Aviation Week on the Lockheed Skunkworks CFR

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

Moderators: tonybarry, MSimon

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

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by D Tibbets »

mvanwink5 wrote:General Fusion has a liquid lead first wall so they have no issue with the high energy fusion 'ash.'
I'm not sure about that. Any first wall, whether, lithium, lead or stainless steel will sputter off material when hit by high energy particles, especially charged particles as they will quickly deposit their KE close to the surface. Neutrons are more penitrating so probably present less of a surface sputtering/ spalling/ flaking/ erosion problem. Lead would be the worst contaminate of the three due to the very high Z, if fully ionized.

The difference for General Fusion, may be the brief operating time. It is an inertial confinement scheme, meaning things happen very fast. There may not be enough time for substantial wall ablated material to mix into the core of the fusing plasma to an extent that it poisons the soup. I could easily be wrong. DPF is also a short pulse time machine, but erosion/ electrode material contamination of the pinch may be a problem. Here the spalling may be more the result of the electron current initiating the pinch. The dynamics are different.

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: Aviation Week on the Lockheed Skunkworks CFR

Post by mvanwink5 »

D Tibbets,
Yes, no question about the same effect of the alphas and neutrons on the liquid lead only the issue is there is no reactor damage as a result. General Fusion has yet to get reactor plasma lifetime, stability, and liquid to plasma dynamics solved and proven to the point of justifying the next step to build the full scale prototype (as far as I know). There does seem to be higher confidence in being able to reach that based on the last interview, but there is the difficulty in investor patience.
Counting the days to commercial fusion. It is not that long now.

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

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by DeltaV »

Regarding Polywell (and inferentially its Lockheed CFR offshoot), back when Rick Nebel still posted here the general consensus, based mainly on his posts, seemed to be that p-11B fusion alphas managed to avoid magrid contact and exited the wiffleball point nodes as loose beams or cones.

Direct conversion devices located outside of the magrid are considered design-optimized when the alphas are decelerated to just above zero velocity at converter contact, meaning that most of their kinetic energy has been converted to electric potential.

So, pending future research results, the magrid + properly designed direct converters + vacuum chamber walls should be safe from excessive alpha erosion.

The trick will be to keep them that way at GW extraction levels.

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

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by D Tibbets »

Indeed good direct conversion eliminates most of the spalling problem with P-B11. Wheather P-B11 of D-D or D-T or D-He3, the magrid is insulated from these charged particles to a large extent. The eharged particles wheather fuel up scattered of fusion derived ions like He4, He3, P, Tritium,etc. from the various reactions are all above the potential well retaining ability, but they still are not hitting the magnets much because of the magnetic shielding- only ExB diffusion allows them to reach the magnet, and with densities below the fusion fuel density and higher energy - thus lower collision driven ExB diffusion, the loads on the magnets is trivial, or at least small. The concerns for the magrid is the neutrons (if present) and the Bremsstruhlung X-rays.. As the ions squirt out of cusps they may be slowed by direct conversion, so the vacuum vessel wall may also exposed to relatively low sputtering loads. It is not a free ride though, there is considerable engineering challenges to make it work. Also the external electrons have to be handled or they also could cause significant sputtering.
The flux of high energy particles may have a relatively small surface area that they are heating up and sputtering, compared to a tokamak first wall. Direct conversion mitigates this considerably. And the greatest vunerable surface by far is the diverter required in a tokamak. I have seen thermal loads of up to several MW/ M2 for a Polywell of generous size (over ~3 meter magrid and perhaps 5 -6 meter vacuum vessel). The diverter in a tokamak might have to dissipated the heat and associated erosion from ~ 40MW per square meter.

Structural damage is not really the issue , certainly not for P-B11 with direct conversion. The Magrids are well protected, and the lining of the vacuum vessel/ first wall can be removed quickly (several days?). Not as convenient as a liquid first wall, but a shut down for several days every few months to few years to pull out a thin shell lining the vacuum vessel is no big deal (I think). Bussard even made arguments for the Riggatron, which would need the internal magnets of that tokamak design replaced every few months, due to the considerable D-T neutron bombardment (circa ~ 1980).

