Reference sketch

Discuss the technical details of an "open source" community-driven design of a polywell reactor.

Moderators: tonybarry, MSimon

hanelyp
Posts: 2255
Joined: Fri Oct 26, 2007 8:50 pm

Reference sketch

Postby hanelyp » Sun May 06, 2012 9:50 pm

A rough sketch of my understanding of the arrangement of a polywell with direct conversion:
Image

Surrounding the familiar toroid cluster magrid is a spherical outer grid. Around that is a shell for the alphas to hit.

The overlaid blue line is a graph of electric potential vs. radius. Voltage is on a logarithmic scale.

An excess of electrons confined by the magrid produce the innermost potential well, which confines fuel ions.

Electron injectors would be placed just inside the dashed line grid. Electrons escaping the wiffleball, unless up-scattered, would have insufficient energy to climb the potential hill to reach the grid, and would be directed back to the magrid.

Fusion product ions escaping the wiffleball would be accelerated a bit between the magrid and the outer grid, mostly miss the outer grid, then be slowed climbing the potential difference to the collector shell.

Fuel could be injected as gas from injectors on the inside surface of the magrid, or from ion accelerators at the outer grid.

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Postby KitemanSA » Mon May 07, 2012 2:23 pm

Yup, pretty much comports with my understanding too.

The blue line can only be a logarithmic scale if there is a positive offset voltage at the "dashed line grid". Even so, the well in your sketch still shows up with an approximately linear proportion to the MaGrid voltage.

If you assume the offset is 10V and the Magrid is 10,000V then the bottom of the well (~1000V) would be 1/4 as deep as the "dashed line grid" depth.

Also, there is supposedly a peak back up in the very center.

wacker.popeln
Posts: 41
Joined: Fri Apr 06, 2012 5:38 pm

Postby wacker.popeln » Wed May 16, 2012 3:44 am

:( hmmm

i cant help it

the magnets are too close together.

lets assume 8 tesla
or just 4

which you would need wiht a bigger denser electronball

it would mean hell on the coils on such a distance

-what would be the ideal distance for a 4 tesla coil and an 8 tesla coil (per material unit)
and
-what does this mean for hte necessary electronball density and hte surrounding catcher
-wiht any cooling installed this cannot work already at these distances

this one is out of proportion in think


opinions?

Tom Ligon
Posts: 1871
Joined: Wed Aug 22, 2007 1:23 am
Location: Northern Virginia
Contact:

Postby Tom Ligon » Thu May 17, 2012 8:04 pm

You have the right general idea if you are just introducing electrons. Ions complicate it a bit.

The central well may be more flat-bottomed and steep-sided than you drew it. The converging ions in the center will partly or even totally cancel out the electron potential well, causing a central cusp. For p-B11 it is important to control the height of the cusp to no more than about 15% of well depth, or bremsstrahlung losses start to mount. This is because the ions start to attract the electrons, which otherwise have very low energy at the bottom of the well.

For the electrons, the well is a "potential hill."

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Postby KitemanSA » Thu May 17, 2012 8:21 pm

The few well depth graphs that I have seen from Dr. B did seem to have a flat bottom, but the graphs were deceptive in that the radius was generally plotted with a logrithmic scale. thus , if it were mostly flat across the inner 30%, it may look more like the middle 75%

But I suspect you have seen many more graphs than I! ;)

Tom Ligon
Posts: 1871
Joined: Wed Aug 22, 2007 1:23 am
Location: Northern Virginia
Contact:

Postby Tom Ligon » Fri May 18, 2012 3:22 pm

Actually, I think we've all seen the same graph. He used it over and over and over again! ;)

krenshala
Posts: 913
Joined: Wed Jul 16, 2008 4:20 pm
Location: Austin, TX, NorAm, Sol III

Postby krenshala » Fri May 18, 2012 4:15 pm

Tom Ligon wrote:You have the right general idea if you are just introducing electrons. Ions complicate it a bit.

The central well may be more flat-bottomed and steep-sided than you drew it. The converging ions in the center will partly or even totally cancel out the electron potential well, causing a central cusp. For p-B11 it is important to control the height of the cusp to no more than about 15% of well depth, or bremsstrahlung losses start to mount. This is because the ions start to attract the electrons, which otherwise have very low energy at the bottom of the well.

For the electrons, the well is a "potential hill."

That makes me wonder if that line would better be drawn flipped vertically. ;)

Tom Ligon
Posts: 1871
Joined: Wed Aug 22, 2007 1:23 am
Location: Northern Virginia
Contact:

Postby Tom Ligon » Fri May 18, 2012 5:13 pm

The last presentation I did I put in a "potential hill" slide. I think that presentation can use some refinement, but it gives someone trying to understand these things some insight.

