Is Veritasium Wrong About Electricity?

Is he right? I'm not so sure.
Last week, Veritasium released a video presenting a thought experiment involving a battery powered circuit connect to a lightbulb with wires extending in 1/2 a lightsecond in each direction. Suggesting that the lightbulb would turn on at a time dictated by the distance between the battery and the lamp, ignoring the length of the wires.

Are you interested in launching a start up company from your research?
Check out my Skillshare course on "How to Start A Company From Your Research":

A few people have asked so I've added the info below. Some/all of these may be affiliate links. If you make a purchase it doesn't cost you anything extra, but a percentage of the sale will help support this channel and my work to bringing entrepreneurship into science.

My gear:

[ap-pvug]

Veritasium featured your video in his new response video.

[ElectroBOOM]

One thing to remember is that the two wires being one meter apart have some capacitance between them, and such capacitance existing right from the start transfers some power to the lamp. as the switch turns on and voltage on the line flips. Veritasium's video is designed with all sort of clues and limitations around the problem that are hard to pick up. But at the end of the if we go with all limitations Veritasium imposes on his problem, he is right.

[radben]

Your video really helped me understand Derrick's and upon rewatching his I have some criticism. He talks about two different effects and doesn't do a good job separating them. The main point of his video is that the propagation of electromagnetic fields along a power line that deliver energy and not a flow electrons as many people believe. The second point, and one he does a poor job differentiating from the first, is that not only do these fields flow along the wire they also flow outward far enough to overcome the 1m air gap and turn on the bulb. Your criticism, and that from some of the top comments here, is that while there may be some current induced in the bulb it wouldn't be anywhere near enough to light it. My biggest criticism of his video is that he implies this is a significant or even the primary flow of electricity to the bulb. He does say it wouldn't be full power (which would arrive in 1s) but he makes them sound like equivalent effects and this conflates the viewer's understanding of this effect with the earlier explanation of how power gets to people's homes.

[vorniy]

"Feel free to argue with me and change my mind. That's totally fine"
One of the most fundamental principles of how science should work and be communicated.

[edwardharvey]

When I was a physics student, it came to me as quite perplexing, the reality of fields, until suddenly it all mostly made sense. It seems Derek at Veritasium has hit an interesting nerve, the difference in the way physicists and electrical engineers see electricity. Tesla thought that electricity could be best served by broadcasting it into the air by which consumers would extract it out of the air. I think Tesla didn’t quite get it, but seems to have been on the right track. Tesla famously lit fluorescent lites, which he invented, wirelessly from the electrical energy he pumped into the room.

[teacherhomieg]

I understood the thought experiment at Veritasium to be more of an argument that the electrons themselves don’t light the bulb up as they move through the filament of the bulb but rather that it’s the energy that flows outwardly that powers the bulb into illumination. I think that was his point. The “lie” part being the rudimentary understanding of current as the flow of electrons that provide the energy to make things work as they flow through wires. Tesla proved you could turn on lights and make things work without wires long ago. Being an 8th grade science teacher, I do agree with your assessment of teachers needing to simplify concepts to an extent for students provided the results are correct. We too undergo retraining and revise what we teach every year.

[ajreukgjdi]

This is my disclaimer: this is my field. However, it's been 4 years since I took the class that would be most relevant here and regularly worked with the formulas needed to predict this counter-intuitive concept. Here's my take on assumptions 3 and 4. First, to combat the faster than light communication scenario you mentioned. I don't think the time relies on the distance between the battery and the light, but the switch and the light. The side of the switch connected to the battery must be equipotential to the battery (assuming the setup has been sitting for any reasonable amount of time). Therefore the electric field strength is exactly the same at each of those 2 points (the battery terminal and connected switch terminal) so closing the switch would be equivalent to just connecting the wire on the other side of the switch to the battery directly. I think it's safe to say the time depends on the distance between the switch and the battery. Since they were negligibly close to each other in the picture we were given, the answer should be the same. However, if a break develops at either end, it will take the field time equal to d/c to get from the break to the bulb. This does not fix another potential faster than light communication issue. Instead of breaking the circuit after being closed, let's break the circuit before being closed. i.e. a switch at the far end of the arm. Now when we close the first switch, by our mental model, the light will either turn on nearly instantaneously or it won't depending on the state of the second switch. Thus we can know the state of the other switch faster than light could carry that information. And I think you covered this one when you broke down assumption 3. The antenna-like properties of a 1/2 Ls long wire would mean the light 'turns on' immediately regardless of the switch 1/s Ls away. However, it will take time for the electric field to propagate to the end of the wire, where charge will start to build up. Like filling a cup, that charge buildup will eventually back-propagate to the battery. when the charge density creates the same potential as the battery terminal, current will cease to flow, and the bulb will slowly dim and turn off. This would also differ from alternating current where the charge never has time to build up to the point where no current flows and so the bulb remains on.

