2021 state of the horse race: ITER vs. dark horses

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

I think it is a mixture of both science and engineering.
Ideally you build ever larger/stronger reactors to measure how the plasma behaves when "scaled up". From this you can defer scaling laws that then allow you to extrapolate to full scale devices. One way to objectively measure whether your scaling laws are holding up, is to look at neutron counts. Is the neutron output increasing as predicted? Good. Is it not, well then you likely have a problem in your model/science, or there is some engineering issue that you need to solve (e.g. plasma contamination which was/is one of the issues that have been plaguing LPPF for the longest time).
IMHO, there are lots of things that can go wrong and you need experiments to identify those. Simulations can only predict things correctly if your model accounts for all these things.

crowberry
Posts: 672
Joined: Sun Sep 08, 2013 6:34 am

Re: 2021 state of the horse race: ITER vs. dark horses

Post by crowberry »

Even for ITER the activities are a mixture of science and technology. TFTR, JET and JT60 showed for tokamaks that the step to Q>=1 is not large, but there still remains a lot of scientific questions to address concerning tritium breeding, disruptions, ELM;s divertors and so on in addition to all the technology development that is also needed.

The DT-results from JET will of course be very interesting to learn about this year.

Lockheed Martin was mentioned earlier in this thread. Their progress and activities can be followed by reading the abstracts that they submit to the Annual Meeting of the APS Division of Plasma Physics. Here is the link to their overview abstract from 2019: http://meetings.aps.org/Meeting/DPP19/Session/BO8.15.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by mvanwink5 »

TAE, based on their published research has two focuses, one on developing and proving their plasma measurements, their so called diagnostic suite, the second is to advance and prove their plasma modeling, including neutral beam and electron heating modeling. These two are strategically tied together and to how big the machine needs to be to gather the engineering data for designing a commercial machine. As a result of those two efforts the size of machine needed to gather the critical design data has changed, and that was the unspoken issue that has changed.

That was the point I was challenging as the size of machines seemed to be used to compare where a company was as compared to others.

No longer is this is a race to build a machine to show it can be done, nor how to do it. For TAE, past machine versions showed this, for instance neutral beam for heating was shown to be the path forward for TAE, needed plasma stability was shown, and machine scaling was shown. This is now about making a commercial machine that can be sold, and it appears there will be more than one company making one.

This is now about cost competitiveness and attractiveness of the machines, which ties into more issues than $/kwh.

The problem with ITER is not IMO the subsystems and their development, but the massive minimum size of their machine, with the sub-problem of safety of operating evermore powerful superconducting magnets.

LM seems to have a military application focus and the speed of the fund flow is tied to how interested the military is in fusion, and that is a function of root bound military leaders and politics. The competition for fusion is fission, and the problem of fission is safety of operation training and waste products, but fission is proven. Based on Crowberry's helpful link, it seems LM 2 years ago was still chasing fusion. Where they are today? Lasers, rail guns, and hypersonic missiles used to be the hot research. Will fusion heat back up? LOL, vicissitudes of Politics.

GF has a newly revealed compression chamber concept which is a change from a hard to manufacture spherical design to a lower cost cylindrical chamber. I would guess this is a result of their hiring of the high pressure fluid scientist and putting together a modeling team. Not much said about the change though, just a short youtube video. Curiouser and curiouser.
Counting the days to commercial fusion. It is not that long now.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by mvanwink5 »

I would suggest that GF is looking ahead, is already sharpening their pencil. This is now about being the most attractive alternative.
Counting the days to commercial fusion. It is not that long now.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

Lockheed, is still pursuing their fusion reactor (though I am still skeptical of their design). I know that from the recent ARPA-E presentation on FUSION Diagnostics First Annual Review Meeting (see thread below).

GF has significantly changed their original design. It now reminds me more of the LINUS concept and what Compact Fusion Systems are doing, though with more pistons and a different target (Tokamak vs FRC).

That should be a warning sign for people. It means that their original concept including their model and simulations were off. But hey, maybe I am over- interpreting things.
One more concern I have is that they are using a tokamak as their target for compression, yet their visualization shows a very tiny center column, more like a spherical tokamak. Actually it looks thinner than the spherical tokamak designs I have seen from the UKEA and even TE. How they are going to make that work is anyone's guess. I mean, that central solenoid has to go somewhere. Also, that central column will get the brunt of the neutron radiation because (at least to my understanding) that is where the density will be at it's highest point during compression.

As for TAE. I am not sure they have sufficiently proven anything. Confinement times are still below what they expected for Norman.
I am actually quite weary of the language they are using because I think it is sort of misleading. There’s a pretty big difference between plasma configuration lifetime (how long a plasma is around) and energy/particle confinement time (how long an given particle is trapped). For example they are using the beams to refuel, so even though they have 30 ms of configuration time, the actual particle confinement time is quite short (<1 ms). That is at least my understanding of the matter. Anyone feel free to correct me. I would be very happy to learn that I am wrong.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by mvanwink5 »

GF has not significantly changed their design, it only looks that way superficially. GF achieves their spherical compression via piston timing. It boils down to cost and lead time to build. Modeling is what is enabling them to use an easy to manufacture, lower cost cylindrical compression vessel.

