41 replies to this topic

[Reedbeta]

No, I can't give a link because the book is not online

Try looking in your library, or a university library if you have access to one.

PlayStationX said:

all i was talking about is just as simple as that: F = m*a = Q1*Q2/r^2

Well, you asked for a *QM* simulation of the hydrogen molecule, and that's what I was saying I don't know about.

If you want to talk about a simulation based on SED, that might be slightly more up a game-developer's alley.

[PlayStationX]

Quote

If you want to talk about a simulation based on SED, that might be slightly more up a game-developer's alley.

id be delighted to, if anyone can decipher formulas and express attraction/repulsion (acceleration) between two particles in some mathematical relation, then i can do all the rest.

similarly to F=m*a = G*M1*M2/r^2 or F= m*a = k*Q1*Q2/r^2, goal is to get acceleration ratio per time-step, after that it gets easy.

Quote

No, I can't give a link because the book is not online

ah, yes.. i noticed afterward. i too recommend Feynman, interesting read no matter what is he talking about and there is lots of it free on the WWW, in one form or another.

anyway, even if this is not the right place i think it cant hurt to talk about physics a bit. i do not think this nuclear science is complicated at all, and you know plenty about it to throw some comments and correct me, which is good enough for me and hopefully someone else too can find this interesting and throw few cents in...


so, let me try to make a conclusion. in the light of my latest findings i can indeed confirm that this "magnetic fields theory" has never been considered, even less simulated. this is mostly due to QM "approximating" all these field forces to some uncertainty function, and unfortunately being so good at it.


one thing is for sure,
if anyone, i will not be the one to get Nobel prize for this "theory", there seem to be more people that have thought about it in much more detail. so, here it is - this is pretty much what i had in mind and hopefully more readable than my blabbering...

Quote

"proposed theory"

According to new interpretation, every atom of hydrogen possesses an electron magnetic moment due to the electron movement. The magnetic moment of nucleus is lower so it is not important in this case. The electron magnetic moment is formed by combination of orbital and spin magnetic moment using known rules of vectors. The covalent bond means that both atoms attract reciprocally due to the magnetic interaction between their magnetic moments.

...As comparison, quantum mechanic is incapable to explain why two opposite spin are lowering the energy of system. In the same time there is a contradiction in actual theory when the electrons are filled on subshell in atomic structure and when a covalent bound is formed.

http://www.elkadot.c...valent_Bond.htm

it just strikes me as very logical...


anyhow, my real and more practical problem is something else. ill share this with everyone in hope to entertain and possibly get a nice and elegant solution to this problem, here is a question for MASTER DEVELOPERS:

- imagine hundreds of bar magnets fall randomly on a floor. eventually they will orient in such way as to make complete system take on the "lowest energy" state. eg. if you put only two bar magnets next to each other they will rotate so that north pole of one is next to the south pole of another.

how would you go about simulating this?

[PlayStationX]

meanwhile, somewhere else...

Quote

Sounds like the Ising model

http://en.wikipedia....iki/Ising_model

hey, thanks a lot! that does indeed sound like it.


now, i have to admit i have no idea how to apply it. let me rephrase the problem slightly so that hopefully we can derive some formula that can be plugged in computer....

lets image only 3 bar magnets sitting on a table randomly spaced. they can not move, but can only rotate around its center, it is 2D situation. there is no friction of any kind, no gravity and the only forces are magnetic forces. here is a picture where "x" is the point of rotation and coordinate center of each magnet, we have "top" magnet, "middle" magnet and "bottom", like this:

[S- x -N]                                topMag: a=90, x=7, y=25

                  [N- x -S]              midMag: a=270, x=18, y=19


      [S- x -N]                          botMag: a=90, x=12, y=4

- input are 3 initial angles and 3 pairs of (x,y) coordinates
- output are the new angles after system stabilize


1.) any idea how to derive equations to fit this case?

2.) for every initial position is there only one solution, even in 3D and with more magnets?

3.) gravity and magneto-electric fields act instantaneously on distance, right? so, is there any way to find out how quick reaction will be - how strong is rotational acceleration each instant for each magnet and how does it depend on distance?

