From Wikipedia, the free encyclopedia
  (Redirected from Talk:Farnsworth-Hirsch Fusor)
Jump to: navigation, search
WikiProject Physics (Rated B-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Physics, a collaborative effort to improve the coverage of Physics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject Energy (Rated B-class)
WikiProject icon This article is within the scope of WikiProject Energy, a collaborative effort to improve the coverage of Energy on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.

Lists of fusioneers should be moved to a separate article[edit]

The list of fusioneers at the bottom of the article needs to be moved to its own page. Enough people have built this device that including a (partial) list of people to build this device falls outside the scope of this page and reads like an achievement list. I have restored the tag that I put up last month, which was removed by an unregistered user. — Preceding unsigned comment added by Bencbartlett (talkcontribs) 16:25, 7 August 2015 (UTC)

The list really is getting a little silly now; people are registering accounts just to add themselves to it. - (talk) 18:46, 20 July 2016 (UTC)

I think we should remove the list altogether and include only the first few who built it, it's getting quite long. ThatGirlTayler (talk) 05:04, 10 March 2017 (UTC)

electron-volts and temperature[edit]

An electron volt is a unit of energy. How can it be equal to 11604 degrees? -phma

Temperature measures the kinetic energy of particles' vibration. E = (mv^2)/2, so a particle's velocity is the most important factor in a material's temperature. All electrons have the same mass, so the mass drops out of the picture. So, an electric field will accelerate an electron to a speed which is proportional to a temperature. By the way, thanks to the person who rewrote it, knew about the history, bremstrahlung losses and got the picture. It's much improved. User:Ray Van De Walker

Specifically, E = (3/2)kT, where T is the temperature in Kelvins and k is Boltzmann's constant. 1 eV is equal to 1.6022e-19 Joules. Rearranging the equation, .
Hmm. So either I'm wrong or the article's wrong, or we're talking about different things. At least it's the same order of magnitude. -mako

Okay, I know what I'm talking about now. An electron volt is defined as the energy an electron gains from falling through a potential of one volt: 1.6e-19 Joules. This can be associated with a "temperature", since the units of k*T is (Joules/Kelvins)*Kelvins = Joules. Rearranging eV = kT,
- mako 06:00, 10 October 2005 (UTC)

It isn't "Kelvins", it is "Kelvin".

why quotes?[edit]

in which the "wall" fields of the reactor were "electrons" or "ions" being held

why all the italics and emphasis quotes? this gives me the impression that we aren't actually dealing with electrons... - Omegatron 22:43, Aug 25, 2004 (UTC)

Fusion power box[edit]

What's up with the "fusion power" box? (replicated here)

Energy source
Fusion power

Extracting net energy from a nuclear fusion reaction. Fusion, if it can be achieved, may be the best option for a truly sustainable or long term energy source because some fusion fuels (deuterium, for example) are virtually inexhaustible and readily available throughout the world.

It's almost totally content-free (even with my additions), it's basically wrong, and it's irrelevant to the page - nobody has any idea how to extract power from fusors, and it's not what they're for. I'll take it out, and if anyone can explain why it should be there, feel free to do so and then put it back. Andrew 22:23, Dec 1, 2004 (UTC)

I believe that farnsworth was trying to get "excess" energy from this in his initial research (IIRC). I'll do some looking around. JDR 22:46, 13 May 2005 (UTC)

focus electrons between nuclei[edit]

What particularly interested Farnsworth about the device was its ability to focus "electrons" at a particular point (for example, between nuclei).

What does the above mean? Again why the quotes? --Gbleem 18:02, 18 Dec 2004 (UTC)

flux ??[edit]

The article currently claims: "Fluxes well in excess of most radiological sources can be made from a machine that easily sits on a benchtop." Unless someone has some references supporting this claim, I believe this sentence should be removed. I am not aware of any real neutron beamline that uses a Fusor (they all use nuclear reactors or particle accelerators). As far as I know, the flux from a Fusor is much much lower than a particle accelerator or reactor. Real beamlines produce >1E14 neutrons/second out of each tube, whereas a Fusor generates maybe ~1E2 neutrons/second. If anyone has some real data that says otherwise I'd love to see it! Kebes 22:16, 27 Apr 2005 (UTC)

From here:
The neutron emissions he achieved (published results on the order of a billion neutrons per second, and unpublished results of around a trillion per second!) would be considered dangerous today.
If you jack the inner grid voltage on this simple little machine up to l0,000 volts or more, and feed deuterium to the system at a pressure a little under 10 microns, it should produce fusion, evidenced by net neutrons I have seen a 17-year-old build a grid that produced 300,000 neutrons a second at 13,000 volts.
And here for a hobbyist version:
Fortunately, the machine described, operating at under 15kV, does not make many neutrons. Operating at the most optimistic output, based on one really good 15-second burst I saw using a larger power supply, the machine should not be able to produce more than about 300,000 neutrons a second, which would require 12 days at 1 meter away before you picked up a dose high enough to even START to worry. This on a machine on which the grid life is probably under 20 minutes if the grid is made of stainless steel. More likely neutron production will be down in the hard-to-detect range of under 10,000 per second. These will radiate off spherically, so only a fraction will hit someone standing to one side.
Not sure if that counts as "real data"... - Omegatron 23:24, Apr 27, 2005 (UTC)

Interesting. That's not much flux! I also found one source that stated 1E8 neutrons/second, and another (seemingly from Los Alamos) that claims 1E10 neutrons/second. Since a Fusor is small, we can assume an operating distance of only 1 meter, but this still gives us only 1E5 n/(cm^2 s). Neutron fluxes from nuclear reactors are now on the order of 2E15 n/(cm^2 s), and spallation sources on the order of 1E17 n/(cm^2 s) (see fig 1 here for info). Considering that the Fusor is 10 order of magnitude worse than (good-quality) nuclear reactors, I don't think it's fair to say that its flux is "well in excess of most radiological sources." If no one objects, I'm going to rephrase that. Kebes 16:33, 28 Apr 2005 (UTC)