[HMan]

Your example with two dipole antennas was perfect. You can also model it as a transformer. Veritasium isn't wrong per se, but his example is all sorts of crooked, and terribly confusing. He doesn't even mention that the geometry of his wires is the only reason he's getting that effect. Or that the bulb will not actually "light up", but rather that a very tiny small fraction of the current will make it to the bulb in 1m/C s... It's just confusing and didn't need to be.

[evrimagaci]

This was a very nice point, I liked your thinking. I am not sure if you allow links, but I urge you to read "Understanding Electricity and Circuits:
What the Text Books Don’t Tell You" by Ian M. Sefton from School of Physics at the University of Sydney (the paper is from 2002, and obviously, one can find a similar discussion in Feynman's Lectures). Derek's video is pretty much a repetition of that (which is also explained by Science Asylum before). I believe the reason why he called it a "lie" is to just resonate with that paper. Thanks for the video.

[das_it_mane]

"Feel FREE to argue with me and change my mind; that's totally fine."

Love that you said something like this AND the way you said it!

[Sunflowrrunner]

I'm not a physics guy, but even I was confused about the faster than light communication problem.

[disgruntledwookie]

It really helps to consider the following (imo)

1. This whole thing has been framed in a misleading/loaded way by talking about circuits from the get go, when there is in fact no circuit to speak of at least within the window of time he's talking about (~1m/c)
2. Locality is a thing. No matter how you dress up your puzzle, the fact is it's impossible for the lamp to "know" about a break in the circuit at some distance d from the origin until at least 2d/c seconds have passed after closing the switch.
3. If the ends of the wire are far enough away that you have to account for the speed of light, then you need to stop thinking like an electrical engineer because this is no ordinary circuit. Changes in applied voltage will take a long time to propagate along the wires, so you aren't getting the usual self-interference. When we attach a wire to the negative terminal of a battery, without completing the circuit, we would say there is no circuit and hence no current. But we would be wrong; there is a current at first, but the electrons have nowhere to go and begin to bunch up against the end of the wire. This produces an electric field to counter the applied field and after some time, an equilibrium is reached and no more current flows. If the length of the wire is L and we (wrongly btw) assume that the signal travels at c, then we ought to expect it take roughly 2L/c seconds for the current to stop flowing. Given a meter of wire... that's 6.7 nanoseconds, so yeah we don't notice it. But if your wire is 1/2 a light-second in length, it will take about 1 second for this equilibrium to be reached. During which time, it is IMPOSSIBLE to determine by any means whether the wire forms a complete circuit, or reaches a dead end. In BOTH cases, the behaviour within the first 1 second will be identical, only diverging afterwards. So whatever energy reaches the lamp before that time (and it does exist for sure), would have done so regardless of whether the circuit was actually closed. It's a tiny amount of energy, not comparable to energy flux that you would have AFTER 1s given a closed circuit.

[wmcewa]

“But I’m very happy to be convinced [challenged], because that’s the whole point of science.”
YES
I wish more people thought this way.

[SuperSaiyanShaak]

I feel like this is something we can easily check with an experiment. Since 30 cm is traversed in about a nanosecond at the speed of light. We can set this up with a distance of 30 cm from switch to lamp and 3 m around the circuit. This should still be measurable with electronic devices and no superconductors or crazy voltages are needed.

