Wikipedia:Reference desk/Archives/Science/2011 September 16

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

September 16

Neutron foundries

How many neutrons are produced regularly at a small, university-sized LEU fission reactor? How many would produced by a reasonably large, but still-university sized tokamak, using today's technology? (I'm obviously not assuming the tokamak reaches breakeven — it need not produce more energy that goes into it, but should still produce some fusion neutrons.) --Mr.98 (talk) 03:33, 16 September 2011 (UTC)

I'm just looking for rough estimates... e.g. 10^x neutrons per hour of peak operation, but it can only run for y hours at a time, things like that. --Mr.98 (talk) 12:10, 16 September 2011 (UTC)

Present thermal neutron flux (i.e. useful for research purposes) at research reactors is limited to around 10¹⁵ n cm^-2s^-1 using fission sources (e.g. the Institut Laue-Langevin. This is increased to just under 10¹⁷ n cm^-2s^-1 for spallation sources (e.g. ISIS neutron source or SNS). It is predicted that fusion sources may be able to produce neutrons with a flux approaching 10²⁰ n cm^-2s^-1. For a review please see Taylor et al. 2007 Science 315 1092. Hope that helps. Polyamorph (talk) 14:56, 18 September 2011 (UTC)

methane green house gas

The sanitary sewer is required to have vents at critical locations in the line to prevent traps from being sucked dry of water. These vents also allow methane to escape into the atmosphere and methane is a green house gas which is burned off at the plant. Is there any reason why sanitary sewer vents can not be tapped in a way that would allow methane to be captured, compressed and then used as fuel instead of being released into the atmosphere? --DeeperQA (talk) 04:59, 16 September 2011 (UTC)

Maybe there's not enough methane there to go through the trouble of capturing it. .... — Preceding unsigned comment added by 88.8.79.204 (talk) 05:30, 16 September 2011 (UTC)

Distance from the plant could also be a factor. On the other hand, it may be possible to flare the methane from the vents to prevent it from entering the atmosphere. 67.169.177.176 (talk) 05:45, 16 September 2011 (UTC)

Sewer gas contains much more than just methane: some of it is toxic and a lot of it smells really bad. You would have to collect it and separate the small part of methane from the rest of it, and then dispose of the remainder which probably isn't economically worthwhile (and environmental regulations are probably stricter on what you extract from sewer gas than what you let naturally float out). --Colapeninsula (talk) 10:04, 16 September 2011 (UTC)

Actually, the article sewer gas says it can be and sometimes is used as a fuel. But the reasons mentioned it's not entirely straightforward. --Colapeninsula (talk) 10:06, 16 September 2011 (UTC)

See this article. Alansplodge (talk) 21:29, 16 September 2011 (UTC)

Cell division and the onset of puberty?

I have a question about mitosis and how it relates (if at all) to the onset of puberty.

As I understand it, every so often, a human, autosomal cell will keep on dividing until it reaches the "Hayflick limit," after which, it simply collapses after a certain time. This is (more or less) what helps prevent cancer as well as what causes people to grow old and die.

I'm curious, though, whether this relates to sexual maturation in any way. Is there a "lesser Hayflick limit" (for lack of a more proper term) that triggers the end of childhood and the beginning of manhood? If so, does this occur in all autosomal cells, or does it only occur in cells of the endocrine system?

May a process similar to "immortalized cell lines"—theoretically, at least—be used to extend childhood to progressively later ages? Did any wikipedia article or any proteomics journal ever illustrate such a hypothesis? Pine (talk) 06:26, 16 September 2011 (UTC)

Puberty is caused by hormones, and we do have meds to block those, which can prolong childhood. I seem to recall a case where this was done with a disabled girl to make her "easier to manage". StuRat (talk) 06:41, 16 September 2011 (UTC)

You may be thinking of this controversial case, Ashley Treatment, which including drugs and surgery. Rmhermen (talk) 13:32, 16 September 2011 (UTC)

Yep, that's the case. StuRat (talk) 18:38, 16 September 2011 (UTC)

In general, I'd say cell division is more 'regulated' than being the 'regulator' of development (i.e. the outcome rather than the effector). Certainly blocking signals regulating development could prolong a pre-pubescent state, as was observed in Eunuchs Jebus989^✰ 10:02, 16 September 2011 (UTC)

Order D

Is there a rational about putting in cordinates the X first and then the Y and the Z? If so, what about 4-D? Exx8 (talk) —Preceding undated comment added 09:58, 16 September 2011 (UTC).

