Hello, all. I am trying to understand the principles of the Polywell from an unsophisticated point of view, so I hope you'll be patient with this question. I read in a message that the paths of perhaps 20 percent of high energy He ions will be obstructed by the magnet housings. Am I guessing right that the high energy from the ions will mostly be absorbed by the housings? And will this energy show itself as heat?
I understand that the magnet faces for a 100 MW reactor will be about 4 meters apart. If the preceding is true, then my rough calculation is that each square meter of housing facing the well must transfer a couple of MW of energy to the magnet coolants, and those coolants must convey 20 MW of heat from those surfaces out of the chamber. If all I am guessing is true, is that a major engineering hurdle? Or is that performance already well-established?
Also, since the physical reactor size isn't much bigger for, say, a 1 GB reactor, the surface area of the inward facing magnet housings won't be much bigger either. But the power to be absorbed will be much higher. Is this relationship a natural limiting factor for maximum reactor size?
Many thanks.
Neil Ferguson
Conveying heat from the magnet housings.
Hello Neil,
Welcome to Talk-Polywell.
Your understandings about heat dissipation are correct. The products of fusion (of D-T etc) are mainly high-energy neutrons which bang into the MaGrid housing and both dissipate as heat and cause transmutation of the surface. This heat will have to be dissipated, and yes, it's a fair bit of heat. When you consider that a superconductor will be running at <20K inside the MaGrid, that makes the heatsinking even more important.
With boron-11 as a fuel (often described here as pB11) the neutrons comprise about 1% of the exiting species. Mostly it's high-energy alphas, which will condense to helium (that's nice) and if you can collect them the right way, will provide direct electric power; but there will still be a great amount of heat dissipated.
Getting rid of this heat, and avoiding neutron spoiling of the MaGrid is what many posts here are about ... because it hasn't been really solved yet.
Regards,
Tony Barry
Welcome to Talk-Polywell.
Your understandings about heat dissipation are correct. The products of fusion (of D-T etc) are mainly high-energy neutrons which bang into the MaGrid housing and both dissipate as heat and cause transmutation of the surface. This heat will have to be dissipated, and yes, it's a fair bit of heat. When you consider that a superconductor will be running at <20K inside the MaGrid, that makes the heatsinking even more important.
With boron-11 as a fuel (often described here as pB11) the neutrons comprise about 1% of the exiting species. Mostly it's high-energy alphas, which will condense to helium (that's nice) and if you can collect them the right way, will provide direct electric power; but there will still be a great amount of heat dissipated.
Getting rid of this heat, and avoiding neutron spoiling of the MaGrid is what many posts here are about ... because it hasn't been really solved yet.
Regards,
Tony Barry
I'm most grateful for your reply. It's the Boron/Hydrogen process that I have been considering. One thing doesn't add up for me yet; no doubt there's still something I don't have right. It's my understanding that the magnet housings are at a much lower positive potential (KV's) than the surrounding "cathode" (MV's). If that is true, then what happens when the high MeV helium ions bang into the housings with such force? They don't just quietly deionize themselves do they? That's what I was thinking of when I suggested a 100 MW plant would need to dissipate 20 MW of heat.
Neil
Neil
Hello Neil,
The nuclear engineers advise me that the alphas will exit from the pB11 fusion region in more or less a random direction; and they will have an energy of about 2MeV; and so the only ways to reduce the energy of impact onto the MaGrid is to either
make the grid have a lower profile (narrower diameter, or elliptical profile coils with the long axis pointing to the fusion centre, or coils that are further away from the fusion centre);
or decelerate the ions by asking them to climb a high electrical potential (which cannot really be done on the MaGrid for other reasons);
or make the MaGrid resistant to alpha impact (surface coatings such as diamond or boron) and then cool the coils.
I understand the coil profile is dependant on the magnetic properties of the generating coils; and the coil distance from the fusion centre is dependant on the magnetic field strength and the resultant wiffleball effect (something to do with the electron gyri radius, but my maths is not up to a better description than that); and the MaGrid surface voltage is governed by other things which I am still unclear on; so really the only thing which we can influence is the surface coating and the cooling.
Yes; that means a heap of cooling, and may in practice be the limit on the output power of the polywell.
Note that MSimon has done a fair bit of investigation on this (and other) subjects pertaining to the polywell; you may wish to read his blog at
http://iecfusiontech.blogspot.com/
Regards,
Tony Barry
The nuclear engineers advise me that the alphas will exit from the pB11 fusion region in more or less a random direction; and they will have an energy of about 2MeV; and so the only ways to reduce the energy of impact onto the MaGrid is to either
make the grid have a lower profile (narrower diameter, or elliptical profile coils with the long axis pointing to the fusion centre, or coils that are further away from the fusion centre);
or decelerate the ions by asking them to climb a high electrical potential (which cannot really be done on the MaGrid for other reasons);
or make the MaGrid resistant to alpha impact (surface coatings such as diamond or boron) and then cool the coils.
I understand the coil profile is dependant on the magnetic properties of the generating coils; and the coil distance from the fusion centre is dependant on the magnetic field strength and the resultant wiffleball effect (something to do with the electron gyri radius, but my maths is not up to a better description than that); and the MaGrid surface voltage is governed by other things which I am still unclear on; so really the only thing which we can influence is the surface coating and the cooling.
Yes; that means a heap of cooling, and may in practice be the limit on the output power of the polywell.
Note that MSimon has done a fair bit of investigation on this (and other) subjects pertaining to the polywell; you may wish to read his blog at
http://iecfusiontech.blogspot.com/
Regards,
Tony Barry
Neil,
I have done some back of the envelope calculations on the subject and I think it is doable with current technology. You are correct about sizing issues as well.
Let me add that Tony has done an excellent job of answering your questions.
I have done some back of the envelope calculations on the subject and I think it is doable with current technology. You are correct about sizing issues as well.
Let me add that Tony has done an excellent job of answering your questions.
Engineering is the art of making what you want from what you can get at a profit.
Yeah, assuming the polywell works the way we hope it does, it becomes a scaling/cost/power issue. We can reduce the heat load by making the machine bigger and/or less powerful (power density, a function of magnetic field strength B^4), but heat load will drop as r^2 while costs probably scale as r^3.
On the plus side, we have about two orders of magnitude to play with before we get to similar costs as for ITER at similar power, assuming we start from Bussard's $200M 100MW estimate.
On the plus side, we have about two orders of magnitude to play with before we get to similar costs as for ITER at similar power, assuming we start from Bussard's $200M 100MW estimate.