Why do injected electrons not fly back out of the well?

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

Moderators: tonybarry, MSimon

Post Reply
deane
Posts: 29
Joined: Sun Jun 14, 2009 11:27 am

Why do injected electrons not fly back out of the well?

Post by deane »

One thing that I've been struggling to understand about Polywell is how the well is formed in the first place.

Electrons are fired in from outside with some initial kinetic energy. As they pass the outer edge of the magrid they will start picking up additional kinetic energy as the magrid forces them toward the center of the well. But that additional kinetic energy should be identical to that needed to climb back up out of the well and leave the grid, shouldn't it? So what keeps the electrons from flying back out?

Or is that the electrons are continuously cycling in and out of the core along the magnetic field lines, but they reach their greatest density in the center, thus creating a net negative charge there?

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

Post by ladajo »

Yes, they oscillate.
The idea to consider is that they are at the slowest and widest at the apogee, and fastest but most focused at the core. The speed change compared to the volumetric change is what drives the "concentration" at the core. Bussard and Nebel have supposed that once the machine is up and running steady state, that it may not need elctron guns, as fuel ionization would be enough to provide the e- needed to keep the Well established. This idea is still not publically proven.

If yo uhunt around on you-tube, as I recall there are some videos of sims showing electron population generation. Also, Happyjack27 did some work on that as well.

The formation of the wiffleball greatly closes the leakage points (cusps). But, those that do leak are suppossed to "recirculate" or "oscillate" back in the cusp to some degree (as argued).

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

Post by D Tibbets »

The electrons are accelerated towards the magrid. But, once they pass the magrid they are not accelerated by the magrid further. They are within the magrid and Gauss's law explains why the charge (potential) on the magrid is now noncontributory to the electrons behavior. The reason why the electons do not immediately fly out of the opposite cusp is because the beam nature of the electrons is very quickly dispersed. How much dispersal and how fast it occurs is debated, but the central negative space charge disperses the electrons from the near center area (not the exact center, that would otherwise reflect the electron back through the cusp it entered from. Once the electrons are dispersed, even by a modest amount it's chances of hitting the Willball shell in any one area is statistically equal to any other area. The electrostatically driven electron escape is thus wholly dependent on the cusp surface area compared to the Wiffleball total surface area. Plain cusp confinement in the Polywell isd claimed to be ~ 60 X. This is simply the effective cusp hole size compared to the total surface area of the magnetic field surface area. The Wiffleball effect changes this ratio to 1: several thousand.
This is a simplified discription, but it represents the net effects of of the physics involved.

The electrons are magnetically contained, despite the electrostatic repulsion.

Once the electrons do escape through a cusp to the outside of the magrid, they again see the potential on the magrid , so they can be accelerated back into the interior just as the new electron gunned electrons are. If the electrons are upscattered too much they will slow as they pull away from the magrid, but they will not stop. They will hit the wall with whatever KE they have left. According to the 2006 [EDIT 2008] patent application, they will not follow a magnetic field line to another cusp because the field lines are designed to reach the wall before they turn around. If this was not the case, some of these upscattered electrons would loop around and approach the magrid due to their own momentum. They would then be accelerated further by the magrid, and renter the Wiffleball with excess energy. This is something that is not wanted. If an escaping electron is stopped by the potential on the magrid, it will then be accelerated to a set speed very near to the accelerating potential on the magrid (all of this probably occurs within a few mm of the midline of the magrid casing, unless the electron speed (KE) is extreamly close to the magrid potential, in which case the distance traveled past the magrid increases progressively.. This serves two (actually three) purposes. It recirculates these escaped electrons at their original speed at ~ zero cost in additional energy. It saves energy , and it also does this while essentially resetting the electrons to their original near monoenergetic state. The upscattered electrons ( above ~ 20 percent of the potential well) are allowed to escape, but much of the energy of these upscattered electrons is recovered by transferring a portion of their KE to the potential energy of the magrid potential. This eliminates undesirable high energy electrons and does so at bargain prices from an energy balance standpoint.

Recall that the potential well will only be ~ 80-85% of the accelerating voltage (potential). I'm unsure of the specifics of why this is, but probably involves less than perfect focus of the electron beams as they pass through the cusps (they are dispersed some) along with the central space charge. Even on the first pass very few of the electrons will be aimed precisely through the exact center of the machine. Instead they are aimed towards a cental sphere with a volume that is some fraction of the total volume within the magrid Wiffleball. Any scattering collision here can divert the electron in almost any direction, but since most of these scattering collisions are occurring near the center, much of the radial vector component of the motion is preserved,

Dan Tibbets
Last edited by D Tibbets on Sun Aug 14, 2011 9:04 pm, edited 1 time in total.
To error is human... and I'm very human.

jcoady
Posts: 141
Joined: Fri Jul 15, 2011 4:36 pm

Post by jcoady »

You can visualize what is happening with this web based polywell 3D simulator if your browser and computer support webgl.

http://members.shaw.ca/johncoady/polywell.html

Just click the "start simulation" button to launch an electron into the polywell. Then click the "display" tab under the "start simulation" button and check the "show electric potential" checkbox. What you will see is a color map of the differing electric potential in the polywell. The electrons are attracted to areas of high electric potential, which would be the yellow, red and violet colors. You can see this a little better by checking the "Electric Potential Mesh" checkbox. Then left click and drag your mouse in the 3D window to view a 3D mesh representation of the electric potential. The electrons are attracted to higher potential and the protons are attracted to lower potential. When the electron leaves the polywell, it is moving toward a lower potential and is slowed down and pulled back towards the higher potential in the polywell. This is analogous to tossing an object in the air and it being pulled back down by gravity.

TallDave
Posts: 3140
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

So what keeps the electrons from flying back out?


One of two (or maybe three) things -- 1) they are either reflected by the magnetic fields, or 2) they find a cusp, and as they exit they see the Magrid which has a more positive charge than the wall, and on average they don't have the energy to make it to a wall.

Or, 3) because some cold electrons don't have the energy to climb out of the cusps to the wall or back up the well, they may get stuck in the cusps until they can diffuse to the Magrid, and as they accumulate there in the cusp they form a local negative potential that then also helps blocks other electrons from exiting the cusps ("cusp-plugging"). I've always envisioned this as a potential "bump" in the landscape of potentials, big enough to block a lot of electrons but not big enough to pull out significant numbers of ions. (Art never liked this notion because he believed it would suck out too many ions, but I think that probably tends to depend on the size/shape of the bump relative to where the ions are confined).

Bussard actually tried negative repeller plates at the cusps, but reported this led to large ion currents. Tuning the balance there may turn out to be important.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...

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

Post by D Tibbets »

jcoady wrote:....
The electrons are attracted to higher potential and the protons are attracted to lower potential. When the electron leaves the polywell, it is moving toward a lower potential and is slowed down and pulled back towards the higher potential in the polywell. This is analogous to tossing an object in the air and it being pulled back down by gravity.
Just to be picky, I would change higher and lower potential to more positive potential and more negative potentials. Of course this is relative to some baseline, generally considered to be ground.

Also, it is further complicated by Gauss Law considerations. Neither the electrons nor the ions see the potential on the magrid, whether it is ground, pos, or negative while they are within the near hollow sphere of the magrid. These contained ions and electrons are in their own little universe from an electric field standpoint, and are dependant only on each other for interactions (plus the applied magnetic field).
There are further considerations about the size of the holes in the magrid and the frequency of the plasma current (which I assume is nearly DC in the basic steady state machine because charged particles are moving in both directions in an even constant flux).

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

Post Reply