Cartesian coordinate system#Notations and conventions maybe? DMacks (talk) 10:32, 16 September 2011 (UTC)

In four dimensions it depends if you're dealing with 3 dimensions of space and one of time (i.e. spacetime) or with 4 space-like dimensions (see Four-dimensional space). 4D space-time coordinates are usually called either (t,x,y,z) or (x,y,z,t) (see Spacetime). In four-dimensional space (x,y,z,w) or (w,x,y,z) are commonly used (e.g. [1]).

As to why we have that ordering, alphabetical order is natural in 2 or 3 dimensions, as when you add z to x and y, but after that it's harder to know what's the logical order. Spatial coordinates are only conventional, i.e. there's no X, Y, or Z axis in nature, and you can swap the axes freely (subject to certain rules to preserve the handedness). --Colapeninsula (talk) 11:12, 16 September 2011 (UTC)

In n-dimensional euclidean spaces, you'll also see standard basis vector notation, which uses numerically ordered subscripts. For 4D, this would be (e_1, e_2, e_3, e_4). SemanticMantis (talk) 15:14, 16 September 2011 (UTC)

Or (e₁, e₂, e₃, e₄) if your brain doesn't process TeX:) DMacks (talk) 15:28, 16 September 2011 (UTC)

In other words, no. There is no reason, other than societal norm and mathematical historical convention that the "horizontal" coordinate should be denoted before the "vertical" component; nor that "x" corresponds to "horizontal," and so on. If you start studying more advanced physics and engineering, you will quickly see that in the general case, we do not always use "x" for "horizontal" and we do not always denote coordinates in this order; we rarely use rectilinear coordinates; and in many fields, we do not even use orthonormal coordinates. For example, the standard basis in elementary image-processing is the x/y cartesian grid: you denote pixel positions in (x,y) pairs. For historical reasons, a computer screen inverts the y-axis from historical mathematical convention; in many 3D toolkits, this means that a right-handed rule is not preserved. In many subfields of image processing, we do not use pixel-location; for example, in image compression, we use the spatial fourier transform to parameterize pixel data; and then we use a lossy compression scheme; so our coordinates are neither orthonormal nor spanning. In a modern video signal, we typically represent a time coordinate; the math we perform on the 2-D image signal is therefore a generalized "3-D" transform; but not in the conventional spatial interpretation. The "ordering" of these coordinates is actually very complicated; but at some point, the data is all serialized to a bit-stream, so if you felt like dragging mathematical idealizations through the muddy mire of engineering details, you could find a description of the "coordinate transform" of any particular pixel (x,y,t) to a (very complicated) linear combination of bitstream-offset coordinates.

Start by reading generalized coordinates. This topic is usually covered in greater detail in the first chapter of any textbook on linear algebra. Nimur (talk) 18:20, 16 September 2011 (UTC)

Huge trees

Do we know about any trees that their frond's thickness is about 20 meters? Exx8 (talk) —Preceding undated comment added 10:41, 16 September 2011 (UTC).

