Anyone seen this yet?
http://pop.aip.org/phpaen/v17/i5/p05251 ... horized=no
Looks like the formal write up from the IEC09 presentation.
http://fti.neep.wisc.edu/iec2009/talks/ ... tthewc.ppt
Well Formation Paper
I've given Famulus a heads up. He is on hiatus until the end of the month I think so it may be a long time (in internet days) before I get a reply.rcain wrote:Good find Ladajo.
Anyone got $28? - should make interesting reading.
Might be nice if Famulus got together wth Joe Khachan's team, cover more ground, more data.
Rock-on Ozzie Fusioneers!
Engineering is the art of making what you want from what you can get at a profit.
Thanks to Betruger for giving me sight of this paper - "The dependence of the virtual cathode in a Polywell on the coil current and background gas pressure" - Carr and Khachan - Physics of Plasmas 24th May 2010.
Wondered if anyone else had given it a read though yet/had any comments.
Though I dont think its revelealed anything ground breaking, it does seem to represent a genuine contribution to our knowledge at this point. I've picked out the following salient points as a summary::
Wondered if anyone else had given it a read though yet/had any comments.
Though I dont think its revelealed anything ground breaking, it does seem to represent a genuine contribution to our knowledge at this point. I've picked out the following salient points as a summary::
does that concur with peoples understanding, or present any surprises?bottom p1: "...Although Bussard and Krall claimed to know the scaling laws, these results have not yet been published in the open peer reviewed literature...."
mid p2:: "Bussard's most successful Polywell (WB6) used the filament method to generate electrons. The Polywell was floated to a high positive voltage to accelerate the electrons and attract them to the general Polywell region, but was not capable of focussing them into a beam. ... This ultimately led to many uncontrollable arcing issues that destroyed Bussard's WB6 device... "
later mid p2:: "By using a cylindrical hollow cathode, the electron beam generation process is greatly simplified and simultaneously negates the need to float the Polywell for electron acceleration. ..."
later p2: "... The hollow cathode was placed 10cm from the Polywell face and the angle of beam incidence could be varied arbitrarily."
top p3: "A single ended Langmuir probe was mounted on a flexible bellows allowing motion along the radial components of the Polywell"
top p3:: "Note that we will use the floating potential as a qualitative parameter to confirm the existance of a virtual cathode and its lifetime. No further interpretation of results can be made at this stage because the electron energy distribution function is uinknown and the discharge may not satisfy the quasineutrality condition since electrons are being injected into the confining space. Measurements of electron density and energy distribution will be left for future work"
Results
mid p3:: "... confirming the presence of a virtual cathode, stable for at least on the millisecond timescale."
mid p3: "Note that there is a nonlinear increasing potential well depth in the range 0-100A, (Class 1 results)... For current s greater than 100A there is a linear decrease in potential well depth, (Class 2 results)... "
bottom p3: "The peak current in class 1 results (are) in phase with the minimum of the potential well. However, class 2 results display a linearly increasing phase shift between the peak current and the potential well minima... Leading to the interpretation that this (100A) threshold is the peak magnetic field above which potential wells are no longer stable"
bottom p3, mid p4:: "A likely explanation of the observed magnetic field threshold is that the magnetic mirror effect will be active from both inside and outside the Polywell.... The idea predicts that there will be a threshold point at which a potential well can no longer form since a substantial portion of the injected electron beam is now reflected and no longer enteres the device... explaining the observed phenomena."
bottom p4: "Potential well formation as a function of injected electron energy was measured at three pressures.... potential wells completely dissapeared at 35mT in all data sets. As pressure is decreased from 25mT to 15mT the potential wells are consistently deeper suggesting that the achievable well depth is dependent on pressure "
top p5:: "... hence the production of ions by the trapped electrons must be considered as a mechanism for reducing the well depth."
mid p5:: "... This strongly suggests that future Polywell designs will need to be capable of supplying significantly greater injection currents if we are to achieve the deep potential well depths required for IEC. "
Conclusions:
bottom: p5:: "We expect that improved performance can be observed in a Polywell of this design by increasing the magnitude of injected electron current and decreasing the background gas pressure. Both of these goals could be achieved by moving the electron gun to an adjacent chamber where differential pumping can be used to maintain a substantial pressure difference. In addition it will be technically easier to generate large electron currents from a plasma source than the alternatives such as thermionic emission. "
The second point seems misleading from my understanding. Arcing in the machine is pressure dependant (Paschin breakdown) and will destroy the potential well, but not the machine. The machine was destroyed when the insulation failed between the wire entering the magrid and grid case(high positive potential. This melted (and probably vaporized) the wire. This was just like what happens when a transformer's windings short out and has nothing to do with what was happening with the plasma.
The comment about the cathodes being hollow? They were auto headlight filaments. I don't know why the defocus of the electron beam (potential well being ~ 80-85% of the drive potential, has to do with potential well maintainance (as opposed to depth). If the electron beams stayed focused without scattering as they approached the center (or were diverged some by the magnetic fields) in the center the beam would exit the opposite cusp on the first pass and there would be no containment.