As for TAE containment was proven before Norman. Norman was to show heating + containment + continued scaling and that is done. Modeling was significantly improved as was diagnostics, all needed to go to their next machine.

Watch the money.

As to LM, funding is the question. If it is all coming from internal, then it is likely the funding level is enough to keep the project from dying while waiting for the military to shift priorities. Good luck, the Navy seems happy with fission. As for their design, yes, they have issues to solve and they don't seem easy, not impossible, but those have been discussed here.

EMC2 is also a cusp machine and modeling results should have been out long ago per Park, but I have not heard anyone say what the results were. Is EMC2 a victim of funding or science or both? No one with inside information seems to be talking about Polywell anymore.
Counting the days to commercial fusion. It is not that long now.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by mvanwink5 »

As to GF's plasma, they built their plasma injector and tested thoroughly compression of the plasmoid created by it. This is old ground.
Counting the days to commercial fusion. It is not that long now.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

mvanwink5 wrote:
Sat Mar 27, 2021 11:03 am
GF has not significantly changed their design, it only looks that way superficially. GF achieves their spherical compression via piston timing. It boils down to cost and lead time to build. Modeling is what is enabling them to use an easy to manufacture, lower cost cylindrical compression vessel.
They changed from a compact toroid to a tokamak. Their compression chamber now has a central column. They changed from spherical compression to cylindrical. I would call those significant design changes. Yes, they did tests with (many) different injectors and determined that the tokamak worked best for some reason. I saw that. What they did not test so far is their new compression design and how the new target reacts to the compression. They also have not tested how the new compression chamber design will handle the environment the mechanical compression will subject it to.
Those are big question marks (for me anyway).
mvanwink5 wrote:
Sat Mar 27, 2021 11:03 am
As for TAE containment was proven before Norman. Norman was to show heating + containment + continued scaling and that is done. Modeling was significantly improved as was diagnostics, all needed to go to their next machine.

Watch the money.
If that is the case, why are they not going for a machine that can demonstrate net gain, at least with D+D and extrapolation?
And I am skeptical of the claim that they have proven confinement. IIRC, Norman was supposed to do 1 second of plasma lifetime and fell short by quite a bit. Confinement is not even in their papers. They are only talking plasma configuration lifetime, which is not confinement time.
mvanwink5 wrote:
Sat Mar 27, 2021 11:03 am
As to LM, funding is the question. If it is all coming from internal, then it is likely the funding level is enough to keep the project from dying while waiting for the military to shift priorities. Good luck, the Navy seems happy with fission. As for their design, yes, they have issues to solve and they don't seem easy, not impossible, but those have been discussed here.
LM's reactor design has grown considerably over the years. They were originally talking about something that could fit on a plane. Now they are talking about 1000 tonne+ range reactor size. Certainly not compact anymore. I think that is why LM is keeping funding low. They themselves are skeptical of the design.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by mvanwink5 »

"They changed from a compact toroid to a tokamak."
GF has tested several plasmoid shapes in search for one that could be adiabatically compressed and remain stable. It is called a test program. They now have a design and built a nearly full scale injector that meets their operation needs and which has been tested. Old news reported here long ago.

Does anyone really know what is going on at LM? LM is known for keeping developments dark.

SJ, your doubts and understandings generally aren't consistent with the companies' and their investors' views nor on publications and presentations. I will point out again, that funding for fusion is very difficult, yet GF and TAE have fully funded and staffed projects, with reputable top management.

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

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

mvanwink5 wrote:
Sun Mar 28, 2021 3:51 am
"They changed from a compact toroid to a tokamak."
GF has tested several plasmoid shapes in search for one that could be adiabatically compressed and remain stable. It is called a test program. They now have a design and built a nearly full scale injector that meets their operation needs and which has been tested. Old news reported here long ago.
I did follow their test program. Anyway, fair enough. We will see where it goes. I am looking forward to their compression tests. That is where the meat is. We will see how these go.
mvanwink5 wrote:
Sun Mar 28, 2021 3:51 am
Does anyone really know what is going on at LM? LM is known for keeping developments dark.
LM last gave a presentation a few years ago. I only know that they are still active because they are taking part in that ARPA-E diagnostics program.

mvanwink5 wrote:
Sun Mar 28, 2021 3:51 am
SJ, your doubts and understandings generally aren't consistent with the companies' and their investors' views nor on publications and presentations. I will point out again, that funding for fusion is very difficult, yet GF and TAE have fully funded and staffed projects, with reputable top management.
We will see. I for one am a bit disappointed by the scope of Copernicus. I was really hoping for something more, I guess. Maybe that disappointment is clouding my judgement. I don't know. Anyway, in both cases (TAE and GF), we will know more in a few years, from what I understand. Let's revisit that topic then.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by mvanwink5 »

Getting the plasma right so that it could be compressed adiabatically and remain stable during compression for GF was tough, but in the end their final plasmoid remained stable long enough that their piston compression could be slower than they thought they needed in the beginning. It turns out a slower compression by the pistons is vital as the liquid metal surface gets frothy and a slower compression rather than their original shock wave concept can help. It is still an issue and the reason they changed their liquid metal compression fluid from lead to lithium.

