Recovery.Gov Project Tracker

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

Post by D Tibbets »

93143 wrote:Ambipolar, not bipolar.

...

Also, if my 'Langmuir onion' multiple-well concept is correct, the MFP only needs to be on the order of the plasma wavelength, not the machine size.

Or I could be out in left field... maybe I should do some actual math on the concept someday...
I do not understand the the 'onion skin' effect except that it might involve standing waves... My best conceptual model is that it is like a relay race, Each ion hands off the baton (KE ) to the next deeper ion (then returns to get another baton). Certainly complex potential wells have been mentioned in various sources, and I suppose this could represent such a layering effect. I think Bussard mentioned something to this effect(?) when he was talking about possible explanations for Robert Hirsch's results, whether something might occur in the Polywell was not excluded, but I recall that he said it was not necessary for it to work.
I'm not sure how this would address thermalization issues of angular momentum and energy spread. Without layering. The dense core contributes to most of the up scattering, and down scattering, while the mantle contributes less due to the lower density compared to the core, and shorter dwell times related to the edge, and with intermediat Coulomb collisionality. With this gradient applied to multiple layers with energy exchange between each layer, the results are much more foggy for me.

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

bennmann
Posts: 241
Joined: Sun May 17, 2009 5:56 pm
Location: Southeast US

Post by bennmann »

Dan, thank you so much for posting those. Of striking note to me is early on in Bussards paper (forgive the weird number icons where formula should be);
If Ec is initially at 10 keV, for example, then
the tail of the distribution will be filled at 60 keV in this
time. Now􀀇 at this higher energy the fusion cross􀀄
section for the fast particles 􀀆of DT􀀇 with bulk ions is
about 20 times larger than for monoenergetic 10 keV
ions alone􀀇 thus their fusion reaction rate will be higher
by about 20􀀆60/2􀀇0.5 􀀃 110 than that of the 􀀆assumed􀀇
initial 10 keV population. These two e􀀊ects plus the
increase in ion speed with 􀀆Ec􀀇0.5 are responsible for the
well􀀄known fact that most fusion reactions in thermal􀀄
ized DT at ca. 10 keV come from regions at 50􀀄60 keV
in the Maxwellian tail.
which may be what you were thinking too as you have used the 60 keV number in recent posts. Cool stuff.

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

Post by D Tibbets »

Further ruminating of density and well depth, a few thoughts.
If the density at tested conditions was higher than expected. This would suggest that the Wiffleball trapping factor was better. This is a representation of the leakage rate of electrons in a given volume of plasma. Bussard, etel generally mentioned a gain of several thousand in the inside density versus the exterior density (which would be proportional to the number of passes before loss- which was quoted as ~ 10,000 passes for electrons). That he did not use a more precise number may suggest that there was some uncertainty in what could be achieved.
To achieve maximum density Beta has to be very near to one. As electrons are fed in at a faster rate than the losses, the pressure (density) increases, and this feeds back to inflate the Wiffleball and improve confinement further. Finally the electron flow needed to maintain the high beta reaches a minimum, but to achieve this goal higher electron currents are initially needed. I think this may imply that with greater than expected densities obtained, the final electron requirements would be less, but only if the density of the plasma was set at a constant level. If the density is higher, then the electron flow must also be higher to maintain near neutrality. I think this means that greater electron flows would be required while the Wiffleball is building (Beta = 1 is approached), but once the B=1 is reached the maintenance electron flow may be less (at least in relation to the contained plasma density). With this convoluted reasoning, this could explain both the need for more powerful electron guns (at the same drive voltage) while at the same time being good news for the capability of the machine to reach positive Q's.

Then again, this may all be irrelevant to the what was said. They may have had optimistic density / Wiffleball trapping results, but their electron injection system was plenty powerful enough in terms of current, and they are indeed merely pushing the voltage up as mentioned as one of the possibilities.


