Talk-Polywell.org

I just came back from a holiday in the Pacific Northwest. While I was in Seattle I spent an afternoon with John Slough. He actually has two labs in Redmond. One is closely associated with the UW, the other, which he calls MSNW in fond memories of the now defunct company that did much of the pioneering work on FRCs, is associated with Helion Energy and the work on PHD (Pulsed High Density).

The basic story line is that every FRC experiment ever done has seen tau scale with the product of trapped flux and B^2. This includes some experiments with the 10 T field he envisions for a reactor, and other experiments with the flux of his design. Because n*tau scales with B^4, if there is some deterioration of confinement with the combined parameters, he expects to be able to still get to a decent Q value by upping the field a bit. (10 T is not extreme, especially in a cylindrical machine, so there is room to play.) Apparently the scaling of the MHD tilting mode has been established in a variety of experiments now, so he should be OK with that, too. It is not clear that the mode is actually stabilized, but the growth rate is reduced enough that it shouldn't hurt him during his short pulse.

It seems that the rest of the FRC community is fixated on producing a steady state device, which forces them to a very different set of parameters. In particular, a steady state FRC would operate in regime that might be tilt unstable, unless you can play some trick like adding a fast ion ring current.

John Slough's philosophy is one that I am sure would appeal to the engineers in this forum. Although it would, of course, be nice to understand what is going on in scientific detail, he doesn't waste time worrying about it. He tries things out, and if it works, he builds on it. For example, they did an experiment where they smashed two FRCs into each other. You might expect that they would squish past each other, or tip, or disrupt, and do any number of deleterious things, but in fact they stopped and stuck, even when the experimenters weren't careful about alignment. I probably would have been so worried about all the bad things that could happen, and maybe even theoretically should happen, that I would never have done the experiment.

The fusion community has an aversion to pulsed devices. They don't even want a tokamak reactor to power up and down in a 24 hour cycle, although that would probably make a lot of sense. They like machines that make a big bang even less. It does put you in a tougher design range where you have to pay a lot of attention to cycle fatigue and arcing and all. (Sometimes it helps, though, like when you can design light weight magnets because they don't have time to move before the force is come and gone.) I say, stay away from pulsed power if you can, but if that's where the physics leads you, good engineering and experience will probably let you make a home there. John is thinking in the 10 Hertz range and a natural power level like 10 MW, but there may be a lot of freedom to choose your operating point.

John said he needs about 3 years and 20 million dollars to build a break-even proof-of-principle. So far he has not been able to interest any venture capitalists, although the time, investment, and potential payoff seem ideal. I think that might be because he is a better physicist than salesman. Aneutronic fusion is impossible, but if you have the nerve to claim that's where you're heading, maybe the physics-ignorant VCs bite more readily. (Rostoker at Tri-Alpha, by the way, not only has tens of millions in private funding, he has some really bright and experienced people like Dan Barnes and Michel Tuszewski working for him.)

I probed John as best I could, but I could find no obvious roadblocks. You at this forum know that spotting potential problems is my strong suit, and my expertise in FRCs should have helped me to do so in this case. Of course, it might just not work for any number of reasons, and John has no illusions about that. In short, this is one I have to stay on top of. If I can entice any of you to join me, maybe we could start a new forum or borrow a corner of this one. And if any of you have friends that are VCs, have them give me or John a call.

Hey Art!

Thanks for this really cool insight! I hope your vacation was great beyond the visit to Helion.
Lets wish John and his team good luck. I dont see why a pulsed device would be such a big problem. Couldnt one just have multiple pulsed devices work in tandem to overcome most of the problems?
Also, if I remember correctly Helions devices are not that big, are they?

Maybe they could put their heads together with the guys at General Fusion. It might be a case of one company has done more with the Liner and the other is more advanced with the Plasma.

_________________
CHoff

I’m glad to hear about this. Is this reactor possible to make into a “clean” fusion reactor or must it in al circumstances be a fusion -fission hybrid?

Several questions:

1) Several of the presentations I have downloaded on John Slough's FRC concept talk about it as part of a fusion/fission hybrid reactor. Can his concept be used for making power exclusively from fusion reactions?

2) Since aneutronic fusion is not possible, would his concept uses D-T or D-D fuel cycle? I assume that the main effect of the neutron bombardment would be the embrittlement of the reactor chamber walls. If so, how often would these have to be replaced and how radioactive would they be? In general, how much of a problem is this?

It seems to me that the VC's may be put off by the perceived radiation problems associated with D-D or D-T fusion, which is why they are investing in Tri-Alpha and not Helion. As you know, radiation is the big bugaboo for all of the greens and anti-nuclear people as well as the NIMBY's in general.

3) $20 million over a 3 year period is quite a lot of money. If the break-even is successful, how much more finance and time would be required to make a commercial plant or IP that can be licensed? If the scale-up is not successful, can it still result in technology IP that can be used for other applications?

4) Would he consider investment from Asian investors (Japanese, Chinese, etc.)? Are there any U.S. federal restrictions on selling this technology into Asian markets?

I'm giving my best guess on these.

Certainly. A hybrid is usually your back-up if your concept does not work as well as you would like as soon as you would like. All other things being equal, I think a hybrid would work better with a linear system than a toroidal one because it is easier to exchange blanket modules.

