reddit: We are nuclear fusion researchers, ask us anything

[Joseph Chikva]

Magnetic deflectors if it weren't for the desire for the wiffleball. Magnetic deflectors STILL for the ions?

Yes, still for the ions if ions with high enough velocity will try to pass through them. Depending on angle some of those ions will be deflected back into the reaction zone.

[mvanwink5]

Joseph, you are bearing in mind that the ions are supposed to be contained by the electric potential well not magnetic confinement?
Best regards

[Joseph Chikva]

Joseph, you are bearing in mind that the ions are supposed to be contained by the electric potential well not magnetic confinement?
Best regards

Yes, in ideal case partial pressure producing by ions (fuel components) at the edge is equal to zero. But permanently growing thermalization will cause non-zero ion partial pressure too. And total plasma pressure is an addition of (ion partial pressure)+(electron partial pressure). This total pressure should be neutralized by mag field pressure B^2/2μ0 Permanently emerging alphas being confined by mag field will fastly thermalize plasma. As 2-10T mentioned by different people is a quite strong field - enough for alphas confinement in case of large enough geometric dimensions of reactor (1.5 m order).

[ladajo]

Alphas were said to get about 1000 passes by Nebel before loss. So very limited confinement.

And why do you persisit in thinking that Thermalizaation will be continually growing? Unceratin at best. More argument exisits to the contrary than for.

[Joseph Chikva]

Alphas were said to get about 1000 passes by Nebel before loss. So very limited confinement.

And why do you persisit in thinking that Thermalizaation will be continually growing? Unceratin at best. More argument exisits to the contrary than for.

1000 passes at what value of mag field? As if I understand correctly, stronger field will provide better confinement. So, more passes. You are right when doubt about continually growing thermalization. As every heating method has its limit of temperature. That limit comes when energy input will equalize with energy losses. In such systems (magnetic traps) as I know ~1 keV (1E107 K) and may be more is conventional temperature. Is that is acceptable for Polywell?
So, we came to the conclusion that stronger mag field causes more thermalization? Will scaling law B^4 be legit in this case? Ask any "reputable" for you man.

[KitemanSA]

Will scaling law B^4 be legit in this case? Ask any "reputable" for you man.

For power production, the answer every time has been yes. What is NOT certain is the loss scaling, and therefor the GAIN scaling. I suspect you are confusing the two, power, and gain, in your thought processes.

[Joseph Chikva]

Will scaling law B^4 be legit in this case? Ask any "reputable" for you man.

For power production, the answer every time has been yes. What is NOT certain is the loss scaling, and therefor the GAIN scaling. I suspect you are confusing the two, power, and gain, in your thought processes.

This is your confusion and not mine. Power production is proportional n^2 any time but n due to many factors is not proportional to B^2. And so, power, gain or whatever is not proportiona to B^4

[mvanwink5]

Joseph is right. In polywell, increase in voltage will also increase density.

[ladajo]

Joseph,
Try these threads:

viewtopic.php?p=18445&highlight=alphas#18445

viewtopic.php?t=1211&start=0&postdays=0 ... ght=alphas

(second page is hwere Dr. Nebel talks about the alphas)

[KitemanSA]

Joseph is right. In polywell, increase in voltage will also increase density.

Without the B to contain the electrons can you GET a significantly greater voltage? At a given beta and B field, wouldn't the density go DOWN with higher V? At the same pressure RETENSION capability (B and Beta remain equal), higher momentum per particle implies LOWER flux capability and thus LOWER density. No?

[KitemanSA]

Will scaling law B^4 be legit in this case? Ask any "reputable" for you man.

For power production, the answer every time has been yes. What is NOT certain is the loss scaling, and therefor the GAIN scaling. I suspect you are confusing the two, power, and gain, in your thought processes.

This is your confusion and not mine. Power production is proportional n^2 any time but n due to many factors is not proportional to B^2. And so, power, gain or whatever is not proportiona to B^4

Every independant source I have ever seen says it is. You state without site or link that it isn't. I will accept their mutual agreement over your unsupported statement.

