I suppose this is OT at this point, but all evidence indicates the earth is NOT a permanent magnet, but rather an induced electro-magnet within the heliosphere. i.e., without the sun's magnetosphere, the earth would not have one...pstudier wrote:I'm not positive about this. The Van Allen Radiation Belts are confined with a permanent magnet called the Earth. The mirror losses cause the northern lights. Don't know the confinement time or density or temperature. Maybe if we filled it with deuterium, we could make the earth into a sun.drmike wrote: PM's just have too much electron loss, so a coil will always be more efficient.
Experiments with solid-state magnets
Tom.Cuddihy
~~~~~~~~~~~~~~~~~~~~~
Faith is the foundation of reason.
~~~~~~~~~~~~~~~~~~~~~
Faith is the foundation of reason.
I believe the sun flips every 22 years and the earth on the scale of 100s of thousands.cuddihy wrote:I suppose this is OT at this point, but all evidence indicates the earth is NOT a permanent magnet, but rather an induced electro-magnet within the heliosphere. i.e., without the sun's magnetosphere, the earth would not have one...pstudier wrote:I'm not positive about this. The Van Allen Radiation Belts are confined with a permanent magnet called the Earth. The mirror losses cause the northern lights. Don't know the confinement time or density or temperature. Maybe if we filled it with deuterium, we could make the earth into a sun.drmike wrote: PM's just have too much electron loss, so a coil will always be more efficient.
Be interesting to see how that all fits together.
Mag fields fall off as 1/r^3. I can't see how the sun's magnetic field could create the earth's field. The sun is only 10^6 times larger than earth, so it's field would be at best 10^6 times larger. But the distance is 10^11 meters, so 10^6/10^33 is really really tiny.
I might have blown a few decimal points, but I just can't see how the sun's field can affect the earth. The solar proton flux does bend the earth's field, and that's been measured - but that's a plasma interaction.
I agree with pstudier - this is like a torroidal coil. Ampere's law says the field is zero outside. Reality is that materials saturate, so there will be some field, but it will be really weak.
That's not to say PM's can't be made to work for fusion confinement - it just says your potentials and field lines have to be different than the Bussard aragement. Charge the magnets negative so the electrons don't run into them and make the path's of the ions miss the magnets because of the electron density. Could be an interesting configuration!
I might have blown a few decimal points, but I just can't see how the sun's field can affect the earth. The solar proton flux does bend the earth's field, and that's been measured - but that's a plasma interaction.
I agree with pstudier - this is like a torroidal coil. Ampere's law says the field is zero outside. Reality is that materials saturate, so there will be some field, but it will be really weak.
That's not to say PM's can't be made to work for fusion confinement - it just says your potentials and field lines have to be different than the Bussard aragement. Charge the magnets negative so the electrons don't run into them and make the path's of the ions miss the magnets because of the electron density. Could be an interesting configuration!
Star magnetic fields
Dr. Mike,
A year or two back, some articles came out on magnetars, which are neutron stars with mind-bogglingly intense magnetic fields, ten to the double-digits Teslas. Some fool pointed out that if a magnetar passed a hundred thousand miles away from you (less than half the distance to the Moon!) the magnetic field would erase your credit cards (that would take on the order of 0.1 T). Like we'd care after the gravitational field slung us out of the galaxy. Earth's field runs about 0.00006 T (60,000 nT) at the surface.
I consider that an example of how fast a star's magnetic field falls off with distance. I checked the numbers and it really does fall off that fast. The Sun's field at 1 AU is pathetic. SOHO produces continuous magnetic field data at its L1 station point, on the order of 50 nT, but that is mostly from solar wind. Solar wind can challenge Earth's field, and I was monitoring a CME a couple of years back that blew Earth's field away down to the atmosphere, but that was strictly due to the current of the CME, not flux from the Sun itself.
http://www.sec.noaa.gov/SWN/index.html
Earth's magnetic field is supposed to be formed by molten iron moving. The iron is way hotter than its curie point, so could not be permanantly magnetized in the usual sense.