4.) imagine now these magnets float in 3D space freely, and lets consider the middle magnet. it has a "choice" to be repelled to the right or to rotate its south pole to the left. will it rotate or will it translate? in other words, how to split acceleration to translation and rotation, what will it depend on if there is no friction and magnets are point particles with magnetic dipole moment?

[Reedbeta]

I believe you can model the situation here using the equations for a magnetic dipole field. From Wikipedia, the vector equation for the far field from a magnetic dipole is:

B = mu_0 / (4 * pi * r^3) * (3 * (M . S) * S - M)

where B is the magnetic field, r is the distance from the center of the magnet, M is the magnet's dipole moment (i.e. a vector pointing along the south-to-north axis, with magnitude proportional to the magnet's strength), and S is the normalized vector from the center of the magnet to the point at which we're evaluating the field.

The equations of motion are then (from the same page):

tau = cross(M,

for each magnet, where tau is the torque on an individual magnet, M is the dipole moment again, and B is the magnetic field due to the other magnets (of course we don't include the field of the magnet we're evaluating the torque for, since it would be infinite).

Since we are in 2D, we can also say that

d^2 theta / dt^2 = length(cross(M, ) / I,

where theta is the rotation of this particular magnet and I is its moment of inertia. In 3D you would have to use a rotation matrix or a quaternion to represent the magnet's orientation and the integration is a bit more complicated.

So, to answer your question #2, yes, I believe these equations of motion give rise to a unique evolution in time for any initial conditions, even in 3D.

As for #3, actually gravitational and electromagnetic fields act at the speed of light, not instantaneously. However, if your magnets are rather heavy and weak, then their motion will not approach light-speed and you can use the instantaneous effect as a good approximation.

For #4, I am not totally certain about this but I would guess that you can actually treat the rotational and linear components independently. In other words, there is no need to "split" the effect between translation and rotation; both motions occur fully. This is analogous to what happens in rigid body mechanics when you apply a force to a point off of the center-of-mass of a rigid body. The force applies fully to the motion of the center-of-mass, but also (as a torque) applies fully to the body's rotation about its center-of-mass. It is counterintuitive, but it is easy to see that it has to work out this way if you look at it from a conservation of momentum perspective.

[PlayStationX]

ok, thanks a lot. that looks very good!

im kind of surprised with these formulas, they smell too good. im really unable to say anything, i think i need to put it in a program and visualize it to get a better feeling.

something tells me there is some problem here, inherently non-solvable, similar to how n-body orbits are unpredictable, but worse. let me do some thinking before i comment, but in any case this looks like great starting point, thanks.

Quote

As for #3, actually gravitational and electromagnetic fields act at the speed of light, not instantaneously.

that would imply that Earth orbits the point where we see sun, some point where sun was, rather than the point where sun mass is.

well, this is not really important as you say and Wikipedia seems to agree with you so i do not really see how to settle this argument, whom to believe? i believe math and geometry, and so if there is such proof i will believe that.


its just that i thought this was accepted knowledge since i also thought the gravity was measured could not be less than many many times the speed of light for some effects to explain.

what i wonder about this situation is - if magnetic fields have limited speed of propagation, would that not make our solution oscillate?

[PlayStationX]

Quote

For #4, I am not totally certain about this but I would guess that you can actually treat the rotational and linear components independently. In other words, there is no need to "split" the effect between translation and rotation; both motions occur fully. This is analogous to what happens in rigid body mechanics when you apply a force to a point off of the center-of-mass of a rigid body. The force applies fully to the motion of the center-of-mass, but also (as a torque) applies fully to the body's rotation about its center-of-mass. It is counterintuitive, but it is easy to see that it has to work out this way if you look at it from a conservation of momentum perspective.

yes, i kind of see what are you referring to, let me try to express my confusion further... this is without considering your new equations, i still need time to grasp that


the part of the trick seem to be that this is point particle, so does it mean that it can rotate instantaneously to "new" position?

[N- x -S]



        [N]
        [X]

say, bottom magnet is fixed and we have our OVERALL attraction, but imagine top magnet is particle at single point 'x', then our attraction is single vector.

obviously south pole of top magnet will stick to [N], but how fast will it rotate (angular acceleration)? how that depends on the distance between the poles? can you throw some equations for this as well or point in Wikipedia to what problem this relates to?

