That's a very reasonable stab at a summary, and is generally correct.
You might like to note the following points;
1) Fusors are already 'choked' up with neutrals. The fusions seen in fusors is mostly fast ions interacting with background neutral atoms. Fast-fast ion fusions in a fusor are not readily detectable. Uni of W claims it has detected a small percentage of such reactions, no-one else has given evidence of such. It is the very fact that fusors are 'choked up' with neutrals that gives them a fighting chance of any reactions at all!
2) The glow in the middle of a fusor are the cold and 'dying' ions that are being recombined with electrons, that then become neutrals again.
3) Polywell must rely on fast-fast ion collisions for any hope of improving on the fusor, and if it is to accomplish this must keep the pressures way way below that of a typical fusor, and remove essentially all of the background neutrals, else thermalisation would kill the reaction.
4) As mentioned above, Polywell is a magnetic confinement version of the ETW machine that uses the inertial of electrostatically confined electrons. ETW never produced any detectable nuclear reactions.
Could Polywell research enable Farnsworth-Hirsch fusors?
Magnetic fields on the central cathode need not necessarily defocus the ions. Depending on configuration, magnetic fields might actually help (eg- the Polywell, could be used to accelerate and focus ions by using the magrid as a negative cathode. In this instance the ion containment would be mostly magnetically based and lousy and the electrons would have to be injected inside the magrid much as the ions are in the actual Polywell).
I'm not sure how many focused fusions can be obtained in the center (a high proportion of which are probably fast-fast ion collisions) in a gas discharge fusor (lots of neutrals). One paper from U. Wisconsin claimed ~ 10% of total several years ago. Another more recent paper claimed ~ 50%. Their detection methods had improved. I don't know what else changed in their setup.
The record for fusion in any type of fusor, I believe, is still Hirsch's model which was built ~ 40 yrs ago (though if any fusion tests withWB8 have been completed, this record may have finally been exceeded). Hirsch's machine used ion guns and presumably operated at densities where neutrals were less prevalent. This is important because evidently the rate of fusion with fast ion-ion collisions increases faster with increased temperature and density than does fast- neutral collisions. This is presumably due to the greater thermalization and resistance to central focusing inherent with neutrals.
Keep in mind that with ion injection (or gas puffing on the edge in a large Polywell, the ions will have a monoenergetic speed (at least initially). In a gas discharge fusor the ions are born randomly anywhere in the fusor . so will experiance different accelerations towards the center (There are several ways that this can be improved- microwaves aimed near the edge would increase ionizations there and an ionization grid could help. Getting rid of the neutrals and using an ion gun is the ideal solution.). Add to that the pesky neutrals which scatter the ions and charge exchange with them and the resultant ion distribution is forced towards a Maxwellian state. How close it comes to actual complete thermalization is uncertain, but it is probably close. Opposing this somewhat (at least in the transverse directions) is the focusing obtained by the grid (the Star mode).
Getting rid of the neutrals and making the grid invisible will allow more monoenergetic ion populations so that you can optimize the drive efficiency so that nearly all of the ions are participating in fusion and few are wasting excess energy on bremsstrulung or escaping confinement. But, even under these conditions the monoenergetic ions will quickly thermalize. It needs the dynamics of the spherical geometry (ions can only upscatter or downscatter in the dense center, little transverse scattering can occur in this area),and Annealing on the edge which restores most of the upscattered and downscattered ions to a narrow energy range. Even if the ions do become thermalized in the transverse direction(angular momentum) so that there is little or no convergence, the radial velocities remain within a narrow range. This is tolerable, though not ideal in the Polywell, at least for D-D fusion.
Dan Tibbets
I'm not sure how many focused fusions can be obtained in the center (a high proportion of which are probably fast-fast ion collisions) in a gas discharge fusor (lots of neutrals). One paper from U. Wisconsin claimed ~ 10% of total several years ago. Another more recent paper claimed ~ 50%. Their detection methods had improved. I don't know what else changed in their setup.
The record for fusion in any type of fusor, I believe, is still Hirsch's model which was built ~ 40 yrs ago (though if any fusion tests withWB8 have been completed, this record may have finally been exceeded). Hirsch's machine used ion guns and presumably operated at densities where neutrals were less prevalent. This is important because evidently the rate of fusion with fast ion-ion collisions increases faster with increased temperature and density than does fast- neutral collisions. This is presumably due to the greater thermalization and resistance to central focusing inherent with neutrals.
Keep in mind that with ion injection (or gas puffing on the edge in a large Polywell, the ions will have a monoenergetic speed (at least initially). In a gas discharge fusor the ions are born randomly anywhere in the fusor . so will experiance different accelerations towards the center (There are several ways that this can be improved- microwaves aimed near the edge would increase ionizations there and an ionization grid could help. Getting rid of the neutrals and using an ion gun is the ideal solution.). Add to that the pesky neutrals which scatter the ions and charge exchange with them and the resultant ion distribution is forced towards a Maxwellian state. How close it comes to actual complete thermalization is uncertain, but it is probably close. Opposing this somewhat (at least in the transverse directions) is the focusing obtained by the grid (the Star mode).
