Talk-Polywell.org

http://www.futurepundit.com/archives/006012.html


David

Sounds like the "micro-reactor" they are planning for Galena Alaska.

For those interested, this reactor was already discussed on this thread and the original design of the Traveling Wave Reactor can be found here.
Here is my takeaway from the new futurepundit and Technology Review articles:
1/ Terrapower have worked on marketing (blurb, video, diagram).
2/ More importantly, they have made several major changes to the design of the TWR:
-Cooling is now performed by liquid Sodium instead of gaseous Helium.
-Reactor control which was supposed to be assured by Lithium thermostating modules is now longer mentioned.
-It is not clear anymore if the reactor is meant to be buried 100m under the ground, below a huge sand heat sink.
Many changes raise many questions and unless more convincing details are published, I am afraid that this interesting project will have lost a bit of its credibility. As an example, unheard of technology is required to prevent corrosion and leakage of the sodium circuit, for a period of 200 years and without any maintenance (if the reactor is buried).
It does bear similarities with Toshiba's 4S reactor planned for Galena, AK. Toshiba's 4S is less ambitious but at least they propose a solution for core control (neutron reflector panels instead of control rods).
Polywell and fusion fans, please notice that the assertive scientist speaking on the video is John Gilleland, the former managing director of US ITER.

As an example, unheard of technology is required to prevent corrosion and leakage of the sodium circuit, for a period of 200 years and without any maintenance

Yes. That particular design factor seems rather iffy to me.

If it could be made to work in any reasonable way the US Navy would have been very interested because it would greatly reduce the weight of their nuclear systems.

Instead of 2,000 psi pressurized water - 20 psi "pressurized" sodium.

I think the US Navy had a go or two at it and then gave up. The liquid lithium required for ITER will have similar problems.

I thought the Alfa class nuclear subs used liquid metal cooling in their reactors. The problem with that was that the cooling metal must never get cold, or the circulation of coolant would stop, with potentially catastrophic consequences. Thats at least according to my very limited knowledge.

Skipjack wrote:I thought the Alfa class nuclear subs used liquid metal cooling in their reactors. The problem with that was that the cooling metal must never get cold, or the circulation of coolant would stop, with potentially catastrophic consequences. Thats at least according to my very limited knowledge.

Ah, yes... First is *seems* like a no brainer that they should use Ambient pressure, liquid metal systems, but perhaps the shutdown/startup requirements are the show stopper. The whole system would take many $Billions of development.

Thanks for the head's up: The Soviets did give it a go! According to Wikipedia: "The practical problems with the design quickly became apparent" and they replaced the reactor design team. Ah, to be a fly on the wall (proficient in Russian, of course).....

Skipjack wrote:I thought the Alfa class nuclear subs used liquid metal cooling in their reactors. The problem with that was that the cooling metal must never get cold, or the circulation of coolant would stop, with potentially catastrophic consequences. Thats at least according to my very limited knowledge.

Corrosion was the bigger problem. Liquid sodium on one side of a heat exchanger and pressurized water (on the order of 500 or 1,000 psi due to the pressure temperature curve of water) on the other didn't work out well.

Helius wrote:

Skipjack wrote:I thought the Alfa class nuclear subs used liquid metal cooling in their reactors. ...

Ah, yes... First is *seems* like a no brainer that they should use Ambient pressure, liquid metal systems, but perhaps the shutdown/startup requirements are the show stopper. ...

I was under the impression it was liquid PbBi.

It was.
I guess corrosion is worth with Pb than with Na. But I would be surprised if the Soviet Navy had plans to leave the ships at sea without a crew for 200 years.
The TWR is supposed to work without operators.

olivier wrote:Polywell and fusion fans, please notice that the assertive scientist speaking on the video is John Gilleland, the former managing director of US ITER.

I worked for John for 6 months in '99. We were trying to make a plasma-based system for partitioning the Hanford wastes. We had an elegant idea (thanks to Tihiro Okawa) and got good technical reviews, but DOE decided not to buy anyway. Before that he worked for Bechtel, so he has some hard-nosed engineering and business experience.
The traveling-wave reactor is interesting. I like the way it has on-site, sealed-core reprocessing. What I don't like is the idea of 30 years of fuel and/or waste in the core. That jacks up both the energy available for causing accidents and the potential radiological consequences of those accidents. Give me a fusion reactor with only a minute or so worth of fuel in the reactor chamber.

Art Carlson wrote:Give me a fusion reactor with only a minute or so worth of fuel in the reactor chamber.

All you need is a good neutron source. I wonder where we can find one...
Is this realistic - putting a nuclear fuel blanket around a polywell to generate fission? No need for sustainable chain reaction, a lot less unburnt fuel.

geleto wrote:

Art Carlson wrote:Give me a fusion reactor with only a minute or so worth of fuel in the reactor chamber.

All you need is a good neutron source. I wonder where we can find one...
Is this realistic - putting a nuclear fuel blanket around a polywell to generate fission? No need for sustainable chain reaction, a lot less unburnt fuel.

Just remember to switch your fueling lines from the protium and boron tanks.

If your don't need to get net power from your neutron source, why not just use a beam-target arrangement? Why use fusion at all, instead of spallation?

Art Carlson wrote:If your don't need to get net power from your neutron source, why not just use a beam-target arrangement? Why use fusion at all, instead of spallation?

Because fusion can give you a lot more neutrons?
Beam-target arrangement has the advantage of making the idea of "a minute worth of fuel" feasible, but is this really needed? With no sustainable chain reaction and the need for a separate neutron source - there are a lot less things you have to worry about.

geleto wrote:

Art Carlson wrote:If your don't need to get net power from your neutron source, why not just use a beam-target arrangement? Why use fusion at all, instead of spallation?

Because fusion can give you a lot more neutrons?

I think a spallation source will give you more bang for the buck. You need something like an accelerator for either a spallation source or a polywell, but you have to satisfy fewer constraints to make the SS beam. For the target, the SS just needs a solid block, e.g. beryllium, not this fancy superconducting magnet cage, where you have to worry about cooling and forces and arcing and such. Most of all, the density of target atoms is many orders of magnitude higher, leading to a more compact neutron source. I don't think you will be dominated by energy costs, but even if you are, I don't think a spallation source requires that much more energy than a Q<1 plasma fusion source.

Art Carlson wrote:I think a spallation source will give you more bang for the buck. You need something like an accelerator for either a spallation source or a polywell, but you have to satisfy fewer constraints to make the SS beam. For the target, the SS just needs a solid block, e.g. beryllium, not this fancy superconducting magnet cage, where you have to worry about cooling and forces and arcing and such. Most of all, the density of target atoms is many orders of magnitude higher, leading to a more compact neutron source. I don't think you will be dominated by energy costs, but even if you are, I don't think a spallation source requires that much more energy than a Q<1 plasma fusion source.

This is a rather simple concept with a straightforward implementation. Why hasn't it been tried many years ago?