chrismb wrote:You've lost me here, can't follow why you make this sequence of assumptions.
The upscattered ions leave through the central cusps and get frogmarched by the excess electrons past the magrid and back. They return with the "right" amount of energy.
The positive ions, if they reach the magrid radius in a cusp will see the positive charge (minus the weaker potential well negative charge) and be accelerated to the vacuum vessel wall. They do not return- they act opposite to the recirculating electrons in thias region. Comments about Gauss's law/ Faraday cages may be confusing. It applies inside the magrid, but not outside.
I don't know how many local electrons will be in this area while they are being recirculated. But, based on some numbers- magnetic containment holds the electrons for ~1000-10,000 passes. Recirculation multiplies this by perhaps 10 to 100X. This implies that the vast majority of electrons at any given time are deep within the Wiffleball. If this density of electrons left behind cannot stop the ion before it reaches the magrid radius, the relatively few local electrons will not have much additional effect, certainly not enough to overcome the positive charge on the magrid.
Also, keep in mind that the cusp flows are not ambipolar according the R. Nebel- ie the more global conditions have enough influence to overcome any tendancy for local ions and electrons to pair up on a one for one basis.
Dan Tibbets
As Art calculated, the same number of ions and electrons leave the cusp, but the ions are travelling slowly and the electrons are travelling fast, so once a lot of ions have dropped back the recirculation jet is electron rich.
The ions that do escape see a strong local negative charge and are frogmarched along with the electrons. The degree of mixing will depend on the distance between the wiffleball and the magrid and should affect how a polywell scales.
Once the ions get past the magrid, only the most energetic can escape the strong local attraction of the electron jet as it oscillates back to the wiffleball.
alexjrgreen wrote:
As Art calculated, the same number of ions and electrons leave the cusp, but the ions are travelling slowly and the electrons are travelling fast, so once a lot of ions have dropped back the recirculation jet is electron rich.
Sorry, but you are confusing yourself here. Just because there is a differential of speed does not mean there is a differential of displaced particles. What is being described here is actually a plasma and it will be 'neutral'. But within a neutral plasma, you can still have a net differential of - and + charges moving through it - obvious, else you'd not be able to pass a current through it!!
In point of fact, what the system will aim to do is shift as much charge (differentially) as it can so as to generate an equal and opposite reaction to the prevailing magnetic field. As such, the magnetic fields in the whole device wil be reduced, their 'field energy' being sucked up by the net particle currents you are describing (if that were to happen, which is questionable).
chrismb wrote:Sorry, but you are confusing yourself here. Just because there is a differential of speed does not mean there is a differential of displaced particles. What is being described here is actually a plasma and it will be 'neutral'. But within a neutral plasma, you can still have a net differential of - and + charges moving through it - obvious, else you'd not be able to pass a current through it!!
But it's not a neutral plasma...
It's quasi-neutral, and electron rich by as many electrons as it takes to fill the well. There's even doubt about whether you can properly call it a plasma.
chrismb wrote:In point of fact, what the system will aim to do is shift as much charge (differentially) as it can so as to generate an equal and opposite reaction to the prevailing magnetic field. As such, the magnetic fields in the whole device wil be reduced, their 'field energy' being sucked up by the net particle currents you are describing (if that were to happen, which is questionable).
The recirculation is real. You can see the scorch marks on the early WB machines and the latest WB7 pic shows it nicely.
alexjrgreen wrote:
It's quasi-neutral, and electron rich by as many electrons as it takes to fill the well. There's even doubt about whether you can properly call it a plasma.
So, now it's electron rich on the outside, as well as the inside?
Can you go over how ions are accelerated, again, please? Me confused...
alexjrgreen wrote:
It's quasi-neutral, and electron rich by as many electrons as it takes to fill the well. There's even doubt about whether you can properly call it a plasma.
So, now it's electron rich on the outside, as well as the inside?
Can you go over how ions are accelerated, again, please? Me confused...
The (negative) well depth is 80% of the (positive) magrid potential.
chrismb wrote:..and I thought you were saying it's electron rich/negative on the outside aswell? Why won't the ions accelerate outwards?
Most of the ions are contained in the wiffleball by the central well.
If they get out of a central cusp (and the ions see a bigger cusp) some of them are eventually pulled back by the well, while others follow the electrons out and back.
If they have enough energy to clear the electrons they accelerate away from the magrid.
chrismb wrote:..and I thought you were saying it's electron rich/negative on the outside aswell? Why won't the ions accelerate outwards?
Most of the ions are contained in the wiffleball by the central well.
If they get out of a central cusp (and the ions see a bigger cusp) some of them are eventually pulled back by the well, while others follow the electrons out and back.
If they have enough energy to clear the electrons they accelerate away from the magrid.
