyses Veritasium's video "The Big Misconception About Electricity" and how energy flows in the Poynting vector in the electromagnetic field OUTSIDE the wire instead of inside the wire.

00:00 - Veritasium's video "The Big Misconception About Electricity"
00:32 - Rection to the points in the video
01:11 - This is a bit MISLEADING!
02:28 - Electron drift
03:51 - Engineers use different tools and theorems
04:27 - Every electrical engineer knows this
05:17 - Everything he says is correct
08:24 - What is current?
09:30 - He doesn't address this in the video. Poynting vectors at DC
11:12 - How the lightbulb works
12:41 - At the physics level, it's correct
14:11 - My only problem with this is...
15:08 - Is it just an academic discussion?
16:17 - The undersea cable is just early transmission line theory
17:20 - So what is the answer to the question?
22:06 - What about skin effect and DC?
25:44 - Let's simulate this and answer the question
29:18 - Transient analysis
33:00 - DC Steady State analysis
34:28 - The quantitative values don't matter
40:24 - What does Richard Feynman think?
« Last Edit: November 24, 2021, 05:43:14 pm by Sredni »
All instruments lie. Usually on the bench.

The following users thanked this post: Amaruk

#### golden_labels

• Frequent Contributor
• Posts: 582
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #1 on: November 24, 2021, 04:11:27 pm »
One thing Derek was not wrong about: that video will spawn a deluge of comments, with everyone correcting everybody else.

Has anyone so far realized that the video ventured deep into the territory of epistemology and metaphysics? A large part of the whole discussion is caused by making assumptions on the meaning of “truth”. To be more specific: confusing models with absolutely true essence, the inherent nature of the matter discussed. Neither of those explanations is more true. At best they may differ in their practical usefulness or how close they are to the most detailed views of the phenomenon. The Poynting vector approach is closer to the more detailed model, but by no means closer to any kind of truth.

Both explanations may be misunderstood and lead to wrong interpretations. It’s not hard to imagine that electrons’ movement play completely no role, after watching just that Veritasium video, if one doesn’t ask a question: how comes magnetic field is non-zero? Also the energy transfer lines (yellow) are reduced to direction only and miss magnitude, which — if considered — would draw a less surprising picture.
Dihydrogen monoxide was responsible for Fukushima, Chernobyl and TMI disasters
Worth watching: Calling Bullshit — protect your friends and yourself from bullshit!

The following users thanked this post: AlienRelics, bdunham7

#### Sredni

• Frequent Contributor
• Posts: 491
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #2 on: November 24, 2021, 04:36:42 pm »
Well, the role of surface charge is very often not mentioned in most introductory books. Chabal and Sherwood have started a new trend in that sense. But one thing is trying to analyze the phenomena with different tools (propagation of fields and redistribution of surface charge, use of transmission line model in the transient phase, even antennas if you will) and another is getting basic physics completely wrong.
To say that at DC the Poynting vector is directed toward the battery, or even towards the wires when the complete circuit has a resistor that will sink the power, is just plainly, unmistakenly and uncontroversially wrong.
It's not a matter of intepretation. It's wrong.

(I corrected the minute in the video were Dave says that)
All instruments lie. Usually on the bench.

The following users thanked this post: Amaruk

#### Amaruk

• Contributor
• Posts: 44
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #3 on: November 24, 2021, 05:11:43 pm »
Dave, you should correct your video, the errors are too big.

I am a big fan of Dave and have watch so many of his videos and I am always amazed by the amount of knowledge shared in them.  Thank you! This video is a bit different though but it was kind of forced upon Dave by his viewers. The result of this is that this video does not feel as solid as the others I have watched... But who cares, we all learn stuff on here and who has not said things that are not correct when put on the spot anyway? These are difficult theoretical discussions so it just shows that Dave is human after all!

#### Kalvin

• Super Contributor
• Posts: 2017
• Country:
• Embedded SW/HW.
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #4 on: November 24, 2021, 05:59:38 pm »
When people are saying "At DC this and that ..." what do they really mean?

- Do they mean the static situation when there is an electric field present but no current is flowing (DC = 0)? In this experiment: Battery connected to the circuit, but no closed circuit present, circuit is in steady state. Because there is no charge moving in the circuit, there is no magnetic field present, thus there cannot be any Poynting vector.

