Wikipedia:Reference desk/Archives/Science/2022 July 26

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 26

Finding an old article in Science et Vie

Is there any way (short of physically travelling to France and searching through their libraries) to find old issues of Science et Vie (the French version, not the Russian one)? The issue I'm looking for is definitely from before 1993 (but definitely not earlier than 1981), and their online archive https://www.science-et-vie.com/archives-par-numero only appears to go back to 2008. 2601:646:8A81:6070:68A5:567C:9B63:FA6E (talk) 02:01, 26 July 2022 (UTC)

Pity, Gallica has most issues from 1914 to 1951, as does this site. This commercial site seems to have many numbers for sale, click the magnifying glass icon to see the cover, which from the mid-seventies lists the main articles. The more recent ones cost less than 2 Euros. 90.252.183.121 (talk) 04:29, 26 July 2022 (UTC)

Three issues from 1990 and 1991 here, for what it's worth.  Card Zero  (talk) 08:24, 26 July 2022 (UTC)

Some more results at the Internet Archive. Alansplodge (talk) 09:13, 27 July 2022 (UTC)

Thanks for the info, 90 IP -- I just found the issue I was looking for at Collectionsmag.com, it was the June 1980 issue (and I also found another one potentially of interest to me, the one from December 1978)! Now, can I buy it as a digital download, or do I have to have the paper version shipped to me? 2601:646:8A81:6070:D949:A3E1:47EA:4FA6 (talk) 09:40, 27 July 2022 (UTC)

thyristor (why doesn't the following work)

hello, I came across a silly LED flasher circuit (see below) in a book and built it. it worked. I also noticed that the two transistors seem to form an equivalent circuit, as commonly depicted, of a thyristor (incl. in our article). However, the circuit doesn't work with an actual thyristor (I tried several - FS0102DA, C106D and some others.) Why is that? Is the two-transistor analogy not accurate? Is there a way to make it work? (there's something nifty about a one-part oscillator) Aecho6Ee (talk) 12:06, 26 July 2022 (UTC)

Extended content

-9V -9V === === | | | |K | - -> .-. ^ -> | | |A | | 270k | LED '-' | | | o------------------+ | | E| | | <| | --- 100n NPN |--------o --- e.g. BC547 /| | | | | | | | | | |/ === o--------| PNP GND |< e.g. BC557 |E | | === GND -9V -9V === === | | | |K K | - -> | | .-. ^ -> | | | | |A <| | | | 270k | LED NPN |--------o G '-' | /| | => | | | | | | | | o------------ G | | |/ | | -------o o--------| PNP --- 100n K| / |< --- _/ | A | ^ | A| | | === === GND GND

Aecho6Ee (talk) 12:06, 26 July 2022 (UTC)

Extended content
+9V +9V === === | | | |A | V -> .-. - -> | | |K | | 270k | LED '-' | | | | | o-------, | | \ o A --- 100n o G  : --- \-----V SCR |  : | o K | | === === GND GND

Philvoids (talk) 09:06, 27 July 2022 (UTC)

thanks, but this doesn't work either... plus, it's a different circuit... Aecho6Ee (talk) 19:41, 27 July 2022 (UTC)

Is it something like this? Note that the capacitor does not go to ground (as the one over here might lead to believe.) I did get that one to oscillate in LTspice, albeit not as much of a LED flasher as its pulse ratio is very low. My brain hurts when I try to translate that into your negative voltage version. 85.76.141.129 (talk) 11:39, 28 July 2022 (UTC)

Oh, and as an aside: if you put nine volts through an LED without a current limiting resistor you should wear protective glasses. The LED makes a funny PING sound as it explodes and sends plastic shrapnel to the ceiling. Been there, done that. 85.76.141.129 (talk) 13:21, 28 July 2022 (UTC)

that's a different circuit, too... here's the one from the book: https://i.postimg.cc/6qP8TmkZ/img1.png . (the book is "311 Schaltungen".) the capacitor does go to ground. The text even mentions a "thyristor effect." The negative-voltage version, however - that was indeed me. This was so there'd be a literal 1:1 correspondence with the two-transistor thyristor model, because only the base of the NPN is connected externally (as the Gate) and the other isn't, in an actual thyristor. (though I imagine there are SCRs where the other junction - C of the NPN to B of the PNP - is connected externally, as a negative-going Gate.) Aecho6Ee (talk) 02:05, 29 July 2022 (UTC)

re 9V across LED - thanks. I imagine it's not a problem in this circuit as the pulses are very short (and far apart.) Aecho6Ee (talk) 02:08, 29 July 2022 (UTC)