[Reedbeta]

PlayStationX said:

something tells me there is some problem here, inherently non-solvable, similar to how n-body orbits are unpredictable

N-body problems in electrodynamics suffer from the same problems as in gravitation (it is pretty much the same equations after all). The problem is not that there are multiple solutions for the same initial conditions - that is not true in classical gravitation either; both theories are deterministic. The problem is that the long-term behavior of the system depends very sensitively on the initial conditions, such that a tiny change initially creates a very big change later on. This means we often aren't very effective in predicting the behavior of the real system due to unavoidable measurement error in the initial conditions. This is called the butterfly effect.

PlayStationX said:

that would imply that Earth orbits the point where we see sun, some point where sun was, rather than the point where sun mass is.

Sure. But the sun isn't moving at anywhere near the speed of light, so the idea that we orbit around the sun is a very good approximation.

PlayStationX said:

well, this is not really important as you say and Wikipedia seems to agree with you so i do not really see how to settle this argument, whom to believe?

The fact that electromagnetic forces act at a limited speed is what gives rise to electromagnetic waves, including radio waves, microwaves, infrared, visible light, UV, X-rays, and gamma rays. If electromagnetic forces acted instantaneously none of these things would exist.

For gravity, the theory of relativity predicts that gravitational waves should exist as well, and there is strong circumstantial evidence for these, although they have not yet been detected directly.

PlayStationX said:

its just that i thought this was accepted knowledge

Your 'accepted knowledge' is about 150 years out of date.

PlayStationX said:

what i wonder about this situation is - if magnetic fields have limited speed of propagation, would that not make our solution oscillate?

Absolutely. However if the magnets are relatively heavy and weak, as I mentioned before, the effects of light-speed delays are very small and can be ignored to a good approximation (the oscillations will be very weak). By the way, if you want to include light-speed delays in the simulation, you can do it as follows: when you evaluate the magnetic field at a point, rather than using the current orientation of each magnet, use its orientation at a time t in the past equal to t = r / c, where c is the speed of light and r is the distance to the magnet.

PlayStationX said:

but how much will it rotate and how fast will it translate? how that depends on the distance between the poles?

Actually the poles of a magnet don't have well-defined locations so you can't easily define a distance between them. I am thinking of the magnets as being physical objects, like little iron bars or something, which means they have a mass and a moment of inertia and therefore take time to rotate. But those parameters (mass, moment of inertia) are independent of the magnetic properties of the object.

[PlayStationX]

Quote

The fact that electromagnetic forces act at a limited speed is what gives rise to electromagnetic waves, including radio waves, microwaves, infrared, visible light, UV, X-rays, and gamma rays. If electromagnetic forces acted instantaneously none of these things would exist.

that sounds all right. is that connected to how our magnets would oscillate?

i never heard of it in that context really, so i'll look it up, but if you can throw some links about it maybe you can save me some time.



as for the rest, i only got more confused after thinking about it, but i'll leave all that for when i make a simulation of it.

[PlayStationX]

Quote

The fact that electromagnetic forces act at a limited speed is what gives rise to electromagnetic waves, including radio waves, microwaves, infrared, visible light, UV, X-rays, and gamma rays. If electromagnetic forces acted instantaneously none of these things would exist.

after carefully studying Wikipedia article on the "Speed of Gravity", i have to say that even thought it states at the beginning that it is that of light, the rest of the article actually gives much more evidence against such possibility.


also, i was unable to find any connection that will point limited speed of field forces would cause radiation. can you explain?

they say radiation is simply caused by acceleration, which has to do with our previous problem that magnetic fields DO NOT accelerate charged particle, while they still DO change their direction?! that is a point where either logic or language fails.


anyway,
i think it is important to make differentiation more clear here, one thing is speed of propagation of electromagnetic WAVE (photon) and possibly, completely another thing is speed of interaction of e/m FIELD.

[Reedbeta]

PlayStationX said:

after carefully studying Wikipedia article on the "Speed of Gravity", i have to say that even thought it states at the beginning that it is that of light, the rest of the article actually gives much more evidence against such possibility.

What do you mean? The "experimental measurement" section of the article clearly says that all current evidence is consistent with the prediction from relativity theory that gravity acts at the speed of light.

PlayStationX said:

also, i was unable to find any connection that will point limited speed of field forces would cause radiation. can you explain?