[Joseph Chikva]

Joseph,
Try these threads:

viewtopic.php?p=18445&highlight=alphas#18445

viewtopic.php?t=1211&start=0&postdays=0 ... ght=alphas

(second page is hwere Dr. Nebel talks about the alphas)

More interesting for me is the following:

The alphas have such a high energy that they shouldn't be affected much by the electric field between the plasma and the magrid, and their density is presumably too small to make much of an electric field on their own. Therefore their losses should be determined by single-particle orbits in the cusp magnetic field. The size of the hole will be determined by the thickness of the alpha sheath, i.e. an alpha gyro-radius, times the linear dimension of the device R (and some moderate numerical factor). That calculation is going to wind up closer to 10 passes than 1000 passes. Rick, can you tell us where your 1000 came from?

As for any ignition fusion machines alphas are the main heating factors. Why they will not heat plasma in Polywell? Are you aware with other mirror machines?
Also permanent repeating beta=1 B**4*R**3 at constant beta does not seem seriously for me. As in the beginning of the same post Dr. Nebel speaks that neither experiment shown desired scaling.

[Joseph Chikva]

Every independant source I have ever seen says it is. You state without site or link that it isn't. I will accept their mutual agreement over your unsupported statement.

Every independant source for laymans. As if such scaling is legit it would be quite universal for any fusion machine. And not only for Polywell. And, so, by incresing of field 10 times you will get 10000 more power.
So, by simple changing of laughable 0.1T short solenoids in WB6 on 2T magnets you would get 240kW of fusion power instead of 1.5mW.
By increasing of radial mimension 10 tomes you would get 240MW.

What do you think, why such scaling does not work for any other fusion approaches?

[D Tibbets]

Ion magnetic containment is certainly a process to be considered in a Polywell. But it is mitigated to a large extent by the electrostatic confinement. The magnetic cusp Wiffleball confinement of electrons reaches ~ a few thousand passes. My understanding is that a simplified surface area ratio of cusp area to total surface area relationship determines the magnetic containment efficiency without needing to delve into more complex analysis. The same ratio will apply to ions which have a high enough KE that the electrostatic confinement becomes miner. Ions with KE much higher than the potential well such as fusion ions experience this situation. Up scattered fuel ions also would behave in this way, except that the electrostatic and magnetic confinement effects are closer and there is some gray area. Bussard recognized that a significant number of mildly up scattered ions would be turned by the magnetic curved magnetic fields and thus perfect confluence was impossible. The key is the ratio of electrostatically turned ions to magnetically turned ions. Based on some general numbers it seams that the vast majority of ions tend to be electrostatically turned. If ions turned magnetically survive for ~ 1000 passes, while the ions electrostatically turned survive for ~ 1,000,000 passes (simple assumption that electrostatically contained ions last longer than even recirculated electrons by a factor of 10 or more- allowing for the claim that from an energy perspective electron losses dominate). This would imply that only ~ 0.1% of the ion population is overcoming the potential well and being turned by the magnetic field per pass, and that ~ 1/1000 of these ions are hitting a cusp and escaping. A possibly important consideration is that these magnetically turned ions consist of mostly up scattered ions, thus there is a selection process where up scattered ions are preferentially lost from the system. This would impeed the growth of the up scattered high energy tail of the thermalization curve. Thus full thermalization could never occur. How much this contributes to ion energy spread is uncertain to me but I'm satisfied that it is significant.

A fusion derived ion may have energy levels ~ 10 to 100 times higher than the fuel ions, so the electrostatic confinement contribution versus the magnetic confinement ratios are reversed. Very few (~none) fusion ions are elecrostatically contained, so their confinement is limited to the magnetic cusp- Wiffleball confinement. As stated this only contains the charged particle for a few thousand passes*. Another important consideration is the MFP of the particles. At ~ 10-100 times the KE, the MFP will increase ~100- 10,000 times. Thus the opportunities for energy exchanging Coulomb collisions are greatly diminished. The Coulomb collisionability and the dwell time is just to small for significant heating of the plasma by the fusion ions. This also applies to the up scattered fuel ions also, though the contribution is much less significant. Note that this is one mechanism to modify the thermalization process. Claimed annealing is the dominate mechanism for those ions close to the potential well energy.

This is why Polywells are not ignition machines. The dwell time of a few milliseconds is not enough time for the fusion ions to interact much with other ions. Compare this to dwell times of many hundreds of seconds in a Tokamak or other similar ignition machine. The laser, Theta pinch approaches have very short confinement times, but the densities are high enough to compensate (the triple product). The Polywell is different in that the fusion energy is extracted through harvesting the KE of the fusion products directly ( through heat or direct conversion) as opposed to further heating the overall plasma and then harvesting the KE of the overall plasma.