We should cover the info in this article, specifically the section "Fusion the Easy Way" and the info about Bussard. I'm going to try to find out what happened with that... - Omegatron 23:34, Apr 27, 2005 (UTC)


  • 1924 - Langmuir and Katharine Blodgett investigate vacuum tubes with concentric electrodes (used in multipactor tubes) - I guess this doesn't need to be included.
  • mid-1950's, P. T. Farnsworth creates "IXL" ("Ion aXeLerator") device - a standard diode tube arranged in a sphere, so positively charged ions are accelerated from the outer positively charged grid towards the inner more negatively charged grid where they miss the grid and collide in the center.
  • 1959 - Elmore, Tuck, and Watson create "EXL" ("Electron aXeLerator") version - the opposite of the above, accelerates electrons into the center, which create a negative "virtual electrode" in the center point. The large number of electrons compared to ions means the positive ions are attracted to this negative virtual electrode. (This confuses me, since they would also be attracted to the negative real electrode once they had overshot the center...)
  • 1967 - Robert L. Hirsch makes "IXL" version - inner grid negatively charged, outer grid positively charged
  • 1969 - Farnsworth-Hirsch fusor at 1010 neutrons per second, which is not explained by the models used by Hirsch (from latest Bussard patent) Thought that it may be a result of using discrete ion guns opposite each other instead of modeling the particles as uniformly distributed (?)
  • 1992 - Bussard patents version with spherical ion acoustic standing waves to create a denser "core" and minimal grids to prevent losso f electrons - Patent 5160695
  • 1995 - "Robert W. Bussard, a physicist who founded the Energy/Matter Conversion Corporation of San Diego (which goes by the clever acronym EMC2) is also experimenting with inertial electrostatic confinement fusion. The U.S. Navy has supported him with about $4 million since 1995, in hopes that this technique will someday provide a compact fusion power source." [1]
  • 2003 - "Robert Bussard and his EMC2 people in San Diego are the ones you need to pray for for your particular dream mecha to come to life. He is the one trying to build a compact gas-cathode IEC Fusion reactor on a 4.4 million dollar grant from the US Navy that he won in March [15th, 2003]." - from some silly thread, but this page ("the official listing of all Federal government contracting opportunities and awards") shows the award for the grant? which, according to this thread is under the project name AGEE.

Art Carlson's summary of Rider's paper

This was relevant to Bussard's version for some reason?

Has a summary explaining the two different types

Rename to plain old "Fusor" or "Fusor tube"?[edit]

It seems the term "Farnsworth-Hirsch fusor" only applies to the 1967 version, and not the the earlier or later versions. Should we rename to just "Fusor" or "Fusor tube" or Inertial electrostatic confinement? - Omegatron 16:35, Apr 28, 2005 (UTC)

Why not just have the Farnsworth-Hirsch Fusor article cover the Farnsworth-Hirsch stuff in depth, and create a higher-level article called Inertial electrostatic confinement fusion which treats all of the IEC fusion topics together, and points to sub-articles (like this one) which can cover each sub-topic in depth? -- The Anome 15:08, May 3, 2005 (UTC)
Sure. But does each individual type of fusor deserve its own article? I was just going to write a short summary of each from what I could glean online. If, for instance, Bussard's section starts to grow big enough, it could be moved to its own article.
There's already Inertial electrostatic confinement, by the way. - Omegatron 15:16, May 3, 2005 (UTC)
I'm in the "just plain fusor" camp myself. After having read the arguements here it seems to make the most sense. Maury 21:07, 3 May 2005 (UTC)
I'm also in the "just plain fusor" camp. The current article name is not representative of the diversity of people that have put work into the idea. I don't think each type of Fusor will require it's own entry... but if one sub-section ever grows significantly, it can be moved to its own page. The more generic idea "Fusor" captures all related devices. And besides, "fusor" is not used in any other context (that I know of) so the additional qualification is superfluous.Kebes 21:15, 10 May 2005 (UTC)
One article for both Fusor and Farnsworth-Hirsch Fusor seems enough. The primary title should then be "Fusor", although that isn't so important since the other one gets redirected anyway. I think I would like to keep the general considerations of the plasma physics in inertial electrostatic confinement and leave Fusor for nuts and bolts and history. The distinction is somewhat arbitrary and a good case could be made for combining them. Art Carlson 07:46, 2005 May 12 (UTC)
I've moved it to Fusor due to this agreement. - Omegatron 13:05, May 12, 2005 (UTC)

Plasma physics[edit]

This is my brief critique of the fusor from the perspective of a (former) fusion research physicist. It is pretty conventional, the sort of worries I would expect any plasma physicist to have, but it could possibly be considered original research. It would be very good if someone who works directly with these things could check it over. It may be that there are valid counter-arguments to some of my objections. Or, of course, even more important problems that I have overlooked. Art Carlson 08:14, 2005 May 4 (UTC)

  1. Debye shielding: It seems that most introductory descriptions of the fusor take a single-particle approach, but at any interesting density we are dealing with a plasma, which can be as different as day and night. In particular, I want to know why a huge current doesn't flow between the electrodes. Maybe they are insulated, maybe they are very thin. Whether or not the current can be suppressed, the electric charges in the plasma will adjust themselves to eliminate most of the electric fields (see Plasma physics#Potentials). So then what electric field accelerates the ions? The only solution I know of (I have encountered analogous problems in the course of my work.) is to have holes in the inner electrode that are smaller than the Debye length, which, if it is poosible at all, makes the electrode very filigree and liable to be quickly destroyed by a fusion plasma.
  2. Bremsstrahlung: The next thing I worry about is radiation losses. Even forgetting about the deleterious effects of the inevitable heavy ions (even carbon would be a problem) sputtered from the electrodes, simple Bremsstrahlung will be prohibitve for any fuel other than D-T (or with considerable optimism, D-D or D-He3).
  3. Non-Maxwellian velocity distribution: When they first fall into the center of the fusor, the ions will all have the same energy (which is sometimes cited as an advantage for various reasons), but they won't stay that way. Simple Coulomb collisions will redistribute the velocities toward a thermal distribution. In particular, some of the ions will get enough energy to leave the potential well of the fusor. I suspect that this process alone is fast enough to make net energy production impossible.