Concerning the potential well depth. By thinking of gas behavior I might have a better handle on the relationship. The ideal gas law (and plasma is an ionized gas) is PV= NkT or PV= nRT. The pressure is dependant on the density (n) and the temperature (T). The voltage (or eV)substitutes for the temperature. You can have a hot gas or a cold gas in a container. The number of particles can be constant but the temperature and resultant pressure can be different. In a Polywell you can have the same number of ions, but at 100 eV not much will happen. But the same number of ions at 80,000 eV is much more interesting. The density and pressure are not the same. When you pump a gas into a container it heats up, this is basic physics. The same occurs when you pump charged particles into a chamber. At some point the density is limited by the final pressure obtainable (before the container bursts, or the pump reaches it's pumping differential capacity). You can cool the container and this lowers the pressure, but of course in a Polywell you want to maintain the temperature, which means the pressure must be higher if you wish to reach useful densities. It would be extremely easy to inject a few high energy electrons into a container and they would create a large potential well as they bounced around until they cooled. The pressure/ potential well decreases as a result. To keep the pressure up you have to put in more energy or replace the cooled electrons with a hot ones. From this rambling it seems that the question of the potential well depth (approaching the KE of the injected electron) depends mostly on the cooling rate. Better magnetic insulation and better Bremsstrulung suppression allows the plasma to stay hot with less input energy (whether in the form of hot electrons or supplemental heating with microwaves). Losses are the key. You cannot keep a room warm with a furnace if the front door is open, unless you use an extremely powerful furnace. You wish to conserve energy to achieve the same final result (deep potential well- warm room) so you close the door, caulk all the windows, and put thick insulation in the walls. A further gain may be achieved if you can utilize some of the heat outside the house. So a heat pump my be beneficial- this would be recirculation in the Polywell.

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

eige1123
Posts: 8
Joined: Wed Jul 28, 2010 2:28 pm

Post by eige1123 »

The higher plasma density in WB-8 prompted the need for higher heating power.

Higher than what? They don't say higher than expected...they just say higher. So, higher... It must be higher than something other than the WB-8...and the most likely candidate for the what is the WB-7.

I think all they're saying is that the WB-8 has a higher plasma density than the WB-7. Also, I'd assume that to save time and money they might reuse parts from the WB-7, such as the electron injectors. So what it all means is that they finally got to the point in their experiments where the WB-7 equipment wasn't properly scaled for the WB-8 and they had to replace it.

Plus, if you look at the first letter of every sentence in the report it can be rearranged to spell "Eureka"! (just kidding)

rjaypeters
Posts: 869
Joined: Fri Aug 20, 2010 2:04 pm
Location: Summerville SC, USA

Post by rjaypeters »

Skipjack wrote:Yeah, I bet Tom is probably biting his tongue right now, trying to figure out what he can say and how he can say that without breaking his NDA.
Well...Tom is probably so hip-deep in lawyers over his previous unauthorized disclosure the WB-8 is "shiny" it wasn't worth the effort to post the list of unanswered questions:

"...What scaling relationships have you found? What have the additional diagnostics shown? Found any truly unexpected results? Any show-stoppers?

Does annealing exist in polywells?

Are polywells only going to be useful for niche applications like high isp thrust, or will they be useful large scale power generators? Which fuels will they be able to use and for which applications?"

..to learn whether there are any answers to these questions.
"Aqaba! By Land!" T. E. Lawrence

R. Peters

ladajo
Posts: 6258
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

Plus, if you look at the first letter of every sentence in the report it can be rearranged to spell "Eureka"! (just kidding)
No, but it does spell, "DTW".
This is very ominous.
The development of atomic power, though it could confer unimaginable blessings on mankind, is something that is dreaded by the owners of coal mines and oil wells. (Hazlitt)
What I want to do is to look up C. . . . I call him the Forgotten Man. (Sumner)

dnavas
Posts: 84
Joined: Sun Nov 11, 2007 3:59 am

Post by dnavas »

eige1123 wrote:The higher plasma density in WB-8 prompted the need for higher heating power.

Higher than what? They don't say higher than expected...they just say higher.
True. They also say that they had to replace the electron source as it was insufficient. Now, if it was expected, why would they have built the machine with undersized bits?

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

dnavas wrote:
eige1123 wrote:The higher plasma density in WB-8 prompted the need for higher heating power.

Higher than what? They don't say higher than expected...they just say higher.
True. They also say that they had to replace the electron source as it was insufficient. Now, if it was expected, why would they have built the machine with undersized bits?
in my sims the electrons had a very tough time getting inside the magnetic containment field. w/a much stronger magnetic field, it's going to be much tougher so you'll need to pump a lot more electrons in to replenish the same number. i'm not sure the scaling laws took that into account.