I think the wedge donut magnet drawing is correct but useless. The torus will form a closed magnetic circuit, with almost all of the flux contained within it. There will be a trace outside, I think in the opposite direction. Permeability of the magnet is not infinite, and there is always some leakage from these, but it will be trivial compared to the needs of a magrid. I worked with something like this when Jeff Kooistra suckered me in to toying with a little mind-bender called the Marinov Motor (a bipolar homopolar motor with the middle of the disk missing, designed to make a fool out of anyone who dared theorize about how it worked, Marinov especially). The thing worked at least partly on field leakage of what was supposed to be a "closed magnetic circuit"
A year or two back, some articles came out on magnetars, which are neutron stars with mind-bogglingly intense magnetic fields, ten to the double-digits Teslas. Some fool pointed out that if a magnetar passed a hundred thousand miles away from you (less than half the distance to the Moon!) the magnetic field would erase your credit cards (that would take on the order of 0.1 T). Like we'd care after the gravitational field slung us out of the galaxy. Earth's field runs about 0.00006 T (60,000 nT) at the surface.
I consider that an example of how fast a star's magnetic field falls off with distance. I checked the numbers and it really does fall off that fast. The Sun's field at 1 AU is pathetic. SOHO produces continuous magnetic field data at its L1 station point, on the order of 50 nT, but that is mostly from solar wind. Solar wind can challenge Earth's field, and I was monitoring a CME a couple of years back that blew Earth's field away down to the atmosphere, but that was strictly due to the current of the CME, not flux from the Sun itself.
http://www.sec.noaa.gov/SWN/index.html
Earth's magnetic field is supposed to be formed by molten iron moving. The iron is way hotter than its curie point, so could not be permanantly magnetized in the usual sense.
I think the wedge donut magnet drawing is correct but useless. The torus will form a closed magnetic circuit, with almost all of the flux contained within it. There will be a trace outside, I think in the opposite direction. Permeability of the magnet is not infinite, and there is always some leakage from these, but it will be trivial compared to the needs of a magrid. I worked with something like this when Jeff Kooistra suckered me in to toying with a little mind-bender called the Marinov Motor (a bipolar homopolar motor with the middle of the disk missing, designed to make a fool out of anyone who dared theorize about how it worked, Marinov especially). The thing worked at least partly on field leakage of what was supposed to be a "closed magnetic circuit"
Howdy Tom,
Thanks for that NOAA link, I love the "total electron content" map. My kids will like the solar image - they mapped the sun during a summer camp - no sun spots at all during the two weeks they looked!
Speaking of field maps, I'm working on some algorithms to map an arbitrary number of coils around a sphere. The geometry is fun, but I'm making too many assumptions about symmetry so odd numbers of coils don't give me what I expect. Hopefully I'll have the math worked out in a week or so.
But once I have the field maps, I need to work on electron space charge. I've got a lot of ideas on how to tackle that, but if you know of any papers that are easily accessable I'd love to know about them.
Thanks for that NOAA link, I love the "total electron content" map. My kids will like the solar image - they mapped the sun during a summer camp - no sun spots at all during the two weeks they looked!
Speaking of field maps, I'm working on some algorithms to map an arbitrary number of coils around a sphere. The geometry is fun, but I'm making too many assumptions about symmetry so odd numbers of coils don't give me what I expect. Hopefully I'll have the math worked out in a week or so.
But once I have the field maps, I need to work on electron space charge. I've got a lot of ideas on how to tackle that, but if you know of any papers that are easily accessable I'd love to know about them.
drmike,
The best exposition on space charge I have seen is here:
http://www.belljar.net/basics.htm
free vacuum tube books. I like Chaffee from 1933 to start.
BTW if you delve into it, they knew enough then to make gas lasers had they had the thought.
The best exposition on space charge I have seen is here:
http://www.belljar.net/basics.htm
free vacuum tube books. I like Chaffee from 1933 to start.
