Robthebob wrote:experimental science without theory is like not doing science...
Experimentation (observation) without theory is called "natural history" and is the bedrock upon which all science is built. You have to have phenomena to explain in theory for science to start.
By the way, as stated before, there is theory; a bit of which is in the Valencia paper; which states that loss will be a function of the total area of the cusps. That area, in the WB6+ series is thought to be controlled by the length of the line-like cusps. Reducing the length reduces the area which reduces the losses... in theory. It was the last bit of theory Dr. B. wanted to prove out before going big. ... Ok, one of the two last bits, the other being the theory that improving sphericity improves output. That is what the higher order unit was for.
The smaller magnets being closer to the center and thus compensating for the fewer amp turns available inside the resultant magnet cans may be true, but that assumes you are going to a smaller radius machine. In the same radius, more magnets means smaller available wire turns. Amps would have to be increased accordingly. Of course if you have a mathematical formula describing the B field, confinement efficiency, recirculation efficiency, machine radius, etc, then any chosen, size, B field, spacing, number of magnets, etc could be used and fitted into the picture.
The problem is the experimental errors/ uncertainty, and engineering difficulty. Smaller sizes complicates magnet construction and possible cooling. Also, out gassing contamination becomes more prevalent. More magnets means more complexity and cost. WB6 at 30 cm diameter and 0.1 T and 12 KV input with dozens of Amps was barely able to produce measurable fusion. nything less will not give useful results (some improvements may have been achieved with WB7 and WB7.1 so mildly smaller sizes might be useful but this is speculation).If you are measuring plasma density and Beta conditions and not fusion, then smaller machines may give useful information.
If you are interested in measuring Wiffleball effects, then by definition you must be operating at near Beta= 1 conditions. Lower Beta conditions may give some trend information and allow for extrapolation.
As for electron leakage outward and entrance, my impression is that they are equivalent, at least in the center of the cusp (Magrid radius). The cusp magnetic field lines are nearly parallel on both sides of the cusp, so it is a two way street. The lack of mirroring of the escaping electrons as they approach the mid line is very nearly equal to the lack of mirroring of entering electrons with the same but reversed vectors. This is, I think, evident from the high efficiency (90%)* claimed for recirculation. The issue may be to get the E-guns and associated focusing mechanism to match this narrow cone for virgin electron entry through the cusps at similar efficiency.
WB6 results were limited, though several weeks were spent with Beta tests and other input and confinement issues before the few tests with deuterium gas and neutron detection efforts. Those results were not released except for the implication that Wiffleball conditions were met. I suspect WB7, and especially WB8 tests have considerably expanded on the data set, especially with regards to non symmetrical variations in the various parameters.
Keeping things 'symetrical' in the since of keeping the same proportions and geometry is what I think Bussard used as a simplification as I think the relationships would be equivalent- electron entry vs escape and recirculation numbers would change but be in proportion. Thus a 300 cm diameter machine with 10 T fields. This maintains equivalent volume/ size and B field relationships so the simple scaling laws can be used without increased complications. Not only is the predicted fusion output scaling consistent, but so is the input loss scaling, and I think, the electron input efficiency given a well defined E-gun performance parameter.
Fusion scales as r^3 * B^4, while losses scale as r^2 * B^0.25. Bo other constants or variables need to be calculated. The question of electron input efficiency at various sizes and B field conditions has been raised, but as above, Ispeculate they are consistant with relative size and B field changes, assuming the same geometry is used. The effect of various geometries on the relationship may need experimentation to pin down.
The square coils versus the round coils is interesting. I originally thought they would be arranged with the coils corners meeting at the magrid corners (still a ~ cube 3D shape). Later I thought the corners of the square coils would meet at the mid line of the next coil. Both arrangements may have advantages. It depends on the relative corner/ line cusp areas and the effects of ExB drift, which I am assuming becomes increasingly important where the magnets are closest together. A short segment where this close approach is reached would presumably reduce ExB losses while a cusp that had long sections in close proximity would have increased ExB losses but with less cusp losses. Again there is a balance to be considered.
A side issue of the square coils is that they are not true squares. They must have rounded edges both because of engineering issues, and to minimize arcing which increases rapidly as the radius of curvature of a surface decreases (becomes more sharp). This was mentioned in the WB4 coils with their square cross section (separate from the issue of keeping the cans conformal to the shape of the magnet windings). The WB6 and WB4 cans represents the two extremes of this engineering and functional consideration. It may be that an intermediate shape (an oval cross section or square cross section with broadly rounded corners may be the best compromise)
Also, keep in mind that the WB4 square cross section coils occupied ~ 25% of the diameter of the machine, while WB6 round coils occupied ~17% of the diameter. This effects the internal can volume that is available for cooling plumbing and windings. The inner edge of the WB4 cans also approached the center of the face point cusps more and I speculate that this would allow for stronger B fields as the face point cusp is approached, with resultant smaller cusp loss areas. In WB6 I think the point cusps were more leaky than the corner cusps, so this might represent a significant gain. The surface area exposed to ExB drift, and x-ray radiation and neutron bombardment would also increase. Experiments or modeling to find the best compromise might be valuable.