The OP's interest in a LED flasher circuit with only one other semiconductor part can be met by a unijunction transistor as shown here (see 10 LED flasher) or by a 555 timer IC (see here for astable circuit). Philvoids (talk) 09:53, 29 July 2022 (UTC)

Try building the exact circuit from 311 Schaltungen. If the book's contributors really built the circuit and made it work, it seems to rely on misfeatures of the components - random capacitances, glitches, and whatnot. Does not seem like a stable way to design a circuit. 85.76.48.243 (talk) 14:24, 29 July 2022 (UTC)

Is there a search engine for weather stations?

I've always wanted the ability to get a list of all weather stations in a big database (i.e. all Wikipedia climate boxes) that meet certain preferably Boolean criteria. For example record annual precipitation ≤254mm AND latitude between 5 and minus 5 AND Koppen BWh. Preferably the results page has many sortable columns like one column to sort list by longitude, one to sort by avg May low and so on. Closest I've found is a site that says (paraphrased) they've data mined gazillions of world weather stations and intentionally dumbed it down so much that even the least technical communications major grandma who can barely use Windows can use it. As a result their genuinely huge database of weather stations in hundreds of countries and dependencies is wasted. Sagittarian Milky Way (talk) 14:20, 26 July 2022 (UTC)

Have you explored Weather Underground (weather service)? --Jayron32 14:45, 26 July 2022 (UTC)

For many years. It'll show a map of stations and some normals and records (i.e. what's the normal and record July 3 high of a specific place) but not let you find things like "what's the lowest avg precipitation with a Koppen classification of oceanic?" or "out of the weather stations with period of record over x decades which of these has the lowest avg annual temperature without recording a frost ever and what country or colony is it in?" or "are there any places where the driest of their 12 monthly averages is wetter than my hometown but it's still tropical wet and dry season according to the Koppen code?". Sagittarian Milky Way (talk) 16:07, 26 July 2022 (UTC)

If Wunderground doesn't have the data you are looking for, perhaps the NWS does? You can find them at https://www.weather.gov/. If you can't find the information at one of those websites, which are designed mostly for the end user and not for specialist researchers, those websites have contact information. They may have firehose data dumps available for researchers who need the full set. You would need to reach out and contact either the Weather Underground organization or the National Weather Service directly. --Jayron32 16:18, 26 July 2022 (UTC)

Hit the reality

How can physicists even figure out which deductions from the formulas of physics describe real results and which are just mathematical artifacts? 2A02:908:424:9D60:0:0:0:99EF (talk) 18:47, 26 July 2022 (UTC)

By experimentation. AndyTheGrump (talk) 18:55, 26 July 2022 (UTC)

And observation. --Stephan Schulz (talk) 20:00, 26 July 2022 (UTC)

Honestly, I feel so stupid. I didn't think of it myself. 2A02:908:424:9D60:0:0:0:99EF (talk) 20:10, 26 July 2022 (UTC)

This qualifies. :) ←Baseball Bugs ^{What's up, Doc?} carrots→ 20:38, 26 July 2022 (UTC)

The relevant article is Empiricism. Alansplodge (talk) 09:08, 27 July 2022 (UTC)

If a given theory proves to work well on a wide range of cases from a relatively narrow space of conditions, it is generally assumed it will also work for further cases within that space. A formula will only be considered a formula "of" physics after it has withstood the test of predictive validity. If a deduction from such formulas turns out not to be a valid description of the real world, physicists get very excited because it means there is new physics to be discovered.  --Lambiam 16:36, 27 July 2022 (UTC)

I wholeheartedly endorse the earlier summaries! Yes, of course - experimentation! This is the core of the Scientific Method and it is what we (as scientists) fundamentally commit to: it is the standard method by which we test our understanding and attempt to separate truth from untruth.