Palm trees and cycads have fronds, which are a type of leaf. But it seems very unlikely that a leaf would be 20 meters thick. Do you mean something else, like the length of the frond, or the thickness of a tree's trunk? Record breaking trees (see "stoutest tree") says the largest trunk is currently the Sunland Baobab, an African baobab (Adansonia digitata) at 10.64 metres diameter; the widest ever was an African baobab 15.9 m wide. --Colapeninsula (talk) 11:27, 16 September 2011 (UTC)

(edit conflict) As Colapeninsula said, fronds are fern, cycad, or palm leaves. If you meant the spread of the leaves on top of the trees, better terms to use are foliage, crown, or canopy. If you meant the trunk, use trunk or main stem. :) Either way, the answer probably lies in clonal colonies. Plants (or fungi) which at first glance looks like a group of several plants or even a small forest, when in reality they are vegetative clones (ramets) of a single organism that may or may not be interconnected underground. The most famous example is Pando (tree) (which may not even be the largest of its species), a clonal grove of a single male quaking aspen that covers a total of 106 acres. For individual trunks of trees, however, there are several sequoias with circumferences (note, this is measured at ground level, and is different from diameter at breast height) exceeding 20 m, see List of largest giant sequoias. A sweet chestnut named the Hundred Horse Chestnut is also recorded to once have a circumference of 57.9 m in 1780, though the trunk have now split to several trunks (though it's still a single individual).-- Obsidi♠n Soul 11:55, 16 September 2011 (UTC)

I always wondered why we want to count clonal aspen as a single organism, but we don't want to count identical twins as a single organism even though they too are clones of each other. Googlemeister (talk) 13:45, 16 September 2011 (UTC)

Identical twins aren't physically connected to each other underneath the soil. As for conjoined twins, you have two distinct central nervous systems, with some shared bodily resources, which makes comparison with plants difficult, let alone anywhere near politically correct. —Akrabbim^talk 13:55, 16 September 2011 (UTC)

(EC) It's not just that the aspen clones share the same DNA. The various ramets transport and exchange nutrients and fluids. "Above ground these plants appear to be distinct individuals, but underground they remain interconnected and are all clones of the same plant." But your comment is still apt. It is difficult to define a notion of 'individual' that applies well across all the myriad forms of life :) SemanticMantis (talk) 13:59, 16 September 2011 (UTC)

The inadequacy of the concept of an 'individual' is perhaps most striking among slime molds. :P -- Obsidi♠n Soul 14:40, 16 September 2011 (UTC)

Studies of people's competence vs their belief of how hard a task is

I'd like to see the results of studies if any like the following have been done. Take an arbitrary task and tell half of the test participants that the task they're about to do has a success rate of 10% (or some very low level) and tell the other half the opposite, and see if the average performance of the people primed to believe the test is easy do any better than the others. I don't have access to any papers open only to subscribing members only, either. 20.137.18.50 (talk) 15:07, 16 September 2011 (UTC)

The articles Overconfidence effect, Optimism bias don't answer your question but may give some clues, maybe for useful search terms.

List of cognitive biases is probably worth looking at for your purposes, Experimenter's bias seems like it could be useful for you. and goes into great detail listing more specific biases - one of these might describe the effect you are thinking about, and be helpful in a search.83.100.239.6 (talk) 23:01, 16 September 2011 (UTC)

Lifeform that has killed the most humans (other than humans)

What lifeform has killed the most humans in human history other than humans themselves. This includes indirect killing through transmitting viruses. Viruses themselves are (arguably) not lifeforms themselves so lets just ignore them. My guess is mosquitoes, am I right? ScienceApe (talk) 17:40, 16 September 2011 (UTC)

Mycobacterium tuberculosis and Vibrio cholera are good bets, maybe Streptococcus pneumoniae too. If you're asking about specific lifeforms, mozzys wouldn't feature but the human-specific members of Plasmodium might. An interesting question, but I don't think humans needed to be excluded, I would guess we kill each other less than other things kill us Jebus989^✰ 17:54, 16 September 2011 (UTC)

Maybe now. There's some statistic in the "Rational Optimist", which I think is that as many as 1/3 of the early human population were murdered (all forms or justifications). Grandiose (me, talk, contribs) 18:10, 16 September 2011 (UTC)

But with the exponential growth of human population. 1/3 of 'early humans' is a drop in the ocean today, and we're talking 'the most humans in human history' Jebus989^✰ 18:18, 16 September 2011 (UTC)

Wars and genocide have been blamed for about 200 million deaths during the last century. That's probably a large enough number to put human violence on the map, though probably not large enough to put it at the top of the list. Of course some of those deaths are also related to various diseases that hang around battle fields. Dragons flight (talk) 23:32, 16 September 2011 (UTC)