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

goldfish
Posts: 4
Joined: Sun Nov 09, 2014 5:09 pm

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by goldfish »

When I saw the Skunkworks design at first I thought it was some kind of magnetic mirror/Polywell hybrid. Now I found this drawing and it turns out to be a spheromak. Search for Skunkworks in the document.
http://fire.pppl.gov/fpa06_woodruff.pdf
Your thoughts?

paperburn1
Posts: 2484
Joined: Fri Jun 19, 2009 5:53 am
Location: Third rock from the sun.

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by paperburn1 »

I am not sure that is the design of current interest.
http://aviationweek.com/technology/skun ... or-details
I am not a nuclear physicist, but play one on the internet.

goldfish
Posts: 4
Joined: Sun Nov 09, 2014 5:09 pm

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by goldfish »

Did Skunkworks ever work on nuclear fusion before 2013? Any links?

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

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by DeltaV »

Another author who didn't get the memo about p-11B and direct conversion:

http://www.forbes.com/sites/amorylovins ... economics/

Wind power. Hah!

Mini-Polywell in the garage.

David_Jay
Posts: 57
Joined: Sat Jan 19, 2008 2:49 pm
Location: Grand Rapids, Michigan

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by David_Jay »

New article on the L-M CFR is up, link:

http://www.physics-astronomy.com/2015/0 ... PXjHk03Njr
not tall, not raving (yet...)

TheRadicalModerate
Posts: 145
Joined: Thu Oct 04, 2007 4:19 pm
Location: Austin, TX
Contact:

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by TheRadicalModerate »

David_Jay wrote:New article on the L-M CFR is up, link:

http://www.physics-astronomy.com/2015/0 ... PXjHk03Njr
Here's the Aviation Week article that's the source.

http://aviationweek.com/technology/skun ... or-details
Overall, McGuire says the Lockheed design “takes the good parts of a lot of designs.” It includes the high-beta configuration, the use of magnetic field lines arranged into linear ring “cusps” to confine the plasma and “the engineering simplicity of an axisymmetric mirror,” he says. The “axisymmetric mirror” is created by positioning zones of high magnetic field near each end of the vessel so that they reflect a significant fraction of plasma particles escaping along the axis of the CFR. “We also have a recirculation that is very similar to a Polywell concept,” he adds, referring to another promising avenue of fusion power research. A Polywell fusion reactor uses electromagnets to generate a magnetic field that traps electrons, creating a negative voltage, which then attracts positive ions. The resulting acceleration of the ions toward the negative center results in a collision and fusion.

GIThruster
Posts: 4686
Joined: Tue May 25, 2010 8:17 pm

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by GIThruster »

Thanks!
"Courage is not just a virtue, but the form of every virtue at the testing point." C. S. Lewis

TheRadicalModerate
Posts: 145
Joined: Thu Oct 04, 2007 4:19 pm
Location: Austin, TX
Contact:

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by TheRadicalModerate »

I keep staring at the patent application for this thing every so often and I think it's a lot simpler than a lot of you are making it out to be:

1) This is just a plain ol' magnetic confinement system, not IEC in any way shape or form.

2) The nod to the polywell they've made is because they're capitalizing on ion recirculation, not electron recirculation. (The electrons recirculate, too, but they're not important to the power balance of the machine.

3) The two interior coils have opposite polarity from the outer coils and generate ring cusps.

4) The recirculation is stuff that leaves one ring cusp and enters through another.

5) In the configuration below, there are actually 3 different magnetic traps, with the central one having only ring cusps and the outer ones having ring cusps and one point cusp each.

6) I think they're counting on the fact that almost everything that escapes from the central trap recirculates, so there's almost nothing that escapes through the point cusps in the magnetic mirror at each end.

7) The supports for the interior coils look like they're solid in the drawings, but if you look at the photos of the interior of the machine they're only "stalks", as they call them in the disclosure. No clue how they're directing the recirculating plasma away from them. I wonder if it could be as simple as charging them positively? You'd have a hefty electron current if you did that, but reinjecting them ought to be pretty cheap.


