I really don't get either the magnetic or electric oscillations.
How do you oscillate the mag field without upsetting the wiffleball?
How do you oscillate the magrid potential with respect to the well potential, when the potential of the electrons and the magrid will go up and down together? You can't inject and 'un-inject' electrons at kHz, so why would oscillating the magrid actually change the electric fields within it? Don't forget Gauss!! You'll then get oscillating cusp losses together with it, that will more than likely accelerate electron loss.
I think oscillating fields for a fixed grid fusor has a lot going for it (and this is what the comments are based on), but for Polywell, I don't really see how it'd work out once you've gone from a real to a virtual electrode.
POPS iec fusion
-
happyjack27
- Posts: 1439
- Joined: Wed Jul 14, 2010 5:27 pm
well oscillating the magnetic field strength was just an idea that i threw out. and that's a good point, it might disrupt the wiffle-ball effects. maybe not so good of an idea.
oscillating the charge, on the other hand, esp at high frequencies (e.g. RF-range) isn't going to affect the magnetic field (well, aside from a small current oscillation, maybe) it will, however, oscillate the voltage potential. and the velocity of protons and electrons are going to be effected differently by the oscillations---by a few orders of magnitude---because they have different masses. one might say they'll have different "slew rate"s.
but as long as the initial voltage bias on the grid is high enough, it's not going to cause any electrons to jump out of confinement. e.g. if you want to oscillate it +/1 a volt, just add a volt of bias to the the magrid to keep everything w/in the same margins. failing the electrons jumping out of confinement or the wiffle-ball being disrupting, the nuclei will be fine.
unless you're talking about the magnetic fields induced by the current from the oscillation of charged particles within the plasma. if the oscillation was fully concentric the fields would all cancel, i believe. and i think even if they're somewhat close, the stray fields would be negligible.
oscillating the charge, on the other hand, esp at high frequencies (e.g. RF-range) isn't going to affect the magnetic field (well, aside from a small current oscillation, maybe) it will, however, oscillate the voltage potential. and the velocity of protons and electrons are going to be effected differently by the oscillations---by a few orders of magnitude---because they have different masses. one might say they'll have different "slew rate"s.
but as long as the initial voltage bias on the grid is high enough, it's not going to cause any electrons to jump out of confinement. e.g. if you want to oscillate it +/1 a volt, just add a volt of bias to the the magrid to keep everything w/in the same margins. failing the electrons jumping out of confinement or the wiffle-ball being disrupting, the nuclei will be fine.
unless you're talking about the magnetic fields induced by the current from the oscillation of charged particles within the plasma. if the oscillation was fully concentric the fields would all cancel, i believe. and i think even if they're somewhat close, the stray fields would be negligible.
That page is nonsense. fluctuating the Magnetic field would do zip, zero, nada... _puossibly_ fluctuating the Electric field would be interesting, but all that would do is affect (fairly slowly, with great hysteresis) the density of the electrons. I suppose this would be a way to get pulses above beta=1, but why they would want to bother is a good question.chrismb wrote:I really don't get either the magnetic or electric oscillations.
How do you oscillate the mag field without upsetting the wiffleball?
How do you oscillate the magrid potential with respect to the well potential, when the potential of the electrons and the magrid will go up and down together? You can't inject and 'un-inject' electrons at kHz, so why would oscillating the magrid actually change the electric fields within it? Don't forget Gauss!! You'll then get oscillating cusp losses together with it, that will more than likely accelerate electron loss.
I think oscillating fields for a fixed grid fusor has a lot going for it (and this is what the comments are based on), but for Polywell, I don't really see how it'd work out once you've gone from a real to a virtual electrode.
Wandering Kernel of Happiness
Well, I think they already do POPS with virtual cathodes. I linked a paper on it a ways back.I think WB could not do any POPS, it would have to be a fixed central electrode.
http://pop.aip.org/resource/1/phpaen/v1 ... horized=no
I wonder if anyone's modelled POPScillation in a wiffleball. It might disrupt the cusp-plugging. Hmmm.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...