Again, you'll find that over the course of human history, that is a drop in the ocean. Smallpox is thought to have killed up to 500 million in 60 years^ref and for longer-term human pathogens the total is incalculably huge Jebus989^✰ 18:45, 17 September 2011 (UTC)

Think of it another way, human violence killed about 3% of the humans who died during the 20th century. Assuming that the long-term rate is at least that high, then there could be no more than ~30 other organisms that were more effective at killing us. Take out deaths from accidents, old age, natural disasters, starvation, cancer, and everything else that isn't an "organism" (which for the current question also excludes viruses like small pox), and the ranking of human violence would climb even higher. I'm not saying that violence tops the list, but it is certainly on the map. Maybe even in the top 10. Dragons flight (talk) 18:48, 18 September 2011 (UTC)

(e/c) Yeah, but early human populations were tiny, and if we're talking raw body counts, we really need to look at the killers of the modern age. Y. pestis would otherwise be worth a mention, but I doubt it makes it very high on the list despite some high impact outbreaks in China and Europe. P. falciparum is another suspect worth mentioning. SDY (talk) 18:26, 16 September 2011 (UTC)

If viruses don't count, I would put my bet on Y. pestis in second after humans. Googlemeister (talk) 14:14, 19 September 2011 (UTC)

table of compunds

Is there a table of compounds that is laid out with both the rows and columns of the tables listing the elements with the row and column intersections or cells showing the compound(s) that are or can be formed by each element? For instance, at the intersection of oxygen and oxygen the compound O₂ would appear. --DeeperQA (talk) 18:42, 16 September 2011 (UTC)

It wouldn't be plausible to construct such a table. Carbon chemistry alone is extremely complex, and the number of meaningful combinations even of just carbon and hydrogen probably numbers in the hundreds. SDY (talk) 18:46, 16 September 2011 (UTC)

There are almost 28 billion different C₃₂H₆₆ structural isomers (and many of those each have many different stereoisomers). DMacks (talk) 18:57, 16 September 2011 (UTC)

The possible theoretical combinations of carbon and hydrogen are almost infinite. The number of them that produce chemicals that anyone actually uses are far smaller, and I'm assuming for a reference table you'd limit it to ones with practical implications. SDY (talk) 19:02, 16 September 2011 (UTC)

SciFinder lists literature references for 142 different chemical entities (==unique CAS#) for that formula (considering elements alone, not different isotopic possibilities). DMacks (talk) 19:57, 16 September 2011 (UTC)

Assuming, for the sake of argument, that everything from 1-50 carbons or so has a similar number of studied (not necessarily useful) chemical entities, that'd still give a number in the 10^4 to the 10^5 range. Guessing that one in ten of those studied entities have some sort of "meaningful" (i.e. practical or common) use, that gives a lower number. Regardless, it was a random and fairly conservative estimate. Even if it were only hundreds the proposed table would be impractical. SDY (talk) 20:10, 16 September 2011 (UTC)

Well, that could be handled by making it a 3D table (or more practically, you could pick at an intersection to get an expanded list of the most common compounds, without worrying about isomers, etc.). Or, perhaps the OP meant to only allow 1:1 combos at each intersection, not variable numbers of each element. They didn't list ozone (O₃), for example. Of course, if you wanted to list common compounds of 3 or more elements, this table would rapidly get out of hand. StuRat (talk) 19:03, 16 September 2011 (UTC)

I realize there are a great number of possible combinations but I was assuming that these could be further divided and handled at the intersecting cell on the basis of numbers of atoms or other elements in the compound through links to a new table. Perhaps a better method would be to list the compounds as the dependent variables and the elements as the independent variables and the bonds (or other relations) in the intersecting cells, using links wherever expansion was necessary. --DeeperQA (talk) 12:56, 17 September 2011 (UTC)