Here are the 3 important figures with their descriptions from the disclosure:
03-10-15 CFR Fig 5.png
03-10-15 CFR Fig 5.png (79.6 KiB) Viewed 11984 times
FIG. 5 illustrates plasma 310 within enclosure 120 that is shaped and confined by center coil 130, internal coils 140, encapsulating coils 150, and mirror coils 160. As illustrated, an external mirror field is provided by mirror coils 160. The ring cusp flow is contained inside the mirror. A trapped magnetized sheath 510 that is provided by encapsulating coils 150 prevents detachment of plasma 310. Trapped magnetized sheath 510 is a magnetic wall that causes plasma 310 to recirculate and prevents plasma 310 from expanding outward. The recirculating flow is thus forced to stay in a stronger magnetic field. This provides complete stability in a compact and efficient cylindrical geometry. Furthermore, the only losses from plasma exiting fusion reactor 110 are at two small point cusps at the ends of fusion reactor 110 along center line 115. This is an improvement over typical designs in which plasma detaches and exits at other locations.

The losses of certain embodiments of fusion reactor 110 are also illustrated in FIG. 5. As mentioned above, the only losses from plasma exiting fusion reactor 110 are at two small point cusps at the ends of fusion reactor 110 along center line 115. Other losses may include diffusion losses due to internal coils 140 and axial cusp losses. In addition, in embodiments in which internal coils 140 are suspended within enclosure 120 with one or more supports (e.g., “stalks”), fusion reactor 110 may include ring cusp losses due to the supports.

In some embodiments, internal coils 140 may be designed in such a way as to reduce diffusion losses. For example, certain embodiments of fusion reactor 110 may include internal coils 140 that are configured to conform to the shape of: the magnetic field. This may allow plasma 310, which follows the magnetic field lines, to avoid touching internal coils 140, thereby reducing or eliminating losses. An example embodiment of internal coils 140 illustrating a conformal shape is discussed below in reference to FIG. 7
03-10-15 CFR Fig 6.png
03-10-15 CFR Fig 6.png (81.14 KiB) Viewed 11984 times
FIG. 6 illustrates a magnetic field of certain embodiments of fusion reactor 110. In general, fusion reactor 110 is designed to have a central magnetic well that is desired for high beta operation and to achieve higher plasma densities. As illustrated in FIG. 6, the magnetic field may include three magnetic wells. The central magnetic well can expand with high Beta, and fusion occurs in all three magnetic wells. Another desired feature is the suppression of ring cusp losses. As illustrated in FIG. 6, the ring cusps connect to each other and recirculate. In addition, good MHD stability is desired in all regions. As illustrated in FIG. 6, only two field penetrations are needed and MHD interchange is satisfied everywhere.

In some embodiments, the magnetic fields can be altered without any relocation of the coils by reducing the currents, creating for example weaker cusps and changing the balance between the ring and point cusps. The polarity of the currents could also be reversed to make a mirror-type field and even an encapsulated mirror. In addition, the physical locations of the coils could be altered.
03-10-15 CFR Fig 7.png
03-10-15 CFR Fig 7.png (35.98 KiB) Viewed 11984 times
FIG. 7 illustrates an example embodiment of an internal coil 140 of fusion reactor 110. In this embodiment, internal coil 140 includes coil windings 710, inner shield 720, layer 730, and outer shield 740. In some embodiments, internal coil 140 may be suspending within enclosure 120 with one or more supports 750. Coil windings 710 may have a width 715 and may be covered in whole or in part by inner shield 720. Inner shield 720 may have a thickness 725 and may be covered in whole or in part by layer 730. Layer 730 may have a thickness 735 and may be covered in whole or in part by outer shield 740. Outer shield may have a thickness 745 and may have a shape that is conformal to the magnetic field within enclosure 120. In some embodiments, internal coil 140 may have an overall diameter of approximately 1.04 m.

Coil windings 710 form a superconducting coil and carry an electric current that is typically in an opposite direction from encapsulating coils 150, center coil 130, and mirror coils 160. In some embodiments, width 715 of coils winding is approximately 20 cm. Coil windings 710 may be surrounded by inner shield 720. Inner shield 720 provides structural support, reduces residual neutron flux, and shields against gamma rays due to impurities. Inner shield 720 may be made of Tungsten or any other material that is capable of stopping neutrons and gamma rays. In some embodiments, thickness 725 of inner shield 720 is approximately 11.5 cm.