- Do they mean that there is an electric field present and there is a [constant] current flowing in the circuit (|DC| > 0)? In this experiment: Battery connected to the circuit, switch closed, and constant current flowing into the load [after transitions have disappeared ie. circuit is in steady state again). Now, as there is charge moving in the circuit creating a magnetic field, there is a Poynting vector, too. The Poynting vector will show the direction of the energy transferred. In this experiment, and if the signs are correct, the direction should be from the battery into the load.

When the conductor is lossless, the Poynting vector is parallel to the conductor.

How about a situation when there is a good conductor, but somewhat lossy nevertheless? If I understood Feynman correctly, the Poynting vector should tilt a little towards the conductor, creating two vector components. The first component is parallel to the conductor, and which is transferring energy towards the load. And the second one, which is into the conductor, representing the losses.

Makes any sense?

What is the situation when we have a resistive wire connected directly on the battery, ie. the load is this resistive wire? This should not be solved using a traditional lumped model, because the load is now distributed along this resistive wire: How the Poynting vector is represented along the wire in this case?

#### jan.met

• Newbie
• Posts: 2
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #5 on: November 24, 2021, 06:42:19 pm »
Hi,

I think answer D for the switch question is wrong. It would be true, if the electric field would build up instantaneouly along the wire, so that current starts flowing immediately through all of the  wire. But the change in the field itself is propagating only at the speed of light along the wire, so that the current build up in the wire takes time. Without current at the bulb, the Poynting vector is zero and the bulb can not receive energy.

Greetings, Jan

#### Sredni

• Frequent Contributor
• Posts: 491
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #6 on: November 24, 2021, 06:51:13 pm »
@Kalvin - "At DC" means in the steady state condition with a current flowing. Now, Veritasium's video is about the transient before DC, but as you correctly stated, in DC with a steady current there are both a magnetic and an electric field.
That are directed as shown by Veritasium and that give the direction of the Poyntin vector shown by veritasium.
So, near the battery the Poynting vector is directed outwards, then it 'bends' to follow the wires and finally plunges into the load.

Your analysis of the importance of the resistivity of the wires is correct.
If the wires have zero resistance, the electric field immediately outside their surface is orthogonal. If they have resistance, the electric field will be slanted. The higher the resistance, the higher the voltage drop, the higher the electric field inside the component. This will cause the Poynting vector to point toward the component, in this case the wire. But you need to take into the picture what else is in the circuit. If there is a load resistor, the current will be limited, so the B field will be lower and the bending in the low resistance wires will be only slight (because compared to the load, they drop nearly zero voltage); in the load, the field is strong and even with the same B field going around it, this will give a Poynting vector directed towards the resistor.

The resistivity of the wires determine how much the Poynting vector is slanted - how much power is dissipated in the lossy wires.
The bigger the conductor, the less the slant.
All instruments lie. Usually on the bench.

The following users thanked this post: Kalvin

#### Sredni

• Frequent Contributor
• Posts: 491
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #7 on: November 24, 2021, 07:08:56 pm »
Hi,

I think answer D for the switch question is wrong. It would be true, if the electric field would build up instantaneouly along the wire, so that current starts flowing immediately through all of the  wire. But the change in the field itself is propagating only at the speed of light along the wire, so that the current build up in the wire takes time. Without current at the bulb, the Poynting vector is zero and the bulb can not receive energy.

Naaa, it's roughly correct.
Current in the wires build up gradually, as the surface charge adapts to the new conditions. When you flip the switch, the charge that had accumulated at the open switch terminals starts to recombine and produce a complex distribution of surface charge that gives the electric field inside the conductor.

Have a look at my answer on SE
https://electronics.stackexchange.com/questions/532541/is-the-electric-field-in-a-wire-constant/532550#532550
in particular the last picture, where the process of "E field forming inside the wire" is shown.
According to this picture (which consider the rest of the circuit so far away as to not influence the charge distribution) we should wait for the perturbation to go halfway to the moon and back to reach the lamp. Surface charge is extremely fast and the perturbation travels nearly at the speed of light.