Rhythmic audience clapping

Is it reasonable to think that a rhythmic audience clapping would keep its exact rate over a long period of time? Is there anywhere I could look for answers on that? Gil_mo (talk) 22:15, 16 September 2011 (UTC)

No, and attending any concert with any musician will substantiate this. Audiences can't hold tempo even with assistance. — Lomn 22:51, 16 September 2011 (UTC)

(ec) There is no such thing as an "exact" rate for a group of people clapping in imperfect synchrony, so the question doesn't really have an answer. You could make the question more reasonable by asking whether the frequency spectrum changes over time, and I'm sure the answer is yes, but I don't see any reason why anybody would have invested resources in doing a formal analysis: there are bound to be a zillion uncontrollable factors that influence audience clapping rates. Looie496 (talk) 22:53, 16 September 2011 (UTC)

For relevant audible frequencies, a clap is very much like an impulse function. A rhythmic train of perfect impulse functions is called a Dirac comb. These are well-known mathematical idealizations and the spectral properties are very thoroughly understood. In fact, our article lists the fourier series (the frequency-domain representation) of a dirac comb:

$\Delta _{T}(t)={\frac {1}{T}}\sum _{n=-\infty }^{\infty }e^{i2\pi nt/T}.$

It's improbable that the sound of a human clap is a true delta impulse; and it's improbable that the timing is perfect; but we can characterize both of those imperfections as types of noise: frequency spread and phase noise (and we could get more sophisticated, if we wanted). Both of these parameters are well-studied as well; they translate to frequency spectral characteristics (basically, a "blurring" of the ideal frequency spectrum). Entire texts - entire libraries - are devoted to the study and parametrization of noise. Nimur (talk) 23:53, 16 September 2011 (UTC)

Looie496, my question was regarding rhythmic clapping, not asynchronous clapping. My question came after listening to a concert after which the audience was clapping rhythmically for over 5 minutes, and the tempo did not change a bit. I understand that for taking a tempo in a laboratory you need a delta, but I'm considering 'tempo' to be derived from the interval between two 'mass centers' of the clapping. That is, the tempo I'd clap to join the crowd. Gil_mo (talk) 09:28, 17 September 2011 (UTC)

Anecdotally, it depends on how musical(ly trained) the audience is. An average audience taking part in a rhythmic clap will speed up, clapping faster and faster. A more musical audience is quite capable of keeping a fairly constant tempo for a while, just as a skilled orchestra, chamber group or choir is capable of keeping a constant tempo. 86.164.76.231 (talk) 14:35, 17 September 2011 (UTC)

It's a coupled oscillator system, effectively, so the applause stays rhythmic until people get tired or something on stage encourages people to clap lighter, faster, slower, etc. Synchronization occurs on its own with even the very slightest feedback effect (such as hearing your neighbors' claps and unconsciously changing your timing, either to stay off-time or in-time (either way results in synchrony)). Such systems are modeled by the Kuramoto equation. SamuelRiv (talk) 02:30, 18 September 2011 (UTC)

Is evolution virtually testable, If so, has it been done?

Thinking about how logical evolution is, I thought is it possible to give computers of a model of the process, I mean the basic things are simple, a changing environment, some species that reproduce and randomly achieve traits while reproduction, and a little bit of time...?--Irrational number (talk) 22:46, 16 September 2011 (UTC)

Not only has it been done, but Genetic algorithms are actually sometimes used for problem-solving in general-purpose applications: see also Artificial life, Evolutionary algorithm etc. AndyTheGrump (talk) 22:55, 16 September 2011 (UTC)

thanks, and has there been new, unexpected species evolving in there?--Irrational number (talk) 23:10, 16 September 2011 (UTC)

Every time a genetic algorithm evolves an unexpected species, a numerical analyst tries to fix it. Unexpected behaviors are practically the norm in complex numerical modeling programs. Sorry, that was a bad pun. Nimur (talk) 23:45, 16 September 2011 (UTC)

These computer programs deal with "survival of the fittest" only in a very abstract sense. There are probably discernable "species" in them, but they aren't anything we'd recognize as living. Tierra (computer simulation) is just programs competing for CPU time, but in the abstract they still "reproduce", "mutate", "die", and "survive." SDY (talk) 23:47, 16 September 2011 (UTC)