Yes (I believe), the cusp flows are electron rich, both with the outward bound escaping electrons and the returning recirculating electrons.. But keep in mind these numbers are much less than the electrons hanging out within the Wiffleball. The ions that escape into the distant cusps (both the central cusps and the corner cusps and/or funny cusps) are tugged along some by the excess local electrons, and then impeaded some by the local electrons that are recirculating beyond the magrid borders. I believe this effect is trivial for two reasons. One is that this effect is small compared to the influences of the potential well and the charged magrid once the charged partical is beyond it. The second reason is that the electrons in the cusps are effectively pulling the ions outward inside of the magrids (electrons generally traveling faster than the ions), and inward when the electrons are recirculating outside of the magrids (electrons traveling slower and reversing). In other words, the effects of the outward and inward flows of the electrons in the cusps cancel out each other from the ions perspective. If anything, the escaping ions may be pulling the escaping electrons a little higher before they fall back due to the effect of the positively charged magrid. Again, I think this effect is trivial due to my assumption that the numbers of escaping ions is much lower than the numbers of escaping/recirculating electrons. This is because the magnetic confinement of the electrons is much poorer than the secondary electrostatic confinement of the ions by the potential well.
Concerning opinions about debye sheaths, ambipolar flows, etc. A. Carlson's interpratations and R. Nebel's interprations are beyond me. Both are knowledgable plasma physisists. I choose to place more faith in R. Nebel's limited revelations, because he has access to the data (and because of my fanboy bias).
the same number of ions and electrons leave the cusp,
Not likely. If the cusps were quasineutral WB confinement wouldn't work the way Nebel and Bussard have described.
Since overall losses aren't anything like ambipolar, there's no reason to think cusp losses would be ambipolar either. Remember, in a reactor we're constantly pumping in 10MW of electrons.
But, based on some numbers- magnetic containment holds the electrons for ~1000-10,000 passes
1e5 (100,000) according to Bussard.
As fast particles will, presumably, be emitting radially, they'll be following the same paths out as everything else. The notion that these wouldn't interact at 'non-central' radii is bizarre and is very wishful wishful thinking.
How much energy do they lose in 1,000 transits, moving at MeV? Rick appears to believe the answer is "very little" but I'd be curious if anyone else has attempted to calculate this.
Shouldn't fusion products have enough energy that, even if they hit the B-field perfectly perpendicularly, that they have an effectively infinite gyroradius and punch right through? And, if they hit a cusp, won't the particle simply go through the cusp and break free as the field lines bend back behind the magrid?
TheRadicalModerate wrote:Shouldn't fusion products have enough energy that, even if they hit the B-field perfectly perpendicularly, that they have an effectively infinite gyroradius and punch right through? And, if they hit a cusp, won't the particle simply go through the cusp and break free as the field lines bend back behind the magrid?
Don't worry none. 'Polywell' appears to be a device that can defeat the laws of physics so long as enthusiasts talk long enough about what they want the electrons and ions to do. Eventually, the electrons and ions will hear the messages and obey (once an actual experiment to measure them happens, of course).
alexjrgreen wrote:The electrons that leave through the centre cusps get recirculated by the magrid. The ions that tag along for the ride mostly get recirculated too.
First, there is no indication that the fuel ions leave in any significant number, nor that they recycle at all. Those fuel ions that have NOT been upscattered stay in the well. Those that have are lost to the chamber wall. Simple. Why complicate it by assuming tag-along behavior and recirculation by ions that seems contrary to physics?
TheRadicalModerate wrote:Shouldn't fusion products have enough energy that, even if they hit the B-field perfectly perpendicularly, that they have an effectively infinite gyroradius and punch right through? And, if they hit a cusp, won't the particle simply go through the cusp and break free as the field lines bend back behind the magrid?
Don't worry none. 'Polywell' appears to be a device that can defeat the laws of physics so long as enthusiasts talk long enough about what they want the electrons and ions to do. Eventually, the electrons and ions will hear the messages and obey (once an actual experiment to measure them happens, of course).
.. i rather thought we were relying on fast fusion products exitting the system - in order to beat thermalisation/neutrality.
surely a lot depends on what asumptions we make about the density and temperature: the statistical distributions of all those collisions occuring at various locations within the structure - in turn depending on the phase of the machine (time). losses at the cusps would seem heavily sensitive to these factors, no?
so, are we are trying first to define/agree the critical operating point, then second, a means of ensuring the system cycles through it/holds it?
The electrons that leave through the centre cusps get recirculated by the magrid. The ions that tag along for the ride mostly get recirculated too.
They can't be pulled out by the electrostatic force and also pulled back in by the electrostatic force.
Electron recirculation makes sense because the electrons want to orbit the Magrid casings. The ions don't, and they aren't going to do it just because the electrons are. They want to do the opposite of whatever the electrons are doing.
