Hi everyone!
I've had an idea and I was wondering if you've heard of something similar, thought of something similar (even better), or in general what you think of it.
I don't have a mastery of the understanding of the Polywell. But let me see if I got the kind of thinking from the start:
Bussard mentioned that the Fusor would be terrific if the accelerating grid were about 100 times more transparent. It appears that this was the reason for the Polywell in the first place, extend the Fusor by figuring out how to eliminate opacity in the accelleration medium. That's it. It is son of Fusor.
So I did exactly the same thing. And I've made my own design. I hope you will agree that even if it doesn't do diddly squat, the thing is simple at least.
I extended Bussard's line of thought, "heck fusor works only with the electric force, the tokamac only uses the magnetic force, lets use both then!". And that's what the Polywell does. It uses both forces.
Both Polywell and my species of Fusor [i'm calling it the MagnoFusor], are tandem field nuclear reactors because they use both magnetic and electric fields in tandem. Looking at the grave yard and walking dead of fusion ideas, most of them seem to concentrate on one or the other.
So here it is:
https://www.youtube.com/watch?v=RPpegulI5Ow
-Scott
[ohiovr]
New Species of Fusor?
Re: New Species of Fusor?
Interesting and well presented video, but.... 
Mentioning vacuum tubes is appropriate as they are all about manipulating plasmas.
The grid transparency issue is a major issue in fusors, but it is not the only issue. A typical glow discharge fusor might top out with a grid transparency of ~ 95%. This means an ion could travel back and forth through the grid about 20 times. Depending on the size of the machine it may need a grid transparency of at least 99.99%, or allow for ~ 10,000 passe sof the typical ion. If the machine is very big, perhaps a kilometer in diameter, the number of passes required is decreased proportionately. So, at least theoretically a typical gridded fusor could reach break even. But only if this grid transparency to ions is the only consideration. You also, have to account for what is happening to the electrons. Also, the effects of non fusing collisions-Coulomb collisions that scatter the charged particles and drive them to Maxwellian energy distributions. The average ion energy/ temperature may be trapped in the grid induced potential well, but up scattered ions will escape,.. And of course the electrons are essentially one pass participants. The Elmore Tuck and Watson version of the fusor reversed things. Electrons are trapped in a potential well from a grid (anode) placed near the edge of the machine. Wishfully, the electrons are electrostatically (potential well ) trapped, and as they converge towards the center they would form a virtual central cathode to then trap ions. The problem again was the grid transparency (to electrons this time. Also, any ions that reached the radius of the anode would then be accelerated (repelled) to the wall. This could not be avoided as the ions thermalized- a certain percentage would always be energetic enough to escape in this manner, as would some of the electrons.
This is starting to move in the direction of the Polywell concept.
The first consideration is to use excess electrons. This produces a deeper potential well, perhaps approaching the strength of the electron injection energy. The electron injection energy can come from two sources. A peripheral anode grid -like in ETW or WB6, or from high energy electron guns outside the grid (in this case the grid (magnet cans) is at ground.
The grid transparency is addressed by not having the electrons pass the grid once trapped, at least that was the plan. This is done by having magnetic shielding. Magnetic shielding is not perfect though. Charged particles penitrate through magnetic fields much like gas diffuses across a room. Except it is done in discreet steps dependent on the particle energy, momentum, magnetic field strength and density driven scattering collisions. Through a random walk process the charged particle moves through the B field one gyroradius distance at a time either deeper or shallower with each collision. As a result the charged particles migrate from greater to lower densities till the magnet surface is hit and the KE of the particle is lost - generally as heat. Ions with their greater weight and thus momentum have gyro radii >60 times greater than electrons, so at the same B field strength, and particle energy the ions diffuse through the B field ~ 60 times faster than electrons. This is one reason why tokamaks have to be so large, so that the reaction volume can be maintained long enough against this loss mechanism. A much smaller machine is possible if only the electrons are contained in this manner (magnetically contained). At least until inertial confinement regimes are entered.
The ions have to be decoupled from the magnetic field. Whether the electromagnetic field or an internal plasma generated magnetic field , the ion diffusion- ExB diffusion, through the B field is too fast, energy is lost too fast in smaller machines. Some would say the magnetic insulation is inadequate in small machines. In the Polywell though, the ions are not magnetically contained (for the majority of the ions for the majority of their lifetime). The excess electrons produce a potential well that not only confines the ions at radii less than the influence of the electromagnet B fields, but also serves to accelerate them to useful high energy towards the center of the machine. The picture is not of ions flying (or spiraling in a magnatized plasma) back and forth along a cyclinder till they fuse, but ions flying back and forth towards a spherically defined center. This may allow for greatly increased densities in the core which is mostly good , but is not required. The magnetic field does come into play for ions that are up scattered and travel to radii great enough to reach the electron confining magnetic field. These ions are mostly turned back towards the center with out as great of an ExB penalty, because they are a minority of the ions and their energy and thus gyroradius are smaller.