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

Post by mvanwink5 »

Is it just by accident that EMC2 has dropped the scaling law testing language and marked final report as yes? It may be that next quarter testing is just to see where the WB-8 device limits are. Just some thoughts.

Best regards
Counting the days to commercial fusion. It is not that long now.

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

Post by D Tibbets »

Some additional replies.

A KE of 60 Kev is an arbitrary number chosen by authors to make some point. I think Bussard's actual target for a demo D-D fusion Polywell was a potential well depth of ~ 80,000 volts. Bussard certainly understood both the physics and engineering issues better than probably anyone (he had an engineering degree before he obtained the physics degree). I assume he selected the parameters of a 3 meter diameter, 10 Tesla B field, and ~ 80 KeV as the best compromise that maximized the fusion rate, the electron losses, the Bremsstrulung losses, the thermalization issues, engineering issues, etc. As for 60,000 KeV . A MIT graduate who was working on a similar concept machine (in some aspects) used 60 KeV as a border where the Coulomb collision thermalization issues became much less significant (of course it also depends on machine size, density, confluence,and other issues like mentioned earlier). This was in a thread where he was countering Cris MB's assertions that Coulomb scattering and thus thermalization could never be overcome.

Concerning the plasma - Electron supply question. EMC2 certainly had scaling predictions for B field strength, and radius for both input and output. They knew the power requirements. It is odd that they would mention increased density while also saying they need more powerful electron input. Granted though, they may have limped along with old, inadequate equipment for budgetary or time reasons. Perhaps they felt the data would be predictive enough that once collected and presented, they could then entice their benefactors to pay for the shiny new equipment. A power supply that could support over 10,000 volts at currents of hundreds of Amps is not a trivial piece of equipment. For WB6 the best power supply they had could deliver the volts, but only at a couple of amps. This was well below 10% of the current needed during the brief steady state run at ~ 40 Amps input. Thus the use of capacitors to work around this limit. I have no idea what power supplies they had for WB7 testing, or if they needed to also resort to capacitors. For WB 8 testing with B of ~ 0.8 T, and size of ~ 60 cm diameter. The input scaling (based on WB6) would result in a need for ~ B^0.25, and radius ^2 increased power or ~ 1.6 * 4 = ~ 6.5 times increased amps or ~ 260 Amps for the high Beta steady state. I don't know how much they would need for the earlier stages. Also, if indeed there was improved density, this would, I think, imply that the B density scaling was improved. This seems unlikely based on the well accepted physics for B scaling. That is why I suggested that the nubs may have been very important. The B scaling was as expected. The difference would be the baseline (WB6) that they started from. If the baseline was optimized, everything else would be changed proportionately.

For electron injection. It is mentioned in the patent application that the distance outside the magrid is important. Too close and the gun acts like a repeller (?) with problems like in WB5, too far away and the electrons do not get in . A number was mentioned, but I don't recall what it was except that it may have been around ~ 1.5 times the radius of the magrid and well aligned with a cusp.

Where will a Polywell be used? Of course this assumes the Polywell works at some level above Q= 1. The use in a space ship would be only if the machine works very well. The mass and thermal loads would otherwise be painful, especially in a D-D Polywell. If the Polywell works with D-D but not with P-B11, the Polywell might still be used in space, but with some considerations. If a D-He3 Polywell works in a space ship- or naval ship, it needs the He3 fuel. D-D polywells supplying grid power could donate the He3 waste product. This could then be used in Polywells where the neutron output would be much less and direct conversion might be doable so that much of the steam plant could be avoided. Significant cooling would still be needed, but much less, and net useful power might be obtainable without the added weight and complexity of a steam turbine. The last fallback application would be a D-T Polywell. If claims hold it would perhaps be more economical that a large Tokamak (if it also could be made to work).

Dan Tibbets
Last edited by D Tibbets on Thu Feb 02, 2012 8:32 pm, edited 1 time in total.
To error is human... and I'm very human.