Length scales[edit]

An important question is how often an ion must slosh back and forth in the potential well before it undergoes a fusion reaction. If the number is too great, we must fear that it will first disappear by some other process, like hitting an electrode or scattering out of the well. The fusion mean free path is λfus = 1/nσfus. If we assume a pressure-limited containment, p = nkT < pmax, equate (3/2)kT with (1/2)mv², and don't worry too much about how we take our averages, we find that λfus is proportional to T3/2/<σv>, so the temperature that minimizes the fusion mean free path for any reaction is similar to that which maximizes the power density (see Nuclear_fusion#Criteria_and_candidates_for_terrestrial_reactions). Let us do a calculation for D-T using the values T = 13.6 keV and <σv>/T2 = 1.24×10-24 m³/sec/keV². Then we have v = sqrt(3*(1.6e-19)(1.36e4)/(2.5*1.67e-27)) m/s = 1.25e6 m/s, and σ = (1.24e-24)/(1.25e6) = 1.0×10-30 m². Let us take a generous 100 atm = 107 Pa as the plasma pressure, so that n = 4.6 ×1021 m-3. Put it all together and we have λfus = 2.2×108 m. That means an ion would have to survive some hundred million reflections to have much chance of undergoing fusion. To be fair, one is investing about 13.6 keV and getting back 17.6 MeV when it works, so even a 2% burnup might be sufficient to produce net power. That would still require an ion to survive around a million reflections. Art Carlson 10:11, 2005 May 5 (UTC)

Time scales[edit]

In terms of time, λfus/v = 176 s = 3 min. For comparison, the ion collision time is given by 1/νi = (4.8e-8 s-1)-1(n/cm-3)-1(λ)-1(Ti/eV)3/2(m/mp)1/2 = 1.1 ms. If the ions are contained long enough to have any serious probability of fusing, then they will be well-thermalized long before that. This estimate assumes classical Coulomb collisions. In fact, the counter-streaming ion beams will be unstable on the ion plasma time scale and thermalize within nanosconds.[2] Art Carlson 10:11, 2005 May 5 (UTC)

Bowl analogy[edit]

I was thinking about the idea that the particles go back and forth through the center the other day and realized it's kind of like lifting marbles up to the edge of a bowl and then dropping them in. They'll make many passes through the center of the bowl at no cost to you, until they happen to hit each other. Good analogy? - Omegatron 14:06, May 4, 2005 (UTC)

An excellent place to start. I am constantly making little sketches of potential structures and imagining ions bouncing back and forth in them. The method is, however, frought with peril. The first habit you have to get into is turning your sketch upside-down when you think about electrons. The bowl that holds the ions will be a hill that electrons slide off. Even that picture is dangerous because the charge density of the ions and electrons influences the potential decisively. Better is to assume the Boltzmann relation for the electrons, which relates the density to the potential. In the case of the fusor, all you really need to know is that you have no control over the potential inside the inner electrode. That means that you have a pie tin rather than a bowl. More important, your inner electrode is in contact with a fusion plasma. It won't last a second. Believe me, I've tried it with Langmuir probes. If you are clever enough to find a way of keeping the plasma off the electrodes, you could probably use the same method to keep the plasma off the wall in any other fusion device. That is, you could make an improved tokamak that would probably work better than the fusor. Finally, look at the calculation I made above. It's one thing if you can give your ion 10, 20 or 100 chances to react. It's another if it needs a million chances, and you have to make sure nothing happens to it before then. Art Carlson 10:11, 2005 May 5 (UTC)
But, don't they routinely keep the plasma away from the electrodes by using the "virtual electrode"? - Omegatron 13:34, May 5, 2005 (UTC)

Is there any such thing as "inertial electrostatic plasma confinement"?[edit]

I just edited this argument into inertial electrostatic confinement. Comments? Objections? Art Carlson 16:38, 2005 May 8 (UTC)

Most experts are skeptical that the IEC concept can ever be used for power production, and some are even skeptical of the basic concept. Most discussions of IEC consider the behavior of a small number of ions in potential structures imposed by electrodes. If the density is greater than about 1012 m-3, however, then the Debye length is smaller than the electrode dimensions. (At T = 13.6 keV, λD would be 90 cm.) A plasma of this density or greater would shield most of the electric field, so that ions would no longer be accelerated. (Making the inner electrode more negative would not affect the nearby plasma potential but would only increase the voltage dropped in the Debye sheath.) It is usually considered a defining characteristic of a plasma that the Debye length be short compared to the physical size of the plasma. In that sense, "inertial electrostatic confinement" should not be called plasma confinement at all.

Something screwy has to be going on here. When I put in n = 1e12 m-3, T = 13.4 keV, and <σv> = 1.24×10-24 m3/s/keV2, I get 230 fusion reactions per cubic meter per second, way below reported value of up to 1e10 neutrons per second. ?!?! Maybe somebody should check my math. Art Carlson 16:52, 2005 May 8 (UTC)

Let me try again. A potential well can hold either electrons or ions, but not both. I might tweak things with double wells, something like this:

__/\     /\__

I need three electrodes. The ions are confined, e.g. by the potential drop between electrodes 2 and 3, while the electrons are confined by the potential drop between 1 and 2. The advantage is that my big volume can be quasineutral and the troublesome regions of net charge are the size of the interelectrode spacing rather than the fusion volume. Anyway, I can't avoid having some volume that has an ion density comparable to that where fusion is happening but an insignificant electron density (or vice versa). I wind up with a limit on the density something like this:

where I don't want my potential difference ΔU to be more than, say 100 kV, and I can't make the size of my electrode structures δ less than, say, 1 cm. This is in fact the same formula I got when considering the Debye length, except that I now allow myself a potential a bit higher than the temperature and a scale length a bit smaller than the machine. Putting these values in, I find

This value, in turn, gives me a maximum of 6.8e11 neutrons per cubic meter per second. If the fusion volume is 1e-3 m^3, this gives about 10^9 neutrons per seond, which is in the right ball park.