BTW if you delve into it, they knew enough then to make gas lasers had they had the thought.
Actually you should be counting the distance to center of the sun not in meters, but in 'sun radii'. 1AU ~= 230 sun radii, so the sun's magnetic field at 1AU should be approximately 12,000,000 times weaker than at the sun's surface, if my math is correct.drmike wrote:Mag fields fall off as 1/r^3. I can't see how the sun's magnetic field could create the earth's field. The sun is only 10^6 times larger than earth, so it's field would be at best 10^6 times larger. But the distance is 10^11 meters, so 10^6/10^33 is really really tiny.
Earth is moving through this magnetic field at over 66,000 miles an hour, so I could expect that this would generate strong currents in the molten iron in earth's core.
Dr. Mike,drmike wrote:Howdy Tom,
Thanks for that NOAA link, I love the "total electron content" map. My kids will like the solar image - they mapped the sun during a summer camp - no sun spots at all during the two weeks they looked!
Speaking of field maps, I'm working on some algorithms to map an arbitrary number of coils around a sphere. The geometry is fun, but I'm making too many assumptions about symmetry so odd numbers of coils don't give me what I expect. Hopefully I'll have the math worked out in a week or so.
But once I have the field maps, I need to work on electron space charge. I've got a lot of ideas on how to tackle that, but if you know of any papers that are easily accessable I'd love to know about them.
I couldn't solve Poisson's equation without a tutor. But I do have a copy of the Langmuir and Blodgette 1924 paper on the spherical reflex diode that was supposedly Farnsworth's original inspiration. I also have the Elmore, Tuck, and Watson paper on the original spherical electron potential well design (already scanned to .pdf, and possibly this one is also posted at Askmar.com). I believe both are out of copyright, if I understand the rules, so maybe there is a way to post them here.
Thanks everybody. I think I'll start writing up my stuff to make it easier to beat me over the head with it. After thrashing around with high level complicated stuff I finally realized how simple the actual problem of setting up coils really is. The process of attempting to describe the theory may well simplify the same way. I'll post stuff on my web page when I get it done.
Too bad I have life to deal with - this stuff is fun!
Too bad I have life to deal with - this stuff is fun!
Hi all,
When I read the suggestion in this thread regarding the use of permanent magnets it got me thinking, so I used femm401 to do a few quick simulations of the whole round magnet idea, using various halbach layouts.
The app I used can be found at http://developer.berlios.de/projects/femm/ it is a free app quite capable of doing 2D magnetics and electrostatics, windows only but works in wine.
At first I just simulated a single round closed halbach array with just 4 elements but due to the femm401 only able to do linear magnetisation I changed to 8 magnet elements. Remember all this was done in air not vacuum so that will make a field strength difference but doesn´t change the general idea.
http://picasaweb.google.com/dewald.piet ... 6164287938
http://picasaweb.google.com/dewald.piet ... 6164287954
I then used 6 of these arrays to try and see how a pm polywell will look, I arranged them in several different configurations.
http://picasaweb.google.com/dewald.piet ... 5157747170
http://picasaweb.google.com/dewald.piet ... 4698478162
http://picasaweb.google.com/dewald.piet ... 5157747186
http://picasaweb.google.com/dewald.piet ... 9452714498
http://picasaweb.google.com/dewald.piet ... 5111722626
http://picasaweb.google.com/dewald.piet ... 0816755314
Each combination reveals an intresting result with distinction advantages and disadvantages. In my opinion a special form of a halbach array could perhaps work but might be superflous. I think a circular array of round magnets with the like poles facing each other is possibly the best permanent magnet solution.
http://picasaweb.google.com/dewald.piet ... 9596461618
http://picasaweb.google.com/dewald.piet ... 9596461634
http://picasaweb.google.com/dewald.piet ... 9596461586
http://picasaweb.google.com/dewald.piet ... 9596461602
Alternatively a special 4 element halbach array could be slightly better due to increased field strength and could possibly minimise the magnetic field lines perpendicular to the electron flow direction between the magnets. This halbach array would consist of 2 small elements magnetised in the axial direction, about 10 arc degrees long. The other 2 elements would have to be magnetised in a circular fashion, filling the rest of the arc.