Is that the paper you are actually thinking of? Doesn't read as if it is about POPS.
The thing is, just play it through as to what the electrons are doing. What makes the potential oscillate is the variation of the magrid to trapped net charge. The electrons are [supposedly] fixed so these ions have to be forced to alter their position wrt the electrons. This isn't a 'plasma oscillation' (POPS) because in the Polywell I thought the ions move [radially] and the electrons don't. So if the ions follow the change of e-field and the electrons don't, but we're told that a change of just 1 in a million can created massive well depths, then what happens if one millionth of the ions get pulled out a bit further and leave those electrons behind? All of a sudden you must end up with a massive well depth that would swap any POPS overlay potential.
It is one thing to not believe in ambipolar diffusion in the wiffleball, but if you are arguing that ambipolar diffusion also doesn't affect electric field oscillations then it seems to me that we really have entered a whole new world of physics, where ions and electrons in the same space can be acted on in entirely separate ways, and ne'er the twain shall influence each other!
The thing is, just play it through as to what the electrons are doing. What makes the potential oscillate is the variation of the magrid to trapped net charge. The electrons are [supposedly] fixed so these ions have to be forced to alter their position wrt the electrons. This isn't a 'plasma oscillation' (POPS) because in the Polywell I thought the ions move [radially] and the electrons don't. So if the ions follow the change of e-field and the electrons don't, but we're told that a change of just 1 in a million can created massive well depths, then what happens if one millionth of the ions get pulled out a bit further and leave those electrons behind? All of a sudden you must end up with a massive well depth that would swap any POPS overlay potential.
It is one thing to not believe in ambipolar diffusion in the wiffleball, but if you are arguing that ambipolar diffusion also doesn't affect electric field oscillations then it seems to me that we really have entered a whole new world of physics, where ions and electrons in the same space can be acted on in entirely separate ways, and ne'er the twain shall influence each other!
Oops that was a little obscure. That was a study of virtual cathode stability for the purposes of POPS, I believe. Here's an overview.
http://fti.neep.wisc.edu/iecworkshop/PD ... /nebel.pdf
On page 18 there's a virtual cathode POPS device.
I think the real problem is space charge limits, but Chacon apparently had a way around that. There's more POPS stuff out there to read.
http://www.ncbi.nlm.nih.gov/pubmed/16090625
http://fti.neep.wisc.edu/iecworkshop/PD ... /nebel.pdf
On page 18 there's a virtual cathode POPS device.
I'm not sure why you think the electrons wouldn't move. They should get reflected around.I thought the ions move [radially] and the electrons don't
It's very purposely e-driven. Ambipolar diffusion would make no sense.It is one thing to not believe in ambipolar diffusion in the wiffleball,
I think the real problem is space charge limits, but Chacon apparently had a way around that. There's more POPS stuff out there to read.
http://www.ncbi.nlm.nih.gov/pubmed/16090625
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...
I think that's the same Murali working with Lerner at LPP (Focus Fusion):TallDave wrote:There's more POPS stuff out there to read.
http://www.ncbi.nlm.nih.gov/pubmed/16090625
http://fti.neep.wisc.edu/pdf/fdm1267.pdf (UWM PhD Thesis, 8.3 MB)
The electrons do move with the ions to a degree, as mentioned by Bussard, and others. Because of the inertial differences between the electrons and the ions, the ions tend to tug the electrons along towards the center. This is how the pure electron square potential well is converted to a parabolic well once ions are added. This is favored (I think) because this tugging of electrons on a local (weakly coupled) scale does not work against the space charge (no net current), while the reverse does. Inertial differences means an electron does not tug on an ion as much (doesnt't change it's path or velocity as much) during the brief times it is close enough to dominate over the space charge. This is what I have argued is the basis for non ambiploar flow in the cusps. So long as the space charge is maintained by injection of excess electrons, the space charge will dominate the ion motions over any local effects.chrismb wrote:Is that the paper you are actually thinking of? Doesn't read as if it is about POPS.