The fact that I arrived at a similar formula looking at the problem two different ways, and the fact that my results do not seem to contradict experimental reports, give me some confidence that I am on the right track. It would mean that

  1. fusors won't ever get much better than they are now in terms of neutron production,
  2. to make useful energy, the ions would have to survive not a million but at least a hundred billion reflections and keep bouncing for half a year, and
  3. the maximum fusion power density (with D-T!) will be about 2 Watts per cubic meter.

If that doesn't damp your spirits, nothing will! Art Carlson 09:28, 2005 May 9 (UTC)

bussard's patent[edit]

I uploaded two pictures from Bussard's patent for use in a future section about him... - Omegatron July 2, 2005 01:26 (UTC)


Bussard 2.png

Help! I tried to edit a small part of the article, to add a little bit of text, and an error message occurred, after which, when the edit was accepted, the whole article was replaced! That was not my intention! Somebody please revert!

Don't panic - it's done --DV8 2XL 18:29, 13 February 2006 (UTC)

neutron source for fission[edit]

Could the fusor be used to provide neutrons for fission and produce power without a chain reaction and produce net power? --Gbleem 13:54, 23 June 2006 (UTC)

Yes. The fusor's only commercial use so far is as a neutron source. I'm doing a three year long research project for high school on such a subject, although not necessarily for inducing a fission reaction. Such a use is certainly within reason. --Liambowen 03:04, 18 July 2006 (UTC)

Merge with Inertial electrostatic confinement[edit]

Strongly For -- Though there have been a few differing incarnations of IEC over the years (usually one variation per research group), to date they all have similar fusion rates for a given current and frankly there is no reason to believe any of them are particularly different. Fusor should redirect to IEC, and gridded IEC (Fusors), penning traps etc should have their own sections. -- Rpf 13:00, 23 November 2006 (UTC)

I've heard of three devices; the IXL, EXL, and polywell types. Are all of these considered "fusors", or only the first? We either need to merge them all into one article or move the stuff that refers to non-fusors into the IEC article and vice versa. I think a merge is probably best. — Omegatron 15:52, 23 November 2006 (UTC)

There is also POPS "Periodically Oscillating Plasma Shere" with Rick Nebel at Los Alamos. IXL and EXL all have varying scales and geometries (certainly not all spherical inner electrode). Some even have multiple electrodes. I believe anything involving DC electrodes should be called a fusor because it still vaguely resembles the Hirsch-Farnsworth stuff (IXL,EXL). Anything that uses a space-charge and/or electric fields to confine and accelerate ions (pops,polywell,fusor) should be a subset of an IEC article. -- Rpf 23:13, 23 November 2006 (UTC)
Do you know of an "official" definition that separates them in that way? Should we keep them as separate articles? — Omegatron 09:21, 25 November 2006 (UTC)

Strongly against -- The devices vary greatly in design and considerations. Each has their own individual history. There are many designs. Consequently, you'll either have a monstrous article, or have to cut valuable detail. What you're proposing is like merging mice, rats, gerbils, etc into "Rodent" because they're similar.

Ok. But what should the Inertial electrostatic confinement article be about then? It has content that belongs here or in Polywell. — Omegatron 21:49, 28 November 2006 (UTC)

The acronym IEC, apparently meaning "inertial-electrostatic confinement," should be defined in each article before it is used there. —Preceding unsigned comment added by (talk) 05:10, 16 March 2011 (UTC)

Strongly against IEC is a branch of fusion research, under which fusors fall. But so do polywells, POPS, the Penning Trap Fusion Experiment and any other fusion machines, designs or efforts which try to use electric fields to heat ions to fusion conditions. Also, does anyone know if pyroelectric fusion counts? How does that work?
No need to respond in 2014 to a failed 2011 proposal. Dicklyon (talk) 16:32, 27 January 2014 (UTC)

a self consistent collisional treatment[edit]

I have strong reservations about including the following statement:

However, this study lacks a self consistent collisional treatment of the ion distribution function in velocity space, crucial for adequately estimating the fusion rate and recirculating power.

In the reference given, this is the only line in 258 pages that so much as mentions recirculating power. The author does not even provide a reference to more detailed work, much less justify the statement himself. I don't remember the details of Rider's derivation, but I believe it was a lower limit that did not require an exact calculation of a particular distribution. If this is the only verifiable criticism we have of Rider's work, then I think we should leave it out altogether. At the very least it must be rephrased to make it clear that it is one voice, not a scientific consensus. --Art Carlson 14:51, 23 November 2006 (UTC)

Todd Rider's paper[edit]

Is Rider's paper really applicable? As it stands, only nuclei are being accelerated by the grid -- not electrons. There are electrons in the grid, but that's hardly what he's talking about. In an "ideal" fusor, the electrons are completely ignored by the fusing ions. -- Rei 20:45, 28 November 2006 (UTC)