But the question asked about deductions! And deductions and inductions are wrapped up in a whole quagmire of philosophical concerns about our trust in the inherent consistency in the explainability of the natural world! Deductions are, so to speak, directly "deduced" from our own observations, and therefore if our logic is sound, we have reason to have confidence in them. A host of postmodernists will tell you why you can't even trust "logic," nor even can you trust "trust," for that matter. But let's put the epistemology of trust aside for a moment, because if we displace our trust in the core concept of coherence, we find ourselves on the wrong side of the teapot argument.

The Plato Encyclopedia of Philosophy has a great review article on the induction problem, with several sections analyzing different embodiments and interpretations of the scientific process. It's pretty heavy stuff - how do we "know" "anything"? How many experiments do we need to perform before we have confidence that the next result is "predictable"? At what point do our infatuations with empiricism foist us into the throes of skeptical solipsisms?

So, yes - we use observation and experiment and the Scientific Method - but great thinkers have expended centuries of effort trying to tease out whether this is really justifiable (or good, or our prerogative, or self-consistent, or maximizing a utitlity function, or serving a higher purpose, or describing the natural order, or ...)

Yeah, it's a bundle of cosmic strings that needs considerable unpacking, and it'd be a slight to great minds if we said it was "simple."

Nimur (talk) 16:10, 28 July 2022 (UTC)

• One of the things to remember is that theory gives us predictions, and observations give us data, and a theory is useful if its predictions match data. Modern scientific philosophy really follows on from the work of people like Karl Popper (esp. his notions on falsifiability), John Dewey (esp. his notions of instrumentalism), and Stephen Jay Gould (esp. his notions of Non-overlapping magisteria). As N. David Mermin succinctly put it "Shut up and calculate!". Modern physics theories, such as quantum mechanics, have often been called "the most successful theory in the history of science" for the accuracy of its predictive power, and yet it's also the most baffling when we get into trying to understand what it means in terms of fitting it in to our "natural" understanding of how the world works. When asking "Is this real", modern science, at its core philosophy, unasks that question. "Reality" is a meaningless idea here; a scientific concept is either useful (because its predictions can be verified any number of times to any arbitrary level of precision) or it isn't (because it doesn't match experimental results). It should be noted that some of these philosophers actually vehemently disagree with each other on the notion of reality and science's role in discovery thereof, but the tension between the views is probably more useful here than deciding who is "right". --Jayron32 17:55, 29 July 2022 (UTC)

Not the topic-starter here, but I agree with the topic-starter. I once did a problem with a physics professor with 2 different approaches and got 2 different answers. The question was what hits a baseball bat harder, a stationary baseball, or a baseball coming towards you at a speed. 1st we did the momentum model which is 50 meter/second, that is, mv. Then, the 2nd approach we did the acceleration approach, which is 50 meter/second^2, that is, mv^2. And he said mv^2 was the best answer. But I asked, does mv^3 mean anything to physicists? Or mv^1.5? He said no. But we're not exactly hitting a baseball bat at exactly 50 mph or 50 meter/second^2 are we? Couldn't someone hit a baseball bat at a momentum more than an acceleration? 67.165.185.178 (talk) 16:25, 30 July 2022 (UTC).

Well, that one is a little more concrete, and a little less philosophical - it's a question that essentially asks "which simplification is more appropriate in this scenario?" The case of conserving energy, vs. conserving momentum, is a standard way to illustrate the difference between elastic collision and inelastic collision; it encourages the student to think about how conservation of energy works in such a situation; it provides a teaching-moment about how there is more to analysis beyond merely plugging numbers into equations.

As far as the question about other polynomials in velocity: sure, these terms mean something, but they don't participate in our standard, simple conservation laws - so they aren't useful (at least, not for solving elementary kinetics problems). There's already much that has been said about "useful models" - and if you're analyzing a baseball interacting with a bat, the most immediately useful models are going to relate to conservation of energy, and conservation of momentum. If you want to analyze in excruciating additional detail, using weirder mathematics, you just might be a physicist.

How much so? So much that when I was a student, the National Science Foundation's REU program provided funding for physics research in baseball. (I do not find such opportunities in the most recent program listings).

You can still find interesting internet articles - Aeronautics expert Peters doesn't watch a game the way a casual fan would; Alan Nathan's site devoted to research on the physics of baseball,...

