Does the Gyroradius of an Alpha fit in a cusp?

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

TallDave wrote:I think you're referring to this:
Design, building and parametric testing of WB-
7 and WB-8, the final two true polyhedral coil systems, with
spaced angular corners, to reduce “funny cusp“ losses at the
not-quite-touching points, and yet provide very high B fields
with conformal coil surfaces. These would be topologically
similar to the original WB-2 and PZLx-1, but without their
excessive unshielded surface losses, and with pure
conformal coils and small intercept fractions. These latter
can be achieved by appropriate spacing between the corner
junctions (typically several gyro radii at the central field
strength between adjacent coils) to allow free circulation of
electrons and B fields through the “funny cusp“ regions,
without direct B field line impact on or intersection with the
coils themselves.
I assume you meant to say the coil casings are spaced, not the cusps. The idea seems to be that since the electrons are going out the cusps anyway, best to keep the casings out of their path.

But remember, the electrons are trying to get to the Magrid; that's why they spiral along the field lines. The alphas, otoh, want to get away from the Magrid. A simulation would be good to see, but I think they'll tend to bounce off the regions of stronger field/charge and head along their merry way.
i _did_ a number of simulations that showed this. the alphas are well contained electro-statically. which i suppose is not all too surprising. you're essentially making an inside-out atom, except the orbiting particles have much higher mass, and the center is 'fuzzy', so to speak, so you don't have quantum uncertainty and interference effects causing the probability waves to cancel out in the inside, and what you get is more like what classical physics predicts the structure of the atom to be (prior to bohr) - except the center has essentially zero volume, so the system doesn't mechanically collapse... etc.

and my sims clearly showed electrons are hot on the outside and cold on the inside, whereas for ions its reversed. the reasons being straightforward and elementary : the electrons are effected by the magnets, which are on the outside. and the ions by the electrons, which are on the inside. by the time an ion makes it to the magrid, chances are it had barely enough energy to get there, and its kinetic energy now is nearly zero and its about to fall back into the core. so like i said the sputtering there is minimal. but dan mentioned the fusion ions - the products of a fusion reaction - and those ofcourse will have very high energies from the get go and their initial direction will be pretty much random. so i'd say the default scenario is you lose as much of their energy to collisions to the thickest part of a the magrid covers a sphere centered at the origin. and that product ion KE is your power output. so if your final power out is 90k and the grid covers 10% of the spherical surface area, that means the grid is absorbing 10k worth of kinetic energy from the product ions. and i imagine in high power systems, that could be pretty substantial. to the point where you'd have to worry about the lifetime of the coil casing and how hot it gets from these collisions. and the latter esp. if you're using superconducting magnets.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

There is some confusion (of course it is not mine). If the alphas- fusion ions in the case of P-B11 fusion., cannot reach the magrid where the magnetic surface is nearly perpendicular as N. Nebel claimed, provided the B field and machine diameter is adequate is straight forward. My ..er... confusion comes with the cusps where the magnetic surfaces are nearly parallel to the center and the magnetic surface is closer to the physical magrid. I assume that the magnetic field strength at the magnetic surface is similar everywhere due to the Wiffleball inflation to Beta=1. Any charged particle will gyrorate about this Wiffleball border until they are scattered deeper into or out of the magnetic domain. This scattering concern for the fusion alphas is unimportant because the Coulomb collisionality at several MeV energies gives MFP's much longer than the diameter of the machine. In areas where the field is perpendicular to the center the alpha is turned and flies back towards the center, but beyond a certain angle as the alpha enters a cusp it will escape with a corkscrew motion through the cusp. Because of Gauss's Law the alpha (or any charged particle)is not effected much (if at all) till it has passed the midpoint of the cusp- magrid minor radius. At that point the distance between the magnets starts widening so if the alpha has made it that far any repulsion now acting against it is not important (or of minor importance) in this regard.

