Talk:Polywell

From Wikipedia, the free encyclopedia

You have new messages (last change).

If I understand correctly, the key to this design is the relative weights, charges, and sizes of the different particles involved. The electrons weigh much less than the ions, and so are carried around in circulation by the magnetic field in a way that forms a quasi-spherical shell inside the device. The ions are much heavier and have more inertia, so they rarely collide with the tiny electrons and aren't affected by the magnetic field, but are attracted to the electrons en masse and crash together after passing through the shell and going towards the center. — Omegatron 22:02, 27 November 2006 (UTC)

Correct.
As for the proposed merger, I would disagree. The Farnsworth fusor has its own article (it's a specific type of inertial electrostatic confinement). Why shouldn't this one? -- Rei 06:16, 28 November 2006 (UTC)
Fusor should be merged, too. If not, all three have redundant content which needs to be split. — Omegatron 13:15, 28 November 2006 (UTC)
Every type of fusor already has their own articles. Why shouldn't this be the case? Why should we jam all of this technical detail, history, etc, for each into a single article? That would be like insisting that mice, rats, gerbils, etc all get grouped into the "Rodent" article.
Polywell is a very different beast than a Farnsworth-Hirsch fusor, even though they're both IEC fusion devices. The physics behind it is very different. It's wrong to shove them together as though they're minor variants on the same design. In short, I stridently object. -- Rei 20:10, 28 November 2006 (UTC)

[edit] Bussard's polywell and lecture

Bussard says a lot, and he says it fast. I'm just an engineer, so I don't know what a lot of these terms mean, but here's my summary of notes from watching the Google lecture. We should cover a lot of this stuff:

Research timeline

  • Has been working on fusors for the past 11 years with his company EMC2
    • Team of 5-10 people for 12 years
    • Funded by DOD/DARPA, so results have been classified. "Embargo" on publishing papers for 11 years.
      • Claims only reason for lack of money is because DOE has charter to research fusion with tokomaks and won't fund other fusion research
  • First test September 1994
  • Now that contract is over? he plans to write papers about the results.
    • First summary paper should be online soon, was published in October in proceedings submitted to the 57th International Astronautical Conference in Valencia, Spain
    • Then hopes to write a 120 page or so paper to explain everything for a bigger journal like Fusion Technology
  • Funding ran out in 2005, "saved" by "Admiral Cohen" to get through 2006 just long enough to get these results
      • Was funding cut? Mentioned being under "Advanced Energy Development" "line item" in Navy's budget, which was killed to make room for Iraq? Or was the budgeted money all used up? Or contract expired?
  • Closed the lab down on 1st November 2005?
    • SpaceDev (SpaceShipOne, James Benson) is interested in fusors for space engines
      • Moved one million dollars of Navy equipment to SpaceDev
      • SpaceDev hired the three best lab people to continue research
    • "Or maybe Google" could fund it, says that he ended up giving the speech there 'by accident', but they do happen to have a lot of money...