Have I ever seen a velocity-cubed term? You bet - all the time, actually. Velocity-to-the-power-of-1.5? Sure, but less commonly... things get weirder and maybe you won't even recognize the same math, done differently. How about velocity-to-the-power-of-transcendental-number? Well, maybe not in a controls-equation, but ... sure. I mean, I read Eureka (when I can understand it)! Besides, like every interaction in the universe, this is fundamentally a scattering problem.

Your takeaway here oughtta be - if you're solving for baseball momentum because it's a homework problem, use the textbook method and solve for the textbook answer. If you're a real actual physicist, you probably aren't as sure about that ... you're probably looking for complications that defy the simple model, and then you're seeking generalities that can simplify those complexities away... because if we bring back the more philosophical discussion, that's how physicists conceive reality: complexity hidden in simplicity, simplicity hidden in complexity.

Nimur (talk) 13:59, 31 July 2022 (UTC)

After sleeping on it - I had some more thoughts on the "v" vs. "v²" vs. "v³" terms... but I know we're going to lose any of our readers who haven't already studied some fairly advanced physics as a preparation. Oh well - "free encyclopedia" where "anyone can participate,"... !

You know, we have a conservation law for momentum, and a conservation law for energy, so that's why we fixate so much on the first and second moments in velocity, e.g. the "mv" terms and the "1/2 m v ²" term.... and in basic high school classes, we take these conservation laws as facts that are empirically observed, but there are some deep, ugly, profound problems with this approach.

First of all, what is energy? What is momentum? We can't measure these things - we have to measure position and time, and then deduce velocity (there's this ugly "deduction" problem again!) And then we compute energy and momentum... and we can define a word (like "energy" or "wakalixes," if we prefer) and we use this noun to capture and encapsule the computed, derived value; and then, it happens that we find these computed values satisfy a conservation law in all of our experiments (... until we look very closely, in very special circumstances, where we are forced to confront additional complexities in the definitions...). So we say we have a new Law of Conservation Of Wakalixes; or perhaps a Relativistically-Corrected Law of Conservation of Noun-Wakalixes... or whatever. But we just say we have this thing, based on the consistency in our observation. The original question - is it "real?" We didn't even open that can of worms yet!

And if we compute other moments (like the third-order term - the velocity-cubed term!) ... all the naming in the world doesn't help us, because we don't find a conservation law for Noun-Naming-The-Third-Order-Moment-In-Velocity. We don't find that conservation law in baseball, and we don't find it in stellar fusion, and to the best of my knowledge, we don't find it in quantum-mechanical formulations, or relativistic formulations, or anywhere. Why in the world... what is so special about the first and second moments that make them profoundly different from all the higher order terms?

And I think the answer came from the ever-forgotten queen of analytical physics - Emmy Noether and her most significant theorem. This comes down to symmetry. "Energy" is conserved because it is a mathematical expression of a symmetry in the universe. The velocity-cubed term is not so conserved. These statements impart no more and no less "reality" to the terms.

Why is the universe symmetric in the second moment, and not in the third? That is finally hitting on like, some serious, Hume-causation-perplexing unanswerable philosophical stuff. That's the core of the question: what is real? What is reality? In physics, we take the observed facts apart, decompose them, transpose them into their absolutely most generalizable mathematical form, and eventually we're stuck with a deeply abstract philosophical question. Why this symmetry, and not that one? Could a different universe have had a different set of symmetries? Is that even possible? What would that even look like?

But to reach this level of understanding, the physicist has to tear through the logical deductive steps - steps that depend on intricate, complicated, detailed mathematics. Look back at the lagrangian expressing Noether's theorem. That bit where we're zeroing out terms is the conservation law. This isn't the way it looks in a high school textbook! These simple lines and squiggles are abbreviations for a lot of years worth of study of math and physics formalism - these are the background details that are floating in your mind as you stare down at that translation invariance representation. Most people can't, or at least won't, be able to do that. So it's really hard for a physicist to express an answer to this question - "what is real?" Well, the way we conceptualize reality is the conjunction of mathematical abstraction and observed fact.

Velocity-cubed is no less "real" - it's just "not present in one particular translation-invariance relation." And this has something to do with baseball. Makes sense? If not, you can look forward to studying it for decades until you reach enlightenment... and somewhere along that route, based on your own individual personality, you'll probably stop looking for enlightenment, and just enjoy the game.

Nimur (talk) 14:40, 1 August 2022 (UTC)