Wikipedia:Reference desk/Archives/Science/2020 October 4

= October 4 =

Electricity flow
If a person touches a +600v DC power line and it is unearthed, will the electrons flow through the person to the ground or will it enter and exit at the same point? 2A01:4C8:494:F9D2:D87C:7EA0:B27C:AE55 (talk) 14:34, 4 October 2020 (UTC)
 * There has to be a voltage difference for current to flow. If the person is grounded, current will flow through them to ground. With appropriate precautions you can touch energized power lines as long as there's no path to ground. Obviously don't do this unless you're trained; a Web search will give you lots of videos of line workers working on energized high-voltage transmission lines by flying up to them by helicopter and equalizing themselves to the line voltage. --47.146.63.87 (talk) 18:18, 4 October 2020 (UTC)
 * Note that in this context "ground" can be anything else the person touches that is not electrically isolated but works to complete a closed electric circuit to the power source. This cannot be the same point as where they touch the power line, because then there is no voltage difference. It does not matter whether this is a +600V or −600V line, and also not whether it is AC or DC. --Lambiam 19:23, 4 October 2020 (UTC)
 * And touching means within a certain distance of completing which varies with air conditions and is several fathoms or meters for high enough voltage lines. Even at several meters there could be corona discharge (not that corona) then at the exact distance artificial "lightning" completes the circuit causing death in milliseconds. Sagittarian Milky Way (talk) 19:37, 4 October 2020 (UTC)

Could you predict if a specific attempt to crush a flying mosquito with a grab will work if you had enough hand, wind and mosquito data?
Or is a human attempt at this this already into the realm of turbulence being chaotic even if you knew every non-air non-mosquito atom as precisely as Heisenberg's uncertainty principle allows and also how the mosquito would react to each possible grab and what the air atoms would've done if you hadn't moved? (I've been lucky and fast enough sometimes so it's possible, grabbing is aiming for center of fist for maximum margin of error, not finger pads) Sagittarian Milky Way (talk) 18:31, 4 October 2020 (UTC)
 * I believe spiders did a similar calculation a while ago, the result was 'NOPE, let's build a trap folks'. Zindor (talk) 18:44, 4 October 2020 (UTC)


 * No, the uncertainty principle gives you an absolute limit on the knowledge of a system, and hence how it would evolve in highly complex examples such as this one, meaning it is just not possible. Even on simple systems it is difficult to predict how they will evolve given enough time, e.g. see the three body problem.
 * However the question remains that if you did know a system completely (as some sort of God-like being), would quantum theory disprove determinism, and would you be unable to predict the system's evolution, given no limit on your power of computation? The jury's still out on that one, but there is an interesting debate at about it. -- Jules  (Mrjulesd) 19:57, 4 October 2020 (UTC)
 * See Maxwell's demon for a related idea. -- Jayron 32 15:33, 5 October 2020 (UTC)
 * On the other hand a mosquito is rather larger than a quantum dot and at its scale I don't think Heisenberg's principle is really relevant. Beside that its fly is quite even and predictable and in the calm air of a closed room with knowledges in the order of one millimetre and 10 milliseconds it is surely possible to consistently hit a flying mosquito or to predict the outcome of an attempt to do so. The largest and fluctuating interval to consider is the starting time of the hand but after the hand has reached some adequate velocity one can say with sufficient certainity if it will hit the mosquito or not. 2003:F5:6F0E:4500:3D1B:A8CB:F3B2:4524 (talk) 22:17, 6 October 2020 (UTC) Marco PB

In the news: Perpetual motion machines may have arrived.
https://phys.org/news/2020-10-physicists-circuit-limitless-power-graphene.html

Best. News. Ever. These published physicists claim their circuit doesn't violate the 2nd law because it does not create an internal temperature differential, however their claim that this machine can do work also means that it can, of course, be used to create temperature differentials and from what? From the circuit's ambient heat due to localized thermal fluxes, which means that decades of Perpetual Motion Machine (of the second kind) skepticism should be swept aside. Not violate the 2nd law? Ha. I can hear, loudly, the strained twisted logic of neatly ranked acceptability bleeding everywhere, as the weight of decades of skepticism is dispelled, although it it may not quite have landed on the chorus yet. So be it. Perpetual motion will be the new norm such that we can be assured that, soon, our newly energized world from the conversion of heat need not get too hot and flooded, nor perhaps will even this universe get too cold in some eternal hell. Low power can be aggregated of course. Real perpetual motion machines will be created. Then again, I have had to be reminded, repeatedly on this desk, that relativity is symmetric. So I could be wrong again. What am I missing or maybe this a situation where we have to be cautious and have to wait all night until these machines are built and working and placed under a tall and decorated tree to draw meaningful conclusions. Regardless of the eventual theoretical outcomes, if this graphene circuit and related work reaches fruition soon, it is very good news at time when we really need some. Have any prominent physicists (do they still exist?) jumped into this on twitter, editorials or elsewhere? Thanx. -Modocc (talk) 19:20, 4 October 2020 (UTC)
 * Machines that use the motion of sea waves to generate power also don't create an internal temperature differential yet can do work. The science news item is a verbatim copy of a press release put out by the PR people of the University of Arkansas and does not offer enough detail to understand what is going on here, but I am confident that nothing beyond a straightforward argument solidly based on accepted physical principles will be needed to show that this does not violate any known physical law – or else physicists would have generated a deafening hubbub, having been much more excited by the prospect of new physics than by any promise of limitless free energy. --Lambiam 19:42, 4 October 2020 (UTC)
 * Sea waves dissipate unless driven. The proposed removal of heat from the circuit does smack of a perpetual motion instead in that the converted energy can always be returned to the circuit within a closed system. -Modocc (talk) 20:17, 4 October 2020 (UTC)