Getting rid of the neutrals and making the grid invisible will allow more monoenergetic ion populations so that you can optimize the drive efficiency so that nearly all of the ions are participating in fusion and few are wasting excess energy on bremsstrulung or escaping confinement. But, even under these conditions the monoenergetic ions will quickly thermalize. It needs the dynamics of the spherical geometry (ions can only upscatter or downscatter in the dense center, little transverse scattering can occur in this area),and Annealing on the edge which restores most of the upscattered and downscattered ions to a narrow energy range. Even if the ions do become thermalized in the transverse direction(angular momentum) so that there is little or no convergence, the radial velocities remain within a narrow range. This is tolerable, though not ideal in the Polywell, at least for D-D fusion.
Dan Tibbets
To error is human... and I'm very human.
The discussion of increasing the density in a magnetically shielded fusor in order to increase the fusion rate has a major problem, which is arcing, or more discriptively described as rapidly increasing conductance of the plasma, to such an extent that high voltages (or potential wells) cannot be maintained. In a fusor at ~ 100 Microns, the maximum voltage that can be obtained may be a few thousand volts. This is because of the increasing recruitment of neutrals being ionized in a cascading manner, or for that matter the increasing numbers (density) of electrons and ions provided directly by guns. This limits fusors to ~ 5-20 Microns if higher voltages are desired (eg:10-100,000 volts).
With recirculation, the Wiffleball is not critical for minimizing losses. But what is critical is the selective charged particle trapping supplied by the Wiffleball effect or something similar.
This essentially allows (with vigorous pumping) to maintain a background density of perhaps 0.1 to 1 Micron outside the magrid. This area is where the current would ground on supports, vacuum vessel wall, etc. The >1000 X trapping factor allows the densities necessary. Without this density gain (of charged particles only) the internal volume would have to be a million times larger to compensate. And assuming losses scale similarly at ~ r^2, then the volume (gain scales as r^3) would have to be huge to reach breakeven, if you ever got there at all. Some magnetic shielding of the periferal anode and/ or central cathode may significantly improve losses, and the working density may even be improved some, and the thermalization properties and confluence issues may be improved. You could continually manipulate the system, untill..... wait, you now have a Polywell...
Dan Tibbets
With recirculation, the Wiffleball is not critical for minimizing losses. But what is critical is the selective charged particle trapping supplied by the Wiffleball effect or something similar.
This essentially allows (with vigorous pumping) to maintain a background density of perhaps 0.1 to 1 Micron outside the magrid. This area is where the current would ground on supports, vacuum vessel wall, etc. The >1000 X trapping factor allows the densities necessary. Without this density gain (of charged particles only) the internal volume would have to be a million times larger to compensate. And assuming losses scale similarly at ~ r^2, then the volume (gain scales as r^3) would have to be huge to reach breakeven, if you ever got there at all. Some magnetic shielding of the periferal anode and/ or central cathode may significantly improve losses, and the working density may even be improved some, and the thermalization properties and confluence issues may be improved. You could continually manipulate the system, untill..... wait, you now have a Polywell...
Dan Tibbets
To error is human... and I'm very human.
Perhaps, if you did not optimize the compromises. That in essence is what the Polywell does. One manipulation may harm the system in a small way, but it helps in other areas in a large way.chrismb wrote:... though it would be 865,000 miles in diameter to get to net energy output?D Tibbets wrote:You could continually manipulate the system, untill..... wait, you now have a Polywell...
Having cusps detracts from an imaginary ideal spherical confinement, but by having the cusps arranged symetrically around this immaginary sphere minimizes the penalty in confluence and electrostatic confinement of the ions(?). Changing the design to allow for improved recirculation, harms the primary electron confinement, but the gain for the system is positive both in terms of power balance, and refreshing / restoring the electrons to their narrow input energy on time scales shorter than the thermalization times. I thinks this may be of increasing importance as the machine size increases.
Having cusps that allow electrons to escape is unavoidable, but that the upscattered electrons exit these cusps more frequently (shorter confinement time) is a benefit for retarding thermalization. Having a positive charge helps to recirculate mundane electrons, but perhaps also important, is that this recovers most of the energy carried by escaping upscattered electrons, so they do not detract too much from the energy balance. Etc, etc....
Dan Tibbets
To error is human... and I'm very human.
Roger, I'm guessing you are referring to continuing work at the U. Wisconsin, similar to this:
http://fti.neep.wisc.edu/iec2009/talks/ ... ianegl.pdf
crismb, I beleive you are thinking of what is (correctly?) referred to as the Hirsch Meeks fusor. This is different from the Hirsch ion gunned fusor where he got ~ 10^10 or 11 Deuterium - Tritium fusions per second.
PS: I'm waiting with baited breath for the release of the presentations from this years conference, like those released from last years conference.
Dan Tibbets
http://fti.neep.wisc.edu/iec2009/talks/ ... ianegl.pdf
crismb, I beleive you are thinking of what is (correctly?) referred to as the Hirsch Meeks fusor. This is different from the Hirsch ion gunned fusor where he got ~ 10^10 or 11 Deuterium - Tritium fusions per second.
PS: I'm waiting with baited breath for the release of the presentations from this years conference, like those released from last years conference.
Dan Tibbets
To error is human... and I'm very human.