The electrons are contained magnetically, their energy is initially from the ETW like grid, but subsequently their energy is only due to their inertia. They will cool through bremsstruhlung radiation, etc. and there is no additional energy input. They stay hot long enough to get the job done- hopefully, otherwise other means is needed to reheat the electrons/ plasma- such as microwaves, or hot neutral beam injection (that would seem to play havoc with the electrostatic ion containment). In the WTW machines the electrons would be reheated/ re accelerated whenever they passed outside the anode grid. In WB6 the appreciation for the inadequacy of electron cusp magnetic confinement it was appreciated, Here escaping electrons were stopped and re accelerated back into the machine due to the magrid anode positive charge, much like the ETW. This is fundamentally different in later EMC2 machines that apparently have a grounded magrid. The electrons are accelerated well outside the magrid and injected through the cusps. This does not allow for reheating of escaping electrons as they loop around to another cusp. There are obviously tradeoffs (think WB5 failure), but I remain curious why they have chosen this path.
All of this boils down to several clever tricks utilized in the Polywell to change the game- Minimizes ion concerns with ExB diffusion, eliminates much of the concerns about grid transparency, and of course the essential Wiffleball consideration. The triple product solution involves the short confinement times of cusp magnetic fields, but improved enough that with the corresponding increased density, and ease (hopefully) of heating, the containment of particles and heat is sufficient for net positive fusion energy output..
Other considerations such as edge stability, cool electrons in the core, ion confluence (central focus) are additional advantages that may allow for advanced fuel fusion, not just baseline D-T fusion.
In your cylindrical design, the plasma is magnatized, there will be significant ExB diffusion issues. There can be edge instabilities, macro instabilities, MHD instabilities (which ever term you prefer). Cusp losses may be less with two ring cusps on either end-and there have been cylindrical solenoid designs with end ring magnets to decrease end losses, but they have not been pursued, presumably because they failed theoretically or experimentally.
One experiment that may be similar to what you are thinking of was done by George Miley. I think it was called the dipole arrangement. A ring electromagnet was placed between two end curved cathodes which injected electrons towards the center. The electrons quickly left due to a negative space charge buildup. A anode was placed on the inner surface of the magnet to limit this. Some electron confinement was achieved, but I think it was far below what was hoped for.
https://www.google.com/search?q=George+ ... Cg#imgdii=_
Dan Tibbets
Mentioning vacuum tubes is appropriate as they are all about manipulating plasmas.
The grid transparency issue is a major issue in fusors, but it is not the only issue. A typical glow discharge fusor might top out with a grid transparency of ~ 95%. This means an ion could travel back and forth through the grid about 20 times. Depending on the size of the machine it may need a grid transparency of at least 99.99%, or allow for ~ 10,000 passe sof the typical ion. If the machine is very big, perhaps a kilometer in diameter, the number of passes required is decreased proportionately. So, at least theoretically a typical gridded fusor could reach break even. But only if this grid transparency to ions is the only consideration. You also, have to account for what is happening to the electrons. Also, the effects of non fusing collisions-Coulomb collisions that scatter the charged particles and drive them to Maxwellian energy distributions. The average ion energy/ temperature may be trapped in the grid induced potential well, but up scattered ions will escape,.. And of course the electrons are essentially one pass participants. The Elmore Tuck and Watson version of the fusor reversed things. Electrons are trapped in a potential well from a grid (anode) placed near the edge of the machine. Wishfully, the electrons are electrostatically (potential well ) trapped, and as they converge towards the center they would form a virtual central cathode to then trap ions. The problem again was the grid transparency (to electrons this time. Also, any ions that reached the radius of the anode would then be accelerated (repelled) to the wall. This could not be avoided as the ions thermalized- a certain percentage would always be energetic enough to escape in this manner, as would some of the electrons.
This is starting to move in the direction of the Polywell concept.
The first consideration is to use excess electrons. This produces a deeper potential well, perhaps approaching the strength of the electron injection energy. The electron injection energy can come from two sources. A peripheral anode grid -like in ETW or WB6, or from high energy electron guns outside the grid (in this case the grid (magnet cans) is at ground.
The grid transparency is addressed by not having the electrons pass the grid once trapped, at least that was the plan. This is done by having magnetic shielding. Magnetic shielding is not perfect though. Charged particles penitrate through magnetic fields much like gas diffuses across a room. Except it is done in discreet steps dependent on the particle energy, momentum, magnetic field strength and density driven scattering collisions. Through a random walk process the charged particle moves through the B field one gyroradius distance at a time either deeper or shallower with each collision. As a result the charged particles migrate from greater to lower densities till the magnet surface is hit and the KE of the particle is lost - generally as heat. Ions with their greater weight and thus momentum have gyro radii >60 times greater than electrons, so at the same B field strength, and particle energy the ions diffuse through the B field ~ 60 times faster than electrons. This is one reason why tokamaks have to be so large, so that the reaction volume can be maintained long enough against this loss mechanism. A much smaller machine is possible if only the electrons are contained in this manner (magnetically contained). At least until inertial confinement regimes are entered.