 * Well they claim it doesn't violate the second law in the paper, which is the primary refutation of the so-called perpetual motion machine of the second kind. If it doesn't violate the second law, then it wouldn't violate the laws of physics, and would therefore be possible. Note that a perpetual motion machine using this would still cool, so it would still need external heat, so it doesn't fit in with conventional notions of a perpetual motion machine without outside interference. Jules  (Mrjulesd) 20:40, 4 October 2020 (UTC)
 * Yes and our article would indeed need a rewrite regarding the 2nd law and PMMSK. It would mean that the 2nd law arguments denying the machines were bogus certainly. It is a bit unclear if the 2nd law gets cremated too, but thermal energy from the circuit can be utilized which was the intent of Maxwell's Demon. --Modocc (talk) 21:02, 4 October 2020 (UTC)

actually the more I think about this the more I feel something bogus is going on. I can't see how it can do as it claims without violating the 2nd law, despite the researchers claiming otherwise. I think that there are three possibilities. I would say 1. is most likely, 2. a possibility, and 3. an extremely low possibility. If the great Richard Feynman claimed this couldn't be done I think it should be approached with extreme caution. Jules (Mrjulesd) 21:33, 4 October 2020 (UTC)
 * 1) The paper is basically hoax material.
 * 2) The circuit does indeed produce a current, but not in the way described.
 * 3) The circuit violates the 2nd law
 * I was also going to ask about how this related to a Brownian ratchet when it popped up on google news. Here's an abstract of the paper with pretty pictures and graphs. Not sure from the abstract, news article, or what they announced earlier in the Bulletin of the American Physical Society if they actually built anything, or just did some numerical simulations. fiveby(zero) 21:21, 4 October 2020 (UTC)
 * Smells like Maxwell's Demon . Which has been demonstrated to be in accordance with 2LOT, not to anyone's great surprise, but probably relief. Greglocock (talk) 21:31, 4 October 2020 (UTC)
 * was discussing something similar i think at Talk:Brownian ratchet back in 2008 w/ pawls vs. diodes. fiveby(zero) 21:34, 4 October 2020 (UTC)
 * Hmm, Numerical simulations show that the system reaches thermal equilibrium and the average rates of heat and work provided by stochastic thermodynamics tend quickly to zero. However, there is power dissipated by the load resistor, and its time average is exactly equal to the power supplied by the thermal bath. is it just a heat engine until it reaches equilibrium? fiveby(zero) 21:43, 4 October 2020 (UTC)
 * This is just Johnson–Nyquist noise. And the reason is does not work is because the load resister (lightbulb) is also putting out that fluctuating voltage. In practice the voltages are around a microvolt, with power around kTB (Boltzmann constant times temperature times bandwidth). Graeme Bartlett (talk) 01:03, 5 October 2020 (UTC)


 * It looks to me from the circuit that the lightbulb is isolated from the circuit by an analog switch. But you are right of course, the diodes also produce Johnson-Nyquist noise which will prevent the circuit from working.


 * Cool article!  This sounds like a Brownian motor to me.  Brownian motors are nanoscale devices that harvest energy not from random thermal motion but from micro differences in temperature, chemical potential or other energy flows, so they don't violate the 2nd law.  Starting with Brillouin, there have been a number of analyses of diode-capacitor circuits as Maxwell's demons,  including two-diode circuits like this.  The conclusion of all of them is that unless there is a temperature difference, or some other energy difference between the parts of the circuit, it can't produce useful energy.--ChetvornoTALK 01:12, 5 October 2020 (UTC)


 * The way diode perpetual motion circuits fail to produce energy is analogous to the way the Feynman ratchet fails. As Graeme said, in addition to the voltage fluctuations produced by the graphene capacitor, the diodes produce their own fluctuating voltages due to Brownian motion, called Johnson-Nyquist noise (this is analogous to Feynman's pawl bouncing).  The tiny voltages produced by the circuit are far below the band gap of the diodes, so from the Shockley diode equation the diodes will be very poor rectifiers; the reverse current due to a negative noise voltage pulse will be almost as large as the forward current due to a positive voltage pulse.   The accumulation of charge on the storage capacitor reverse biases the two diodes, so there will be a constant leakage of current off the capacitor backward through the diodes.  If the whole circuit is at the same temperature, and there is no other hidden source of current, the currents in both directions through the diodes should be equal, so there will be no net accumulation of charge on the storage capacitor.  However if the graphene has a slightly higher temperature than the diodes, it will produce larger voltage fluctuations than the diodes, the device will act as a heat engine and the capacitor will charge.--ChetvornoTALK 01:12, 5 October 2020 (UTC)