In some embodiments, inner shield 720 is surrounded by layer 730. Layer 730 may be made of lithium (e.g., lithium-6) and may have thickness 735 of approximately 5 mm. Layer 730 may be surrounded by outer shield 740. Outer shield 740 may be made of FLiBe and may have thickness 745 of approximately 30 cm. In some embodiments, outer shield may be conformal to magnetic fields within enclosure 120 in order to reduce losses. For example, outer shield 740 may form a toxoid [sic]
I wonder how--or if--they're going to do ash scraping? Maybe they just evacuate and refill periodically.

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

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by D Tibbets »

I beg to differ!

Having said that I suppose I should read the patent...
Could you provide a link?

The reasons I disagree , based on limited understanding, is some of their statements and images of the plasma. The central three ring magnets, at least in one iteration of their machine has two obvious ring/ line cusps. The additional ring magnets on the ends do modify the pattern of recirculation somewhat. There are also two point cusps , and these feed into two end confinement regions with each again with a ring cusp - that is continous with the central ring cusp, and a point cusp towards the ends of the machine. That the central assembly point cusps feed into another confinement region does not negate the dynamics of charged particles escaping/ transiting between regions, but it does change what happens to them subsequently. Even with out more than the three basic ring magnets, the end point cusps are continuous externally to the two ring/ line cusps. In a simplified view this might be viewed as a contained plasma with embedded magnets. Recirculation of the plasma- electrons and/ or ions is probably more accurate than suggesting they are contained though. With the additional external magnets 'guiding' the recirculation,edge instabilities may become an issue. Provided the plasma in these regions is substantially less dense than in the primary core region, this may be tolerable.

They have mentioned a potential well and that is only possible with electrostatic containment by definition. The potential well is not only a method to accelerate a charged particle, it is a way to contain them, and expel them at the same time. If one is contained- pos. ions, the other must be expelled- neg. electrons. The Polywell tries to expel the excess electrons while electrostatically retaining the ions. There is no way to get around this. But, if you can magnetically confine the electrons, you can avoid this conundrum. The confinement of electrons is magnetic, the ion confinement is electrostatic- and is secondary to the potential well/ electrostatic conditions created by the magnetically confined electrons. You could reverse the process by confining the electrons electrostatically by magnetically confining excess ions. This is of course undesirable for several reasons. First it is the ions that fuse. You want them to be at the bottom of their potential well so that they are fastest in the center and so that the quasi spherical symmetry allows for ion confluence/ density maximum in the center. And, yes, the Lockheed machine is quasi spherical- at least by my understanding. The species that is magnetically contained will of course escape confinement by several means. For electrons this loss is primarily through the cusps. Limiting these cusp loses and / or recirculating the electrons can save considerably on the energy costs needed to inject new / replacement hot electrons. Don't forget that electrons are also lost through ExB transport across the magnetic fields irregardless of other loss mechanisms. If your cusp losses/ recirculation is good enough, the ExB losses can come to dominate the losses. The important point here is that this cross over point for electrons is much different than for ions. The ExB losses are ~ 60 times or less than for ions. The ExB losses are directly proportional to the gyroradius of the particles, and as the ion gyro radius is at least 60 times more... Even excellent ion magnetic cusp and / or recirculation efficiency would lose out to ion ExB losses long before adequate Lawson criterion were met. The only way to avoid this is by going big. This is a primary reason why Tokamaks must be so big. They do not restrict the ExB transport of ions because the ions are not held away from the magnetic field by electrostatic processes. The plasma- ions have to be forced to travel further before fully penetrating the magnetic fields. If you prefer, you can also say the ion gyro radius is small on the magnetic edge because the ions are at the top of their potential well and are cold in this region. Arguements about high Beta creating a sharp transition on the edge and thus limit initial progression of ExB diffusion may play a role, but I suspect it is minor if existent at all. For a Polywell to be small- meters instead of tens of meters, the ions must have only trivial losses through ExB (pronounced E cross B) transport. The only way to do this is to decouple the bulk of ions from the magnetic field and the only way to do this is with an electrostatic field- centrally directed potential well. Electrostatic ion confinement is absolutely essential, unless you are inventing new physics.