The ions have to be decoupled from the magnetic field. Whether the electromagnetic field or an internal plasma generated magnetic field , the ion diffusion- ExB diffusion, through the B field is too fast, energy is lost too fast in smaller machines. Some would say the magnetic insulation is inadequate in small machines. In the Polywell though, the ions are not magnetically contained (for the majority of the ions for the majority of their lifetime). The excess electrons produce a potential well that not only confines the ions at radii less than the influence of the electromagnet B fields, but also serves to accelerate them to useful high energy towards the center of the machine. The picture is not of ions flying (or spiraling in a magnatized plasma) back and forth along a cyclinder till they fuse, but ions flying back and forth towards a spherically defined center. This may allow for greatly increased densities in the core which is mostly good , but is not required. The magnetic field does come into play for ions that are up scattered and travel to radii great enough to reach the electron confining magnetic field. These ions are mostly turned back towards the center with out as great of an ExB penalty, because they are a minority of the ions and their energy and thus gyroradius are smaller.
The electrons are contained magnetically, their energy is initially from the ETW like grid, but subsequently their energy is only due to their inertia. They will cool through bremsstruhlung radiation, etc. and there is no additional energy input. They stay hot long enough to get the job done- hopefully, otherwise other means is needed to reheat the electrons/ plasma- such as microwaves, or hot neutral beam injection (that would seem to play havoc with the electrostatic ion containment). In the WTW machines the electrons would be reheated/ re accelerated whenever they passed outside the anode grid. In WB6 the appreciation for the inadequacy of electron cusp magnetic confinement it was appreciated, Here escaping electrons were stopped and re accelerated back into the machine due to the magrid anode positive charge, much like the ETW. This is fundamentally different in later EMC2 machines that apparently have a grounded magrid. The electrons are accelerated well outside the magrid and injected through the cusps. This does not allow for reheating of escaping electrons as they loop around to another cusp. There are obviously tradeoffs (think WB5 failure), but I remain curious why they have chosen this path.
All of this boils down to several clever tricks utilized in the Polywell to change the game- Minimizes ion concerns with ExB diffusion, eliminates much of the concerns about grid transparency, and of course the essential Wiffleball consideration. The triple product solution involves the short confinement times of cusp magnetic fields, but improved enough that with the corresponding increased density, and ease (hopefully) of heating, the containment of particles and heat is sufficient for net positive fusion energy output..
Other considerations such as edge stability, cool electrons in the core, ion confluence (central focus) are additional advantages that may allow for advanced fuel fusion, not just baseline D-T fusion.
In your cylindrical design, the plasma is magnatized, there will be significant ExB diffusion issues. There can be edge instabilities, macro instabilities, MHD instabilities (which ever term you prefer). Cusp losses may be less with two ring cusps on either end-and there have been cylindrical solenoid designs with end ring magnets to decrease end losses, but they have not been pursued, presumably because they failed theoretically or experimentally.
One experiment that may be similar to what you are thinking of was done by George Miley. I think it was called the dipole arrangement. A ring electromagnet was placed between two end curved cathodes which injected electrons towards the center. The electrons quickly left due to a negative space charge buildup. A anode was placed on the inner surface of the magnet to limit this. Some electron confinement was achieved, but I think it was far below what was hoped for.
https://www.google.com/search?q=George+ ... Cg#imgdii=_
Dan Tibbets
To error is human... and I'm very human.
Re: New Species of Fusor?
Its very simple. The magnetic field confines ions along the z and y axis. The electric field accelerates and confines ions on the x axis. No need for heaters of any kind. If the magnetic force in this situation is twice as powerful as the electrostatic field, collisions won't reach an energy that will escape the magnetic field. The ions are all on independant corkscrew paths (otherwise Mannhart's machine wouldn't work). Not all collisions worsen the case for their confinement. Sometimes ions will collide toward the center of the z and y axis.
Both the electric field and the magnetic fields are handling the same number of dimensions in this concept ([z,y],[x, energy])
How could the situation be more ideal?
Re: New Species of Fusor?