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

D Tibbets wrote:
Concerning the potential well depth. By thinking of gas behavior I might have a better handle on the relationship. The ideal gas law (and plasma is an ionized gas) is PV= NkT or PV= nRT. The pressure is dependant on the density (n) and the temperature (T). The voltage (or eV)substitutes for the temperature. You can have a hot gas or a cold gas in a container. The number of particles can be constant but the temperature and resultant pressure can be different. In a Polywell you can have the same number of ions, but at 100 eV not much will happen. But the same number of ions at 80,000 eV is much more interesting....
i think in a polywell, PV=nRT is not exactly an accurate formula as the T refers to entropy of particles travelling random walks - a chaotic mess of billard balls, essentially. the P comes from the electromagnetic force per unit area that the billard balls, through their bouncing around, apply against a surface.

in a polywell, however, you don't have that kind of thermalization. you have two completely independent components: radial and axial. and while your radial "temperature" is very high if you measure it in terms of KE and forget about how "thermalized" it may or may not be. your axial "temperature" - as in the mean axial KE of particles - is very low. and so, thus, too, is the absolute variance of the axial component of the KE.

so now you have a very non-thermal system, and when talking about pressure you're a lot better off going down to first principles and realizing here you're just talking about the outward radial electromagnetic force of the ions against a spherical surface. now as far as magnetic flux goes, since their inertia is almost all tangential to the surface, there's virtually no magnetic flux from the ions. so now we're electrostatic and electrodynamic, i.e. voltage and current. and current is from voltage difference and em-field, and the ions reactions to em-fields is negligable compared to that of voltage, so we can disregard that component, leaving only voltage to determine current. .... long story short, as far as the ions are concerned, the pressure incident on a concentric surface is just the voltage flux through that surface.

as far as the electron temperature is concerned, well they're really cold in the center. so their electric pressure is pretty much the static electric field, which is pretty constant. the counterbalancing magnetic pressure is more complex to calculate directly. in any case if they're reaching electron densities higher than expected that is definitely good news. that means both that confinement is better than expected and that the well depth is higher than expected. and as ions are confined by the well, that also means ion density is higher than expected, and fusion yield is proportional to density squared.... so yeah, higher electron density = very good.

i'd fathom it's just both:
what i said in a prior post about harder to get electrons in a higher mag field.
they're running higher mag fields so now they need to scale up the electric current as well.

and in anycase means they're scaling up the mag field.

Roger
Posts: 788
Joined: Fri Jul 06, 2007 2:03 am
Location: Metro NY

Post by Roger »

ladajo wrote:
Another consideration. Nebel mentioned that the Nubs in WB7 were hot spots, implying that they were more significant for ion and /or electron impacts /loses than anticipated. Going to wall standoffs may have helped due to the removal of cusp repeller effects effects similar to the repellar plates in WB5, which were large loss pathways for ions. With mild magnetic shielding of the nubs, the magrid potential would accumulate electrons here, and these would compromize the central electrostatic ion confinement. Thus lower density would result. The absence of these nubs near the mid plane of the magrid magnets could lead to greater ion confinement and density. Since the resultant ion density is greater, the contained electron density must also be greater, since the plasma is almost neutral. This would require greater electron input current, all other things being equal. I think that this would result in increased capacity electron guns/ magrid potential maintenance being required without invoking poorer electron confinement. Or perhaps not , my reasoning is nebulous here.
Interesting line of thought Dan. I did not run that one out in my first take. If they did go to standoffs, you may be on to a part of it.

If I'm following correctly....
I would expect that any design that removed nubs would remove some loses? No?

SO when designing/building WB-8 there should have been an anticipated improvement as far as loses are concerned because of no nubs.

We'rent we all thinking that no nubs would see less loses?
I like the p-B11 resonance peak at 50 KV acceleration. In2 years we'll know.

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

Post by D Tibbets »

Roger wrote:.....

If I'm following correctly....
I would expect that any design that removed nubs would remove some loses? No?

SO when designing/building WB-8 there should have been an anticipated improvement as far as loses are concerned because of no nubs.

We'rent we all thinking that no nubs would see less loses?
Yes and no. If WB7.1 used standoffs from the wall versus the nubs on WB7 they may have had some experimental data on the relationship. But some perhaps correct speculations follow.