* Note that recirculation is dependant on two factors, the lack of mirroring of the recirculating electrons (from either direction). Also, the escaping electron must not have been up scattered to a speed greater that that of the centrally acting potential (positive charge on the magrid). If the electron is too fast upon escape it will be slowed by the magrid potential, but not reversed. This is a good thing as up scattered electrons need to be eliminated and this recovers a significant portion of these up scattered electrons potential energy- the Magrid is acting as a direct energy converter in this regard.
The point is that I do not know how much of the quoted recirculation efficiency is due to this beneficial high energy electron loss mechanism. If a portion of the electrons were not up scattered the recirculation efficiency may be closer to 100%.
I was just mentioning the fact that without theory, the experiment can only really determine if square faces are better than circular faces and if higher order is better than lower order shapes. (really if that, because size and b field scaling might not function as we expect, and since very little work has been done on geometry... it's not unfair to assume the worst) It cant determine how much better.
It's impossible to predict the benefit (if changing shapes will result in benefit) of changing shapes will provide as you build bigger machines.
D Tibbets wrote:The square coils versus the round coils is interesting. I originally thought they would be arranged with the coils corners meeting at the magrid corners (still a ~ cube 3D shape). Later I thought the corners of the square coils would meet at the mid line of the next coil. Both arrangements may have advantages. Neither of these configurations is correct. This is a cuboctahedron. The square planform magnets would encompass the red faces. The yellow faces would be the virtual opposite pole magnets.
It depends on the relative corner/ line cusp areas and the effects of ExB drift, which I am assuming becomes increasingly important where the magnets are closest together. A short segment where this close approach is reached would presumably reduce ExB losses while a cusp that had long sections in close proximity would have increased ExB losses but with less cusp losses. Again there is a balance to be considered. Given that neither was correct, there is no balance to consider.
A side issue of the square coils is that they are not true squares. They must have rounded edges both because of engineering issues, and to minimize arcing which increases rapidly as the radius of curvature of a surface decreases (becomes more sharp). This was mentioned in the WB4 coils with their square cross section (separate from the issue of keeping the cans conformal to the shape of the magnet windings). The proposed square planform magnets have a round minor cross section. WB4 does not apply. The rounded corners are good in that they allow a gap between magnets which allow for recirculation.
Robthebob wrote:I was just mentioning the fact that without theory, the experiment can only really determine if square faces are better than circular faces and if higher order is better than lower order shapes. (really if that, because size and b field scaling might not function as we expect, and since very little work has been done on geometry... it's not unfair to assume the worst) It cant determine how much better.
It's impossible to predict the benefit (if changing shapes will result in benefit) of changing shapes will provide as you build bigger machines.
I am confused as to your point. There is theory. And determining whether the square planform is better and by how much should tend to prove or disprove the theory. I fail to see the issue. I guess, the higher order polyhedron would need to have some discussion as to how properly to scale the windings, but round vs square seems simple enough.
I see, well that's good that there is detailed theory.
okay, I think I may have close to enough to proceed, but not now yet, because I have to do other unrelated things.
Anyways, so I was thinking, for every configuration, size, b field strength, and electron gun current there is a max capacity where the loss rate is equal to the retain rate (I'm assuming this equilibrium occurs very quickly) . There is however a true max capacity of electrons a machine can confine which depends only on config, size and b field strength; as you increase the electron gun current, you get close to the true max cap, and the machine can no longer hold anymore, so there's a minimum electron gun current needed to achieve this. Certain setups ranges and above (config, size and b field) can achieve the wiffleball effect.
If I remember correctly, wb effect depends only config, size and b field and if the electron gun current is high enough; dont need ions in the system?
Only if you plan to populate (e-) by neutral stripping.
Wiffleball is about (e-) confinement. Make the most efficient big (-) possible in the center for power in. This is proven according to EMC2.
This is the point Joe misses time and again. Magnetically confine (e-), Electrostatically attract ions.
The development of atomic power, though it could confer unimaginable blessings on mankind, is something that is dreaded by the owners of coal mines and oil wells. (Hazlitt)
What I want to do is to look up C. . . . I call him the Forgotten Man. (Sumner)
KitemanSA wrote:As far as I know, you do NOT need ions to make a wiffleball. Corrections anyone?
I admit that I am uncertain about the dynamics of Wiffleball formation. There is a Bussard paper describing the generation of a Wiffleball through two possible approaches, but the details are minimal.