Research results

  • Says tokomaks will never work, Bussard was one of the people who got them started
    • "Stars aren't toroidal", "billions of dollars to discover that it's not any damn good", and other choice quotes
    • Better to simulate center-pointing force like gravity in stars; uses electric fields to push towards the center instead of a toroid
  • Says "like marbles in a well", so my wanting to mention an analogy to "marbles in a bowl" is not a half bad idea, is it?
  • Mentions burning nuclear waste from fission reactors with a D-T system while simultaneously producing power, to drop nuclear waste storage time from 4000-9000 years to 40-90 years
  • Older designs are IXL or EXL
    • IXL (ion acceleration) energy's lost and grid melts. 90-95% transparency is the best they can get, but no grid is transparent enough.
    • EXL (electron acceleration, two guns) inversion of IXL. Gets rid of electron interception but replaces it with ion interception
    • Hirsch has on his desk the prototype that reached 10^10 fusion reactions (per second?) on D-T
      • Ion guns facing each other
      • 10^-6 total energy gain, due to grid loss problem and collision with walls
    • quasi-spherical magnetic fields
      • trap injected energetic electrons to form spherical negative potential well
    • There are no magnetic monopoles. Bussard patented a device in which the magnetic fields are on the edges of a polyhedron with an even number of faces around every vertex so alternate faces are north-south
    • Electrons are trapped by polyhedral magnetic field in a spherical configuration, ions are dropped into electron shell and trapped by electrons. Has to do with weight difference of electrons and ions?
    • fusion ions trapped in this spherical well
      • focussed through central region (1/r^2)
      • oscillate across core until reacted
  • Previous designs like two magnets facing each other had line cusps around the edge, with two magnets facing each other there's an equator of loss
    • Alternating polyhedra design has lots of point cusps instead of line cusps, which have lower losses. "Polywell" magnetic polyhedral grid
    • System acts like a spherical colliding-beam device
    • Fuel gas is input at potential well edge
    • Fusion products escape to system walls
  • Maxwellian distribution problem
    • His fusors do not use a Maxwellian distribution. Not maxwellian equilibrium plasmas like Tokomak
    • Non-local thermodynamic equilibrium
  • Boron with a charge of +5 falling into a 100 kV well will reach 500 kV; doesn't need a 500 kV well
  • Needs to be slightly ion-rich to create a virtual anode instead of a neutral core
    • There's significant wiggle room between ion-rich enough to create fusion and too ion-rich that it 'blows the well'
  • The major problem they were facing is electrons striking the apparatus without a magnetic shield. they solved this a year ago.
  • Another problem is arcing
  • Neutrals can be ionized with microwaves and "gauss lines"? (used a modified consumer microwave oven) to allow them to be controlled
  • Without big enough power supplies, used capacitor bank to run wb4 for a few milliseconds
    • in 2005
    • 12 kV drive, 10 kv well depth, D-D fusion 10 kv
    • Achieved 10^9 fusions/second for a quarter of a millisecond, based on count of three neutrons
  • Things they know it needs:
    • No metal surfaces open to the electrons
    • Needs to be recirculating
    • All coil containers need to be conformal to magnetic field shape, so electrons are not striking the magnets, need gaps between magnets

Future research

  • SpaceDev
  • Also considering easier to build neutron-emitting systems to retrofit existing fossil fuel plants
  • Navy is interested in p-11B for electric submarines
  • Proposed project is estimated at 150 (D-D) to 200 million (pB11), five years to complete a working reactor
  • Size would be 1.5-2 m for D-T, 2-2.5 m for p-B11, no larger needed
  • Physics problems are gone, engineering problems and money are the only things left to deal with, though engineering is 10 times as expensive as physics
    • "2 megavolt output"
    • "200 kV standoffs"
    • Future devices will not be circular coils, but will be optimized for the shape of the magnetic field?
  • Would first spend $2M?:
    • creating two more small machines like the last one created (WB6?)
    • Then run those results by a review panel before proceeding to the rest of the $200M?

Others working on fusors:

  • George Miley of the University of Illinois, working with grid-based fusors
  • Gerry Kulcinski and John Santarius of University of Wisconsin also working with grids, trying to improve Farnsworth's design

There are also letters online ostensibly written by him: [1] [2] — Omegatron 20:31, 29 November 2006 (UTC)

Thank you for the summary. I am listening to the lecture now. Two things jump out. First, Achieved 10^9 fusions/second for a quarter of a millisecond, based on count of three neutrons. Isn't this a bit of an extrapolation? Second, if they really have something, why don't they have a website? None of the charts in the video are legible. Paul Studier 23:14, 29 November 2006 (UTC)
Websites concerning products are for trying to sell them. Bussard is trying to raise funders, which isn't quite the same thing. A quick google search looking for scientific review of Polywell shows what little is out there so far as being favorable.[3]. As for the neutron count, I'd have to listen to the talk again. That doesn't sound right -- 10e9 fusions/second for a quarter millisecond would mean the release of 2.5 million neutrons. Surely their detection rate couldn't be that low, could it? -- Rei 00:15, 30 November 2006 (UTC)
Well, seeking funders is the same as advertising. I would like to just see the charts he put up in the video. As for 3 neutrons, Bussard mentioned that the detectors were several feet away, the fusion event was short and it generally takes several cm of solid material to stop a neutron. If I recall correctly, neutron detectors have a couple cm of area and contain gaseous boron trifluoride, which is not very dense. So the conversion factor is plausible. However, 3 neutrons from one event is not enough to convince me. Still this justifies an article because I believe that even hoaxes should be documented. Paul Studier 04:46, 30 November 2006 (UTC)


First, Achieved 10^9 fusions/second for a quarter of a millisecond, based on count of three neutrons. Isn't this a bit of an extrapolation? As for the neutron count, I'd have to listen to the talk again. That doesn't sound right -- 10e9 fusions/second for a quarter millisecond would mean the release of 2.5 million neutrons. Surely their detection rate couldn't be that low, could it?

