AlphaPhoenix I bought 1000 meters of wire to settle a physics debate.
1,334,747 views Dec 17, 2021 6,827 comments.
I constructed the Veritasium electricity thought experiment in real life to test the result.
If you were watching my community posts a month ago, the day that Derek over on Veritasium posted his video about electricity misconceptions, you saw me obsess over that problem a bit too much and immediately use it as the excuse I've been looking for for years to own my own oscilloscope. Instead of two light-seconds of wire, I used about 3 light-microseconds of wire, but it was PLENTY to resolve exactly what is happening in this circuit. I hope you enjoy the analysis!
Thanks to Derek at Veritasium for his blessing to make a real-world version of his gedanken experiment. If you haven't seen his video yet, you might want to go watch that for context, and I also highly recommend ElectroBOOM's video on the topic and EEVBlog's video on the topic. Electroboom's video has some simulated scope traces extremely close to what I saw IRL, and a REALLY fantastic animation (8:27) of him waving an electron around in his hand, shedding magnetic fields as it moves (Even though I ignore magnetic fields in this video - I'm trying to think of a test to find out if they matter).
Veritasium https://youtu.be/bHIhgxav9LY
ElectroBOOM https://youtu.be/iph500cPK28
EEVBlog https://youtu.be/VQsoG45Y_00
Pinned by AlphaPhoenix 1 month ago (edited) COMMENTS AND CORRECTIONS:
Thanks to Derek at Veritasium for his blessing to make a real-world version of his gedanken experiment. If you haven't seen his video yet, you might want to go watch that for context, and I also highly recommend ElectroBOOM's video on the topic and EEVBlog's video on the topic. Electroboom's video has some simulated scope traces extremely close to what I saw IRL, and a REALLY fantastic animation (8:27) of him waving an electron around in his hand, shedding magnetic fields as it moves (Even though I ignore magnetic fields in this video - I'm trying to think of a test to find out if they matter).
CORRECTIONS TO THIS VIDEO:
The most important thing I believe I ignored in this video is the actual, physical distribution of charge in the switch-side wire while the current is starting up. How much charge travels AT the advancing wavefront and how much charge gets stuck along the wire in between the fuzzball I drew and the battery will depend on the physical size of the wires and how close they are to each other, setting their capacitance.
This charge distribution also DOES NOT look the same on both sides of the switch, although I drew it that way for simplicity.
In a later experiment (next video) my mind melted a bit as I measured the resistors on both sides of the battery and found the current going through them is different.
It doesn't change any of the logic I presented in this video, but it makes some diagrams less than perfect.
It's possible that cross-inductance between the wires contributes to the effect, using almost exactly the same diagram except the wires are connected by a magnetic field rather than an electric field. I couldn't figure out how to decouple these effects day-of, so I'm still thinking on how to test. Hopefully more to come there.
I'm sure there will be loads more - please leave comments about what I screwed up.AlphaPhoenix I bought 1000 meters of wire to settle a physics debate.Veritasium reckoned that the Poynting Field would light up his bulb soon after 3.3 ns, & that it would shine brightly ever after (& 1 second later a bit brighter when the main current arrives).
AlphaPhoenix (Brian) mentions the Poynting Field at 11:40, but i think that Brian duznt care much for Veritasium's Poynting explanation for electricity. Brian merely said that his 1000 m X confirmed that Veritasium was correct that the bulb would turn on (& stay on) well before the main current arrived.
And Brian conceded that his 1000 m X could not verify Veritasium's 3.3 ns delay. Brian's switch he says takes over 20 ns to work (whatever that means). Brian is using a mickey mouse oscilloscope, ie only 100 MHz, which can't see finer than 10 ns. He needs at least 1000 MHz, about $4,000. He can buy a used 20,000 MHz for $4,000, this can see 1/20th of a nanosecond.
Anyhow, Brian invented his own (unique i think) explanation for his early 0.2 V of current (which later climbed to 1.7 V). At 13:40 Brian says that when he flips his switch….
(1) His battery starts pumping electrons from one side to the other, & (2i) the negative wire gets a negative charge, & (2ii) the positive wire gets a positive charge, & (3) it creates a wave front of electrons pushing electrons along the negative wire, & (4) the pushing is modulated by photons of the em field which (5) travel at the speed of light, & (6) the pushing wave travels along the wire at approx the speed of light, which (7) creates a pocket of concentrated negative charge at the wavefront (going left), & likewise (8 ) we have a pocket of concentrated positive charge in the positive wire going right, & (9) electrons in the top (battery) wire interact with electrons in his bottom (bulb) wire, & (9) electrons sitting on the bottom wire near the bulb are free to move, & (10) they are pushed (repelled) & pulled (attracted) & pass through the bulb, giving (11) (not a lot of) current, but (12) it is almost immediate, (13) via the charge imbalance reaching across the air gap with electric fields, (14) without the far ends knowing. (15) Brian mentions two possibilities for the initial current at the bulb, (15i) capacitance tween the wires, & (15ii) inductance, & he says he will ignore inductance today, & he gives the above explanation for the capacitance effect. (16) Brian does not mention the Poynting Field or the Poynting Vector in his explanation.
My comments are as follows (here i am trying to explain Brian's ideas together with conventional ideas)(compared with my ideas)….(1a)
No. He does not have a battery, he has a 5 V DC source off his AC supply.
(1b)
No. The 5 V does not start pumping when the switch is flipped, it is pumping all the time. Hence the negative wire already has a negative charge (across to the switch) before the switch is flipped, & the positive wire already has a positive charge (around to the switch).