 * In a prototype nano circuit like this there are all kinds of side effects and trace sources of current that could provide the energy that the investigators reported. The presence of the battery in the circuit is of course a big question mark; have the investigators eliminated the possibility that it is providing the energy?   There are also contact potentials, electrolytic effects, surface leakage currents.--ChetvornoTALK 01:12, 5 October 2020 (UTC)


 * I vaguely recall Brownian motors being investigated some years ago and had wondered if this was an improved version of one of those. Apparently not.  Thank you for the replies.  --Modocc (talk) 02:36, 5 October 2020 (UTC)


 * The last sentence of the abstract of the Phys. Rev. E article is: "Excellent agreement is found between experiment and theory." The article was peer-reviewed, so the theory would not violate any laws of physics, and therefore the experimental results – whatever they are – do not either. If something is not possible according to our current understanding of the laws of physics, this cannot be what is reported on as having been experimentally observed by the authors. --Lambiam 07:06, 5 October 2020 (UTC)


 * In case people missed it, while the journal article may be behind a paywall, there is a preprint version is still on arXiv [//arxiv.org/abs/2002.09947]. Assuming Copyright policies of academic publishers is accurate, I think the author could redistribute the final version on their website; or update the version on arXiv (or eslewhere) to the final version by themselves if they wanted to, but I didn't see any signs of this. Still if someone does find a version that seems to not be an external copyvio, feel free to link it. I don't know what differences there are between the preprint and final version. Nil Einne (talk) 18:28, 5 October 2020 (UTC)
 * P.S. I did come across [//hardforum.com/threads/physicists-build-circuit-that-generates-clean-limitless-power-from-graphene.2001989/post-1044753559] although I have no idea if that criticism is valid and am not sure if it's on the preprint or final version. The equation one in particular is the sort of thing that should have been picked up in peer-review assuming it's a clear error but..... I also found [//physics.stackexchange.com/a/583795/276416] although that's definitely commenting only on the preprint version in part probably because it the final version wasn't linked. Nil Einne (talk) 19:12, 5 October 2020 (UTC)


 * Thanks, Nil Einne. I read the paper.  Wow.  Although I couldn't understand the math, it seems to be claiming that the circuit can harvest power from the peculiar Brownian motion of the graphene membrane at equilibrium (constant temperature).
 * "Our model provides a rigorous demonstration that continuous thermal power can be supplied by a Brownian particle at a single temperature while in thermodynamic equilibrium, provided the same amount of power is continuously dissipated in a resistor."
 * There is no mention of any other energy gradient providing power, as in a Brownian motor. This seems to me to violate the 2nd Law, although the paper says it doesn't:
 * "Here, coupling to the circuit allows electrical work to be carried out on the load resistor without violating the second law of thermodynamics."
 * The circuit they actually used in the paper is different from the one shown in the animation: it consists of the graphene capacitor in series with a battery, load resistor, and two diodes connected in parallel, back to front. As far as I can understand, the way they claim it works is: the graphene membrane has two states, concave and convex curved, and snaps back and forth randomly between these two states, driven by Brownian fluctuations.  The graphene is one plate of a capacitor charged by the battery, so each time the membrane flips, it creates a (relatively) large voltage step, $$\Delta V = Q/\Delta C$$, which drives a quantity of charge through the circuit, through the diodes and resistor, heating the resistor.  The purpose of the diodes is to block the Johnson-Nyquist noise voltage pulses produced by the load resistor, which would otherwise transfer an equal amount of power from the resistor to the graphene.  The parallel diodes have a larger resistance at low bias voltage than at higher voltage, so they act as a (leaky) potential barrier, selectively blocking the small noise voltage from the resistor but letting the large voltage pulses from the graphene through.  Thus the resistor absorbs more power from the circuit than it generates, dissipating net heat power.  The resistor would get warm and the graphene would cool down, violating the 2nd Law.  --ChetvornoTALK 18:43, 6 October 2020 (UTC)
 * So this appears to be a bit like a heat pump. The energy dumped to the resistor comes in part from the battery charging the capacitor. Graeme Bartlett (talk) 22:42, 9 October 2020 (UTC)

It seems to me that nobody is accounting for the fact that diodes (which are part of this graphene circuit) have a significant voltage drop across the junction in the conduction direction (0.6V for silicon diodes) and it is unlikely for the potential generated by the graphene electrode to overcome that barrier. ~Anachronist (talk) 18:48, 10 October 2020 (UTC)