Sounds like a very ballpark estimate to me. He said the detector was on the other side of the room, so it's only going to intercept a portion of the neutrons produced, but a count of 3 seems like not enough data points to accurately measure anything. Could neutrons prefer to shoot off in certain directions if the plasma is not perfectly spherical, or are they completely unimpeded by everything?
Also, he says his fusors had 100,000 times as many fusions as Farnsworth or Hirsch, yet it doesn't seem to be that high, and even if it is, others have reached higher numbers since.
'“Just plugging it into the wall, I think I produced 105 neutrons per second,” Hirsch recalls. (His more carefully controlled trials in 1967 yielded more than 1010 neutrons per second, a benchmark that has yet to be beaten by the modern versions of this device.)'
"New fusors based on Hirsch's design were first constructed in the later 1960s. Even the first test models demonstrated that the design was a "winner", and soon they were producing production rates of up to a billion per second, and has been reported to have observed rates of up to a trillion per second."
Here's some info on measurements and calculations.
I've had to mention this many times already, but I guess that I have to mention it again. Bussard claimed 100,000 times the neutron production rate under the same well depth and driving conditions as Farnsworth. Sure, one can build a bigger fusor or put more current into it to get a higher neutron production rate, but that's not an apples to apples comparison.
As for a count of 3, yes, you'll have a very wide confidence interval, but it is statistically meaningful. If I interviewed a a million people and three reported that that they had Condition X, I could conclude with a good degree of confidence that Condition X was not a one-in-a-billion event. -- Rei 16:41, 30 November 2006 (UTC)
Aha! — Omegatron 21:23, 30 November 2006 (UTC)

None of the charts in the video are legible. I would like to just see the charts he put up in the video.

I already wrote to the address listed on fusor.net and asked for copies of the slides or images. No response.
I guess we have to wait for the papers to be published, though he said he already published a summary for the International Astronautical Conference proceedings. Why can't I find it?
The abstract is available at http://www.iac-paper.org/user/download.php?id=4338&type=abstract&section=congressbrowser Chandon 06:09, 4 December 2006 (UTC)

Still this justifies an article because I believe that even hoaxes should be documented.

It's our duty to report neutrally on anything notable, regardless of whether it really works, is a hoax, or is a good idea that fails to produce good results. — Omegatron 13:51, 30 November 2006 (UTC)

From [4] Neutrons produced from the D-T reaction are emitted isotropicly (uniformly in all directions) from the target. Neutron emission from the D-D reaction is slightly peaked in the forward (along the axis of the ion beam) direction. The source described here is a commercial neutron source that accelerates ions which hit a solid target. Since the ions in the fusor are pretty much going in all directions then the distribution should be isotropic even if the plasma itself is not symmetrical. Paul Studier 20:47, 30 November 2006 (UTC)

As for 3 neutrons being significant, I would have to know the details of the detector. Did they detect the neutrons only during the quarter millisecond that the plasma was operating, or over a much longer time period? What is the background rate from cosmic rays, etc. for the time period? As I recall, neutron detection was controversial with both cold fusion and bubble fusion. This thing is smelling like a hoax to me. I would be very interested in seeing the slides. Paul Studier 20:56, 30 November 2006 (UTC)

changing from a cubic construction to a polyhedral one.

A cube is a polyhedron... — Omegatron 10:39, 3 December 2006 (UTC)

More references:

Bussard, R. (1991). Some Physics Considerations of Magnetic Inertial-Electrostatic Confinement: A New Concept for Spherical Converging-Flow Fusion. Fusion Technology, 19(2), 273-293.

KRALL, N. (1992). The polywell™: a spherically convergent ion focus concept. Fusion technology, 22(1), 42-49.

Robert W. Bussard's Legislative Proposal to Congress

Retrieved from "http://en.wikipedia.org../../../p/o/l/Talk%7EPolywell_c156.html"