[TurpInTexas]

For RF engineers, this is a no-brainer scenario as this is transmission line theory 101 and this example is essentially a trick question. When the switch closes, the voltage pulse generates a magnetic field starting at the switch, and also encompasses the adjacent parallel conductor causing an opposite current to flow, but this only applies to a changing magnetic field. Since this is DC, once the field stabilizes the magnetically induced current in the adjacent wire will cease but by that time the electron flow will power the lamp. In an AC scenario the changing magnetic field keeps the light on. In a scenario where the wires are separated and mutual magnetic is kept to a minimum, then we fall back on the round trip delay calculation, barring any capacitance issues that contribute to velocity factor (yes, like optical fiber, wires have velocity factors too). Thanks for the video, I thoroughly enjoy the thought experiments and getting folks engaged in discussions about science.

[chriswoollacott]

You can understand this by considering the parallel wires as transmission lines of characteristic impedance Zo. Zo is a function of the inductance of the wires and the capacitance between them. When the battery is first switched on there will be an almost immediate step increase in current defined by the battery voltage and the characteristic impedance of the transmission lines plus the lamp resistance. The lamp will light but only dimly. After a delay equal to the delay of propagation to the end of the line and back again, an increased voltage will be applied across the lamp and it will be brighter. As the transmission line is not connected to a matched impedance at either end, reflections will go back and forth and the voltage across tge lamp will gradually step into its final battery level.

The characteristic impedance of a zero resistance line, Zo, is the square root of the ratio of the inductance per unit length to the capacitance. If Zo is very high, the lamp may well not light up until the signal returns but, if it is very low, it may light up almost immediately. It all depends on the geometry.

Of course, all of this occurs due to the magnetic and electric fields for it is they that define the transmission line inductances and capacitances.

[StrollHikes]

I like this analysis. The key to understanding Veritasium's video is to realize that he spent the majority of his time describing how a radio works, not how practical power transmission works. (This is also Not My Field.) If you have two absolutely ginormous radio antennas only one meter apart, yeah, sure, you're going to get some serious coupling between them, and they don't even need to be connected at the ends. Or another way to think about it is, if the online calculator that I found can be believed, Derek's thought experiment describes the construction of a ~1, 200 uF capacitor and when you first turn that sucker on you're absolutely going to get some current flow across the "plates" as the capacitor charges up. Quite a lot of current, actually. But the line-of-sight power transfer stops as soon as the capacitor is charged and if you want continuous power transmission, you're definitely going to have to wait for EM propagation along the length of the wire, otherwise causality is violated. I don't think the Veritasium video was wrong, precisely, but he ended up answering a different question than the framing of the thought experiment led people to expect, so ultimately I think it's a confusing piece of work.

[dr.palsonp.h.d]

The Poynting vector theory is correct, there would be a very small amount of power being delivered right away, but to get any significant amount of power, the the electrical wave would have to propagate for a longer time than 1m/c. also in terms of a capacitor (that electroboom says) the equation for the voltage would vary with time, and therefore there would be some displacement current aswell in the other line (also time varying). so that would be power delivery in terms of V*I as well.

This is all taught in any fields/waves and antenna design course so framing it in that way I think would make it easier to understand why that is the case.

To frame this as a circuit does seem a bit misleading--because the video makes it out to be that the electrical power flows through the air and not through wires -- which is somewhat true but not as magical as the video makes it seem, although it does add to the effect and makes for a good engaging video, so I can see the reasoning behind it.

[cdscan]

This is a great experiment because it asks us to think clearly about what is actually happening, vs blindly following what we are taught.

[rasg]

When Derek first posted an image of this problem on his channel, my gut told me that regardless of current, resistance, voltage, etc it had to do with the speed of light or the speed that the information of the switch being closed would reach the light bulb. I was shocked when he stated that it would almost be instantly because as you said it would break everything we know about relativity. I'm glad to know I'm not alone. Hopefully Derek will do a follow-up video answering this issue