I remember a comment by Tom Ligon with regard to WB5 saying "What was good for the electrons was bad for the ions and vica versa" I was wondering whether anyone ever tried to shield the repeller plates with earthed plates with slits at the cusps at a distance in front of the repeller plates that was large in comparison to the thickness of the repeller plates. These earthed plates would then respond to the nearby negative voltage of the repeller plates by accumulating positive charge which in turn would screen out the field from the repeller plates that could attract the ions. The electrons emerging from the cusps would in turn be insulated from these earthed plates by the magnetic field and would stream through the slits instead. Once inside the slits they would be exposed to the unscreened negative potential of the repeller plates and would slow down and turn back.
Has this been tried on WB5 already?
Earthed Plate screening for WB5 repellors
Let's see if I got this right: instead of just having straight-up repellers, you'd add another grid or something between them and the core to produce a field to repell the ions. So the potential would be more positive at this screen/grid, and it would turn the ions back, while the electrons would pass through and then be reflected by the larger negative potential created by the repeller plates beyond the ion screen.
I think the problem with that idea is that your ion screen will attract electrons and soak them up. But it does have the advantage that it would remove electron recirculation, so the confinement coils could go outside the vacuum box. That would certainly simplify construction.
I think the problem with that idea is that your ion screen will attract electrons and soak them up. But it does have the advantage that it would remove electron recirculation, so the confinement coils could go outside the vacuum box. That would certainly simplify construction.
The potential would be earth at the screen/grid and strongly negative at the electron source/repeller plates. However if something at earth potential is placed near something at negative potential, positive charges are drawn up into it.
If the grid was placed a number of electron larmor orbits away from the cusps then the degree to which it would suck up electrons would be limited by cross-field transport. Since an electron larmour radius is quite small it could still be placed near enough to the cusps to screen out the negative charge of the repeller plates.
In addition to this you could place electron scrapers closer to the repeller plates in flux space at a negative potential that is slightly less negative than the repeller plates or the electrons, that way you could scrape off the electrons while they were at low kinetic energy before they hit the earthed grid. This would not reduce particle losses but would, on the otherhand greatly reduce energy losses.
If the grid was placed a number of electron larmor orbits away from the cusps then the degree to which it would suck up electrons would be limited by cross-field transport. Since an electron larmour radius is quite small it could still be placed near enough to the cusps to screen out the negative charge of the repeller plates.
In addition to this you could place electron scrapers closer to the repeller plates in flux space at a negative potential that is slightly less negative than the repeller plates or the electrons, that way you could scrape off the electrons while they were at low kinetic energy before they hit the earthed grid. This would not reduce particle losses but would, on the otherhand greatly reduce energy losses.
Not getting this thread.
None of the WB-X machines have been closed-box designs.
Adding a negatively charged plate still wouldn't change the fact that magnetic lines of force would penetrate it. It might slow down those electrons, but it would still fail to eliminate electron losses at the cusps. In Bussard's Results and Final Conclusions report, he wrote
None of the WB-X machines have been closed-box designs.
Adding a negatively charged plate still wouldn't change the fact that magnetic lines of force would penetrate it. It might slow down those electrons, but it would still fail to eliminate electron losses at the cusps. In Bussard's Results and Final Conclusions report, he wrote
So I'm not sure why trying to fix a design Bussard abandoned as impractical has any interest here.No closed box machine can ever yield net fusion power; open recirculating MG machines and systems are required. This is an immutable result of the determination of losses of electrons in experiments, that show that losses to surfaces that are NOT magnetically shielded must be kept to less than 1E-5 or so of the cusp axis flow of electrons in the WB effect at beta = one. This is impossible for two reasons: (a) it is not practically possible to cover all but 1E-5 of the entire surface of a box containing the interior plasma, with magnetic oils that protect all of this surface, and (b) even if this were possible, it is not possible to protect against losses directly along the cusp axes to the end plates that bound each cusp. These intrinsic losses are inherent in the magnetic topology of a closed box system and forever prevent this from operating at small losses.
If you look at Tom Ligon's comments on the PXL-1 in this thread you can see that they worked with a closed box machine that had some success. I can't remember where, but I'm fairly certain it was suggested that a negative charged element be placed in the cusps to repell electrons and prevent cusp loses. Using this method, the cusp losses in the closed box machine could be eliminated without using recirculation. Thus the closed box could become viable.