What's your picture? Are there no electrons within the grid? Then your density will be damned small due to space charge. Even if you only consider ion-ion collisions at the center, there will be diffusion of the velocities toward a Maxwellian. And the collisions that occur near but not at the center will cause diffusion in space. I don't see any reason that Rider's conclusions wouldn't apply. --Art Carlson 21:07, 28 November 2006 (UTC)
There are electrons in the grid. In an ideal fusor (no collisions with the grid and completely ionized fuel), the only type of collisions you have are between ions. Even if the ions were a perfect Maxwellian distribution, with no electrons mixed in with them (only on the grid), where are the significant bremsstrahlung losses to be coming from? Unless I'm mistaken, Rider's paper is about a plasma containing both electrons and nuclei, dissociated but intermixed. Feel free to correct me if I'm wrong. -- Rei 22:43, 28 November 2006 (UTC)
OK. I thought you were talking about maintaining a non-Maxwellian distribution. You are apparently concerned instead with the bremsstrahlung losses. (He covers a lot of ground.) What do you mean "There are electrons in the grid" but "the only type of collisions you have are between ions"? If you are contemplating the case that there are no electrons at radii less than the radius of the inner grid, then calculate how high the ion density can be before their mutual repulsion will prevent any more from getting in. You may not have any bremsstrahlung, but you won't have a thimblefull of fusion either. (And then we have to start talking about what the ion collisions do.) --Art Carlson 08:35, 29 November 2006 (UTC)
Lets back up for a second: you agree with me, that in an "ideal" fusor (no collisions with the grid and completely ionized fuel) that there won't be any bremsstrahlung losses, right? Okay, now lets accept that such an ideal fusor is impossible  :). How close can you get to that ideal fusor? Do you know? There have been continual improvements in the design (whether it can ever even approach break-even is questionable, but that's not the issue here).
Most critically, however, is that Rider's paper is not about losses in a situation where there's a huge imbalance in ratio of atomic nuclei and electrons in the fusing plasma (a non-neutral plasma), and thus it shouldn't apply to the Farnsworth-Hirsch fusor. There are electrons in a Farnsworth-Hirsch fusor, but they're geometrically constrained, not mixed in. The plasma is non-neutral.
Most key among Rider's assumptions, which the fusor violates, is on page 30: In calculating Bremsstrahlung rates, the plasma is assumed to be quasineutral. I don't know how one can describe the fusor's plasma as being quasineutral. It's not even close. Rider later addresses "fusion without electrons" (p. 135-136), but doesn't describe the method used in the fusor. Namely, Rider discusses the Brillouin limit for containment, which means magnetic confinement of the ions. Farnsworth-Hirsch fusors use no magnetic fields for ion confinement (only electric), and Polywell uses magnetic fields only to geometrically constrain electrons, not the ions (the ions are still constrained electrostatically).
Consequently, I can only conclude that Rider's paper is simply not applicable to the Fusor. Do you agree? I do have to admit, though, that I love Rider's apology at the top of his Acknowledgements page.  ;) -- Rei 19:57, 29 November 2006 (UTC)
Rider doesn't take his time on non-neutral plasmas because it is so easy to prove that they are not interesting. Have you done your homework? --Art Carlson 22:29, 29 November 2006 (UTC)
That's a non-sequiteur. The discussion is about whether Rider's paper proves that the Farnsworth-Hirsch fusor cannot break even on heavier ions. As the paper is about quasineutral plasmas, it does not, and thus, such claims in the article are erroneous. Do you deny this?
As for your statement, you ought to back that up. Rider briefly discusses magnetic confinement, and I'll readily agree that magnetic confinement of non-neutral plasmas is uninteresting because of the Brillouin limit. IEC devices don't use magnetic confinement of ions, so it's a moot point. -- Rei 22:58, 29 November 2006 (UTC)
I'm not talking about the Brillouin limit. I'm talking about Gauss's law. Your edits make it sound like nobody has ever thought about taking the electrons out of the plasma. If you want to keep your changes in (which I don't think is a good idea), then you need to talk explicitly about the physics and limitations of non-neutral plasmas. --Art Carlson 09:09, 30 November 2006 (UTC)
I haven't edited out Rider's paper yet. However, since not yet have you contested my assertion that Rider's paper doesn't deal with the situation covered in the fusor, I will. Since you're raising a *new* issue, I'll replace it with a link to the IEC section on the limitations of Gauss's Law. All want is to not have a false statement (that Rider's paper makes IEC impossible) in the article. I'm not trying to argue that the fusor has the potential to break even. -- Rei 16:51, 30 November 2006 (UTC)
Update: I just removed the link that I added to the "maximum pressure" calculations that you did, because I (and apparently others as well) are concerned about your non-peer-reviewed "original research" and how well it conforms to what's actually going on in an IEC reactor. If we can clear that up, I'll put the link to your work back in. -- Rei 17:23, 30 November 2006 (UTC)

Rie, your argument appears to be based on two points; 1) that the issue of concern is Bremsstrahlung losses, and 2) that Ryder's critque doesn't apply because it's non-quasineutral. However, you haven't offered any evidence to actually support either claim. Quite the opposite, at the start of the thread you question whether not the Fusor really is quasineutral, and then dance around the point with these "ideal fusor" confusions. Nor do you ever address the question of whether or not the Bremsstrahlung losses could be calculated without the quasineutral assumption, yet this is a hidden requirement in your line of reasoning.

To answer your very first question, "Is Rider's paper really applicable?", the answer is an unqualified "yes". The paper is called "A General Critique of Inertial-Eletrostatic Confinement Fusion Systems". The paper addresses both the Fusor and the Bussard devices directly, with diagrams. Of course the later context suggests you're really asking whether or not it applies due to some sort of "mistake" in the calculators due to the quasineutral assumption. However if you examine the Spacial Profiles area of the paper, you will note that he addresses this issue directly, and shows mechanisms for electron mixing. Yet you recently stated "As the paper is about quasineutral plasmas". The paper is not "about quasineutral plasmas", and mentions them only in a theoretical section before directly addressing the problem in the lengthy discussion that follows. I cannot help but ask you the same question Art did, have you done your homework?

Based on what I have seen so far I will be inclined to revert any removal of Ryder's work based on these arguments. If you have a real point-by-point argument that covers these issues, I'm all ears. But in the meantime, removing material based on nothing more than the fact that Art hasn't replied yet (which is precisely the justification you offer above) strikes me as a breach of good-faith editing. Maury 23:12, 30 November 2006 (UTC)