The thing is, just play it through as to what the electrons are doing. What makes the potential oscillate is the variation of the magrid to trapped net charge. The electrons are [supposedly] fixed so these ions have to be forced to alter their position wrt the electrons. This isn't a 'plasma oscillation' (POPS) because in the Polywell I thought the ions move [radially] and the electrons don't. So if the ions follow the change of e-field and the electrons don't, but we're told that a change of just 1 in a million can created massive well depths, then what happens if one millionth of the ions get pulled out a bit further and leave those electrons behind? All of a sudden you must end up with a massive well depth that would swap any POPS overlay potential.
It is one thing to not believe in ambipolar diffusion in the wiffleball, but if you are arguing that ambipolar diffusion also doesn't affect electric field oscillations then it seems to me that we really have entered a whole new world of physics, where ions and electrons in the same space can be acted on in entirely separate ways, and ne'er the twain shall influence each other!
I will assume this process is fast enough that it would accommodate POPS type frequencies.
Dan Tibbets
To error is human... and I'm very human.
i'm not sure we can dissmis, so easily the possible affects of modulating the mag field.
i do not think the affect will be zilch (precisely), though nor do i think it will necessarily benefit; in the simple case it might well prove detrimental to eg. stability.
but, the mag fields could be used to provide 'magnetic'/inertial impulse to the plasma/ion/electron populations (/geometry). (mageto-acoustic or Alfven perhaps).
with the right inputs (eg. Heaviside func) / boundary conditions, it seems conceivable there might be some statistical 'synchronised' density effects we could take advantage of - without blowing the coils apart (relaxation oscillator perhaps).
perhaps something to play with after we've see the basic model working at its best, sometime soon, we hope....
i do not think the affect will be zilch (precisely), though nor do i think it will necessarily benefit; in the simple case it might well prove detrimental to eg. stability.
but, the mag fields could be used to provide 'magnetic'/inertial impulse to the plasma/ion/electron populations (/geometry). (mageto-acoustic or Alfven perhaps).
with the right inputs (eg. Heaviside func) / boundary conditions, it seems conceivable there might be some statistical 'synchronised' density effects we could take advantage of - without blowing the coils apart (relaxation oscillator perhaps).
perhaps something to play with after we've see the basic model working at its best, sometime soon, we hope....
The problem with modulating the magnetic fields , if it did work on the time scales needed is that this would defeat at least some of the purpose of the superconductor - has to be constant if you want to minimize input magnet drive power. A supplemental copper wire magnet might get the job done. But an alternating magnetic field means constant friction on the wires as they move about within the magnet. This is what is believed to have caused the insulation failure in WB6.
Dan Tibbets
Dan Tibbets
To error is human... and I'm very human.
-
happyjack27
- Posts: 1439
- Joined: Wed Jul 14, 2010 5:27 pm
i don't mean alternating the field. i understand that would probably just tear the coils apart, not to mention pose problems for the confinement. i mean just modulating it a little bit, by just a fraction of the overall current. keeping the polarities the same just changing the strength a little.D Tibbets wrote:The problem with modulating the magnetic fields , if it did work on the time scales needed is that this would defeat at least some of the purpose of the superconductor - has to be constant if you want to minimize input magnet drive power. A supplemental copper wire magnet might get the job done. But an alternating magnetic field means constant friction on the wires as they move about within the magnet. This is what is believed to have caused the insulation failure in WB6.
Dan Tibbets
as regard modulating the bias voltage on a polywell, it probably wouldn't hurt to modulate the ion/electron injection and ionization (through RF/micro waves) to match the modulation of the bias. e.g. so that ions/electrons are only injected when the bias is at its lowest.