This suggestion was ruled out because these negative elements would distort the potential well and pull out ions and neutralize them. Jmc's idea is simple enough: just put a barrier in there to keep the ions away from the repeller. This is a basic principle of vacuum tubes, but with ions rather than electrons.
This suggestion was ruled out because these negative elements would distort the potential well and pull out ions and neutralize them. Jmc's idea is simple enough: just put a barrier in there to keep the ions away from the repeller. This is a basic principle of vacuum tubes, but with ions rather than electrons.
Vacuum tubes are designed on the basis that electron losses be kept under 10%. We have to do 4 orders of magnitude better than that.Solo wrote:If you look at Tom Ligon's comments on the PXL-1 in this thread you can see that they worked with a closed box machine that had some success. I can't remember where, but I'm fairly certain it was suggested that a negative charged element be placed in the cusps to repell electrons and prevent cusp loses. Using this method, the cusp losses in the closed box machine could be eliminated without using recirculation. Thus the closed box could become viable.
This suggestion was ruled out because these negative elements would distort the potential well and pull out ions and neutralize them. Jmc's idea is simple enough: just put a barrier in there to keep the ions away from the repeller. This is a basic principle of vacuum tubes, but with ions rather than electrons.
The vacuum tube stuff can only be a guide.
In any case cusp losses are not a problem they just increase recirculating power. They can also be mitigated by increasing the B field.
Engineering is the art of making what you want from what you can get at a profit.
@scareduck
Even open box machines have fieldines that touch the vacuum vessel wall, but because they are earthed and outside the magrid by the time the electrons hit them they have praticall zero kinetic energy.
WB5 was a closed box machine. Look at the photograph, the magnetic coils are on the outside. In fact, WB5 was the second last of the WB-X machines built by Bussard. So his decision that a closed box machine can ever yield nuclear fusion was a relatively recent one, and his enthusiasm for WB6 is based of five shots, a total of nine neutrobn counts all detected by the inner most detector in a grossly under-diagnosed plasma. I see any conclusive proof for this comment he made and don't think it should be taken as gospel. Placing repeller plates at the ends of the cusps is the same as the magrid idea in that once the electrons reach them they havbe practically zero energy and therefore constitute a negligable loss.
@MSimon
We do have to do 4 orders of magnitude better, but in my suggestion, the earthed screen will in no place touch the cusps electrons will have to travel via cross field transport to reach it and if the magnetic field is strong enough and (fingers crossed) the cross field transport is not turbulent then this maybe sufficiently insulated from them.
A plus of using repeller plates rather than an inner magrid is that you can cover everywhere where the cusps touch with shielded repeller plates and not have any corners. Corners are catastrophic at the corners electrons will meet solid surfaces of even an open box machine at full velocity.
If you read this:
http://ecow.engr.wisc.edu/cgi-bin/getbi ... nl0107.pdf
You will clearly see the Bussard did not eliminate the corners in WB-6 he merely reduced them, looking at the photograph you can see the corners of the device as clear as daylight. However by reducing the spacing between the coils at the corners, he made the effective area pressented by the line cusps where they intersected the corners as low as possible.
WB6 isn't the be all and end all of the Polywell story, there are still new ideas out there to be explored.
Even open box machines have fieldines that touch the vacuum vessel wall, but because they are earthed and outside the magrid by the time the electrons hit them they have praticall zero kinetic energy.
WB5 was a closed box machine. Look at the photograph, the magnetic coils are on the outside. In fact, WB5 was the second last of the WB-X machines built by Bussard. So his decision that a closed box machine can ever yield nuclear fusion was a relatively recent one, and his enthusiasm for WB6 is based of five shots, a total of nine neutrobn counts all detected by the inner most detector in a grossly under-diagnosed plasma. I see any conclusive proof for this comment he made and don't think it should be taken as gospel. Placing repeller plates at the ends of the cusps is the same as the magrid idea in that once the electrons reach them they havbe practically zero energy and therefore constitute a negligable loss.
@MSimon
We do have to do 4 orders of magnitude better, but in my suggestion, the earthed screen will in no place touch the cusps electrons will have to travel via cross field transport to reach it and if the magnetic field is strong enough and (fingers crossed) the cross field transport is not turbulent then this maybe sufficiently insulated from them.
A plus of using repeller plates rather than an inner magrid is that you can cover everywhere where the cusps touch with shielded repeller plates and not have any corners. Corners are catastrophic at the corners electrons will meet solid surfaces of even an open box machine at full velocity.