Certainly. Imagine a piece of rope. Say you and your friend hold the ends of this rope and pull it tight between you, like you were going to have a tug of war or something. Then you start moving your end of the rope up and down. You'll create waves in the rope that you can see travelling down the rope toward your friend's end. If you cannot visualize this I highly recommend actually trying it.

This is a good analogy to what happens when electromagnetic radiation is emitted. The rope is like the electromagnetic field, and you moving it around is like an accelerating charge. The motion of the charge creates a disturbance in the field that moves away (in all directions) from the charge but maintains its shape as it does so.

You can see that this depends on the field having a finite speed of propagation. If the field everywhere changed instantly when a particle moved, there would be no opportunity for waves to be created. It's as if instead of having a rope you had a rigid metal bar. If you moved one end of it, it would instantly move the other end as well, and never change its shape.

PlayStationX said:

magnetic fields DO NOT accelerate charged particle, while they still DO change their direction

Actually, in physics, acceleration means a change in velocity, and since velocity includes both speed and direction, a change in direction IS in fact an acceleration and causes radiation. This is why, for instance, particle accelerators create synchrotron radiation from a particle moving in a magnetic field.

PlayStationX said:

i think it is important to make differentiation more clear here, one thing is speed of propagation of electromagnetic WAVE (photon) and possibly, completely another thing is speed of interaction of e/m FIELD.

They are one and the same thing for electromagnetic waves in a vacuum. However, for other kinds of waves such as acoustic waves (sound waves), or electromagnetic waves in a dielectric medium (such as glass or water), the distinction can be important. See: phase velocity vs. group velocity.

[PlayStationX]

Quote

What do you mean? The "experimental measurement" section of the article clearly says that all current evidence is consistent with the prediction from relativity theory that gravity acts at the speed of light.

the way whole article is written leans toward your interpretation.

still, "experimental measurement" is portrayed as questionable and pretty much everything else, like 80% of the article, points to greater speed and inconsistency with experiments if it is limited to speed of light. on the other hand, most of that refers to historical interpretations and some particular famous scientist from the "past".

so, let me describe again - it can hardly change anyone's opinion


anyway, let me ask you why planets do not radiate gravity waves and collapse into the sun, just as electrons are predicted to radiate and fall in nucleus?

Quote

This is a good analogy to what happens when electromagnetic radiation is emitted. The rope is like the electromagnetic field, and you moving it around is like an accelerating charge. The motion of the charge creates a disturbance in the field that moves away (in all directions) from the charge but maintains its shape as it does so.

You can see that this depends on the field having a finite speed of propagation. If the field everywhere changed instantly when a particle moved, there would be no opportunity for waves to be created. It's as if instead of having a rope you had a rigid metal bar. If you moved one end of it, it would instantly move the other end as well, and never change its shape.

how about e/m fields are "instantaneous" like metal bar and e/m waves are like sound waves inside that metal?

Quote

They are one and the same thing for electromagnetic waves in a vacuum.

i guess what im saying is that there is a rope, an aether, which is made of e/m fields, they are everywhere and there is no such thing as vacuum in that respect. e/m waves are just that, some deformation/vibration pulsating in the "sea" of e/m fields. this does not mean e/m fields act instantaneously, just makes difference between "medium" and "localized state of medium".


well, im not quite sure if there is any point talking about this really, i wish there could be some agreement on what kind of reference we can both agree on, so then we can simply quote that reference and finish the argument.


as for spinning magnets,
im still struggling with torque equations, i have no idea how to compare results to the real-world? how to make the test case?

isn't there some java applet with rotating bar magnets?

Quote

Actually, in physics, acceleration means a change in velocity, and since velocity includes both speed and direction, a change in direction IS in fact an acceleration and causes radiation. This is why, for instance, particle accelerators create synchrotron radiation from a particle moving in a magnetic field.

"a change in direction IS in fact an acceleration" - that is exactly what i am saying. so, it is absolute contradiction to say magnetic field can change direction, but can not accelerate.

its just one of those things we have to accept, its not like we have a choice here as with "speed of gravity", which is not really important since our equations work without that part anyway.


the other option is to simply give a new name for this type manifestation.

