Eureka! The net number of electrons in WB6

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

Moderators: tonybarry, MSimon

Post Reply
mattman
Posts: 459
Joined: Tue May 27, 2008 11:14 pm

Eureka! The net number of electrons in WB6

Post by mattman »

EUREKA.

Hello all,

The number of net electrons in WB-6. I must have done this calculation a dozen times. You can use Coulomb's law or Guasses Law to find the number. They essentially find out how many net electrons would be needed to make a 10 KV drop at a given distance. The problem is: what is the distance? I always assumed it was from the edge of the gas puffer to the center. I do not know this distance. I have gone to great lengths to estimate it from photos.

Today I had a Eureka moment. Bussard could not measure the voltage drop in the center. He probably did this with previous experiments. He would have most likely need to shove a Langmuir probe into ring center. But for WB-6, this was not done. His data reflects this. It never shows a plot centered around 10 KV. He must have estimated it. The 10 KV is taken from where the gas ionizes, to ring center. This is a distance I do know.

The distance is ~8 inches plus or minus three quarters of an inch. That means there was between 1.2 and 1.5E12 net electrons inside WB-6

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

Post by D Tibbets »

I don't follow you arguments. What is the 8 inches? Is this the distance from the center? If, so that would be a radius of 8 inches or 20 cm,that is greater than the radius of the magrid. If you are referrint to the diameter of the plasma quasisphere, that would be possible, though quite close to the inside surface of the magnets.

The video-
http://www.youtube.com/watch?v=7f5d-bRgieI

has nice grafics but it has some major flaws. The most obvous is the placement of the gas puffers well outside of the Magrid radius. Actual placement needs to be very near the Wiffleball border within the Magrid radius. It does not need to be at a cusp, it might actually be better to be placed away from a cusp. The neutral gas needs to have low velocity which means its Coulomb scattering cross section is huge. It quickly scatters in all directions. Outside of the Magrid this would quickly flod the vacuum chamber to high densities (perhaps ~ 1/100,000 th of an atmosphere and arcing (glow discharge would occur and this destroys any chance of forming a potential well inside the magrid. The slow neutral gass needs to be released at the Wiffleball border (perhaps several cm inside the radius of the magrid) where it is quickly ionized by the crowd of hot electrons contained within the magrid. This is quick but not instatanous, so some of the neutral gas molecules/ atoms may scatter back out of the magrid or traverse the magrid and enter the exterier vacumm chamber. There it is aviable for arcing and effective shut down of the machine if quantities are two great. There are two ways to minimize this. One is to have robust (some think impossible) levels of vacuum pumping. The other is to ionize more of the gas before it can escape the magrid volume. This will naturally occur with larger sized machine. It takes more time to transit a longer distance, thus the time dependent ionization process can ionize a greater percentage of the gas, which leaves less that can escape.

I have wandered if gas injection into the vacuum chamber at pressures of less than ~ 1/100,000 th of an atmosphere could lead to increasing numbers of ions being trapped as the neutrals drift through the Magrid, and if this would gradually build up the concentration in the magrid volume as the neutrals entering would be ionized and could not escape. This would be a simple solution if it worked and it would avoid several problems. But, that this is not done argues that it is not appropriate. The Wiffleball formation is dependent on several processes happening in a relative short time frame and I suspect this is incompatible with a gradual ion build up through this process. Though, I still wonder....

Your assertion that the potential well was not measured in WB 6 is suspect. That they reported that a potential well of ~ 10,000 Volts was achieved with an electron input voltage of ~ 12,000 volts implies that it was measured. Remember that WB6 was tested for several months before the final fusion of deuterium tests was undertaken. Also, the generation of deep potential wells with available electron injection methods was the holy grail (along with related confinement) of the Polywell project. That this was claimed for WB6 while WB5 failed, implies that this parameter had to be measured. I know that Languor probes had been used by EMC2, that this was abandoned in WB6 testing seems unlikely.

What was the density of electrons (and ions within 1 ppm) in WB 6? I am not sure. It seems I saw mention of ~ 10^13/cc. That would be ~ 10^19 / M^3. That may be too much . Scaling supposedly provided densities of ~ 10^22 in a demo machine with a B of ~ 10 T. With the density following B^2 scaling that would put WB 6 at a density of ~ 10^18/ M^3.
Density was measured by Bussard by photometry. Nebel said this was suspect and that they thus used laser interferometry for WB7. The two measurements reportedly gave comparable results.

Calculating density may be possible, but only if you understand all of the relationships. Electron input of perhaps ~ 45 Amps, effective confinement times of a few milliseconds. Using ~6*10^19 electrons/ Amp second * 45 Amps / 0.002 seconds / volume. I think this gives a result of ~ 2*10^21 /0.002 s = 10^23/ V = 10^23/ assumed ~ 8,000 CC Wiffleball volume (20 cm cubed)= ~10^19 electrons/ cc. While this seems to give a compatible answer within an order of magnitude, some of the parameters are only estimates.

