Interested in a fusion webinar?

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

Torulf2
Posts: 286
Joined: Fri Sep 21, 2007 9:50 pm
Location: Swedem

Re: Interested in a fusion webinar?

Post by Torulf2 »


hanelyp
Posts: 2261
Joined: Fri Oct 26, 2007 8:50 pm

Re: Interested in a fusion webinar?

Post by hanelyp »

Plugged D Tibbits' proposal into my code and got
Image
Assuming the estimate of plasma pressure distribution is reasonable we have a nice nearly spherical well when run in wiffleball mode. And 2 full line cusps.
The daylight is uncomfortably bright for eyes so long in the dark.

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

Re: Interested in a fusion webinar?

Post by D Tibbets »

Pretty much, though the thingies sticking through the end magnets seem intrusive. If there ends up being plasma guns at these point cusps, I suspect they would not extend much, if any, beyond the midplane of the end magnets (much like in the Mini B machine. that Dr Parks described). This illustrates a potential problem with this design, there are only two point cusps available for injection of plasma, ions, or electrons. I suppose the two line cusps could be used, though I am uncertain of how injectors would be situated.

Also, I think plasma guns locatated in such ways is fine for a research machine, but has serious consequences for a working reactor, both in terms of recirculation, direct conversion, repellar plate interference (like in WB5, etc.). This is one area where I think neutral gas injection is attractive. It need not be through a cusp and thus does not directly interfere with cusp behavior. Again, it boils down to electron injection capacity/ efficiency in order to allow this approach.

One feature that is attractive to me is the fan shaped cones of the two line cusps- this would seem to be easier to surround with direct conversion grids compared to the 8 spiky corner cusps (or more with a higher order polyhedra) and 6 point cusps. I fantasize about one line cusp fusion product escape cone being focused for rocket thrust, while the other is direct converted for power. Any combination of forward or reverse thrust, or minimal thrust with maximum power harvesting might be possible- useful for your space ship acceleration and deceleration while keeping your deflector/ armour located at the front of the ship, oriented in the direction of motion.

And, the left line cusp is illustrated, but the right is absent/ feeble.

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

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

Re: Interested in a fusion webinar?

Post by D Tibbets »


Here is a shameless and poor modification of Toruf2s image to better emphasize my appreciation of the cusp shapes. It is essentially a 2 dimensional overlay of the right side. It shows a more central plasma concentration and illustrates (again poorly) the convex surfaces towards the plasma. The end magnets push more towards the center when the B field strengths are adjusted to achieve central B field null. It obliterates the three dimensional appearence of the line cone shape, but then my imaging capabilities fall far short of Torulf2s.

Image


It is interesting that the B field null or minimum can range from a dumbell along the longitudinal axis to a sphere to a torus (ring) along the perpendicular axis depending on the relative strengths of the end and central magnet. This would allow for some interesting variations in the plasma cloud which might make for some real time variations in the plasma geometry which might have some useful properties (?).

[EDIT]I should say - variations in the central and end magnets strength and/ or seperation. Obvously the magnets would be as close together as possible to minimize the width of the line cusps, while preserving as much internal volume as possible.

The end magnets probably should be smaller in major radius than the central magnet as is illistrated. This allow for better direction(?) of the line cusps, perhaps allowing for a smaller diameter vacuum chamber and advantages in direct conversion grids, along with better central focus. Also the central magnet may not need to have a circular minor cross section. An oval with as much as 1.5 times the radius in the longitudinal direction may be permissible (based on my modeling and ruminations). This allows for greater internal volume while maintaining narrow line cusps (ie:marrow magnet separation).

Dan Tibbets
Last edited by D Tibbets on Fri Jul 25, 2014 5:43 am, edited 1 time in total.
To error is human... and I'm very human.

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

Re: Interested in a fusion webinar?

Post by D Tibbets »

Another image showing B field modeling and some of my ruminations...

Image

The three ring design can be stacked though, by using one end magnet for the start of the next array.This eliminates one escape point cusp, and also further limits input opportunities.

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

mvanwink5
Posts: 2143
Joined: Wed Jul 01, 2009 5:07 am
Location: N.C. Mountains

Re: Interested in a fusion webinar?

Post by mvanwink5 »

Tom Ligon wrote:If I had a clue that they were interested in fusion back when I was working with them on the Morphing Air Vehicle, I would have dropped a hint about my EMC2 experience and maybe have gotten my foot in the door. And then I'd have TWO NDA's keeping me from saying anything.
I would not be surprised if LM takes an active interest ($) in EMC2 progress, would be a nice modification to their program and a home run for the skunk team / EMC2.
Counting the days to commercial fusion. It is not that long now.

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

Re: Interested in a fusion webinar?

Post by D Tibbets »

If the Skunks Works machine is a derivative of the Polywell concept as I believe, then Dr Park's demonstration of an unequivical Wilffleball can only reenforce their position- assuming they had not already satisfied this limiting criticism.

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

DeltaV
Posts: 2245
Joined: Mon Oct 12, 2009 5:05 am

Re: Interested in a fusion webinar?

Post by DeltaV »

Looking at Dan's mod of Torulf's pic, it would seem that electron recirculation/oscillation (prior arguments on T-P about which dominates in a Polywell...) would be mainly along a distorted torus enveloping the central coil.

For propulsion applications, it might make sense to inject electrons along the axis through one point cusp. A slightly weaker field for the coil on the opposite end would encourage fusion products to exit in that direction.

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

Re: Interested in a fusion webinar?