As mentioned, it is not only the electron losses. If there was a WB5 repeller like effect, the ion losses would increase with the nubs. Because of the mass difference of the electrons and the ions, if ions are leaving at a higher rate (due to the space charge effects of a repellar plate), they may tug more electrons with them. Note that the reverse process of electrons tugging ions along during the brief time they are close to each other is much less significant. This is the same process (except in the opposite direction) that Bussard briefly described in the Google talk where he mentioned that the ions tugged the electrons along to a modest extent and thereby converted the pure electron square potential well into a more elliptical (actually parabolic) potential well. The relationships would be dynamic and perhaps difficult to predict. If this was the case then electron losses might be greater, but ion losses would be even greater yet (some of the electrons may still be recycled). This would translate into lower ion / near neutral plasma densities at the same input electron input. As this condition was present in WB6 and perhaps to a slightly less extent in WB7 ( they may have placed the nubs a little further outside the midplane of the magrids), then extending the predictions to another system without nubs may have given a low estimate of Wiffleball trapping factor at any given Beta.

The other considerations are that even if WB7.1 had wall standoffs, the distance outside of the magrid in a ~ 5 cm wide magnet can would be 2.5 cm, while the distance outside the mid plane of a 10 cm thick can (assumeing WB8 was twice the diameter with the same can minor radius proportions) would result in the stand offs being ~ 5 cm outside the mid plane of the magrid cans. That might make a significant difference(and the cans may have been even thicker if the WB4 can dimensions were used (~ 25% minor/ major radius vs. the ~ 17% of WB6 and WB7). Also, the mention by M. Simon that they may try coating the standoffs with an insulating ceramic would allow a negative charge to accumulate on the standoffs through electron impacts, and thus perhaps inhibit further electrons from grounding on them before they were reversed by the positive potential on the magrid. These would be electron repellars, but again the distance outside of the magrid may be the dominate factor. In additional a negative induced charge on the standoffs (there are still excess electrons over ions escaping the Wiffleball by at least the 1 PPM ratio and probably a much larger ratio) may attract the ions laterally to the standoffs, so that they do not have as much effect on the escaping electrons (do not inhibit recirculation as much- separates the ion electron pairs so any coupling is reduced). Such convoluted reasoning can perhaps lead to significant differences in performance from seemingly trivial differences in design, or not, depending the depth of theory understood, or of course the trump of experimental data.

Another perhaps important consideration, if indeed greater than predicted plasma densities were obtained. The greater Wiffleball trapping factor would result in a greater internal to external ration of charged particles. It has been discussed in other threads how demanding the vacuum pumping will be in a Polywell to prevent external to the magrid arcing, while maintaining internal densities that results in useful fuion rates. These vacuum pumping concerns would be reduced by even a modest inprovement in the traping factor.

It might seem confusing to say that the nubs reduced ion containment, and perhaps electron recirculation, and yet their elimination resulted in greater density, which implies better ion containment and also electron confinement/ recirculation. This would seem to indicate that less electron injection power would be needed. But this applies only if the density is a constant. If the density increases proportionately greater than the input savings, the net result would be greater input needs. Another way of saying this (I think) is that is they cranked the electron injection to the maximum the supply could deliver, and anticipated a density of X and a Beta of 0.99, they may have gotten a density of X*1.5 or 2 and a Beta of 0.9. There would be room for even better performance if they get more power in order to hit the adjusted target of Beta=0.99. *


*Note that this reasoning feels fragile, but I cannot not yet put a finger on any flaws.


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

Betruger
Posts: 2321
Joined: Tue May 06, 2008 11:54 am

Post by Betruger »

You gents reckon these bits of info are anywhere near enough to interest Dr Carlson for a few replies?

Skipjack
Posts: 6805
Joined: Sun Sep 28, 2008 2:29 pm

Post by Skipjack »

You gents reckon these bits of info are anywhere near enough to interest Dr Carlson for a few replies?
The few new facts, just fuel for speculation. I dont think Art will appear for that.
I wished he would come here and talk a bit more about what he has been doing lately in regards to FRCs. I know that his interest in FRCs was reignited by his visit to Helion Energy. I still think that some people should get together and do some lobbying for them like they did for Polywell and EMC2. Helion might not have such lofty goals, but their work is solid and they could achieve something in the very near term. I am still confused why they have not gotten any funding yet, compared to General Fusion, e.g.

Post Reply