DD is 30 times harder to burn than DT (Lawson criterion), and the power density is 70 times lower. The benefit is that only 66 rather than 80% of the energy is released as neutrons, and you don't have to breed tritium. In an FRC you would probably have to more than double the magnetic field to get a burn (n*tau scales as B^4), which is probably possible. The cylindrical geometry makes it much easier to replace the plasma-facing components and blanket, so the neutrons should be less of a problem with an FRC compared to a tokamak. You might be talking an exchange of the first wall every couple months in an FRC rather than after several years in a tokamak, but the geometry makes it possible to do that. I would definitely stick with DT.

VCs don't like projects smaller than 20 million. Too much work for the payoff. I think I read 100 million more for a prototype. I suppose several years, too. I lost some sleep last night worrying about IP. The idea is public, but there are bound to be tricks to be learned before it works and these can be patented. In addition to the usual alternative applications of breeding fissile fuel and burning nuclear waste with the neutrons, there may be some applications with directed energy weapons or space thrusters.

I'm sure he would take money where ever he can find it. It might be hard to get him to move away from Seattle, though. I can't imagine any export restrictions other than the usual embargos to rogue states.

As you say, Art, it is great that folks actually run the experiments without a principle regard for when the theories say it's a no-go-er. It's not really a 'scientific' approach, but it is surely an engineering one, and one which I [hopefully clearly] never cease to garner support for.

The main motivation, though, seems to focus on the presumed n*tau = f(x,B^4) but it's already said that the theory breaks down to explain why the experiment runs at all. It can't have it both ways, either we're into territory where the theory applies, or the n*tau = f(x,B^4) theory doesn't apply! So I can't really see how a useful discussion on theory can be perpetuated for this experiment and we will have to wait for experimental results, just as with Polywell.

I guess the corollary of these two points is to say that it is an experiment to pursue, but there is nothing that particularly suggests B^4 scaling except for hopeful thinking - much as Polywell has - though, in fairness, hopeful thinking is a trait of humans that got us out of the trees.

The Polywell has delivered indisputable evidens that speaking to the public on google video might result in funding.

Indisputable, you say. Please, do let me know what this indisputable evidence is, I am not sure I have come across it. Clearly, EMC2 haven't either, else would not be proposing another 2 years of experimentation to refine 'nuanced' results to some form of conclusion.

It's not really that the FRC results contradict any theory (as opposed to rumored polywell results). It's more a question of murky areas where it is not clear what effects have to be taken into account. Concerning MHD tilt stability, the experiments show the stability boundary to be where the average gyroradius is a few times smaller than the plasma radius, with a factor thrown in for the aspect ratio. That sounds plausible, regardless of the details of the theory, and there are theories that show this scaling. For transport, experiments show that the profile likes to have a gradient scale length of a few ion gyroradii. Again, this seems fundamentally plausible, and is in rough agreement with theories based on the lower hybrid drift instability. (I thought I had killed this theory with my doctoral thesis, but that experimental result doesn't trouble John. Until we have an alternative theory that is able to explain transport, he'll stick with this one and assume the waves found some way of hiding from me.) I suppose the tau ~ B^2 must drop out of this result with some simple scaling arguments, but I can't reproduce it now. The other two factors of B come simply from pressure balance, so n*tau ~ B^4 is a lot more than wishful thinking. It is a robust experimental result backed up by theoretical arguments, even if the details of the theory are sketchy.

Very cool! I wish I could have visited that lab! But you probably made alot more use of the experience.

I think having a forum or a subsection of this forum devoted to FRCs would be a great idea. I suppose Joel Strout would need to OK the subsection.

Art, what do you think about the prospects of a steady-state FRC reactor? It looks like the next step is to create higher-flux (or higher-field?) FRCs so that nbi can work, or to use a bigger RMF driver. Seems like there's plenty to do, and the equipment should be relatively straightforward. And diagnostics/basic theory is important too.

I'll definitely apply to UWash for grad school! (Yikes, I should really start on those applications soon....)

I'm in favor of getting to practical fusion in any way that works. Love the idea of discussing FRC in a corner of this website, but Joe is the one to make that decision. In any case, Glad you are back. You liven the place up ! Of course we're all waiting for more data on Polywell, so having another design to talk about in the meantime would be great in my opinion. (Then we can wait on data from it ! )

Talk-PolyFusion anyone ?


David

Dr. Bussard and Dr. Nebel both thanked bloggers for their contribution to getting EMC2 refunded. The basis of the blogger work for the refunding was the Google Video.

That is as close as I can come to evidence.

_________________
Engineering is the art of making what you want from what you can get at a profit.

Interesting to see there's more FRC stuff going on out there. I've always thought FRCs had almost as good a shot as Polywell.



Nah, it just depends on the firm. Some VCs want something they can IPO, but lots of others are fine with selling on the private market, so they go as low as a million or two. It's not impossible to find VC money in the $250K ballpark.

http://en.wikipedia.org/wiki/Venture_capitalist


Tsk tsk.



There's definitely an ignorance problem. VC is full of smart people looking for the Next Big Thing to invest in. The problem with nuclear fusion is that commercially speaking it (of course) just doesn't exist as a field, so there's a real shortage of people who have much knowledge about it to inform their decisions, and (of course) there's a whole minefield of "unknown unknowns" between current theory and productization. So the VC guys will tell you this is basic research territory.

Where do you get this idea? What conclusions are you saying WB-7 was intended to reach that it hasn't?

The conclusions from WB-7 were sufficient to lead to a far more powerful machine than Bussard ever worked with from 1991-2007. That says WB confinement has been confirmed and now we're looking at how losses (and perhaps power, if one questions whether the previously measured fusion is beam-beam) scale with B. The proposal also asks specifically for a reactor design.