First off, A General Critique of Inertial-Electrostatic Confinement Fusion Systems is not the paper that I was discussing (and removed). Lets not bait and switch here. The paper that was referenced was Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium. I have not read A General Critique of Inertial-Electrostatic Confinement Fusion Systems, and make no claims about what it says and what it doesn't.
In Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium, page 30, we get the assumption of a quasineutral plasma. Check it out. For this assumption to hold, the fusor must use a quasineutral plasma. This is impossible; for there to be any electrostatic confinement, the plasma must be non-neutral.
I started out discussing the fact that in an ideal situation, there will be no electrons in the plasma. How do you interperit this as me questioning "whether not the Fusor really is quasineutral"? Quite to the contrary, I was pointing out that IEC fusors strive for just the opposite -- a completely non-neutral plasma. If we had a quasineutral plasma -- defined by Wikipedia itself as "a very good approximation to assume that the density of negative charges is equal to the density of positive charges over large volumes of the plasma" -- electrostatic confinement would be impossible. Consequently, this assumption of Rider's paper is fundamentally violated. Yes, no fusor will be ideal. There will always be some electrons. But pretending that the plasma is quasineutral is a transparently false assumption.
As for this other paper, I'll have a look at it, but it does sound applicable. I'll go ahead and tentatively support adding this other paper into the article. But I will oppose adding mention of the first paper, because IEC fusion violates its assumptions. -- Rei 16:44, 1 December 2006 (UTC)
There are several papers by Rider: his MS thesis, PhD thesis, and journal-published versions of both. "General critique" is the MS thesis, and "Fundamental limitations" builds on the earlier work.
About your quasineutrality argument, pointing to page 30, Rider only says that the bremsstrahlung calculations are done under that assumption. I can't see how you can ignore the following sentence: "The thesis will later return to these assumptions and consider systems which violate them". - mako 04:17, 2 December 2006 (UTC)
See earlier in this conversation. He has a section called "Operation without electrons". However, the only confinement he considers is magnetic, not electrostatic. He dismisses fusion without electrons on account of the Brillouin limit limiting magnetically confined non-neutral plasmas to poor densities.
''Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium, the paper that was cited on here before, really wasn't applicable to the fusor. A General Critique of Inertial-Electrostatic Confinement Fusion Systems is. Obviously, some disagree with his conclusions or think that they fall into some of his many exceptions (for example, Bussard, the University of Illinois team, etc). However, this is published, peer-reviewed material, and is a good fit for the article. -- Rei 06:53, 2 December 2006 (UTC)
He does mention IEC: see the paragraph in 4.1.17 starting "One might try to exploit...". Cite 86 is his "General critique".
I think you're sidestepping Rider's conclusions that non-Maxwellian setups require a prohibitive amount of recirculating power (chapter 3). - mako 12:27, 2 December 2006 (UTC)
Surely you'll agree that a conclusion is only valid if the assumptions used to derrive it are accurate. So, the only issue at hand is whether the assumptions are met. The general assumptions obviously are not met, so his analysis of exceptions is all we have to look at. Naturally, "operation without electrons" seemed to be a sensible place to start. If you have a better place, I'll look there. However, 4.1.17 is not a sensible place to start, because there is no 4.1.17 in Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium, the paper whose presence in this article I objected to. 4.1 ends at 4.1.12.
Perhaps you mean 4.1.7, fourth paragraph, which starts out with the same text. This section is still about containment magnetic fields; see ref. 85 which starts out this paragraph, The Brillouin Limit and Beyond. There's also no context clues to suggest that he's switching to containment with electric fields, so there's really no reason to believe that he is. Referring back to his original paper would this be referring to the parts in his original paper on IEC fusion would be thus referring to his calculations in that paper which affect such a system.
Why include a paper that, at best, would supply a premise relative to this article through an indirect reference? Lets just use his original paper on IEC fusion. -- Rei 23:12, 2 December 2006 (UTC)

← (My mistake, I meant 4.1.7.) A "spherical ion focusing system" sounds like IEC. Ref. 85 specifically mentions IEC ("infinite Brillouin ratio", "magnetic field vanishes everywhere", last page), and 86/87 certainly deal with IEC. "Dense fusion core", "unrealistically high fields"; it sounds like IEC to me. He's ruling it out based on real-world constraints. I guess this is the crux of the matter. We're talking about real fusors, not perfect ones. Do you agree that Rider's conclusions apply to IEC under current technology? Would you be satisfied if we made that distinction?

Let me also take a different tack. Rider specifically says that his results in "Fundamental limitations" rule out any kind of IEC breakeven. He's the one making the claim. It certainly meets WP:V. According to WP:NPOV, we put it in. If someone doesn't agree, they write something that meets V and we put it in. - mako 02:14, 3 December 2006 (UTC)

The Brillouin limit is about magnetic confinement. Specificly, the Brillouin limit defines the maximum density of a non-neutral plasma under magnetic confinement with a given field strength. This is not IEC fusion. It doesn't matter if it's a "spherical ion focusing system"; if it's magnetic confinement, it's not IEC, by definition. Even if it "sounds like IEC" to you, it's not.
No, I don't agree that Rider's conclusions in this paper apply to IEC under curernt technology. Even "current technology" IEC fusions have non-neutral plasmas. They're not perfectly ionized, certainly, and so there will be some Bremsstrahlung losses, but they're distinctly *not* quasineutral (if they were, electrostatic charges wouldn't drive the plasma). Rider's calculations in this paper are on quasineutral plasmas. Therefore, they simply don't apply.
Now, there *is* a paper -- by Rider -- that does all sorts of calculations about IEC devices: A General Critique of Inertial-Electrostatic Confinement Fusion Systems. This does belong in this article.
Note that Rider does not prove, in either article, that non-Maxwellian distributions are impossible. Rather, he takes on the current incarnations of several methods, in simplified form, and shows that their most basic form is impossible without unexpected behavior occurring. He actually goes on, in his general critique, to propose many possibilities and possible changes that could maintain non-Maxwellian energy distributions. Most of his proposed involve methods use resonance to filter out ions of non-optimal energies and then accelerate the filtered out ions. It's really an interesting paper. The math is beyond me, but the content is a good read. -- Rei 06:46, 3 December 2006 (UTC)
He's talking about exceeding the Brillouin limit. Note the "Inertial Electrostatic Confinement" in the title of each ref.
  • "spherical ion focusing system with a dense fusion core [85]"
    • 85. Leaf Turner and D. C. Barnes, Brillouin limit and beyond: A route to inertial-electrostatic confinement of a single-species plasma.
      • p801: To best understand the infinite Brillouin ratio, we shall take the extreme case in which the magnetic field vanishes everywhere. This is a case of inertial-electrostatic confinement (IEC)...
  • "unrealistically high fields ... [86,87]"
    • 86. T.H. Rider, A General Critique of Inertial-Electrostatic Confinement Fusion Systems.
    • 87. W.M. Nevins, Can Inertial Electrostatic Confinement Work Beyond the Ion-Ion Collisional Time Scale?
      • Should be obvious what these two papers cover.
You cite references in order to support your own statements. Do you really think Rider would back up statements on magnetic confinement with IEC papers? That simply makes no sense. What else could he be talking about in that paragraph, other than IEC?
I think you're hung up on this quasineutral thing. Again, he never says the paper exclusively covers quasineutral plasmas, he's only using that assumption for bremsstrahlung calculations. You can certainly analyze non-neutral plasmas quasineutrally; bremsstrahlung is a reasonable case. Did you notice that Rider makes the same assumption in "General critique", the paper you are so confident about including?
I never claimed that non-Maxwellians were impossible to obtain. I did suggest that Rider's claims meet V and NPOV policy, though. - mako 04:27, 4 December 2006 (UTC)