I am of the opinion and I could be wrong that perpendicular field lines (to the electron flow recirculation flow) between the arrays need not be a problem, since the electrons have a lot of energy in that vicinity (I assume from what I understand how this thing works) and are travelling near perpendicular to those field lines, therefore traversing those field easily without being redirected at the magnet´s north or south pole.
If this setup could be modeled in ephi with some electrons circulating and possibly adding an unipolar negative charge to the middle of the array to try and create a ¨whiffle ball¨ this could yield some intresting results to see if this could work.
Permanent magnets do have some advantages in small simple setups, requiring less input power, less complexity (not necessarily fitting the american big is beautiful mindset though
).
I hope my understanding of how Dr Bussard´s poylwell works is more or less correct, I´m no plasma physicist only a mechanical design engineer.
For anyone intrested in what I have done I will gladly email you the .fem files upon request, pm me with your email address.
Dewald
PS I´m from South Africa, currently working on my countries infamous nuclear project.
When I read the suggestion in this thread regarding the use of permanent magnets it got me thinking, so I used femm401 to do a few quick simulations of the whole round magnet idea, using various halbach layouts.
The app I used can be found at http://developer.berlios.de/projects/femm/ it is a free app quite capable of doing 2D magnetics and electrostatics, windows only but works in wine.
At first I just simulated a single round closed halbach array with just 4 elements but due to the femm401 only able to do linear magnetisation I changed to 8 magnet elements. Remember all this was done in air not vacuum so that will make a field strength difference but doesn´t change the general idea.
http://picasaweb.google.com/dewald.piet ... 6164287938
http://picasaweb.google.com/dewald.piet ... 6164287954
I then used 6 of these arrays to try and see how a pm polywell will look, I arranged them in several different configurations.
http://picasaweb.google.com/dewald.piet ... 5157747170
http://picasaweb.google.com/dewald.piet ... 4698478162
http://picasaweb.google.com/dewald.piet ... 5157747186
http://picasaweb.google.com/dewald.piet ... 9452714498
http://picasaweb.google.com/dewald.piet ... 5111722626
http://picasaweb.google.com/dewald.piet ... 0816755314
Each combination reveals an intresting result with distinction advantages and disadvantages. In my opinion a special form of a halbach array could perhaps work but might be superflous. I think a circular array of round magnets with the like poles facing each other is possibly the best permanent magnet solution.
http://picasaweb.google.com/dewald.piet ... 9596461618
http://picasaweb.google.com/dewald.piet ... 9596461634
http://picasaweb.google.com/dewald.piet ... 9596461586
http://picasaweb.google.com/dewald.piet ... 9596461602
Alternatively a special 4 element halbach array could be slightly better due to increased field strength and could possibly minimise the magnetic field lines perpendicular to the electron flow direction between the magnets. This halbach array would consist of 2 small elements magnetised in the axial direction, about 10 arc degrees long. The other 2 elements would have to be magnetised in a circular fashion, filling the rest of the arc.
I am of the opinion and I could be wrong that perpendicular field lines (to the electron flow recirculation flow) between the arrays need not be a problem, since the electrons have a lot of energy in that vicinity (I assume from what I understand how this thing works) and are travelling near perpendicular to those field lines, therefore traversing those field easily without being redirected at the magnet´s north or south pole.
If this setup could be modeled in ephi with some electrons circulating and possibly adding an unipolar negative charge to the middle of the array to try and create a ¨whiffle ball¨ this could yield some intresting results to see if this could work.
Permanent magnets do have some advantages in small simple setups, requiring less input power, less complexity (not necessarily fitting the american big is beautiful mindset though
).I hope my understanding of how Dr Bussard´s poylwell works is more or less correct, I´m no plasma physicist only a mechanical design engineer.