[Reedbeta]

PlayStationX said:

still, "experimental measurement" is portrayed as questionable

I would disagree with that. The consensus today among professional physicists is that the current experimental evidence is not in conflict with the prediction of gravitational waves. There is also evidence, for instance the orbital decay of the Hulse-Taylor system, that would be very difficult to explain without gravitational waves. Plus, the prediction comes straight from general relativity (GR), which is an extremely well-tested theory of gravity over 90 years old. GR has done so well in the past that no one really expects it to be wrong about this subject.

That being said, it is true that we have not yet directly detected a gravitational wave; however, this does not falsify of the idea because of the limitations of current experiments - specifically, that they can only detect high-frequency gravitational waves, which if they exist would be relatively rare. There is an experiment, LISA, in the planning stages now, which would be able to detect low-frequency gravitational waves as well.

PlayStationX said:

pretty much everything else, like 80% of the article, points to greater speed and inconsistency with experiments if it is limited to speed of light.

I think you may be confusing the historical development of gravity theories with the current consensus. The various theories discussed in the article are not to be taken as being equally valid. GR is the premier theory of gravitation today and it predicts that gravity acts at light-speed. The fact that gravity acts instantaneously in Newtonian gravity (for instance) should not be taken as conflicting with GR - rather, Newtonian theory is an approximation that is only valid when discussing weak gravitational fields and slow-moving objects, in which case instantaneous gravity is a very good approximation.

PlayStationX said:

anyway, let me ask you why planets do not radiate gravity waves and collapse into the sun, just as electrons are predicted to radiate and fall in nucleus?

According to GR, they actually do radiate gravitational waves. However, they would lose energy very, very, slowly, so that it would take an extremely long time (trillions of years, at least) for there to be any noticeable effect on the orbit. However if you look at orbital systems like binary stars, such as the aforementioned Hulse-Taylor system, the fact that the stars are much more massive and closer together than the Earth and Sun makes them radiate much more strongly. And we can actually measure the decay of their orbit due to emission of gravitational waves.

PlayStationX said:

how about e/m fields are "instantaneous" like metal bar and e/m waves are like sound waves inside that metal?

This is inconsistent, because if an object is truly rigid you can't have sound waves in it. After all, sound waves are just mechanical movement of atoms relative to each other, and in a truly rigid object the atoms can't move relative to each other.

Of course, real objects are never truly rigid, but in metal the sound waves travel very quickly, thousands of miles an hour.

PlayStationX said:

i guess what im saying is that there is a rope, an aether, which is made of e/m fields, they are everywhere and there is no such thing as vacuum in that respect. e/m waves are just that, some deformation/vibration pulsating in the "sea" of e/m fields.

Yes, this is accurate.

PlayStationX said:

as for spinning magnets,
im still struggling with torque equations, i have no idea how to compare results to the real-world? how to make the test case?

Heh, unfortunately I can't help you with that one. I'm a theoretician.

[PlayStationX]

Quote

I would disagree with that. The consensus today among professional physicists is that the current experimental evidence is not in conflict with the prediction of gravitational waves.

i agree that applicability of that theory is good enough basis to believe that the rest, less proven, aspects of it could be taken as true.

it is only my feeling that all the trouble about great unification of theories comes out of lack in understanding about field forces, so im simply refusing to believe what does not come intuitive to me, until there is some more practical evidence. by 'practical' i literary mean "possible to simulate".


that's it then, we settled it.

just those magnets then... awwwwwww. equations, algorithms, ideas are welcome.


meanwhile, this is what im trying to get my head around...

those new equations do have cross product, so there is some "right hand rule" going on there and there is something being perpendicular to something else.


however, from my real-life experience i think that i saw magnets attract each other in straight lines? do they not?

i mean what am i to expect when i place two magnets in free space like this:

[N- x -S]          [N- x -S]

should they connect in a straight line like opposite charges would, or should they "wiggle" and kind of spiral toward each other?

i mean, how do i know whats bug and whats not... hahaha!

[Reedbeta]

Yes, those magnets move toward each other in a straight line. In this case since their directions are aligned the cross product will be zero and so there is no torque, and no rotation.