The magnetic field seems to be the lesser understood of the two fields. Even that article I referenced said they didn't really understand exactly what it was doing to charged particles. Positive is attracted to negative, and like charges repel, this is easy to comprehend. However magnetic fields do weird things to charged particles. But there seems to be some kind of pressure a magnetic field asserts on a particle to slow its movement perpendicular to field lines. Its not the same thing as air pressure, or hydraulic pressure. Though electric fields have a similar sensation to those kinds of pressures. I am thinking the magnetic field is like a damping field to charged particles wanting to traverse its field lines. This does not explain their loopty loop patterns however.
Re: New Species of Fusor?
I would say that the ions will traverse field lines towards the cathode, but how fast? This would be an important consideration in confinement. It maybe that big diameter tubes are more ideal, but you still got to fill that space with a magnetic field. It would be logical that the magnetic field gets stronger, the closer it gets, to the coils. So additional diameter may not be really important.
Re: New Species of Fusor?
It seems that when immediate results are not had, try more powerful magnets. Polywell started with ceramic magnets, and bigger and bigger electromagnets, and then maybe to super conducting cryogenic magnets. Is the magnetic strength inline with the accelerating voltage? What is the equivilant of 50KeV in terms of a magnetic field? 50 KeV is not very hard to reach. So why these exotic magnets? It would be a quench fest if you tried to use that with a HEAT ENGINE.
Re: New Species of Fusor?
Reminds me of a http://en.wikipedia.org/wiki/Penning_trapohiovr wrote:Its very simple. The magnetic field confines ions along the z and y axis. The electric field accelerates and confines ions on the x axis.
The daylight is uncomfortably bright for eyes so long in the dark.
Re: New Species of Fusor?
Interesting. Well the penning trap does use both fields. If you do a google image search on penning trap it looks quite a bit different inside.hanelyp wrote:Reminds me of a http://en.wikipedia.org/wiki/Penning_trapohiovr wrote:Its very simple. The magnetic field confines ions along the z and y axis. The electric field accelerates and confines ions on the x axis.
https://www.google.com/search?q=penning ... B550%3B700
Also it doesn't seem to intentionally accelerate ions and the confinement pattern looks a little different.
There appears to be an electrode on the caps, and an electrode on the cylinder wall, so it does appear to be different. The cylinder electrode in a penning trap traverses most of the machine, where it is hoped that the electrodes in the MagnoFusor are seen by ions as point locations. There will be drift towards the cathode. The question is, would it be too fast?
Re: New Species of Fusor?
There be no virtual electrode, no cloud of electrons, to force your ions to swarm in and out of. This keeps bremstralung losses down hugely, not to mention thermalization. Electrons might have a small mass but it is still considerable. Containing electrons seems to be quite a problem in the Polywell. The virtual cathode concept is one way to make a fairly transparent acceleration medium. However, it is a lot easier sounding than in practice....D Tibbets wrote:Interesting and well presented video, but....
Mentioning vacuum tubes is appropriate as they are all about manipulating plasmas.
The grid transparency issue is a major issue in fusors, but it is not the only issue. A typical glow discharge fusor might top out with a grid transparency of ~ 95%. This means an ion could travel back and forth through the grid about 20 times. Depending on the size of the machine it may need a grid transparency of at least 99.99%, or allow for ~ 10,000 passe sof the typical ion. If the machine is very big, perhaps a kilometer in diameter, the number of passes required is decreased proportionately. So, at least theoretically a typical gridded fusor could reach break even. But only if this grid transparency to ions is the only consideration. You also, have to account for what is happening to the electrons. Also, the effects of non fusing collisions-Coulomb collisions that scatter the charged particles and drive them to Maxwellian energy distributions. The average ion energy/ temperature may be trapped in the grid induced potential well, but up scattered ions will escape,.. And of course the electrons are essentially one pass participants. The Elmore Tuck and Watson version of the fusor reversed things. Electrons are trapped in a potential well from a grid (anode) placed near the edge of the machine. Wishfully, the electrons are electrostatically (potential well ) trapped, and as they converge towards the center they would form a virtual central cathode to then trap ions. The problem again was the grid transparency (to electrons this time. Also, any ions that reached the radius of the anode would then be accelerated (repelled) to the wall. This could not be avoided as the ions thermalized- a certain percentage would always be energetic enough to escape in this manner, as would some of the electrons.
This is starting to move in the direction of the Polywell concept.
The first consideration is to use excess electrons. This produces a deeper potential well, perhaps approaching the strength of the electron injection energy. The electron injection energy can come from two sources. A peripheral anode grid -like in ETW or WB6, or from high energy electron guns outside the grid (in this case the grid (magnet cans) is at ground.