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

mattman
Posts: 459
Joined: Tue May 27, 2008 11:14 pm

Re: Eureka! The net number of electrons in WB6

Post by mattman »

Dan,

I just saw this.

I will print it out, read it, and get back to you.

It means allot to get constructive feedback and it is seemingly impossible to be error free.

mattman
Posts: 459
Joined: Tue May 27, 2008 11:14 pm

Re: Eureka! The net number of electrons in WB6

Post by mattman »

Dan,
On the gas puffers: I think we mostly agree.

The difference seems to be in ion loss due to having too much energy.

I often think you have more information than I do. I have to extract information.

I am not working with a machine, so I might be wrong. I freely admit this. Here is my justification for puffer placement.


1. The puffer location is taken from a photo of WB6.

Image

2. We both agree: Ionization must happen inside the rings. If they don’t, they will fly in the wrong direction.

Image

3. If the electron cloud really is a “14-point star” structure, there ought to be electrons at the corners.

Image

4. If electrons have a bell curve of energy around 2,500 eV, then some should be hotter than the 16 eV needed to ionize the gas.

Image

5. Bussards presentation gives the dimensions of WB6. I assume these distances are in meters.

Image

6. Based on this blueprint, you can extract rough distances. The rings are about ~12” in outer diameter, ~8” in inner and are ~15.5 apart. The dimension in the film and posts are probably slightly off – every time I do this calculation the numbers seem to change slightly. But the distance should be – ballpark - 8 inches.

Image

7. Bussard surely used a Langmuir probe in past work with WB6. But in the November 2005, tests he could not have. He states that WB6 had a 10 KV well.

8. The well must have been estimated from the ionization point to ring center: a distance of about 8 inches. Put that information into Columb law. There is an estimate of 1.5 and 1.2E12 net electrons.

Image

9. The ion forms about 8 inches from the center. The ion falls towards the center. The farther away it is, the longer it takes to reach center. This means it is accelerated for more time. This means it builds up more kinetic energy. This improves the odds of a collision resulting in fusion. Hence, there is a logical reason for putting the puffers at the corners.



A couple of final thoughts:

A. Has the navy tried putting puffers at opposite corners? This may help with head on collisions.
B. Bussard does not heat the gas. So it comes in at room temp (~0.02 eV) before it ionizes up to over 16 eV.
C. The puffers need to be non-conductive if they are inside the wire cage.
D. You point out that making an ion with too much energy will probably lead to ion loss. I agree with this. I do not think we yet know what the “right” energy is.
E. I could buy the argument that the puffers can be well outside the cage. I would need time to think this over.

mattman
Posts: 459
Joined: Tue May 27, 2008 11:14 pm

Re: Eureka! The net number of electrons in WB6

Post by mattman »

Though, as I am finishing this post, I am looking at a photo of WB6 and I am not seeing capacitor emitters along the side.

... I may need to re-think this, yet again.

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

Re: Eureka! The net number of electrons in WB6

Post by hanelyp »

I recall mention of electron injectors facing the cube corner cusps on WB-6. On the magrid itself, facing inward, seems to me a good place for gas injection.
The daylight is uncomfortably bright for eyes so long in the dark.

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

Re: Eureka! The net number of electrons in WB6

Post by D Tibbets »

Based on the picture and discussion, I guess that the e-guns are the four extensions with bumps on the end pointing at the four top corner cusps. The e -guns were car light bulb filaments and the bumps would be consistent with the bases of these bulbs (glass removed). The gas puffer may be the tube approaching from about the 10 O Clock position and reaching to the mid plane of the magrid. It looks like it may be a glass tube and this would be consistent with the gas puffer introducing the gas near the Wiffleball border.

The quoted text may be misleading, out of context. There were multiple Beta tests, presumably at different densities and voltages and B field strengths. The numbers may reflect one of these tests but that may not reflect the only combinations leading to Beta= one conditions.

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

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Re: Eureka! The net number of electrons in WB6

Post by KitemanSA »

OMG, I can't believe I'm saying this, but... what he said; except the four bumps are over the virtual point cusps... :D

mattman
Posts: 459
Joined: Tue May 27, 2008 11:14 pm

Re: Eureka! The net number of electrons in WB6

Post by mattman »

Dan,

Just to be clear.


Image


So we have 4 electron emitters and 1 glass tube puffer.


It makes sense. The machine needs LOTS of electrons per ion, to work.

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

Re: Eureka! The net number of electrons in WB6

Post by hanelyp »

As I understand it, once the machine reaches operational conditions the electrons and ions are nearly in equal numbers. But there remains an excess of electrons sufficient for a large potential difference between the plasma and magrid. The total number of electrons is far greater than the number which would produce the potential difference absent the ions.
The daylight is uncomfortably bright for eyes so long in the dark.

mattman
Posts: 459
Joined: Tue May 27, 2008 11:14 pm

Re: Eureka! The net number of electrons in WB6

Post by mattman »

hanelyp,

I do not know for sure.