As Coulomb forces severely limits the concentration of a plasma with only one charged species - positive or negative alone, it seems unlikely that a pure electron Wiffleball could form. From an understanding that only about 1 part per million deviation in net neutrality is possibly without going to ridiculous voltages, only a very low density of pure electrons could be confined. Nearly balancing amounts of positively charged ions must be provided, either through ionization of the neutral gas by hot electrons, or by ion guns. The dynamic addition of both species needs to maintain a potential well (slight electron excess) while allowing for a progressive increase in internal pressure that ends at the balance point of Beta = 1 and maximal obtainable Wiffleball effect. My impression is that this is not difficult (provided your power supply can overcome the early cusp confinement electron losses before the higher electron magnetic confinement efficiency of the Wiffleball can manifest) but the details may be tricky.
Note that the ion magnetic confinement benefits from the Wiffleball effect also. The confinement time may increase from the cusp confinement of perhaps 60 passes to up to several thousand passes (taken from the patent application). But with ions, the collisional ExB losses build up rapidly as the density increases, and I suspect this would become a dominate loss mechanism before full Wifflleball pressures/ densities are reached. This would limit the obtainable Beta to some intermediate level in a neutral plasma irregardless of the output of the power supply. Thus the use of 1 ppm excess electrons to create a potential well/ electrostatic ion containment that mostly negates ion ExB drift concerns. And it also has the added benefit of decreasing the ion cusp losses relative to electron cusp losses. This, I think, counters one of the criticisms of A.Carlson (from several years ago) concerning equal ion and electron cusp flows Which would apply if the plasma was neutral.
D Tibbets wrote:The square coils versus the round coils is interesting. I originally thought they would be arranged with the coils corners meeting at the magrid corners (still a ~ cube 3D shape). Later I thought the corners of the square coils would meet at the mid line of the next coil. Both arrangements may have advantages. Neither of these configurations is correct. This is a cuboctahedron. The square planform magnets would encompass the red faces. The yellow faces would be the virtual opposite pole magnets.
So, still still 6 real magnets, as in a cube, but the corners are truncated clear to the center of the square magnet border. The truncated corner virtual magnets are thus much larger and the square real magnets effectively rotated 45 degrees. I'm not sure my description makes sense to you, but I do now appreciate what you are saying about the configuration.
Note that the ion magnetic confinement benefits from the Wiffleball effect also. The confinement time may increase from the cusp confinement of perhaps 60 passes to up to several thousand passes (taken from the patent application).
Dan, what are you saying here? Please clarify.
As I understand, Ions are attracted, not "confined". If they do not fuse, and escape via the vacuum maintenance system, then there is the opportunity to recover and recycle them as fuel.
Power In is a function of driving (e-) and the well. Net Power is a result of sufficient fusion events to generate more power than used to make the well. It seems that you are losing sight of the basic concept to some degree.
Magnetically Confine Electrons, and thus confined Electrons create a large negative potential in the center of the magnetic fields.
Limit Electron Loss Rate via Wiffleball Dynamic (Magnetic Cusp Pinching).
Positively Charged Fuel Ions are Electrostatically Attracted to the Big Negative.
Ions seeking center collide with each other creating head to head and offset events.
The development of atomic power, though it could confer unimaginable blessings on mankind, is something that is dreaded by the owners of coal mines and oil wells. (Hazlitt)
What I want to do is to look up C. . . . I call him the Forgotten Man. (Sumner)
D Tibbets wrote:So, still still 6 real magnets, as in a cube, but the corners are truncated clear to the center of the square magnet border. The truncated corner virtual magnets are thus much larger and the square real magnets effectively rotated 45 degrees. I'm not sure my description makes sense to you, but I do now appreciate what you are saying about the configuration.
This is Dr. B's basic patent configuration, except that it is made with realistic magnets. I have trouble imagining how there could be a confusion.
Ladajo,
In theory you are correct. In practice, the magnets will help with confinement a bit too, especially with those up scattered ions that would otherwise have just barely enough energy to escape the well. They add to the plasma pressure a bit.
D Tibbets wrote:So, still still 6 real magnets, as in a cube, but the corners are truncated clear to the center of the square magnet border.
A cube that has been "truncated clear to the center" has been "rectified" and is called a "cuboctahedron". That is what I have been calling it all along. If you did not understand the term you could have looked it up in Wikipedia.
KitemanSA wrote:Ladajo,
In theory you are correct. In practice, the magnets will help with confinement a bit too, especially with those up scattered ions that would otherwise have just barely enough energy to escape the well. They add to the plasma pressure a bit.
I guess I was getting at the idea that it is not a primary mechanism or concern. Just something to be aware of.
The development of atomic power, though it could confer unimaginable blessings on mankind, is something that is dreaded by the owners of coal mines and oil wells. (Hazlitt)
What I want to do is to look up C. . . . I call him the Forgotten Man. (Sumner)