The grid transparency is addressed by not having the electrons pass the grid once trapped, at least that was the plan. This is done by having magnetic shielding. Magnetic shielding is not perfect though. Charged particles penitrate through magnetic fields much like gas diffuses across a room. Except it is done in discreet steps dependent on the particle energy, momentum, magnetic field strength and density driven scattering collisions. Through a random walk process the charged particle moves through the B field one gyroradius distance at a time either deeper or shallower with each collision. As a result the charged particles migrate from greater to lower densities till the magnet surface is hit and the KE of the particle is lost - generally as heat. Ions with their greater weight and thus momentum have gyro radii >60 times greater than electrons, so at the same B field strength, and particle energy the ions diffuse through the B field ~ 60 times faster than electrons. This is one reason why tokamaks have to be so large, so that the reaction volume can be maintained long enough against this loss mechanism. A much smaller machine is possible if only the electrons are contained in this manner (magnetically contained). At least until inertial confinement regimes are entered.
The ions have to be decoupled from the magnetic field. Whether the electromagnetic field or an internal plasma generated magnetic field , the ion diffusion- ExB diffusion, through the B field is too fast, energy is lost too fast in smaller machines. Some would say the magnetic insulation is inadequate in small machines. In the Polywell though, the ions are not magnetically contained (for the majority of the ions for the majority of their lifetime). The excess electrons produce a potential well that not only confines the ions at radii less than the influence of the electromagnet B fields, but also serves to accelerate them to useful high energy towards the center of the machine. The picture is not of ions flying (or spiraling in a magnatized plasma) back and forth along a cyclinder till they fuse, but ions flying back and forth towards a spherically defined center. This may allow for greatly increased densities in the core which is mostly good , but is not required. The magnetic field does come into play for ions that are up scattered and travel to radii great enough to reach the electron confining magnetic field. These ions are mostly turned back towards the center with out as great of an ExB penalty, because they are a minority of the ions and their energy and thus gyroradius are smaller.
The electrons are contained magnetically, their energy is initially from the ETW like grid, but subsequently their energy is only due to their inertia. They will cool through bremsstruhlung radiation, etc. and there is no additional energy input. They stay hot long enough to get the job done- hopefully, otherwise other means is needed to reheat the electrons/ plasma- such as microwaves, or hot neutral beam injection (that would seem to play havoc with the electrostatic ion containment). In the WTW machines the electrons would be reheated/ re accelerated whenever they passed outside the anode grid. In WB6 the appreciation for the inadequacy of electron cusp magnetic confinement it was appreciated, Here escaping electrons were stopped and re accelerated back into the machine due to the magrid anode positive charge, much like the ETW. This is fundamentally different in later EMC2 machines that apparently have a grounded magrid. The electrons are accelerated well outside the magrid and injected through the cusps. This does not allow for reheating of escaping electrons as they loop around to another cusp. There are obviously tradeoffs (think WB5 failure), but I remain curious why they have chosen this path.
All of this boils down to several clever tricks utilized in the Polywell to change the game- Minimizes ion concerns with ExB diffusion, eliminates much of the concerns about grid transparency, and of course the essential Wiffleball consideration. The triple product solution involves the short confinement times of cusp magnetic fields, but improved enough that with the corresponding increased density, and ease (hopefully) of heating, the containment of particles and heat is sufficient for net positive fusion energy output..
Other considerations such as edge stability, cool electrons in the core, ion confluence (central focus) are additional advantages that may allow for advanced fuel fusion, not just baseline D-T fusion.
In your cylindrical design, the plasma is magnatized, there will be significant ExB diffusion issues. There can be edge instabilities, macro instabilities, MHD instabilities (which ever term you prefer). Cusp losses may be less with two ring cusps on either end-and there have been cylindrical solenoid designs with end ring magnets to decrease end losses, but they have not been pursued, presumably because they failed theoretically or experimentally.
One experiment that may be similar to what you are thinking of was done by George Miley. I think it was called the dipole arrangement. A ring electromagnet was placed between two end curved cathodes which injected electrons towards the center. The electrons quickly left due to a negative space charge buildup. A anode was placed on the inner surface of the magnet to limit this. Some electron confinement was achieved, but I think it was far below what was hoped for.
https://www.google.com/search?q=George+ ... Cg#imgdii=_
Dan Tibbets
From a desired result point of view, electrons are a means to an end, but for the most part, they are in the way, and are a pain in the ass to contain.
The cathode and anode in the MagnoFusor appear to be point sources to the ions at a distance, closer they get to the center, the more they are going to want to drift to the edges. Starting from the exact center will help matters a great deal.
There should be no cusp losses, because there are no cusps. It is simply impossible for the ions to get out of the machine via the ends. What is more likely to happen is that the ions get pulled into the cathode anyway as a long drift pattern. Sudden perpendicular motion caused by collisions (failing to produce a fused ion), gets dampened. The product of this absorption is what exactly? I would guess that it would be a radiated photon. The stronger the field, the less perpendicular motion, and the higher energy photon released. If these result in an x-ray, then I would assume the perpendicular change in distance would be very very small. But if it is a radio wave, then it could be kind of sloppy. It could be that these glancing angle collisions result in ions that loose most of their forward momentum and they no longer have enough energy to fuse. Its up to the ions that are sill moving fast to hit these dead ions for fusion to happen to them before they get sucked into the cathode.