My personal favorite AL programs are GenePool and Darwin Pond, both by Jeffrey Ventrella. They're free, visually appealing, and easily understandable by even young children. They're both swimbot programs and illustrate how sexual selection, environmental conditions and adaptations to it can steer the course of 'evolution' of the swimbots' "genes". It's limited because of the hardcoded nature of all the specific 'traits' however (they only have 15 'genes'). They're also all 'herbivores', i.e. no individual eats another, so there is no predator-prey 'arms race' evolutionary pressure. It's more accurate to view them as simulations of genetic drift that can happen within a species.

Another interesting easily accessible AL program is Creatures. It's actually a game with a twist. Had a lot of fun with this as a kid. Each individual creature has inheritable genes that control appearance, biochemistry, and behavior; each of those further affecting how the creature reacts to things around it in life. They can also mutate and can be inherited. Unlike swimbots their genes are far more complex. To an extent, they generate unpredictable mutations in a permutative sense, as all 'genes' are of course preliminarily defined (they're actually discrete strings of AI that utilize variables). An example is a mutation which affected the production a certain chemical, producing 'alchohol' continuously instead and making that particular creature, perpetually congenitally drunk. Their long lifespans however (in some cases, mutations can render them effectively immortal), doesn't make them ideal for showing selection pressures.-- Obsidi♠n Soul 03:30, 17 September 2011 (UTC)

There's a popular legend among computer scientists that I've heard many times but I can't track down to the source. (A link would be aprecieated.) The story goes like this :

It seems that some scientists did an evolutionary simulation to see if they could evolve simulated creatures with new and interested methods of locomotion. They set up their evolutionary simulation so that the beings that could walk the farthest before they got stuck would be considered "fit" and be used to populate the next "generation of beings". They set it up and sure enough their simulated beings were wiggling around on the simulated ground. They let it run for some long period of time and came back to it. When they looked into the simulation, they found that all their beings had evolved into extremely thin, and extremely tall cylinders. Why? Because if you're a mile tall you can move half a mile forward simply by falling over.

So this story, even if purely legendary, demonstrates a way that evolutionary algorithms can cause "unexpected" results, but not in the wonderful way the researchers were hoping. The simulation just found a way to "cheat" that the programmers hadn't anticipated.

I, personally, had a similar experience using evolutionary algorithms to create AIs opponents for a racing game I made in college. The AIs started off pretty stupid. They went fast, but more often than not they'd crash and burn. I turned off the graphics and let them run thousands of races overnight, the next day after class I came back and I played a game against the AIs I had grown. I was naively expecting that they had all become unbeatable, ultra-skilled, super-racers while I'd slept. Well, I guess I had made the penalty for a collision too severe, because all the AIs were piloting around the track in a single-file line at about 1/20th throttle. They would occasionally jockey for position in areas where the track was very wide, but even then they'd do it very, very slowly with exaggerated caution. Needless to say, I was disappointed. APL (talk) 04:06, 17 September 2011 (UTC)

The first example (falling over) seems like a defective design, since creatures that tall couldn't exist. However, you 2nd simulation seems like it worked, to me. Given the constraints on racers where safety is much more important than winning, that might be exactly how they would behave. In fact, I think they do behave much like that in real life. StuRat (talk) 04:29, 17 September 2011 (UTC)

Heh, what you described sounds like breve, but I dunno about the anecdote though. And yeah, the "unexpected" results are often bugs. In the previous example of Creatures, the immortality gene (colloquially known as the 'Highlander gene'), was one such thing. And it led to overpopulation and was fixed in subsequent versions, however, it again reappeared for wholly different reasons. Immortality, it seems was the logical conclusion of a world where resources never run out.-- Obsidi♠n Soul 04:41, 17 September 2011 (UTC)

If you are trying to accurately model the exact process of evolution which occurred on Earth with your program, it probably won't work. That's because we would need to know absolutely everything about their environment, like all the predators and prey, with precise numbers, as well as their own DNA, precisely, to predict accurately how it might mutate and evolve. StuRat (talk) 04:34, 17 September 2011 (UTC)