Post by D Tibbets »

I'm not sure what you mean by a torus electron containment distribution. The electron containment would be from reflection from the Wiffleball surface just like a Polywell. The lowest B field in a low Beta situation would be in a torus/ sphere/ dumbell depending on relative central and end magnet strengths, but this does not necessarily match the electron distribution. In Bussard's comments pure electron plasmas have electrons spending most of their time near the wiffleball border, with large angular momentum components (square potential well). Only with the ions do the electrons assume a more parabolic potential well distribution. Near the Wiffleball border the shape is a spiky ball in a Polywell, and a funny flared cylindrical shape in the 2 ring machine, at least at low Beta. At high Beta the system much trend towards a more spherical shape (I think).

I think the central focus of minimal B field relates more to the symmetrical expansion of the Wiffleball border as it forms. This near spherical expansion allows for the near Beta=1 condition without "stresses"- B field bulging asymetrically, which would result in B field cusp local failure before full global Beta=1 conditions are reached.

The central focus / confluence of ions is beneficial as it increases fusion at the same edge pressure conditions, something which may be nessisary for P-B11 fusion. Also while fusion is increased, Bremsstruhlung is decreased proportionately, within limits with central confluence. I think allowing weaker cusp confinement on one side may harm this central confluence. I doubt the input benefits or direct conversion versus direct fusion product exhaust would be helped much. The exception to this may be if these variations can be timed on scales similar to the plasma frequency, probably more than a few MHtz. I doubt the B fields could be changed that fast practically, though local microwave heating, etc and Nebel POPS type effects might allow for various interesting global and/ or local interactions.My idea for selective direct power conversion versus directed thrust is that this is controlled external to the magrid. The conversion grids are organized for either energy harvesting and/ or thrust vectoring, ideally with the same structure and only switches being changed.

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

DeltaV
Posts: 2245
Joined: Mon Oct 12, 2009 5:05 am

Re: Interested in a fusion webinar?

Post by DeltaV »

D Tibbets wrote:I'm not sure what you mean by a torus electron containment distribution.
I did not say containment. I said recirculation. Vital for Polywell and (I'm assuming) also for Lockheed. So any electron recirculation trajectories (outside the plasma ball) would tend toward the surface of an imaginary, distorted "torus" enveloping the central coil. The two circular cross-section, semi-conical cusps define that surface near the plasma ball and the most probable electron recirculation trajectories define it in the outer regions, whether the salvaged electrons loop all the way around or return to the cusp they exit.

Whether doable or not, having the power exit mostly in one direction has vehicle design benefits, regardless of propulsion method* (high-Isp/low-thrust exhaust of non-p-11B fusion products, or low-Isp/high-thrust arcjet using direct conversion of p-11B alphas for relativistic electron beam heating or direct alpha heating of reaction mass).

*Excluding Woodward-Mach Effect or some other propellantless approach, where you would not necessarily want a plasma asymmetry.

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

Re: Interested in a fusion webinar?

Post by D Tibbets »

With a greater major diameter of the central magnet, the line cusp exit cone is mostly towards the end. Any charged particles would mostly (?) follow the field lines either around the central magnet or around the end magnets. I suspect that most may be around the end magnets. If around the central magnet, then a torus distribution of these charged particles is reasonable. If around the ends, more of a dumbbell with inversion?()- a warped torus perhaps. In either case this recirculating cloud would follow these patterns only if there are no external modifiers- like chamber wall, external grids or grounding surfaces. Bussard emphasized the importance of recirculation only involving reentry through the same cusp- looping around to another cusp is bad because that allows for progressive upscattering of electrons leading to runaway drive voltage requirements. A upscattered electron is lost to external grounding surfaces befor it can complete a loop. Non upscattered electrons are stoped and reversed through the same cusp by the positive charge on the magrid. Note that the energy of the upscattered electrons is recovered to the extent of the voltage of the magrid, so only the excess upscattered portion of the energy is lost to the system.

This brings up an other item that confuses me.Bussard claimed that recirculation was a key component of achieving overall electron confinement to the extent that Q >1 is possible (or at least Q great enough to be useful). Yet WB8 and 'Mini B' seem to have high voltage e- gun, with the magrid at ground. This would seem to eliminate most of the possible recirculation. Any electron that escaped would not be decellerated by the magrid and would be lost with it's full escape energy. The exception would be looping around to another cusp. But this brings up the upscatter problem. Unless there is some clever way of moderating these escape electrons energy, perhaps with external grids, the run away voltage problem would manifest. This was one of the arguements that A. Carlson used to insist that the Polywell was non viable. Recirculation results in a near 10X improvement in net effective elecron confinement. Abondoning this through using a grounded magrid seems counter productive. As such, I don't know why EMC2 seems to have followed this route of high voltage e- guns.

Mini B has a ~ 7,000 volt E-gun. WB 8 in the picture has a similar appearing "E-gun " sticking out the side. Some voltage on the E-gun is obviously needed for extraction purposes and initial focusing, but it seems that a high positive potential on the magrid would aid focus further and help prevent electron beam spread before entry (higher speed results in less time for beam spread) and subsequent mirror rejection by the B fields. A push- pull optimization would seem to be superior to a push only approach. WB6 was essentially a pull method only while Mini B was a push only. Also, without a positively charged magrid rejected electrons mirror away from the magrid without decelerating, the energy is lost from the system. The exception would be looping around continuously until somehow the electron penetrates a cusp (if the electron is on a field line that mirrors outside the magrid, it would continue to do so indefinitely, at least in a collisionless plasma, in the actual collisional plasma in these machine the interactions are much more messy). But, even if that could work it results in greater external electrons which may effect the potential well (think of WB5), and also arcing concerns.

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

Torulf2
Posts: 286
Joined: Fri Sep 21, 2007 9:50 pm
Location: Swedem

Re: Interested in a fusion webinar?

Post by Torulf2 »

A newer version due to Tibbets ideas.
https://www.flickr.com/photos/77233366@N08/14608188299/

Post Reply