The basic question is the combination of magnetic strength and distance from the Wiffleball edge to the physical magnet surface in the two comparative areas- the cusp and non cusp regions. The gyrofrequency is dependent on the particle energy and the turning magnetic energy. In the cusps there is a zero magnetic strength area in the mid line of the opposing magnetic fields. If the geometry- nature of how the alpha enters the cusp results in most of the alphas entering on this very weak ( and very small width)magnetic area, the gyrofrequency may be so long that the alpha does not have much chance to travel along a corkscrew path before it is through the cusp. This would relieve the concern. I'm not sure how this would be represented mathematically. It is dependent on very rapid fall off of the magnetic field in the center of these ' funny' cusps. This would be consistent with the magnets being very close together. The things that conflict with this solution is that the magnets are not lines but have a actual real diameter, thus the maximum field is not at the coil casing surface, but deeper within the can- at the center. Then you have to consider the additional can thickness needed for cooling. Unless it is a Bitter design these structures will be outside the magnet windings, and the Bitter arrangement would only help some.

Also, the electron picture is representative. I have not seen any concern for the electrons hitting the magnets except in the cusps. That implies to me that they stay well away from the magnets in other portions of the Wiffleball surface area, but not in the 'funny' cusps. There is electron exposure to the magrid casings away from the cusps, but this is acceptable and unavoidable ExB drift. The KE of the electrons will be perhaps ~ 10-20 times less than the fusion alphas, and the gyroradius is additionally smaller due to the much less mass/momentum of the electrons. The alphas are a different story. My impression is that these edge or funny cusp alpha surface exposure is small (but not absent) enough to be a minor irritant, or the higher energy and tendency for the alpha to exit in the center of the cusp where there is little B field and thus very little effect (turning) of the alpha, so that it maintains a more linear path. This latter reasoning seems likely wrong and I have little confidence in its reality or applicability.

Of course the electrons also hit the magnets, through nubs or other unshielded surfaces ( has to be less than 1/10,000 th of the total surface area, and there are further electron losses due to ExB drift. That alphas may reach inadiquatly shielded surfaces is not a direct problem for fusion efficiency, it is more of an engineering concern for local heating and spalling , and the proportion of alphas that reach areas where direct conversion can be applied.

Of course D-D or D-T fusion also produces high energy charged particles, in addition to perhaps higher energy neutrons. so they would have similar or worse spalling and local heating concerns. Even if my concerns about the narrow cusps applies, the P-B11 reactor may have net fewer concerns in this regard than D-D of D-T machines. It is just that the the direct conversion may not be as efficient as anticipated. Or, perhaps that is part of why direct conversion is already predicted to be ~ 80-90% efficient instead of 99-100% efficient.

Dan Tibbets
To error is human... and I'm very human.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

DTibbets wrote: Because of Gauss's Law the alpha (or any charged particle)is not effected much (if at all) till it has passed the midpoint of the cusp- magrid minor radius.
I've never been quite sure I can buy that given that the Magrid isn't all that spherical. I suppose one could calculate it to be sure, but does it really seem reasonable that a charged particle sees no force from the Magrid in the interior even very close to the midpoint of the funny cusps where the corners of three coils meet?
Roger wrote:I thought Rick said alphas exit out the 8 tri faced cusps, the truncated corners of the cubes.

?
Did he say that? I thought he just said cusps. I'll dig around.

...as best I can tell he only said they exit from a cusp, but interestingly he did talk about the alphas hitting the magnets.

-------------------------------------------------
Posted: Tue Apr 07, 2009 12:34 am Post subject: viewtopic.php?p=18445&highlight=#18445
...
4. As for Mr. Tibbet’s questions relating to alpha ash, these devices are non-ignited (i.e. very little alpha heating) since the alpha particles leave very quickly through the cusps. If you want to determine if the alphas hit the coils, the relevant parameter is roughly the comparison of the alpha Larmor radius to the width of the confining magnetic field layer. I’ll leave that as an “exercise to the reader” as well.