But the surface charge is also influenced by nearby portions of the circuit. As you can see from the different configurations the electric field can assume near the surface of a conductor (and highlighted in the simulation shown in that answer - from a paper reference by Veritasium, as well).

In fact, the charge that is redistributing near the switch produces an electric field that travels across the 1 meter space between the wires to the lamp and the nearby portion of wires. This will try to shape the electric field inside those conductors to the final steady state, but it still lacks the contribute from the charge redistribution that is traveling towards the moon and back.
But nevertheless, the local surface charge near the switch is affecting (after 1/c seconds) the surface charge near the lamp (and consequently the field inside the wire in that portion).

What will be the magnitude and the direction inside the wire in that transient? I am not sure, it might be complicated. A simulation might be required. And you need to take into account that you cannot use the inductance per unit length when the steady state is reached. Probably only the internal inductance counts until the full current is flowing in all of the circuit... I would love to see a simulation.
« Last Edit: November 24, 2021, 07:10:41 pm by Sredni »
All instruments lie. Usually on the bench.

#### firewalker

• Super Contributor
• Posts: 2409
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #8 on: November 24, 2021, 07:33:42 pm »
The capacitor explanation is somehow misleading? What if the lamp was also 300000 km away from the battery? The capacitor is still there.

Alexander.
Become a realist, stay a dreamer.

#### Kalvin

• Super Contributor
• Posts: 2017
• Country:
• Embedded SW/HW.
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #9 on: November 24, 2021, 08:14:39 pm »
In my opinion using a simple transmission line-model is sufficient for explaining Veritasium's experiment. In addition to that, knowing that it would take 1m/c seconds (3.3ns) before the change of electrical/magnetic field propagates through the distance of 1m will provide the answer.

The capacitor explanation is somehow misleading? What if the lamp was also 300000 km away from the battery? The capacitor is still there.

Alexander.

Using a capacitor as an explanation is not a good model, because it will not explain for example why the lamp/load will not get the full power right after the switch is closed in Veritasium's original circuit. A transmission line and its characteristic impedance will provide better and more realistic model and correct answer.

Answer to your question: It would take approx. one second or so before the lamp will be lit / current starts flowing through the load. The actual delay is depending on the cable's signal propagation speed.

Pls find my simulation below. I run the simulation for 20 seconds in order to visualize the reflections, and how the circuit is gradually converging towards the steady state.

Edit: Please note that I have used a 50 ohm transmission line as the model in my simulation. In Veritasium's experiment the characteristic impedance of the transmission line will be different, and the transient current through the load will be different, but the principles and the explanation will remain the same.
« Last Edit: November 24, 2021, 08:37:54 pm by Kalvin »

The following users thanked this post: AlienRelics

#### Kalvin

• Super Contributor
• Posts: 2017
• Country:
• Embedded SW/HW.
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #10 on: November 24, 2021, 09:02:32 pm »
For completeness: Here is a simulation for Veritasium's experiment using a 50 ohm transmission line and a 1 Kohm transmission line. It can be seen that when the characteristic impedance of the transmission line is high, it will take much longer time before the system converges to the steady state.

The following users thanked this post: AlienRelics

#### Unixon

• Frequent Contributor
• Posts: 381
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #11 on: November 24, 2021, 09:16:15 pm »
Dave, how could you put the ground symbol on the wrong side of the switch?
This breaks the definition of the ground reference point.

#### Kleinstein

• Super Contributor
• Posts: 11057
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #12 on: November 24, 2021, 10:17:15 pm »
For completeness: Here is a simulation for Veritasium's experiment using a 50 ohm transmission line and a 1 Kohm transmission line. It can be seen that when the characteristic impedance of the transmission line is high, it will take much longer time before the system converges to the steady state.

Is there something like a 1 KOhms transmission line ?  I though there is a kind of natural limit at some 370 ohms.

#### bdunham7

• Super Contributor
• Posts: 4945
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #13 on: November 25, 2021, 01:25:36 am »
Is there something like a 1 KOhms transmission line ?  I though there is a kind of natural limit at some 370 ohms.

The impedance of free space is 377 ohms or something, but yes you can have characteristic impedances higher than that without any problem except the increasingly improbable dimensions (small wires, large spacing).
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.