Magnetic rules -true brushless DC motors

I'm trying to recall some rule on magnetic forces (ie lorentz and laplace forces) when applied to the construction of an electric motor (ie a rotating machine)

Here's a typical setup -- mounted laterally on the edge around a rotating disc are N bar magnets all aligned the same way (parallel to the axis of rotation, call this "Z" plane) - this setup would be pretty much the same (in terms of shape of the magetic field) as a cylindrical magnetic (poles on circular faces) with the centre hollowed out, with the centre line co-axial with the axis of rotation. Clearly (?) the magnetic field lines will be like a bar magnet, except some with go from N to S through the centre of the hollow cylinder.. Does that make sense so far.?

The next step is to have a DC current in a loop at one end, with the loop in an radial plane. eg the plane defined by the "Z" and "Y" axis is such a plane. I was interested in the net force between magnet and current carrying wire due to "Laplace force" - I seem to remember that there was a (named) rule that states (or shows) that there can be no net force (assuming a constant Permeability (electromagnetism) exterior to the magnet and wire..) However it seems that since one half of the wire which produces an opposite force to the other half (due the the current being reversed due to the loop) is farther away the forces (and torques) will not balance.. - ie the net force will be in the direction given by the interaction between the part of the wire nearer to the magnet.

If there is a net force then this would be a simple brushless DC motor - which I though was not possible under these conditions?

Can someone either explain why there is no net force.. or if the opposite applies - please link to a page/book that describes "true" brushless DC motors in more detail. Thanks.83.100.239.6 (talk) 22:51, 16 September 2011 (UTC)

Try Brushless DC electric motor. It works just as you described. --Heron (talk) 14:16, 18 September 2011 (UTC)

wtf? Brushless DC electric motor describe a electronically communtated motor - absolutely nothing like my description. First sentence Brushless DC motors .. are electric motors powered by direct-current (DC) electricity and having electronic commutation systems - first sentence, the article..Imgaril (talk) 16:01, 18 September 2011 (UTC)

If I'm reading your setup correctly, the current loop will produce a force and therefore a torque, see right hand rule, but this field alone, without commutation, is not dynamic enough to produce a motor (for instance, a second stationary magnet substituted for the current loop will give the same negative result). But I'm thinking you might be alluding to the homopolar motor although it is not brushless (for instance, see the brush contacts used on the homopolar generator disks). The magnet (and screw) of the homopolar motor that is pictured in the article spins when the DC current traverses the magnet material where the current itself is imparting its momentum to the magnet (this has been done even with single molecules); but if the wire that is shown touching the magnet's edge is moved to where it touches either the screw or the lower battery post there is no current passing though the magnet and also no continuous spin is produced; again the static magnetic field of the current loop produces a small torque just as any magnet would, but it does not produce the motor's spin. --Modocc (talk) 22:22, 18 September 2011 (UTC)

Yes it is 'similar' to the brushed homopolar motor.. in vague principle ie that it makes torque through the left hand rule rather than magnetic reluctance or other effects used in many motors .. though would be much less efficient in its use of the available magnetic flux - since in a homopolar motor the magnetic field lines go through the current carrying disc in one direction only, whereas in this example the field will start in one direction +Z and turn to -Z at somepoint so that it can return to the opposite magnetic pole) - the issue I have is that it seems quite possible to get a net torque.

I don't think a permanent magnet would do anything (torque wise) - the magnetic field is effectively cylindrically symmetrical - a permanent magnet would just make a 'pull' or 'pushing' force - not torque.

The principle is based on the left hand rule - for a current travelling in the X direction not symmetrical about the axis of rotation eg only in positive X (the magnet is N/S alligned in Z axis) then the force will be constant in the Y direction - this makes a rotational torque. Of course there must be an equal amount of current going in the -X direction for a loop of wire - but this is farther away - my gut reaction is that the forces should balance - but I imagine that it might not be.