Mirror reflection (actually bounce may be the more appropriate term, I think) might end up with the back and forth mirroring (or bouncing) occurring outside the magrid. But, this ignores the (I assume) strong and dominate effect of the positive potential on the magrid. Any electrons outside the magrid get sucked back in unless they hit something first(unless they have been upscattered to energies above the drive potential- in which case you don't want them anyway- in that case the electron is not recycled, but some/ most of it's kinetic energy is recycled). The magnetic field lines act as focusing or directional controls on the electron flight paths, but does not resist the inward acceleration of the electrons due to the positive charge on the grids. Magnetic fields do not accelerate charged particles. The above situation might change if the magrid was grounded and the electron guns were at high negative potential, but with the positive charged magrids, the potential of the grid is what provides the recirculation force, not mirroring or bouncing of the electrons along the magnetic field lines. If you assume a large portion of the internal electrons are at the magnetic border (Wiffleball border), they may be traveling along field lines and bouncing or mirroring back before they reach the tightest portion of the cusp. If they manage to pass this point they would tend to mirror outside the machine (or hit something). But, again, this ignores the positive potential on the magrid which I assume dominates. If the electrons are ricocheting off the Wiffleball border without more than a partial gryo orbit then the picture of the Wiffleball with tiny openings providing for the trapping ratio is more useful. I suspect it is a combination of the two, and I have no idea which dominates. At least when ions are introduced and electrons are dragged towards the center so that an elliptical vs the initial square potential well forms, I suspect the ricocheting model may be more useful/ dominate.
Dan Tibbets
The comment about the cathodes being hollow? They were auto headlight filaments. I don't know why the defocus of the electron beam (potential well being ~ 80-85% of the drive potential, has to do with potential well maintainance (as opposed to depth). If the electron beams stayed focused without scattering as they approached the center (or were diverged some by the magnetic fields) in the center the beam would exit the opposite cusp on the first pass and there would be no containment.
Mirror reflection (actually bounce may be the more appropriate term, I think) might end up with the back and forth mirroring (or bouncing) occurring outside the magrid. But, this ignores the (I assume) strong and dominate effect of the positive potential on the magrid. Any electrons outside the magrid get sucked back in unless they hit something first(unless they have been upscattered to energies above the drive potential- in which case you don't want them anyway- in that case the electron is not recycled, but some/ most of it's kinetic energy is recycled). The magnetic field lines act as focusing or directional controls on the electron flight paths, but does not resist the inward acceleration of the electrons due to the positive charge on the grids. Magnetic fields do not accelerate charged particles. The above situation might change if the magrid was grounded and the electron guns were at high negative potential, but with the positive charged magrids, the potential of the grid is what provides the recirculation force, not mirroring or bouncing of the electrons along the magnetic field lines. If you assume a large portion of the internal electrons are at the magnetic border (Wiffleball border), they may be traveling along field lines and bouncing or mirroring back before they reach the tightest portion of the cusp. If they manage to pass this point they would tend to mirror outside the machine (or hit something). But, again, this ignores the positive potential on the magrid which I assume dominates. If the electrons are ricocheting off the Wiffleball border without more than a partial gryo orbit then the picture of the Wiffleball with tiny openings providing for the trapping ratio is more useful. I suspect it is a combination of the two, and I have no idea which dominates. At least when ions are introduced and electrons are dragged towards the center so that an elliptical vs the initial square potential well forms, I suspect the ricocheting model may be more useful/ dominate.
Dan Tibbets
To error is human... and I'm very human.
re: the causes of Bussard's WB6 machine failure - yes, historically it was due to coil short, not arcing; although arcing problems were also documented earlier-on IIRC. However, that is all rather beside the point.
DTibbets - where do you get 'well depth = 85% drive voltage' from - i dont see that quoted in the paper anywhere? Only a measured max well depth <= 250v produced by variable cathode voltage up to 20kV at variable current of <=20mA.
So, this is a lot smaller experiment with very limited (though to my mind very worthwhile) objectives.
His main points to me are:
1) Critical issues with injection and its downstream effects. Interestingly, Rick Nebels recent shopping list for posh ion injectors addressed this area also.
His point is, its quite difficult to get (unfocused) electrons in though the magnetic holes (cones) - which makes total sense to me.
Varying injection angles would make another interesting experimental setup - i recall some theoretical work on it somewhere already (Bussard's i think).
2) The nonlinearity of the 'magnetic threshold' for instantiation and sustainance of the potential well.
This it seems to me is pretty much as expected, though perhaps someone with greater mathematical facility than me would like to demonstrate how.
Experimentally, this 'feature' seems to me worthy of further exploration. It is fundamental to the device, and must also play a part in scaling. It is also accessible still to 'low-budget' research.
Perhaps Dr's Carr & Khachan will look at this in more detail in their next phase of work along with ion densities and velocity space.