Bussard's Google talk[edit]

[Bussard's Google Talk]

The fusion reactor that he's looking for money for is obviously a decendant of the Fusor. The ions are still confined using electrostatic confinement but the light electrons are now confined using a magnetic field rather than the grid. It's a thousand times easier to confine electrons than it is to directly confine the ions because of the mass difference so even when the electrons are thrown out a near lightspeed they can be captured by the magnetic field and routed back into the 'core'. With an overabundance of electrons confined in the 'core' the ions don't really "want" to leave (ie electostatic confinement) so you get a ball of plasma rather than a 'star' as the ions evacuate.

There will of course still be large numbers of electrons lost, anything pointy or even just metal will appear to suck them right out but from the video it appear these losses can be kept down low enough to be replenished easily. The result; reaction rate going up at the SEVENTH POWER of the reactor size is definitly worthwhile!

So I think this should be added to this page in a more explicit manner but I can't think of a good way to do it without removing half the 'this cannot be used for fusion' parts. 08:55, 10 December 2006 (UTC)

You're going to need a better argument than "from the video it appear these losses can be kept down low enough". --Art Carlson 08:23, 11 December 2006 (UTC)
Exactly, all we have is Bussards reputation as I can't find the overview paper that's been published and he hasn't completed a full paper yet. However, this guy has one hell of a reputation and he's literally saying that it will work if build the "core" in this way (the rest of the equipment needed is 'just' a large fusor). Wikipedia obviously cannot say the same thing unless we see it working (whatever the guy's reputation - perhaps on his page) but I do believe this page should acknowlage the linkage. Not to worry I'll think about it some more. 09:33, 12 December 2006 (UTC)
Just found the Polywell page; that will do nicely. 09:42, 12 December 2006 (UTC)

Essential references missing[edit]

I've added three {{fact}} for bold claims without sources given.

  • The link to Daimler-Chrysler Aerospace
  • The existence of the company NSD-Fusion (in contrast to the existence of the website) and its ability to deliver products
  • The buildability as amateur project. 20kV doesn't look nearly sufficent to initiate fusion and that's about the order of magnitude, where buildability as amateur project ends.
    • There's a "neutron club" on for people who have achieved fusion. — Omegatron 14:49, 20 December 2006 (UTC)

Pjacobi 08:24, 20 December 2006 (UTC)

The DASA fusor work, FusionStar, is fairly easy to find references to, although the original is gone because DASA was split up into parts and merged into DC and EADS. You can see "real" mentions in things like [5] for instance. I don't believe it is offered any more though, and the original DASA pages are gone, along with their domain. Maury 13:35, 20 December 2006 (UTC)
Better than nothing, but not a first tier reference. Over at de:Portal_Diskussion:Physik#Farnsworth-Hirsch-Fusor de:User:KaiMartin has done some calculation and finds the published number missing the fusion threshold by a factor of 30. So we are wondering whether to delete the entire article (on de: at least). --Pjacobi 14:35, 20 December 2006 (UTC)
I'm not sure I'm properly parsing any of this. The reference in question seems to be more than enough evidence that DASA was indeed funding a fusor research project. Perhaps the article could be slightly re-worded to make that clear, but it's hardly a honker as it is. Then you go on to talk about some calculation, but I am completely mystified what this has to do with DASA, if anything. The sentence following that one only deepens the mystery. As it reads it appears you are saying that due to an unreviewed post on a wiki talk board, you believe the "whole fusor thing" is some sort of hoax and are going to delete the fusor article in its entirety. Is this what you are saying? Maury 12:51, 21 December 2006 (UTC)
The question of deletion is discussed on de:, not here. Different deletion criteria apply for different Wikipedias.
Regarding your parsing problem: I have different, essentially unrelated concern:
  1. A definitive DASA reference should unearthed
  2. Is NSD-Fusion an actually operating company (and does it deliver fusors)
  3. How much scientific attention got the entire approach (where does it score on the mainstram-minority-fringe scale)
  4. Is the technology really in the reach of the hobbyist?
  5. Why can it work, when a naive calculation gives a center of mass energy 30 times lower than threshold?
Being able to answer these questions would any case make article better, don't you agree? And in answering (5) I don't propose doing original research, but finding the answer in the published sources.
14:11, 21 December 2006 (UTC)

My concern here is that some of these issues seem so minor that questioning them seems pendantic. I would be equally confused by a post suggesting the article on "bicycle" should be removed because it contains the unreferenced claim that Trek made a bike called the 5000. Whether or not there is a bike called the 5000 has no effect on the reality of bicycles as a whole. This example might seem trite, yet it appears that this is the logic being used to suggest the deletion of the German version of the fusor article, unless I'm seriously misunderstanding your point (which is certainly possible).

I will adress your points directly:

(1) seems more that well supported. But the real question is why do you need a better one? Do you really doubt that DASA might have provided some finding to a university researcher for some small-scale fusor research? You tagged it and called it a "bold" claim, and I am trying to understand what you feel is so bold. (2) who cares? If you don't think this claim isn't supported enough, remove it. It doesn't effect the article as a whole. (3) are you asking whether or not mainstream scientists "believe" in the fusor? If that is the question, then the answer is "yes" and it rates "mainstream". (4) well obviously yes. TEEN GOES NUCLEAR If you have some more stringent definition of what you mean here, you'll have to more clearly state it. (5) I can't read German very well though, but the "calculation" you mention appears to be one persons back-of-the-envelop musings. If you can translate it into English I'd appreciate it.