The following users thanked this post: AlienRelics, notadave

#### Kalvin

• Super Contributor
• Posts: 2017
• Country:
• Embedded SW/HW.
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #14 on: November 25, 2021, 02:32:11 am »
For completeness: Here is a simulation for Veritasium's experiment using a 50 ohm transmission line and a 1 Kohm transmission line. It can be seen that when the characteristic impedance of the transmission line is high, it will take much longer time before the system converges to the steady state.

Is there something like a 1 KOhms transmission line ?  I though there is a kind of natural limit at some 370 ohms.

Yes, it is possible to build practical 450 ohm or 600 ohm open-wire ladder line transmission lines: https://www.qsl.net/co8tw/openline.htm.

Parallel feeders go back to the beginnings of radio. By 1930, the "two-wire untuned feeder system" was a standard ARRL Handbook feature. The Jones Radio Handbook of 1937 provides a table of line losses showing the advantages of open-wire feeders (a 440-Ohm line in the table) over lower impedance twisted-pair feeders (p. 70). The use of 600-Ohm lines was fairly standard, using a spacing of about 6". "To reduce radiation from the feeders to a minimum, the two wires should not be more than 10 to 12 inches apart." (The Radio Amateur's Handbook, 7th Ed., ARRL, 1930, p. 162) Rarely did hams exceed the 6" spacing.

I wasn't really paying attention to whether 1 kohm characteristic impedance is practical or not with, because in this context we are talking about theoretical, lossless 300 000 km transmission line anyway.

But yeah, I should have probably used 450 ohm or 600 ohm transmission line impedance in my last simulation.

The following users thanked this post: AlienRelics

#### David Hess

• Super Contributor
• Posts: 14327
• Country:
• DavidH
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #15 on: November 25, 2021, 04:06:53 am »
Dave did not mention it exactly, but while the transmission lines are charging, which powers the bulb immediately, they are the equivalent of a resistance equal to their characteristic impedance.  I have occasionally run across circuits which took this into account, like sampling circuits which use a transmission line instead of a capacitor.

It seems that Mehdi (Electroboom) is going to do a video about it also.

The following users thanked this post: AlienRelics

#### golden_labels

• Frequent Contributor
• Posts: 582
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #16 on: November 25, 2021, 10:41:23 am »
The capacitor explanation is somehow misleading? What if the lamp was also 300000 km away from the battery? The capacitor is still there.
Then the wires in the Battery-C-Lamp-C loop would be 300Mm long. Dave has clearly mentioned it multiple times, that this is a situation with them being 1m long. If that would be 300Mm, then the signal would have to propagate over those 300Mm-long capacitors and it would increase time.

My gripe with Dave’s approach is that it’s skipping one important detail. The elements in there are infinitesimals. Which means that while response is immediate,(1) the current and voltage are also infinitesimals. In that case it’s a bit of a stretch to say the load receives power. This is the reason why increasing decreasing time step in a numerical calculation decreases the pulse duration: that pulse has no length.
____
(1) Subject to propagation time limits.
« Last Edit: November 26, 2021, 06:36:40 am by golden_labels »
Dihydrogen monoxide was responsible for Fukushima, Chernobyl and TMI disasters
Worth watching: Calling Bullshit — protect your friends and yourself from bullshit!

• Contributor
• Posts: 48
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #17 on: November 25, 2021, 02:28:57 pm »
Does not anyone disagree with the statement and picture:
"E-field and B-field are in phase." ?
I have seen that statement three times on youtube and no one disagrees:
3:40
5:40
6:30

#### nctnico

• Super Contributor
• Posts: 22819
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #18 on: November 25, 2021, 02:48:23 pm »
In the end it is highly academic and a lot of the fine details are left out. In an experiment of the proposed size the rotational speed of the earth is also going to play a role depending on the orientation of the wires..
« Last Edit: November 25, 2021, 02:53:40 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.

#### bdunham7

• Super Contributor
• Posts: 4945
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #19 on: November 25, 2021, 03:30:59 pm »
In the end it is highly academic and a lot of the fine details are left out. In an experiment of the proposed size the rotational speed of the earth is also going to play a role depending on the orientation of the wires..