The drift is slow, collision motion peperindicular is a lot more sudden, but these events are a lot rarer.
Mannharts machine used very strong rare earth magnets and I think this is primarily to shrink the gyroradius of the emitted electrons (I like to call them thermions to be old fashioned) more than to keep them happily going along field lines.
Re: New Species of Fusor?
I got my wire windings going the wrong way, 'sposed to be wrapped like a solenoid
http://www.gonefcon.com/flymanaged/magnets1.htm
I'll update the video this weekend maybe.
http://www.gonefcon.com/flymanaged/magnets1.htm
I'll update the video this weekend maybe.
Re: New Species of Fusor?
Some mechanism to accelerate ions is fine, but you cannot ignore electrons. Any electric field that acts on positively charged ions does the opposite on electrons. As has been mentioned , to achieve useful fusion levels (more than micro watts of power), you are going to have to have large concentrations of ions. 10^18-10^22 per cubic meter is reasonable, depending on the device. A Penning trap confining some exotic ion may have a density of 10^3 particles per cubic meter. There is a huge difference. In the polywell you may have 1,000,001 electrons for each 1,000,000 ions. You cannot have a large concentration of a positive charge without having a nearly equal concentration of negative charges (1 part per million is doable, but not much more). Look up Brillion limit. This is a limit for charge seperation. Only low density pure ion or electron plasmas can be maintained without the required confinement energy going through the roof. Look up Coulomb explosions.
ExB diffusion through magnetic fields is a very big consideration. This is dependant on the gyroradius of the charged particles of interest, the strength of the B field, and the density squared. The squared relationship means that for every 10 fold increase in density, the Coulomb collisions and thus ExB diffusion rates increase ~ 100 fold. ExB diffusion is a minor concern at a density of perhaps 10^13 particles / M^3, but it increases ~ one trillion fold for densities of ~ 10^19 particles/ M^3 (mixture of almost equal portions of electrons and positively charge ions is obtainable). This is a density you might see in a tokamak. Because of this painful increase in ExB losses as you approach useful densities, thus Bussard's quote (I don't recall who he is quoting) that magnetic fields are no dam,,, good for containing ions. There are work arounds, one is to go to large machine sizes (more distance for the ion to diffuse through before loss), another is to use non neutral plasma with a ~ 1 ppm excess of electrons. The electrons confine the ions electrostatically, so the magnetic confinement limits only applies to the electrons with their gyroradii of <1/60th that of ions. As such the ExB loss rates are ~ 1/ 60th less rapid. This does not work for a neutral plasma for reasons that I will not go into.
You cannot magnetically or electrostatically contain a pure ion plasma of useful density without gargantuan magnnetic and/ or electrostatic fields. For perspective an electrostatic field that contains an almost neutral plasma may require 100,000 Volts. To illistrate, the same density of plasma that was dominantly one species of charge might require 10,000,000,000 Volts. The magnetic confinement picture is more fuzzy for me, but suffice it to say, the magnetic insulation would break down and the plasma would penetrate the field quickly, unless the B field was also gargantuan in strength or size.
Bremsstruhlung radiation is primariy a funtion of electron speed , density and the density of the relatively stationary ions that they interact locally with. Also the ion Z (or positive charge of the ion + 1 for D, +5 for B11). Since the density of ions and electrons must average about the same, you cannot have a pure ion plasma of useful density, the electrons must be present and thus bremsstruhlung is always present. The Polywell with its near spherical symmetry, and potential well results in some interesting electron and ion dynamics that may significantly change the bremsstruhlung picture. This is involved with agruments by Rider and others about the viability of small dense fusion machines that burn any fuel other than D-T, or at best D-D.
The Polywell is a tiny nudge in the direction of a non neutral, equal positive and negative charge balance, yet this tiny nudge creates the electrostatic field from excess energetic electrons that establishes a potential well that can confine ions at reasonable temperatures and density. Even then, A. Carlson and others have pointed out that the the potential necessary to contain almost all of the thermalized ions could grow to many millions of volts. Some cheating of the ion thermalization , and perhaps electron thermalization process is also required for the machine to work. There are mechinisms that probably do this in the Polywell, but that is another discussion.
Noting that ions can be knocked in either directions is certainly true. The important point is that this process is random, sometimes called a random walk process. Just as with gas diffusion, this leads to a movement of particles from a high density region to a lower density region. with the magnetically confined plasma, the process occurs in discreet steps based on the gyroradius for each collision. As such the charged particles move from higher density (the core or desired confinement volume) to the magnet wall, there the ions grounds and becomes neutral, effectively removing it from the system. The diffusion continues until an equilibrium is reached. As the concentration is essentially zero immediately adjacent to the magnet surface, the diffusion will continue on a logarithmic scale till only one charged particle is left. A mathematical description might look something like :
Rate of loss = Density squared *gyroradius / distance to reach magnet surface.