I have a question in return though, because I am still thinking much of my confusion is a language issue. So let me try to make this succinct. Pjacobi, do you personally feel that the fusor is not capable of producing limited amounts of fusion and the neutrons that result from them? IE, do you feel that the fusor is basically a scam? Please be clear on this; after two messages I am more confused than ever about your position. If you do doubt the existence of fusion events in the fusor, can you explain why? Is it, as you seem to suggest, due to that single post on the German wiki?

Maury 22:47, 21 December 2006 (UTC)

Answering your questions:
  • From the view sources, I was able to find in limited time, I came to the conclusion the inertial-electrostatic confinement works but is generally considered impractical, see e.g.
  • I'm unclear whether it is really in amateur's reach, because of said calculation (which is indeed of the back-of-the-envelope type) and the markedly higher voltages stated in above research paper, compared with the values given at
  • OTOH, the work of Zambelli ( looks believable. But it would be perfect to find a souce and use it in the article, explaining why it starts producing neutrons at such low voltages.
  • Regarding the commercial side of this, I'm skeptical, but you can also call this paranoid. I've seen too much strange ideas looking for investors on Wikipedia.
Pjacobi 14:12, 22 December 2006 (UTC)
Skepticism is good, but paranoia is not. Why don't you believe this? There are several bonafide PhD researchers studying this stuff either at schools or under defense contracts. Yes, it really does fusion. "Plasma focus" neutron sources are based on the principle. Yes, amateurs can really use it to do fusion. There's a list of 20 people on who have measured neutrons. I'm not seeing anything bogus here. — Omegatron 06:02, 25 December 2006 (UTC)

I just got an e-mail from George H.Miley. He confirmed that DASA was indeed funding part of his research and were indeed planning on using it for a commercial fusion source. They eventually decided not to pursuit it, but George continues to working field. He had heard that some of the DASA people left, notably John Sved, but had not heard of NSD. I consider this to be more than enough confirmation of the story. Maury 01:11, 27 December 2006 (UTC)

Thanks for all your efforts. I feel all my points are addressed, with the possible exception, regarding the operation of NSD-Fusion. --Pjacobi 19:10, 27 December 2006 (UTC)
Agreed; I'll see if I can get in contact with them. At this point in time I believe these are indeed the ex-DASA people, but I'll wait to hear back first. Maury 21:28, 27 December 2006 (UTC)

Here's an amateur project by a high school student. According to this article, it does not produce net power output, but it is a true fusion reactor of the Farnsworth type. Have fun adding it. -- 18:10, 16 February 2007 (UTC)


Demining#Nuclear_detection says "A better neutron source is to use a sealed tube electrostatic D-T neutron generation tube". Is this a fusor? — Omegatron 05:08, 13 August 2007 (UTC)

I would say no. A Neutron generator is an accelerator powered by a DC voltage of tens of thousands of volts, with a solid target. To my knowledge, no one has claimed that they could ever produce energy. They are commercially available now and a good description is at [6]. Paul Studier 05:37, 13 August 2007 (UTC)

Addit reference[edit]

IEEE Spectrum 03.09 (March 2009) issue Pages 20-21 by Paul Wallich, feature title: "hands on" (lower cased), article title "Fusion on a Budget", describes the Fusor.

Elmore, Tuck and Watson Never Built a Fusor[edit]

Hey guys, ETW fusors are a misnomer. They only modeled these machines. They never built one. Here is their original paper: William, Elmore C. "On the Inertial-Electrostatic Confinement of a Plasma." Physics of Fluids ser. 2 (1959): 239-46. Web. I think we need to somewhat disconnect the ETW design from the fusor and shift the article to fusors being built by amateurs and professionals today. These are all Hirsch machines, but we need to clear up that confusion as well. A Fusor is a negative cage inside a positive cage in a vacuum chamber. That is what is being built around the world. The path to this machine is less important, compared with information about the machine that is working now.

Photo of Taylor Wilson[edit]

Taylor Wilson is the most well know amateur fusioneer. He did win a science fair for his construction of a fusor, he was on CBS news (3/30/2013) for the work and he did present this to the president at an event at the white house. So why has his photo been taken down from the amateur fusor section of this page? — Preceding unsigned comment added by (talk) 14:10, 13 August 2013 (UTC)

Amateur Builder Additions?[edit]

This one might be interesting enough to add, Conrad Farnsworth. Here's his story: AE7EC (talk) 02:59, 11 January 2014 (UTC)

Has Robert Murray-Smith Gotten Neutrons?[edit]

Look, there are plenty of people who are working on fusors. But, this wikipedia article should limit itself to amateurs who have actually done fusion, actually detected neutrons. This is still not a small list of people (somewhere in the region of 75, or higher) and it is also a bit dicey, since these people are often self-reporting. Richard Hull maintains a list and he places rules, there are different hoops you have to get through, first vacuum, first glow and first nuetrons, ect... and the folks have to demonstrate progress along this path, to move forward. Amatuers should be listed if they get covered by the local news, or win a science fair, or a win a well known science scholarship, because they detected neutrons created by fusion. — Preceding unsigned comment added by WikiHelper2134 (talkcontribs) 15:05, 28 January 2014 (UTC)

No mention of safety[edit]

The article has a glaring omission by not describing the safety issues concerned with the building and operation of a Fusor. In addition to potentially-lethal levels of high-voltage electricity, there are issues concerning x-ray and ultra-violet radiation exposure when a unit is operated at high neutron-flux levels.

A simple search reveals that to date there are more than TWELVE HUNDRED posts that invoke the word “safety.” What is interesting is that, near as I can tell, there is no “FAQ” set that specifically addresses the safety considerations that should be taken in to account when building and operating one of these things.


algocu (talk) 14:01, 29 September 2016 (UTC)