(1c) If the end wires have been cut then the positive wire/charge must end at the cut. Which makes me wonder what the
white trace would look like if that end was not cut & if only the left end was cut (i think the
trace would look weird), & what would be Brian's explanation for the
weird trace result (i think that he would have trouble trying to make his theory fit).
(1b again) The positively charged wire has a concentration of positive charge at the switch (the switch is a capacitor). And a concentration of positive charge near the bulb, in the length opposite the negatively charged wire (the wires are a capacitor).
(1d) Ok, now we flip the switch. Electrons already pushing on the switch now flow through the switch away from the -2.5 V terminal. After a short time this exit of electrons is felt back at the terminal, & electrons then start leaving the terminal, to replace the electrons going to & through the switch. A short time later this exit of electrons from the negative terminal is felt at the +2.5 V positive terminal, & electrons start entering the positive terminal from the positive wire. The 5 V source can't (initially) pump electrons at full flow because the positive wire is (initially) depleted (& to some extent the negative wire is initially over saturated). This shortage of electrons at the positive terminal will not be fully remedied until electrons have managed to flow from the switch around through the bulb & around to the positive terminal (ie 1 full lap of the circuit).
(1d) Indeed the
green trace showing the voltage loss across the resistor near the positive terminal starts at zero volts & then slowly climbs gradually (in a lumpy way), & then does not reach its max voltage until a time corresponding to 1.2 full laps of the circuit. Brian says nothing about the
green trace. His silly pumping idea should show the
green trace starting at full current -- or more logically starting above full current, & slowly dropping to the steady full current as the distribution of charge along the wires reaches steady state, including the usual ups & downs due to any circuit related reflexions.
(4a)
No. I reckon that an em field is not made of photons (not important today).
(5a)
No. The em field does not travel at the speed of light. Or, yes, it does travel at the speed of light, but, the speed of light (& we assume the speed of em radiation) propagates at i think 10 m/s in copper (whereas i am sure that Brian assumes 300,000,000 m/s).
(6a)
No. In view of (5a) above, how can the wave travel at almost the speed of light, or, yes, it might travel at the speed of light, but, the speed of light in copper is i think 10 m/s (whereas i am sure that Brian assumes 300,000,000 m/s).
(7a)
No. In the light of (1abcd), Brian's pocket of concentrated negative charge starts at the switch not at the terminal. And, initially it is matched by a pocket of depleted negative charge starting at the switch & going to the terminal. And after it reaches the terminal there will be a new (small) burst of concentrated negative charge going away from the terminal. But, because the switch is close to the terminal, this complication is i suppose trivial.
(7b)
No. Brian's concept of a pocket of charge at the wavefront is not realistic. The wavefront gives a leading edge, but the negative charge extends all of the way back to the terminal.
(7c)
No. Brian thinks that he needs a pocket of negative charge well left of the bulb to repel electrons in the bottom wire to the right towards the bulb. A pocket of charge is not needed. Any kind of general distribution of negative charge in the top wire will repel electrons in the bottom wire, & common sense tells us that some of these will go right (for a while), no pocket needed.
(8a)
No. We do not have a pocket of concentrated positive charge going right in the positive wire. Or, yes, we do, but, the
green trace shows us that any such effect on the positive side of the circuit would be zero at first, & would take a long time to grow. But, the
white trace does not show any evidence of that kind of growth. And, the pocket of positive charge is supposed to contribute say a half of the current through the bulb, ie a half of the
white trace voltage, hence the
white trace should definitely have growth (but duznt). Or, putting it another way, the
green trace should not have growth (but it duz).
(7d) Similarly to (8a) if Brian showed us his
trace for the resistor near the switch, i reckon that the current would have a quick spike & then fall, & after that grow in a similar fashion to the
green trace.
(9a)
Yes. Brian says that electrons sitting
on the bottom wire near the bulb are
free to move. Yes, it is the (conduction) electrons sitting
on the surface of the wire (not in the wire) that are free to move. That is a key part of my own idea.
(10a)
Yes &
No.
Yes, electrons are individually pushed (repelled) & pulled (attracted) & pass through the bulb. But, on the left half of the circuit, the overall charge in the top wire squeezes electrons out of the bottom wire, some going left (for a while), & some going right (through the bulb). And on the righthand half of the circuit the overall positive charge in the top wire attracts electrons in the bottom wire, & electrons near the bulb flow to the right (for a while), pulling other electrons through the bulb behind.
(11a)
Yes. There will be some (not a lot of) current through the bulb, before the main current arrives.
(12a)
Yes. The initial small current is almost immediate. However Brian's oscilloscope can't tell us exactly what happens in the first few nanoseconds, hence he can't actually answer the Veritasium gedanken question.
(15a) Brian mentions two possibilities for the initial current at the bulb, capacitance & induction. We could add radio as a separate class.
(15ia)
Yes. Brian ignores inductance, because capacitance is the main culprit.
(15ib)
No. Brian reckons that he explains the initial capacitance effect, but he duznt. His explanation has little resemblance to capacitance.
(16a)
Yes. Brian quite correctly ignores Poynting in his explanation (unlike Veritasium who loves Poynting). Brian could have emphasized that Veritasium's Poynting explanation does not explain even one electron of what happens. Brian could have explained that Veritasium genuinely reckoned that there would be a very significant initial Poynting electric current. As it turns out there is indeed a significant initial current, but because of capacitance, not Poynting.
So, Brian's X pt1 did a good job, & we are all eager to see pt2. And me myself i want to see the trace for the resistor near the switch.