If you read this:
http://ecow.engr.wisc.edu/cgi-bin/getbi ... nl0107.pdf
You will clearly see the Bussard did not eliminate the corners in WB-6 he merely reduced them, looking at the photograph you can see the corners of the device as clear as daylight. However by reducing the spacing between the coils at the corners, he made the effective area pressented by the line cusps where they intersected the corners as low as possible.
WB6 isn't the be all and end all of the Polywell story, there are still new ideas out there to be explored.
JMC,
The problem with electron repeller plates is that they absorb alphas. That means two things - cooling and output loss.
In any case - ultimately the screen will not be a "screen" in the final device. It will be a grid used to establish a potential gradient.
Electron losses from the "screened" volume will have to be low because the collector will be generating 2 MeV electrons. That is a loss mechanism and a heat load on the outer walls or collector grids.
The problem with electron repeller plates is that they absorb alphas. That means two things - cooling and output loss.
In any case - ultimately the screen will not be a "screen" in the final device. It will be a grid used to establish a potential gradient.
Electron losses from the "screened" volume will have to be low because the collector will be generating 2 MeV electrons. That is a loss mechanism and a heat load on the outer walls or collector grids.
Engineering is the art of making what you want from what you can get at a profit.
Yes, you're right about WB-5, and I should have checked that. Problem is, we do not have access to his calculations under which he made the determination that WB-5 was an impractical machine. I would take his word over yours, no offense.jmc wrote:WB5 was a closed box machine. Look at the photograph, the magnetic coils are on the outside. In fact, WB5 was the second last of the WB-X machines built by Bussard. So his decision that a closed box machine can ever yield nuclear fusion was a relatively recent one, and his enthusiasm for WB6 is based of five shots, a total of nine neutron counts all detected by the inner most detector in a grossly under-diagnosed plasma. I see any conclusive proof for this comment he made and don't think it should be taken as gospel. Placing repeller plates at the ends of the cusps is the same as the magrid idea in that once the electrons reach them they havbe practically zero energy and therefore constitute a negligable loss.
Roger:
That picture is PXL-1, WB5 is on page 9 of the link I sent.
Scareduck:
If you have heard more about WB5 than I have then please tell me I'm interested to find out every piece of information about that machine I can get my hands on. My level of knowledge at present is that they tried using repeller plates to stop the electrons from hitting the solid surfaces at high energy, the problem with open exposed repeller plates is that slow electrons will linger by them leading to an unscreened cloud of negative charge that distorts the potential surfaces hampering ion convergence and also drawing out ions from the well. In an open box device, the cloud of negative charge linger outside the magid, this allows capacitative effects from the magrid to screen the electron cloud from the ions.
Another way of doing this would seem to me to be to put and earthed grid with slits at the cusps between the repeller plates and the wiffle ball. Matbe Bussard has already tried this on WB5 and demonstrated that it didn't work, if he did then even I, would take his word over mine. However I have heard nothing so far to suggest that this was tried. If it wasn't tried then its not a matter of anyones word against anyone elses, its just an idea that hasn't been tried yet.
Bussard knew more about Polywells than probably anyone else on the planet at the time but even that wasn't very much. Remember he spent 10 million dollars building HEPS which completely failed, and the second last machine he built was a closed box design. That doesn't sound to me like an exhaustive refutation that closed box machines can never work.
This is still a field that is very much underexplored. There are still plenty of things to try and find out.
MSimon:
The temperature in fusion plasmas are 10's to 100's of Kev, the Magrid is likely to be 100's of KeV, Alphas are several MeV, their going to hit the vacuum vessel whether you like it or not. All you can do is ensure the things they hit can take some punishment. This screening grid does not have to be a grid, it could be a thick earthed metal plate with cooling pipes going through and a hole or slit in the middle where the cusps are located which leads to te repeller plates.
Whether you wish to call it a screen or a grid is not that important. The main thing will be that electrons escaping through the cusps will be turned around, while the potential well confining the ions will not be distorted by the electron cloud lingering at the repeller plates.
The most important question for this design will be, "is the crossfield transport going to be sufficiently large to lead to unacceptable electron losses to the plate?"
That picture is PXL-1, WB5 is on page 9 of the link I sent.