Not considered in this discussion is MHD instabilities, edge instabilities, macro instabilities, turbulance (all different labels for the same thing) which is another monster that has to be placated one way or another.
Dan Tibbets
ExB diffusion through magnetic fields is a very big consideration. This is dependant on the gyroradius of the charged particles of interest, the strength of the B field, and the density squared. The squared relationship means that for every 10 fold increase in density, the Coulomb collisions and thus ExB diffusion rates increase ~ 100 fold. ExB diffusion is a minor concern at a density of perhaps 10^13 particles / M^3, but it increases ~ one trillion fold for densities of ~ 10^19 particles/ M^3 (mixture of almost equal portions of electrons and positively charge ions is obtainable). This is a density you might see in a tokamak. Because of this painful increase in ExB losses as you approach useful densities, thus Bussard's quote (I don't recall who he is quoting) that magnetic fields are no dam,,, good for containing ions. There are work arounds, one is to go to large machine sizes (more distance for the ion to diffuse through before loss), another is to use non neutral plasma with a ~ 1 ppm excess of electrons. The electrons confine the ions electrostatically, so the magnetic confinement limits only applies to the electrons with their gyroradii of <1/60th that of ions. As such the ExB loss rates are ~ 1/ 60th less rapid. This does not work for a neutral plasma for reasons that I will not go into.
You cannot magnetically or electrostatically contain a pure ion plasma of useful density without gargantuan magnnetic and/ or electrostatic fields. For perspective an electrostatic field that contains an almost neutral plasma may require 100,000 Volts. To illistrate, the same density of plasma that was dominantly one species of charge might require 10,000,000,000 Volts. The magnetic confinement picture is more fuzzy for me, but suffice it to say, the magnetic insulation would break down and the plasma would penetrate the field quickly, unless the B field was also gargantuan in strength or size.
Bremsstruhlung radiation is primariy a funtion of electron speed , density and the density of the relatively stationary ions that they interact locally with. Also the ion Z (or positive charge of the ion + 1 for D, +5 for B11). Since the density of ions and electrons must average about the same, you cannot have a pure ion plasma of useful density, the electrons must be present and thus bremsstruhlung is always present. The Polywell with its near spherical symmetry, and potential well results in some interesting electron and ion dynamics that may significantly change the bremsstruhlung picture. This is involved with agruments by Rider and others about the viability of small dense fusion machines that burn any fuel other than D-T, or at best D-D.
The Polywell is a tiny nudge in the direction of a non neutral, equal positive and negative charge balance, yet this tiny nudge creates the electrostatic field from excess energetic electrons that establishes a potential well that can confine ions at reasonable temperatures and density. Even then, A. Carlson and others have pointed out that the the potential necessary to contain almost all of the thermalized ions could grow to many millions of volts. Some cheating of the ion thermalization , and perhaps electron thermalization process is also required for the machine to work. There are mechinisms that probably do this in the Polywell, but that is another discussion.
Noting that ions can be knocked in either directions is certainly true. The important point is that this process is random, sometimes called a random walk process. Just as with gas diffusion, this leads to a movement of particles from a high density region to a lower density region. with the magnetically confined plasma, the process occurs in discreet steps based on the gyroradius for each collision. As such the charged particles move from higher density (the core or desired confinement volume) to the magnet wall, there the ions grounds and becomes neutral, effectively removing it from the system. The diffusion continues until an equilibrium is reached. As the concentration is essentially zero immediately adjacent to the magnet surface, the diffusion will continue on a logarithmic scale till only one charged particle is left. A mathematical description might look something like :
Rate of loss = Density squared *gyroradius / distance to reach magnet surface.
Not considered in this discussion is MHD instabilities, edge instabilities, macro instabilities, turbulance (all different labels for the same thing) which is another monster that has to be placated one way or another.
Dan Tibbets
To error is human... and I'm very human.
Re: New Species of Fusor?
Are there supposed to be electrons in the fusor?D Tibbets wrote:Some mechanism to accelerate ions is fine, but you cannot ignore electrons. Any electric field that acts on positively charged ions does the opposite on electrons.
Re: New Species of Fusor?
Yes, Yes, Yes!
A plasma is collection of charged particles. The total negative charged particles (electrons primarily), and positive charged particles (ions) must be very close together in numbers. Otherwise the plasma essentially blows itself apart. Coulomb pressure quickly builds to impressive values with only small differences in charge balance. Even when you want a non neutral plasma, such as in the Polywell, you can only push the imbalance a small amount. In the Polywell, the difference is only about one part per million. That means that if you had 1.000000 *10^20 ions in a given volume, you would have 1.000001 *10^20 electrons.