Posted: Mon Apr 06, 2009 8:08 pm Post subject: viewtopic.php?p=18441&highlight=#18441p ... te&p=18441
The ions also show some magnetic confinement under reactor conditions. Run the numbers. The ion Larmor radii are also smaller than the device size, as are the alpha particles' Larmor radii.

Posted: Mon Apr 06, 2009 4:06 pm Post subject: viewtopic.php?p=18432&highlight=#18432p ... te&p=18432
The alphas make about 1000 passes before they exit through the cusps. They leave at essentially full energy.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

TallDave wrote:The alphas make about 1000 passes before they exit through the cusps.
Why are you so sure that they will make 1000 passes before escaping and much earlier will not occur something like this: http://en.wikipedia.org/wiki/Spicule_%2 ... physics%29 ?

Betruger
Posts: 2321
Joined: Tue May 06, 2008 11:54 am

Post by Betruger »

Dunno about "so sure" but that order of magnitude came from Dr Nebel IIRC. Next question is - what does he know that you don't?

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

Betruger wrote:Dunno about "so sure" but that order of magnitude came from Dr Nebel IIRC. Next question is - what does he know that you don't?
With all respect for Dr. Nebel, there is not an argument: "He has told".
Earlier I spoke about possibility of occurrence of instabilities responsible for particles losses. And got an answer like that: as particles in Polywell are in potential well, they can not be lost via instabilities.
Now I am asking: is Sun’s plasma in a very deep gravitational potential well?
Nonetheless does its significant part loss or no?
What Dr. Nebel said?

ladajo
Posts: 6258
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

Joseph Chikva wrote:
Betruger wrote:Dunno about "so sure" but that order of magnitude came from Dr Nebel IIRC. Next question is - what does he know that you don't?
With all respect for Dr. Nebel, there is not an argument: "He has told".
Earlier I spoke about possibility of occurrence of instabilities responsible for particles losses. And got an answer like that: as particles in Polywell are in potential well, they can not be lost via instabilities.
Now I am asking: is Sun’s plasma in a very deep gravitational potential well?
Nonetheless does its significant part loss or no?
What Dr. Nebel said?
Joseph,
Dr. Nebel based his statements on his experimental observations of WB7 and WB7.1. There is no argument to be made, only reporting of observations.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

ladajo wrote:Joseph,
Dr. Nebel based his statements on his experimental observations of WB7 and WB7.1.
Did you see? I did not.
But instead I heard something about particles losses.

Dan Tibbets states that losses occur only because of imperfection of magnetic field near the cusps, small sizes of reactor and I “stuck” on instability.
Now I have read about instabilities on the Sun with deeper potential well and am asking again: are they possible in Polywell?
Where Dr. Nebel says that no?

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

Joseph Chikva wrote: Now I have read about instabilities on the Sun with deeper potential well and am asking again: are they possible in Polywell?
Where Dr. Nebel says that no?
strictly speaking everything is possible, esp. when you take quantum physics into account. the question is whether the likilihood is substantial enough to merit design or control changes.

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

happyjack27 wrote:
Joseph Chikva wrote: Now I have read about instabilities on the Sun with deeper potential well and am asking again: are they possible in Polywell?
Where Dr. Nebel says that no?
strictly speaking everything is possible, esp. when you take quantum physics into account. the question is whether the likilihood is substantial enough to merit design or control changes.
Instabilities are not a subject of quantum physics.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

Dan Tibbets states that losses occur only because of imperfection of magnetic field near the cusps, small sizes of reactor and I “stuck” on instability.
Yes, this is what decades of experiments with mirror machines tends to indicate.
Now I have read about instabilities on the Sun with deeper potential well and am asking again: are they possible in Polywell?
Where Dr. Nebel says that no?
I'm fairly certain Rick never addressed solar flares in relation to Polywells, probably because the comparison doesn't make much sense. But thanks for a laugh.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