I like using simple analogies. For this one I suggest looking at electrons as pieces of gravel spewing from a gravel gun. A single piece will arc up nicely to whatever height it can achieve from a particular initial velocity working against gravity, and make a nice parabolic arc to land back on the ground with the same speed, just down. But at the top of the arc the vertical speed reaches zero.

Keep spewing gravel at one point, and a pile starts to build. The pile grows until it hits he peak of that parabolic arc, and no higher. The pile spreads out as gravel slumps down the side.

That's the potential hill, and one thing it does is show that the electrons at the top are at low kinetic energy.

Then turn the thing around and see it from the ions point of view. Then see what happens as the ions pile up.

When talking to kids, there are some nice analogies to skateboards and a half-pipe, too. Kids understand skateboard dynamics instinctively.

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

Postby D Tibbets » Fri May 18, 2012 11:02 pm

Without drawing, I will try to describe my impression. There is some variation depending on conditions- before or after ions injected, etc. But basically within the magrid is a square ( before ions) electron dependent potential well, after ions the well is more parabolic, perhaps with several shelves and a central virtual anode (~ 15% decrease in the negative potential). At the magrid the potential changes quickly from negative to positive due to Gauss Law and the charge on the Magrid. transiting this region is escaping electrons and fusion ions at millions of eV. These charged particles will be fanning out along field lines exiting the cusps depending on their energy and momentum. The electrons will be curving most and so can be picked off by grounding plates surrounding the cusp.At greater radii plates will be negatively charged and protected magnetically either by the Magrid B field and/or by local B fields. Gauss Law prevents acceleration of the positive ions till they pass them, then they are decelerated and thus curve more so that they can ground on further plates. The process continues till most of the KE is harvested. The number of suceeding plates/ grids depends on how much effort you can apply to pick off the incremental energy of the mixed energy ions. With P-B11 a two grid system may pick off most of the energy of the ~2.4 MeV alphas, and the second grid picking off most of the remaining 2.5 MeV of the more energetic alpha. I suspect the grids will be open cones surrounding the face point cusps and the corner cusps as this might integrate with the existing Magrid magnetic fields best.

I suspect a cross section of the electric charge outside the magrid may be more stepwise ( and ~ 5-10 times the voltage on the magrid). Serous considerations of Gauss Law , magnetic field interactions, etc. will be required to design an efficient system. The DPF has the advantage that the fusion ions are in a single beam and pulsetile. This considerably simplifies things. The Polywell may have a minimum of ~ 14 less collimated beams- one for each cusp. The fusion ions escaping in the 'funny cusp' regions may not lend themselves to direct conversion only with considerably greater effort. And I suspect these ions may hit the magrid anyway as their gyro radii would seem to be greater than the spacing between the magnets.

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

wacker.popeln
Posts: 41
Joined: Fri Apr 06, 2012 5:38 pm

Postby wacker.popeln » Sat May 19, 2012 6:09 pm

this has implications for a realigned fusion spot. if its not a round ball then the event horizon as fusion spot would determine the angles fo the injectors. no doubt.

mvanwink5
Posts: 1808
Joined: Wed Jul 01, 2009 5:07 am
Location: N.C. Mountains

Postby mvanwink5 » Sat May 19, 2012 6:22 pm

wacker.popeln wrote:this has implications for a realigned fusion spot. if its not a round ball then the event horizon as fusion spot would determine the angles fo the injectors. no doubt.
Sorry, but fusion spot and event horizon doesn't square with the physics behind polywell as I understand it. Maybe you could could illuminate with your understanding of the physics and with some description of what you think is going on inside the "whiffle ball," which is not actually a ball as one would think of a ball as spherical.
best regards
Near term, cheap, dark horse fusion hits the air waves, GF - TED, LM - Announcement. The race is on.

wacker.popeln
Posts: 41
Joined: Fri Apr 06, 2012 5:38 pm

Postby wacker.popeln » Sat May 19, 2012 9:01 pm

the electronball is nothing but an accelerator

the self confining design of polywell is smart enough if it works

since hte core attracts the nuclei you inject accelerates them
the problem is you catch electrons on the way
or too many cumulate and discharge - soyou lose a part of your electron ball

my idea was simple
lets shoot the nuclei that are barely influenced by electrons through it and hit teh other nuclei before hte electronball in the event horizon of hte electron confinement
this way it would save also bremsstrahlung

it was a thesis of mine.
check teh start of topic
"realign the fusion spot" including graphic