The only time you could have a large proportional excess of one charge is with extremely dilute plasmas. You might manage a non neutral plasma of 1 million ions and one electron. That would have about the same coulomb pressure as the above example. The problem is that if you was generating 100 MW of fusion at the first density, at the second density you would only be generating ~ 100 MW / square of 10^13 or ~ 1 billionth of 1 billionth of 1 billionth the amount of fusion. That would not be very useful. You would be lucky to achieve 1 nanowatt of fusion power with an ion density of ~ 100,000,000,000 and electron density of 99,999,000,000.
This means that if you contain the ions perfectly with zero containment of electrons , then losses would be very close to 50% of the input energy. In the Polywell the electron losses dominate (much above 50%) compared to ion losses because the ion containment, while not perfect, is much better than the electron containment. The electrostatic containment of ions is better than the magnetic containment of electrons. This need not always be the case though. In WB5 the efforts to better contain the electrons by adding electron electrostatic containment mechanisms (cusp plugging with cold electrons) worked for the electrons, but worsened the electrostatic containment of the ions. No ground was gained. It is a delicate balance between the opposing effects that contain electrons and ions. Basically, electrons are contained better than ions with magnetic fields. By having excess electrons (1 part per million) the ions are instead contained by electrostatic forces (virtual cathode). It is a Goldilocks situation. The different considerations are balanced to reach the best and hopefully adequate solution.
Some appreciation for the magnitude of the Coulomb force that can be achieved with small charge separation may be gained from this link:
http://hyperphysics.phy-astr.gsu.edu/hb ... lefor.html
Dan Tibbets
You might ask why cusp magnetic containment for electrons is used instead of a closed magnetic system like a tokamak. Well, you need to also consider density, temperature control, ExB diffusion and edge stability issues to achieve the required triple product. The questions are different, but the final answer is the same.
Dan Tibbets
A plasma is collection of charged particles. The total negative charged particles (electrons primarily), and positive charged particles (ions) must be very close together in numbers. Otherwise the plasma essentially blows itself apart. Coulomb pressure quickly builds to impressive values with only small differences in charge balance. Even when you want a non neutral plasma, such as in the Polywell, you can only push the imbalance a small amount. In the Polywell, the difference is only about one part per million. That means that if you had 1.000000 *10^20 ions in a given volume, you would have 1.000001 *10^20 electrons.
The only time you could have a large proportional excess of one charge is with extremely dilute plasmas. You might manage a non neutral plasma of 1 million ions and one electron. That would have about the same coulomb pressure as the above example. The problem is that if you was generating 100 MW of fusion at the first density, at the second density you would only be generating ~ 100 MW / square of 10^13 or ~ 1 billionth of 1 billionth of 1 billionth the amount of fusion. That would not be very useful. You would be lucky to achieve 1 nanowatt of fusion power with an ion density of ~ 100,000,000,000 and electron density of 99,999,000,000.
This means that if you contain the ions perfectly with zero containment of electrons , then losses would be very close to 50% of the input energy. In the Polywell the electron losses dominate (much above 50%) compared to ion losses because the ion containment, while not perfect, is much better than the electron containment. The electrostatic containment of ions is better than the magnetic containment of electrons. This need not always be the case though. In WB5 the efforts to better contain the electrons by adding electron electrostatic containment mechanisms (cusp plugging with cold electrons) worked for the electrons, but worsened the electrostatic containment of the ions. No ground was gained. It is a delicate balance between the opposing effects that contain electrons and ions. Basically, electrons are contained better than ions with magnetic fields. By having excess electrons (1 part per million) the ions are instead contained by electrostatic forces (virtual cathode). It is a Goldilocks situation. The different considerations are balanced to reach the best and hopefully adequate solution.
Some appreciation for the magnitude of the Coulomb force that can be achieved with small charge separation may be gained from this link:
http://hyperphysics.phy-astr.gsu.edu/hb ... lefor.html
Dan Tibbets
You might ask why cusp magnetic containment for electrons is used instead of a closed magnetic system like a tokamak. Well, you need to also consider density, temperature control, ExB diffusion and edge stability issues to achieve the required triple product. The questions are different, but the final answer is the same.
Dan Tibbets
To error is human... and I'm very human.
Re: New Species of Fusor?
Wouldn't the electrons be blow away from the cathode, and sucked into the anode in the fusor? Forget about polywell we are talking about the classic fusor.
Re: New Species of Fusor?
http://tommccarthyprojects.com/how-the-fusor-works/ohiovr wrote:Wouldn't the electrons be blow away from the cathode, and sucked into the anode in the fusor? Forget about polywell we are talking about the classic fusor.