Concerning the comment about plasma pressure- the plasma pressure versus the magnetic pressure apparently is complex. The 2008 patent application describes the necessary magnetic pressure/ strength necessary to contain the electrons and subsequently the ions.

* The likelihood of the fusion ion escaping to the exterior through a cusp or hitting the magnet through cross field transport is indeed dependent on the relationship between machine radii and magnetic field strength . Nebel mentioned ~ 3.5 Tesla fields in a several meter wide machine.
Considering J. Roger's simulations where 1-2 Tesla fields and ~ 12-15 meter diameter machines were required to reach breakeven it may be that the B to R ratios may be consistent for both breakeven and for fusion ion (alphas) cusp escape dominance. If you reach breakeven (through a combination of magnetic field strength and machine radii), you may already have satisfied the requirements for alpha direct conversion possabilities.

Dan Tibbets

[D Tibbets]

Every independant source I have ever seen says it is. You state without site or link that it isn't. I will accept their mutual agreement over your unsupported statement.

Every independant source for laymans. As if such scaling is legit it would be quite universal for any fusion machine. And not only for Polywell. And, so, by incresing of field 10 times you will get 10000 more power.
So, by simple changing of laughable 0.1T short solenoids in WB6 on 2T magnets you would get 240kW of fusion power instead of 1.5mW.
By increasing of radial mimension 10 tomes you would get 240MW.

What do you think, why such scaling does not work for any other fusion approaches?

That is indeed B^4 R^3 scaling. I have seen this in multiple texts and papers. Even knowledgable critics have accepted this scaling. It is basic to plasma fusion science. What is in contention is if this baseline scaling applies, there may be some modifiers in select situations (like the presence of neutrals the type oc collisions that dominate- beam- beam, or beam - neutral or target, etc.). But, the primary contention is the cost side of the scaling. It seems that the B^0.25 loss scaling is accepted (?). The R^2 loss scaling question is more contentious. My understanding is that this was the primary purpose of the WB8 tests. WB7 tests would have shown the accuracy of the B4 scaling over a limited range.
Making the blind assumption that the scaling has been confirmed (perhaps with some modifiers) in WB8, that still leaves the question of this scaling being consistant at larger scales. Bussard's desire to go directly to a several meter wide machine would have avoided this uncertainty, though additional up front money would have been needed. For WB6 this was 0.1 T in 15 cm radius yielding 1 mW of fusion power at a cost of ~ 500 KW. Magnetic field input strength ignored as superconductors are assumed, just like the Tokamak. WB 7 may have modified this baseline set of values. If the gain scaling is accurate, I suspect the 1mW value may be ~ correct. But the input value may have been different. Nebel,s mention of the problems with nub heating may imply that the input baseline costs may be significantly less. Certainly WB4 with 'huge nubs' was improved upon with the 'small nubs' of WB6. WB7.1 may have improved the situation even more, though I suspect to a much smaller amount. Also keep in mind that WB6 output estimates suffered from a large uncertainity, perhaps ~ 1 mW +/- 0.3 mW. With more tests and better instermantation, WB 7 may have optimistically given values of 1.2 +/- 0.1mW without invalidating WB 7 results. This could be significant from a limiting engineering consideration.

The output scaling is no different than that for a Tokamak. The difference is that the baseline is much different. The input costs for the Polywell and the Tokamak are the limiting factors. And to reach breakeven the Tokamaks are already at ~ 5-10 Tesla (?) The B fields in ITER are not going to be much greater than in JET (?). Assuming the B fields are increased 2X, then B^4 scaling would imply an output ~ 16X greater. That would be ~ 17MW increasing to ~270 MW. Increase the size some and the scaling is consistent with predictions of B^4 R^3. , just as with Polywell. I don't know the results for earlier Tokamaks but I suspect this scaling is consistent*. What has bedeviled the Tokamaks is uncertainty in the cost scaling and the tremendous efforts to overcome or compensate for them. Hopefully the Polywell does not suffer this fate.

*I'm referring to the performance of the Tokamaks when their magnetic fields are behaving, and not the time averaged output when the disruptive macro instabilities manifest. This is another claimed advantage of the Polywell, in that the magnetic fields are very well behaved.

Dan Tibbets