I think I may have grasped the idea now - any magnetic field line passing through (ie into) the current carrying loop (will generate a left hand rule force) , but , must also exit the loop to return to the opposite pole - this will generate an opposite force. I'm fairly certain that the overall effect (including distance effects etc) should be zero - I though there was a relatively simple mathematical principle that would verify/prove this - but can't seem to find or recall it. I'm sure it relates or derives from one of Gauss's laws, but haven't got the skill to prove it to myself - can anyone help with this. If someone with a higher IQ score than me can say that "Gauss's theorem makes true brushless DC motors impossible" and link to a source that would be a great help..

(Though I also imagine that soft magnetic materials might be used to generate an overall torque by 'guiding or intercepting' the magnetic flux into a certain path to avoid the counterbalancing opposite torque due to the current loop, and field line loop.) Has anyone got a clue about this as well - either proofs or disproofs? Imgaril (talk) 23:35, 18 September 2011 (UTC)

I've not come across an example of such a motor, and I'm not skilled enough in the field to assert that there are none, but of course, without rotation taking place, the magnetic and gravitational torques will balance each other, and this static balance would be true even with the push and pull of regular magnets. As far as I am aware, there is no distinction between a electromagnetic field of a permanent magnet (the field lines of these are rarely uniform anyway) from an equivalent electromagnet's field, thus I don't likewise see a reason to expect a net torque producing a continuous rotation. Perhaps, someone more familiar with Gauss's law than I will have something relating to this to add though. --Modocc (talk) 00:46, 19 September 2011 (UTC)

Thanks. I think I see what you meant about the bar magnet -the loop (as a whole) is indistinguishable from a bar magnet - and I agree that a bar magnet will not work. so. yes. that helps me confirm my gut feeling

I'm clutching at straws with gauss's theorem - I vaguely recall a named theorem - it's obviously not Earnshaw's theorem, but I don't think the result is directly explained by Gauss's law for magnetism either. Perhaps there is another named theorem? (or I imagined it)Imgaril (talk) 10:41, 19 September 2011 (UTC)

But how...

Why/how does the background begin waving? http://www.youtube.com/watch?v=EnrwrwMfNSs 65.92.5.209 (talk) 23:49, 16 September 2011 (UTC)

Simple answer is that the rotor is going faster when moving in the direction of the craft and slower when it is on the other side going back towards the rear. In the advancing mode, it experiences more lift – and rises: whilst in the retrograde it loses lift and bends down. The makes the horizon appear to rise and fall. --Aspro (talk) 00:04, 17 September 2011 (UTC)

But at one point the image inverts itself, how does that happen? 65.92.5.209 (talk) 03:38, 17 September 2011 (UTC)

The helicopter is probably changing directions and even flying inverted throughout the video. APL (talk) 03:42, 17 September 2011 (UTC)

Even in a single frame, the horizon seems very curved. The only explanation I can think of is that the camera uses some kind of raster scan, taking its picture from one end to the other, with vertical rows as seen on the YouTube video. Because helicopter blades constantly turn as they rotate, it would be at one angle at the beginning of the shooting of one individual frame, and at another angle, even 180 degrees opposed, at the end of it. No guarantees on this guess, however. Wnt (talk) 14:27, 17 September 2011 (UTC)

Yes, the exact explanation is going to be a bit more complicated. If one looks at the stationary shot, the angle of view is what one would expect from a normal lens. At speed however, you will notice that the angular distances between the sun and its reappearance is far, far less than 360 deg. This means that the camera is recording the forward and retrograde path of the rotor in one 'linearly' compressed frame due to the rotation being so much faster then the scan rate. Remember too, the pitch of the rotor is changing during each revolution. In forward flight, the advancing blade has to pitch down and the the other blade has to pitch up so to achieve the same lift on both sides of the chopper – for obvious reasons. It appears that the camera pitches with the blade that its mounted on. This will increase the apparent curve of the horizon. Must have been tricky getting the counter-balance right on the opposing rotor. --Aspro (talk) 16:46, 17 September 2011 (UTC)