I can say (-) and (+) like to mix together.

So adding ions makes it hard to concentrate all the (-) charge in one place. This will kill the voltage drop in the center.

I think we want high, high numbers of electrons for every ion in the center.

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Re: Eureka! The net number of electrons in WB6

Post by KitemanSA »

You want enough electrons to make the well, then you want to add as many electrons as protons to keep the well. This can be done poorly by puffing neutral gas. But that leaves a bunch of cold ions that need heating quickly.

Oops, I meant cold electrons... :oops:
Last edited by KitemanSA on Fri May 03, 2013 3:29 am, edited 1 time in total.

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

Re: Eureka! The net number of electrons in WB6

Post by hanelyp »

You want the ions cold near the edge of the wiffleball, where most of the injected neutral gas will be reduced to ions and electrons. The potential well gives the ions the energy they need. The abundance of cold electrons introduced along with the ions isn't so good.
The daylight is uncomfortably bright for eyes so long in the dark.

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

Re: Eureka! The net number of electrons in WB6

Post by D Tibbets »

I'm not certain exactly what the dynamics of the potential well is. Is it simply the concentration of electrons inside the Wiffleball? Is it related to the averge energy of the electrons? Is it the radial movement of the electrons? Is it all or parts of the above?

Certain things seem to be that electron numbers optimally exceed the ion (Z=1) by ~ 1 part per million. That means that if there are 1.000000 *10^13 ions per cc, then there are 1.000001 *10^13 electrons per cc. This ratio has been calculated in the past and a 1 ppm difference is consistent with the voltage needed to maintain the negative charge excess. Much higher and the required voltages quickly builds past a million volts. The negative charge excess is a key element of the Polywell, and apparently this seemingly small excess gets the job done without requiring ridiculous voltages.

As far as electron temperatures, there is a cross section curve for ionization collisions just as there is for Coulomb collisions or fusion collisions. Apparently for hydrogen this curve peaks near 100 eV. Thus most of the ionizations will release free electrons at this ~ energy so long as the injected hot electrons exceed this energy. At first it seams these secondary, relatively cold electrons could not be heated much by the hot electrons before an equilibrium is reached.

But, if the injected hot electron current is ~ 40 Watts or ~40Coulombs/s * (6*10^19 electrons / Coulomb), or ~ 2*10^21 hot electrons injected per second, and the neutral gas puffer derived electrons are created at a current equivalent to ~ 1 Watt or ~ 6*10^19 electrons per second, then there are ~40 hot injected electrons for each cold secondary electron.
This allows for the average energy to be very close to the injected electron energy. I think it would be ~ 40 electrons at 10,000 eV +1 electron at 100 eV / 41 total electrons = 41 electrons at an average energy of ~ 9,750eV. These numbers are estimates but reflect my understanding of the reported input electron current and the relative numbers of secondary electrons that match the numbers of ions produced per second. As the ions have life times of at least 10-100 times that of the electrons,the injected electron energies dominates the total average electron energy by this proportion. With ion guns the dynamics would change some, but not by much.

A cross check of the ion production/ introduction in WB6 is assuming that the ions had ~ 40 times the life time of the electrons. If the life time of the electrons was ~ 0.2 milliseconds, the ion life times would be ~ 8.0 milliseconds. Electron recirculation does not change this ratio, though it does modify the energy input balance as it is 'recirculating ' the energy/ work needed. These life time comparisons are consistent with my estimated input rates necessary to maintain these numbers with WB6 claims, though I recall that the ion life times were reported as closer to ~ 20 milliseconds (which would improve these comparisons further).

Measuring the potential within the Polywell was done with Langmuir probes placed at various radii within the machines. Tom Ligon was employed at EMC2 at one point and he has related the challenges of implementing this measurement methodology. In the Google talk, Bussard related the shape of the potential well with electrons only and with the mixture of electrons and ions. The shapes were square and parabolic respectively (with the added central vertual anode due to ion convergence). This has been seen in research from other researchers involved with IEC fusion as well.

As far as density in WB6 , I think these numbers can be applied to obtain an estimate. Using the ions, if the input current was ~ 1 Watt or ~ 6*10^19 particles/ second, and the life time was ~ 20 milliseconds, the the maintained density would be ~ 6*10^19 particles / second * 0.02 seconds lifetime = ~1.2 *10^18 particles. This was distributed over a volume of ~ 24,000 cubic centimeters (very rough guesstimate of the internal Wiffleball volume inside the 30 cm diameter magrid) = ~ 10^14 particles per cc. This is reasonable to reported densities within an order of magnitude. All of these numbers are interrelated and consistent.

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

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Re: Eureka! The net number of electrons in WB6

Post by KitemanSA »

Oops, I meant cold electrons. :oops:

Post Reply