Perhaps something Famulus could get his teeth into?
We still require some solid theoretical work to back this up I think.
3) Criticality of background pressure. This has been observed and explained before, iirc, and is expected based on what i've understood of our discussions on the forum.
His idea of a split chamber ( + plasma source) with differential pumping seemes like a good idea.
DTibbets - where do you get 'well depth = 85% drive voltage' from - i dont see that quoted in the paper anywhere? Only a measured max well depth <= 250v produced by variable cathode voltage up to 20kV at variable current of <=20mA.
So, this is a lot smaller experiment with very limited (though to my mind very worthwhile) objectives.
His main points to me are:
1) Critical issues with injection and its downstream effects. Interestingly, Rick Nebels recent shopping list for posh ion injectors addressed this area also.
His point is, its quite difficult to get (unfocused) electrons in though the magnetic holes (cones) - which makes total sense to me.
Varying injection angles would make another interesting experimental setup - i recall some theoretical work on it somewhere already (Bussard's i think).
2) The nonlinearity of the 'magnetic threshold' for instantiation and sustainance of the potential well.
This it seems to me is pretty much as expected, though perhaps someone with greater mathematical facility than me would like to demonstrate how.
Experimentally, this 'feature' seems to me worthy of further exploration. It is fundamental to the device, and must also play a part in scaling. It is also accessible still to 'low-budget' research.
Perhaps Dr's Carr & Khachan will look at this in more detail in their next phase of work along with ion densities and velocity space.
Perhaps something Famulus could get his teeth into?
We still require some solid theoretical work to back this up I think.
3) Criticality of background pressure. This has been observed and explained before, iirc, and is expected based on what i've understood of our discussions on the forum.
His idea of a split chamber ( + plasma source) with differential pumping seemes like a good idea.
'Background pressure'!!!
What's all this talk about background pressure? Isn't it bloomin' obvious that this thing ain't gonna take off with neutrals flapping around? How many can it tolerate? Next to none, I though the general agreement was? "Keep pumping the heck out of it", was Tom Ls quote I think, or something similar?
What's the point in working out that decreased background pressure improves the performance, as do more electrons? ...Straight from the school of the blindingly-obvious!!
Might aswell check whether eating pizzas near by has an effect. Ah, no, it didn't seem to make a difference... but, hey, let's try this again tomorrow just in case our measurements are out.... then we can move on to beer and after that we'll try the going-down-the-strip-bar-before-running-the-experiment test, to see if that makes a difference.
What's all this talk about background pressure? Isn't it bloomin' obvious that this thing ain't gonna take off with neutrals flapping around? How many can it tolerate? Next to none, I though the general agreement was? "Keep pumping the heck out of it", was Tom Ls quote I think, or something similar?
What's the point in working out that decreased background pressure improves the performance, as do more electrons? ...Straight from the school of the blindingly-obvious!!
Might aswell check whether eating pizzas near by has an effect. Ah, no, it didn't seem to make a difference... but, hey, let's try this again tomorrow just in case our measurements are out.... then we can move on to beer and after that we'll try the going-down-the-strip-bar-before-running-the-experiment test, to see if that makes a difference.
Ah, the Aussies have published.
The electron gun info is certainly interesting, maybe there's something in there that will help Rick.
Good to get more confirmation of virtual cathode stability.
I'll have to dig through this further.
The electron gun info is certainly interesting, maybe there's something in there that will help Rick.
Good to get more confirmation of virtual cathode stability.
I'll have to dig through this further.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...
i'm sorry to hear that ): you could always consider selling your body and soul to the mod - it would have the added benefit you'd then be forbidden to talk to us - the perfect excuse.
seriously, on the pressure, they did at least state a figure - >15mTorr and to be fair they needed some for their cylindrical cathode arrangement, plus their gauge was only precise up to 10E-4 Torr. at least we now know we can still pursue some useful research within such a regime.
ps. did anyone calculate/plot the theoretical rate of production of (slow) neutrals in the core?
seriously, on the pressure, they did at least state a figure - >15mTorr and to be fair they needed some for their cylindrical cathode arrangement, plus their gauge was only precise up to 10E-4 Torr. at least we now know we can still pursue some useful research within such a regime.
ps. did anyone calculate/plot the theoretical rate of production of (slow) neutrals in the core?
I believe he got it from the Valencia paper. And in this case, the drive voltage is the voltage between the MaGrid and the chamber wall (or very close to that) which was also the "gun" voltage.rcain wrote: DTibbets - where do you get 'well depth = 85% drive voltage' from - i dont see that quoted in the paper anywhere? Only a measured max well depth <= 250v produced by variable cathode voltage up to 20kV at variable current of <=20mA.
The fact that these folks only got 250v suggests to me that their MaGrid wasn't very strong in either the Ma or the Grid department.
By the way, did they charge their chamber to the same voltage as their gun? If not, they would just be shooting thru the MaGrid to the chamber wall. Seems a waste.
But then I didn't read the paper, just these posts, so I may be way off base here.