Scareduck:
If you have heard more about WB5 than I have then please tell me I'm interested to find out every piece of information about that machine I can get my hands on. My level of knowledge at present is that they tried using repeller plates to stop the electrons from hitting the solid surfaces at high energy, the problem with open exposed repeller plates is that slow electrons will linger by them leading to an unscreened cloud of negative charge that distorts the potential surfaces hampering ion convergence and also drawing out ions from the well. In an open box device, the cloud of negative charge linger outside the magid, this allows capacitative effects from the magrid to screen the electron cloud from the ions.
Another way of doing this would seem to me to be to put and earthed grid with slits at the cusps between the repeller plates and the wiffle ball. Matbe Bussard has already tried this on WB5 and demonstrated that it didn't work, if he did then even I, would take his word over mine. However I have heard nothing so far to suggest that this was tried. If it wasn't tried then its not a matter of anyones word against anyone elses, its just an idea that hasn't been tried yet.
Bussard knew more about Polywells than probably anyone else on the planet at the time but even that wasn't very much. Remember he spent 10 million dollars building HEPS which completely failed, and the second last machine he built was a closed box design. That doesn't sound to me like an exhaustive refutation that closed box machines can never work.
This is still a field that is very much underexplored. There are still plenty of things to try and find out.
MSimon:
The temperature in fusion plasmas are 10's to 100's of Kev, the Magrid is likely to be 100's of KeV, Alphas are several MeV, their going to hit the vacuum vessel whether you like it or not. All you can do is ensure the things they hit can take some punishment. This screening grid does not have to be a grid, it could be a thick earthed metal plate with cooling pipes going through and a hole or slit in the middle where the cusps are located which leads to te repeller plates.
Whether you wish to call it a screen or a grid is not that important. The main thing will be that electrons escaping through the cusps will be turned around, while the potential well confining the ions will not be distorted by the electron cloud lingering at the repeller plates.
The most important question for this design will be, "is the crossfield transport going to be sufficiently large to lead to unacceptable electron losses to the plate?"
Is it even possible to do direct conversion with a closed box? I doubt it, and if it's not, you're back to a thermal plant and rockets are out the window (except maybe medium-power deep space drives).
I can't see the advantage in forcing a closed-box configuration to work (except as a learning exercise, which may be warranted) - if closed-box works, surely an open configuration would work?
Cusp losses in an open machine are not going to hit the outer wall at any significant rate. Due to the charge on the grid (which is the only reason the electrons are at keV energies in the first place), they won't even get past the electron emitters unless they're upscattered. All you have to do is make the innermost trap grid far enough away from the magrid that electron losses to the walls are negligible.
I can't see the advantage in forcing a closed-box configuration to work (except as a learning exercise, which may be warranted) - if closed-box works, surely an open configuration would work?
Cusp losses in an open machine are not going to hit the outer wall at any significant rate. Due to the charge on the grid (which is the only reason the electrons are at keV energies in the first place), they won't even get past the electron emitters unless they're upscattered. All you have to do is make the innermost trap grid far enough away from the magrid that electron losses to the walls are negligible.
Well no actually. The alphas are going to be decelerated by the collector grid(s). That is the plan for getting the energy out.jmc wrote: MSimon:
The temperature in fusion plasmas are 10's to 100's of Kev, the Magrid is likely to be 100's of KeV, Alphas are several MeV, their going to hit the vacuum vessel whether you like it or not. All you can do is ensure the things they hit can take some punishment. This screening grid does not have to be a grid, it could be a thick earthed metal plate with cooling pipes going through and a hole or slit in the middle where the cusps are located which leads to te repeller plates.
Whether you wish to call it a screen or a grid is not that important. The main thing will be that electrons escaping through the cusps will be turned around, while the potential well confining the ions will not be distorted by the electron cloud lingering at the repeller plates.
The most important question for this design will be, "is the crossfield transport going to be sufficiently large to lead to unacceptable electron losses to the plate?"
To that end we wish to minimize the obstructions to alpha travel. Zero size grids if we could get them.
BTW electrons escaping from the cusps are turned around by the positive charge on the MaGrid. Electrons are allowed to oscillate from reaction space to "dead" space. They recirculate. Losses are kept sufficiently low by population annealing in the appropriate zones.
That is one of the beauties of the BFR. The efficient use of nothing.
The electric field keeps the ions in and brings the electrons back.
Engineering is the art of making what you want from what you can get at a profit.