Because the ions are electrostatically confined by excess electrons and the electrons have to be magnetically confined to maintain the picture without humongous electron currents being needed. This implies that the escaping charged particles must be primarily electrons and efforts to minimize losses through excellent cusp geometries and recirculation of escaped electrons (or at least recovery of their energy efficiently) is the name of the game. The Lockheed's approach of layering the loss channels may help in this regard. It introduces other questions- like continued ExB electron losses, and unstable magnetic surfaces towards the electrons in some regions. Also, as the electrons must be the dominate species in these non core regions, the density of the electrons must be kept low otherwise the Coulomb repulsion would be unmanageable. This implies that in these outer regions the primarily magnetic confined electron losses are what is being recovered. The ions, with their more efficient electrostatic confinement, are mostly absent from this picture.

If electron are cusp confined for 100,000 passes or more, then escape and are mostly recovered through recirculation, the relative electron currents/ densities in the core (inside the primary magnet rings) is much greater than the external recirculating electrons of new injected electron densities. This allows for maintaining high internal densities that provide for useful fusion rates in relatively small volumes, while minimizing overall ExB losses,and perhaps very importantly minimizing edge instabilities exterior to the core, as the magnetic curvatures cannot be maintained in these outer regions. The much smaller charged particle densities (primarily electrons) in these non fusing external to the core areas help to minimize the losses through these two mechanisms. It is a case of having your cake and eating it too.

Without a potential well and the other basic properties envisioned in the Polywell, I do not see any logic to the Lockheed design.

Basic properties of Polywell are:
1) Quasi spherical geometry

2) Central virtual cathode generated by (tiny) excess of hot electron injection

3) Resultant decoupling of magnetic confinement of ions because of the electrostatic potential well created by the magnetically confined excess electrons

4) Resultant elimination of significant ion ExB losses

5) Electron cusp losses dominating over electron ExB losses, but good enough with or without recirculation to allow for adequate energy balances

6) The above requiring much better cusp confinement than is achievable with typical mirror confinement schemes

7)The above being achieved by high Beta operation which leads to so called Wiffleball operation which not only provides adequate electron magnetic confinement (perhaps with some degree of electron recirculation through the same cusp or adjacent cusps depending on design decisions). This not only allows for adequate electron magnetic confinement, but at densities that not only allows for meeting Lawson criteria, but does so at densities that allows for useful fusion yields in relatively small packages (compared to low Beta machines like Tokamaks).

8) And finally, not to be depreciated, the important maintenance of favorable B field curvature relative to the bulk of the plasma so that edge instabilities / macro instabilities are of trivial concern, even at the increased densities possible.

Things like recirculation efficiency, electron injection efficiency, bremsstruhlung issues, various engineering concerns are all important or even critical additional considerations.

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

TheRadicalModerate
Posts: 145
Joined: Thu Oct 04, 2007 4:19 pm
Location: Austin, TX
Contact:

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by TheRadicalModerate »

Dan--

Here's the main patent. There are others mentioned up-thread, but I haven't gone through those.

The patent never mentions a potential well, although it does refer to a magnetic well. (I just did a search through the text.)

tombo
Posts: 334
Joined: Fri May 09, 2008 1:10 am
Location: Washington USA

Re: Aviation Week on the Lockheed Skunkworks CFR

Post by tombo »

I like it.
I see 5 coils making 2 pairs of cusps with their equatorial loss planes bent & joined together so that none of the field lines lead to the outside.
Then I see 2 coils helping bend the end-most loss plane over to meet its partner.
Then I see the 2 end coils squeezing down on the axial losses.
The cutaway in the patent shows these end coils as much larger than the article version and with a small ID.
I think it is simply a magnetic bottle, not an IEC machine.

The problem I see is:
The patent claims that the field is everywhere MHD favorable. (although there may be some weasel words in there)
But:
In figure 6 there are 2 unlabeled regions shown with horizontal crosshatching that clearly show MHD unstable shapes.
i.e the plasma boundary and field lines are concave toward the plasma. This creates an incipient "aneurysm".
I think they are going to have problem there. Maybe they are counting on the beta being small enough at that point to prevent problems an with the "stalks".

I see various coils labeled 4.3MA, 7MA, 1MA, 1MA & 10MA. Is that 4.3 MegAmp-turns?
-Tom Boydston-
"If we knew what we were doing, it wouldn’t be called research, would it?" ~Albert Einstein

Post Reply