Yes, you might get a bit more voltage than you bargained for, enough to vaporize your entire experiment.  I'm sure what will happen is a scaled-down version of this will be set up that is claimed to be the equivalent.  And at that point it will become obvious that the experiment demonstrates nothing of much interest or value, but that won't matter to the target audience.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.

#### Sredni

• Frequent Contributor
• Posts: 491
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #20 on: November 25, 2021, 04:04:30 pm »
Ben Watson did a very nice simulation of the fields, that shows how the propagation starts from the switch and reaches the lamp first, then propagates along the wire to the left and right end, senses the kind of termination there (open, short, load? you cannot know until your fields get there and back), then bounces back and after a few bounces you reach steady state.

Response to Veritasium - Electricity Propagation Time Problem

There is also Eric Bogadin's take in a recent video by Feranec that is more transmission-line centered. The point is that voltage and current are just approximations and we need to fold back to fields.

My personal note: what is most interesting is that the fields that reach the lamp after a few nanoseconds are the same irregardless of the far ends being shorted or opened (it remains to be seen if - when considered in that particular topology the characteristic impedance is the same as when the generator in on the left and the load is on the right... Is that a long and narrow line or a short and wide one?)

On second thought, this needs to be clarified: the fields after a few ns would be the same if the far ends are opened simultaneously with the switch closing. If the far end were already open, the charge accumulation would not be exactly the same at both the switch terminals (a different initial condition).
« Last Edit: November 25, 2021, 05:24:03 pm by Sredni »
All instruments lie. Usually on the bench.

The following users thanked this post: Fungus

• Regular Contributor
• Posts: 205
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #21 on: November 25, 2021, 04:29:06 pm »
Here is another excellent analysis by Eric Bogatin from Teledyne LeCroy:

https://youtu.be/Lp_b8gQpxW8

#### Bud

• Super Contributor
• Posts: 5444
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #22 on: November 25, 2021, 05:37:33 pm »
I remain of the opinion that the problem has nothing to do with " transmission line" approach and has to be explained in classic electromagnetic theory terms. If anyone based his explanation on  "parallel wires 1 meter apart" , or " X mm trace on an FR4 PCB, the line impedance and such, if they instead wrap the wires around the Globe in a circular shape rather than running them in parallel, their answer to the challenge may differ.  This will right away indicate this approach is wrong.

• Regular Contributor
• Posts: 205
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #23 on: November 25, 2021, 06:28:03 pm »
Same phenomenon can be described multiple ways. For example, motion of planets and satellites in Solar System can be described both with Newton’s Law of Universal Gravity and General Relativity (and a future Quantum Gravity Theory that is yet to be discovered). It is well known that both Newton’s Law and GR are wrong (the former does not explain many experiments and observations, the latter epically fails at quantum scales with variances on the scales of 10 to 120th power or so). Yet still Newtonian Law is good enough for JPL and NASA engineers in majority of the cases.

Same here. Classical Electrodynamics describes reality better than Ohm’s Law, while being classical approximation of Quantum Electrodynamics. Still, in majority of the cases, electrical engineers are better off with Ohm’s Law. Good luck to anyone wishing to describe a simplest electrical circuit with QED.
« Last Edit: November 25, 2021, 06:30:08 pm by vad »

• Regular Contributor
• Posts: 205
• Country:
##### Re: EEVBlog 1439 - Analyzing Veritasium's electricity video
« Reply #24 on: November 25, 2021, 06:48:23 pm »
I remain of the opinion that the problem has nothing to do with " transmission line" approach and has to be explained in classic electromagnetic theory terms. If anyone based his explanation on  "parallel wires 1 meter apart" , or " X mm trace on an FR4 PCB, the line impedance and such, if they instead wrap the wires around the Globe in a circular shape rather than running them in parallel, their answer to the challenge may differ.  This will right away indicate this approach is wrong.

It does not matter if you run wires straight in parallel or wrap around the globe, it does not matter if the length of the wire is 300,000 km or 30 meters. You will see displacement current in the opposite wire in d/c seconds, where d is the shortest distance from the switch to the opposite wire in meters, c is light speed in the medium in m/s.

The initial displace current will be as high as voltage divided by intrinsic impedance (e.g. for wires that spaced at significant distance, the impedance will be close to 377 Ohms in air).

The following users thanked this post: AlienRelics

Smf