[PlayStationX]

Quote

Yes, those magnets move toward each other in a straight line. In this case since their directions are aligned the cross product will be zero and so there is no torque, and no rotation.

are you saying that if there was some initial angle they would not connect in a straight line? its just that i never experienced it, shame i do not have any magnets here.

i know they wiggle when they repel, but i never noticed they wiggle when they attract. this leads to question if the line of attraction between metal and magnet is a straight line or not?


what is the difference?

take a neutral metal molecule and take some charged non metal molecule with the same mass. how trajectories of non-metal ion and neutral metal molecule differ as they are attracted and travel towards magnet?

what part of metal molecule is responsible for attraction - electron/proton charge, or their magnetic moment?

[PlayStationX]

i was trying to answer my own questions when i stumbled over something that will throw us few steps back.

Quote

http://en.wikipedia....haw%27s_theorem

Earnshaw's theorem states that a collection of point charges cannot be maintained in a stable stationary equilibrium configuration solely by the electrostatic interaction of the charges. This was first proven by Samuel Earnshaw in 1842. It is usually referenced to magnetic fields, but originally applied to electrostatic fields. It applies to the classical inverse-square law forces (electric and gravitational) and also to the magnetic forces of permanent magnets and paramagnetic materials (but not diamagnetic materials).

the way i read this is that n-dipole-body system does not actually have unique solution... or, it means solution oscillates and if it oscillates(regardless of any speed limits), then is it regular in amplitude or is it some chaotic oscillation?

i still think this is not the same kind of chaotic, but predictable problem as with n-body orbit. whether chaotic or kind of oscillatory, i think n-dipole system actually could be non-deterministic, which could explain the need and practicability of QM and uncertainty.



next, i think we need to separate these distinct case:

1.) magnet dipole - magnet dipole
2.) magnet dipole - electric charge
3.) magnet dipole - metal molecule
4.) magnet dipole - charged metal molecule

the problem with all of them is that it is really impossible to separate magnetic field from electric, so lets take 1st two case as example...

#1 - we take some equation to calculate dipole-dipole magnetic attraction, but as soon as they start moving towards each other they will produce some electric (radiation too perhaps) property, which is not accounted for in original equation we started with.

#2 - we take some equation to calculate dipole-charge attraction, but as soon as charge starts moving it will produce magnetic moment, which is not accounted for in original equation we started with.

#3,#4 - similar to above...


obviously im very confused about it all, much more then i was few days ago when i thought i know a lot about magnets... it turns out - i know nothing

help!

[alphadog]

Just a quick suggestion: Try real physics textbooks and journals, instead of piecing together stuff from Wikipedia...

[PlayStationX]

Quote

Just a quick suggestion: Try real physics textbooks and journals, instead of piecing together stuff from Wikipedia...

are you saying you know the answer but you will not tell us?


Wikipedia does contain knowledge from the textbooks and to my knowledge they actually try to keep it updated and that they have considered many of the textbooks before they wrote the articles.

in that respect Wikipedia should/could be more objective and complete reference than the printed "real" textbooks.


in any case, i do have quite a few real physics textbooks as well as access to the rest of the Internet that is not Wikipedia, including all the scientific journals, articles, peer-reviewed published papers as well as non-published...


i can not find the answer, so can you please tell us?

[alphadog]

No. I would not be able to spit out a model from memory.

I'm just saying that you may be better served by going to high-level textbooks because the level of detail and understanding you want or need is not going to be in Wikipedia. Wikipedia is good for "summary" information, and as a "kick off" place to start your deeper research.

At least, that's what I would do.

I had another thought: you may want to search for "demo scene" forums too. These folks frequently adapt scientific concepts like fluid dynamics, Conway's Game of Life, etc, into graphical representations...

[PlayStationX]

Quote

I had another thought: you may want to search for "demo scene" forums too. These folks frequently adapt scientific concepts like fluid dynamics, Conway's Game of Life, etc, into graphical representations...

thank you.

any information that can bring me closer to answer is welcome. can you give a link of that forum please?


i still think that im asking some very basic questions. i mean half of these could be answered by anyone who have some magnets at home.

the other half i thought i would have no trouble finding info on Wikipedia, even less on the whole world wide web, i simply ask how magnets attract - in spiral or straight line. i thought i knew, i thought they told me that even back in high-school, but i now realize i really have no idea about it.


i mean, im only asking if your fridge magnet spirals as it travels to your fridge door?

mine seem to go straight line, but they are so weak and i can not judge what happens in that small distance, such short time.