TallDave wrote:
Dan Tibbets states that losses occur only because of imperfection of magnetic field near the cusps, small sizes of reactor and I “stuck” on instability.
Yes, this is what decades of experiments with mirror machines tends to indicate.
So, mirror machines provide better confinement than for example toroidal?
Thank you for laugh.
TallDave wrote:
Now I have read about instabilities on the Sun with deeper potential well and am asking again: are they possible in Polywell?
Where Dr. Nebel says that no?
I'm fairly certain Rick never addressed solar flares in relation to Polywells, probably because the comparison doesn't make much sense.
So, there is a possibility of instabilities occurring (solar flares, solar wind) in the Sun and not in Polywell? Is there in the Sun more deeper potential well or no?
Thank you for laugh again.
AlexK wrote:Generally, once beam particles are above ~50 kV, thermalization happens at a rate that causes ion/energy losses which are smaller than the energy production rates due to fusion, unless there are instabilities (such as the 2-stream or Weibel). This is quite a general result.
viewtopic.php?t=3022&postdays=0&postord ... n&start=15

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

So, mirror machines provide better confinement than for example toroidal?
No one said that (mirror machines have cusp losses, which again has been known for decades), they're just more stable against some classes of instability that tend to plague toks. But thanks for the further hilarity.
So, there is a possibility of instabilities occurring (solar flares, solar wind) in the Sun and not in Polywell? Is there in the Sun more deeper potential well or no?
The Sun is neither electrostatically nor magnetically confined to any significant degree, it has laughably small power production for its mass, and spicules are not composed of alphas, nor does alpha sputtering of the Magrid have much to do with plasma stability -- but again, very amusingly irrelevant to the question of the alphas' Larmor radius.

Humor aside, some of us were trying to have a constructive discussion here about whether fusion product alphas will hit the Magrid on the way out a funny cusp, so probably best I cease responding to this. If you want to start a thread on how the Sun is like a Polywell, please feel free, and I wish you the best of luck with that. :)
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...

Joseph Chikva
Posts: 2039
Joined: Sat Apr 02, 2011 4:30 am

Post by Joseph Chikva »

TallDave wrote:
So, mirror machines provide better confinement than for example toroidal?
No one said that (mirror machines have cusp losses, which again has been known for decades), they're just more stable against some classes of instability that tend to plague toks. But thanks for the further hilarity.
AlexK wrote:Generally, once beam particles are above ~50 kV, thermalization happens at a rate that causes ion/energy losses which are smaller than the energy production rates due to fusion, unless there are instabilities (such as the 2-stream or Weibel). This is quite a general result.
viewtopic.php?t=3022&postdays=0&postord ... n&start=15[/quote]
So, FPGeneration observed at least two types of instablities. And you are wrong.
TallDave wrote:The Sun is neither electrostatically nor magnetically confined to any significant degree,
What's the difference of confinement method? Neither electrostatically nor magnetically but confined gravitationally. Instability accelerates particles till energies exceeding the depth of potential well. And good buy.
Good luck you too.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

DTibbets wrote:As far as sputtering, the elecrtrons have relativelty low momentum, aso even though the may hit the magnets due to cross field trasnport while at their higest energies the effect will be minor compared to ions.
I'd never heard of electron sputtering raised as an issue, but I think they're expecting the Magrid to absorb something like 5-10MW of them. Not sure what if any materials issue that raises, or if this sort of question comes up in any other sort of device (I can't think of any offhand). Maybe Simon or Tom knows?
happyjack wrote:i _did_ a number of simulations that showed this. the alphas are well contained electro-statically.
I'm assuming you meant the fuel ions were contained, as opposed to the alpha fusion products, which should definitely not be contained :)

Did you ever simulate alphas from the p-B11 reaction? It would be interesting to know if you came up with a similar number of passes and similar expectations re their energy when they exit.
happyjack wrote:so i'd say the default scenario is you lose as much of their energy to collisions to the thickest part of a the magrid covers a sphere centered at the origin
Apparently the collision cross-section at their energy is too low for that according to the EMC2 guys. But I think that can be calculated, given that we know their energy spread and roughly the operating density of a reactor. Or, hell, we could probably calculate it for WB-8.1.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...

Post Reply