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

Postby D Tibbets » Mon May 21, 2012 10:14 pm

I still do not appreciate your terminology. The 'electron ball' is an accelerator, for the ions introduced at the edge, once inside the electron ball has little effect on the ions (with some adjustments depending on the shape of the potential well). The ions that have been accelerated towards the center in isolation (non collisional plasma)to form a tiny condensed core- point like volume, works, but with useful numbers of ions (collisional plasma) a vertual anode forms and this acts as a decelerator or diverter that prevents a point (event horizon?) limits the size of the core to larger volumes. If the ions converged (were focused) to a tiny , almost point like volume in the center (or anywhere else for that matter) the density of positive chagred ions would be so great that virtual anode would not be 15-20% of the potential well, but very much greater than the potential well. This would destroy the potential well very quickly.The 2008 patent application did mention a ' black hole' like effect where the ions condensed to such a tiny volume that the ions would have a zero chance of escaping before fusion. This was a speculative assertion though with the recognition that reaching such conditions was impossible, at least without accelerating the ions to many millions of eV which would cost to much energy and would open up another new set of problems (like photodisentigration). A pulsating inertial confinement machine might do this but only for very brief period of time. Things like two stream instability ,etc. precludes this working profitably except in those machines where the inertia of the ions and very high densities are the dominating considerations (ie: a bomb). Polywell is a steady state machine, so that the density, temperature and confinement time working together allows for the claimed positive Qs. What you are conceiving may be more like inertial confinement fusion. These can only operate for short time intervals like bombs, or laser inertial confinement fusion. While not impossible, the difficulty and cost of this approach is represented by the National Ignition Facility, and perhaps Dense Plasma Focus (opposite ends of the scale). The idea is that the much higher densities compensate for the much lower confinement time. Also,the polywell may actually work with almost no confluence of the ions. This is similar to the Tokamak, except the increased density again makes up for the shorter confinement times. Again the triple product consideration.

As for a focus of ions outside the Wiffleball I don't think this is possible with a quasi spherical machine. A cylindrical device could certainly accelerate ions, but not have a central focus. This would be equivalent to a Polywell with out any convergence. A Tokamak would fit this description. The difference is that the Pollywell can reach densities ~100-1000 times greater, and this directly determines the necessary confinement time needed for adequate fusions to occur. The triple product consideration is critical, and all fusion approaches uses tradeoffs to attempt to achieve this balance without expending too much energy. A Beam- Beam approach does not work, because a dense enough focus cannot be achieved without expending too much energy. The Polywell may operate in a similar manner, except it cheats. If the ions do not fuse they scatter outward in a spherical fashion from near the center and are recycled for thousands of additional attempts, and importantly do this without needing much additional energy input. If this could be done in a cylindrical fashion the multipass confluence would not be as great as the center is defined as a line along the cylinder instead of a point (spherical volume) in the center of a sphere. The density gains are less.

As for injection of electrons or ions from external guns, they have to be aligned well with the cusp axis or more of the charged particles will be mirrored back without entering the machine. This costs more and may limit the internal densities obtained compared to the external densities- which is a limiting factor.

There are other considerations that apply, but basically the Polywell traps ions at higher density (whether there is convergence or not) than Tokamaks can, and through clever tricks do so so at acceptable costs. This results in useful fusion rates at much smaller machine costs and size.

Assuming all the various approaches work, I wonder if the Polywell will be a more condensed power producer than say a DPF. The latter may have the advantage if only a few MW is needed as one machine could produce this. But for a few hundred MW, a single Polywell may suffice, while perhaps ~one hundred DPF machines would be required. Compared to Tokamaks this consideration may not apply as Polywells can be scaled up quickly, while DPF cannot. And the baseline Tokamak is already too large for typical grid applications.

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

wacker.popeln
Posts: 41
Joined: Fri Apr 06, 2012 5:38 pm

Postby wacker.popeln » Sat Jun 09, 2012 5:21 am

lets take a look at the polywell from a different angle

and lets assume, that it discharges because it catches electrons.

lets assume the electrons s q u e e z e the nuclei together
wouldnt this mean the whole concept is different

that the low tesla rate is a problem
or that the amount of electrons increasing squeezing more nuclei together would make hte machine more effective?

thats what i mean

WHAT I F THE POLYWELL SQUEEZES THE NUCLEI WIHT HTE ELECTRONBALL instead of assuming its just an acceelration ball

-how many tesla would we need fior a certain size of elctron ball
-HOW MANY ELECTRONS DO WE NEED FOR A CERTAIN AMOUNT OF NUCLEI
which leads us to necessary teslas then.

right?

wacker.popeln
Posts: 41
Joined: Fri Apr 06, 2012 5:38 pm

Postby wacker.popeln » Sat Jun 09, 2012 5:24 am

isnt this a
************************************************************
NUCLEI SQUEEZER machine that uses electrons to squeeze nuclei simply -

which has implications on how many electrons you need per nuclei amount - which has implications on the tesla field
**************************************************************
this is a complete new equation line
can be diagonal line or an exponential function

Image

http://s12.postimage.org/il2lumfwt/polywellquestion.png


Return to “Design”

Who is online

Users browsing this forum: No registered users and 0 guests