For anyone intrested in what I have done I will gladly email you the .fem files upon request, pm me with your email address.
Dewald
PS I´m from South Africa, currently working on my countries infamous nuclear project.
http://www.neodymium-magnets-uk.com/Made_to_order.html
I wonder what sort of sizes/amounts and price one might be looking at for this.
I have to say, I quite like the idea of trying to build a small one myself, I was going the fusor direction, but this in some ways looks easier to build.(Plus I have always had a thing for magnets..)
Incidently, I finally caught up reading the entire thread at nasaspaceflight about this, and it further encourages me to try and build a small continuous operating model ultra cheaply.
I wonder what sort of sizes/amounts and price one might be looking at for this.
I have to say, I quite like the idea of trying to build a small one myself, I was going the fusor direction, but this in some ways looks easier to build.(Plus I have always had a thing for magnets..)
Incidently, I finally caught up reading the entire thread at nasaspaceflight about this, and it further encourages me to try and build a small continuous operating model ultra cheaply.
Most shelf ones are square, if one could get horse shoe ones with the straight ends chopped off, possibly.
I once ordered 100 very small ones: 2x3x10 mm, cost me about a half a euro a piece in todays money for a halbach brushless motor I wanted to build. For this type of application one wants bigger magnets.
For a test setup one could combine the appropriate shelf and non shelf ones. I don´t understand enough of plasma physics to be able to say if this can work, that´s why I was hoping to do an ephi simulation to see if it could.
I suggest heading to the yahoo group and looking at endrek´s ephi simulations.
I once ordered 100 very small ones: 2x3x10 mm, cost me about a half a euro a piece in todays money for a halbach brushless motor I wanted to build. For this type of application one wants bigger magnets.
For a test setup one could combine the appropriate shelf and non shelf ones. I don´t understand enough of plasma physics to be able to say if this can work, that´s why I was hoping to do an ephi simulation to see if it could.
I suggest heading to the yahoo group and looking at endrek´s ephi simulations.
I've ordered myself some 700 Neodymium magnets, 6mm,(1/4") cube ones, Dark Nickel (Ni) Plated . Grade N48, cost 9 UK Pence each.
Can anyone tell me whats best to clean them with if (Well, they are all coming damaged, job lot half price because of/etc.) they are damaged ?
Whats an easy way to DIY at home replate them once they have been damaged/cut/filed/etc. ?
I wonder if gold plating them might be a good idea, thoughts ?
(I thought perhaps if they was visible, gold would look nice to the media, and suffer less from outgassing in a vacuum.)
What methods of joining them together might be suitable for working in a vacuum ?
(I had wondered about stainless steel brackets, and I'd assume they would need covering too as well to protect them ?)
Would they need active cooling as well ? (water cooling ?)
Do they need to be inside the glass bell jar vacuum chamber, or could they function through it from outside ?
Also, what would be a good way to look at the magnetic field they generate, iron filings ?
And, what methods exist to get iron filings off magnets ? (stronger electro magnet ?)
Can anyone tell me whats best to clean them with if (Well, they are all coming damaged, job lot half price because of/etc.) they are damaged ?
Whats an easy way to DIY at home replate them once they have been damaged/cut/filed/etc. ?
I wonder if gold plating them might be a good idea, thoughts ?
(I thought perhaps if they was visible, gold would look nice to the media, and suffer less from outgassing in a vacuum.)
What methods of joining them together might be suitable for working in a vacuum ?
(I had wondered about stainless steel brackets, and I'd assume they would need covering too as well to protect them ?)
Would they need active cooling as well ? (water cooling ?)
Do they need to be inside the glass bell jar vacuum chamber, or could they function through it from outside ?
Also, what would be a good way to look at the magnetic field they generate, iron filings ?
And, what methods exist to get iron filings off magnets ? (stronger electro magnet ?)