It seems the remaining question is how many pistons do they need, and it seems like it is more than they want if their latest 11 second youtube video is their latest thinking for it shows multiple pistons being struck for each steam driven piston. It may also be a trade off for changing their vessel to a cylinder from a more costly sphere.

I guess that the point I was apparently having difficulty is expressing is that the issue these days is not who is first, but who has the most attractive fusion island and support auxiliaries. There are going to be perhaps several machines for utilities to chose from. I would include your favorite Helion in that list, and probably some small Tokamaks, as well as GF and TAE. I don't have a feel though on which will be getting the lion share. Retrofits is a different market than new machines so there are different target markets.

It looks like Helion has a cleaner design, but has other trade offs. It is small enough to be attractive to all utilities. GF has a drop-in configuration for steam plant retrofits which is a huge sell. TAE assuming they can push their machine temperature to fuse Boron and Hydrogen avoids key safety issues with tritium, and that is their big goal for Copernicus, to shed enough insight to justify their Boron machine, Da Vinci. Fusing Boron is an attractive feature for new machines.

Anyway, fusion is no longer if, but when, and whose machine is the most attractive for what market.

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

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

mvanwink5 wrote:
Sun Mar 28, 2021 9:38 am
Anyway, fusion is no longer if, but when, and whose machine is the most attractive for what market.
I agree with that. The thing is that with how things are going now, the overnight costs are going to be a big driver for who wins. Renewables are really cheap to build and operate (but have the disadvantage of being unreliable). The current market trend seems to be towards renewables with gas peaker plants for backup. If you can not startup as quickly as a gas turbine, you have to compete with renewables on overnight costs.
Another advantage of being able to compete with peakers is that you do not have to run your reactor at all times, which can be an advantage for some more maintenance heavy designs. You can still be profitable since electricity prices from peaker plants are significantly higher than from baseload plants.
So either your reactor is very cheap to build and maintain or it can startup and shut down quickly.
D+T reactors will have to rely on steam plants and those have the problem that it takes a while to fire them up and get them to full performance. How much they will be able to load follow is another question. Can you operate the reactor at reduced output?
I think for some of these issues the answer can be to be smaller. Less than 100 MW or so. That way, you can activate and deactivate multiple units to load follow more easily. I might be wrong, but I believe that a smaller steam cycle is also faster to startup than a big one. So there may be another advantage there.
Concepts that can use direct conversion and can start up quickly are also very attractive in this regard. I am quite certain that Helion's reactor will (eventually) be able to startup within seconds and I assume that LPPF's reactor will be able to do same.
I don't know how this would go for TAE's reactor since it is steady state and rather big. I assume and hope that they will be able to compete with gas turbines in terms of startup time (minutes), but I am not sure. Any minute that you take longer, is a minute that you need to support with grid scale batteries. If you have to do that, then overnight costs go up quickly.

How would TE, GF, ZAP, Commonwealth and the other D+T concepts do in this regard? IIRC, ZAP and TE are aiming for smaller units in the 100 MWe range, but traditional Tokamaks, even those with high field magnets likely can not be economic with that output level. AFAIK, ARC will have to be a 500MWth+ machine. TE has a leg up in that regard, since Spherical Tokamaks ramp up Q quicker with size than traditional Tokamaks can. So they can stay smaller and still have a decent Q. ZAP has the advantage of being pulsed. So they can theoretically ramp up quickly. Again the steam cycle will be the limiting factor for how quickly they can do that.
A ZAP reactor core is really cheap to build, especially if you mass manufacture them, but likely the electrodes will need to be replaced often due to erosion. Right now, no one knows how often that will have to happen. The maintenance can be tricky in a neutron activated D+T environment (and disposal can be costly). In that case, being small is certainly better too.
TE will be roughly the same output level. How much will it take to build a production version of that reactor? How much will maintenance be?
IF ZAP's design works, TE might have a hard time competing.

How does General Fusion fit into all this? I don't know. Is going to be cheap to build? Is maintenance going to be low?
Thoughts?

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

So the latest news for TAE is that Da Vinci, their first machine to demonstrate PB11 fusion is now set to begin operation in 2028 (see TAE thread for source).
Copernicus is supposed to demonstrate "viability of net energy" by 2024/25

Prospective length for the commercial reactor vessel: 80 meters
Power out: 1,200 MWth and 350 MWe which seems oddly pessimistic at less than 30% conversion efficiency. I would expect a reactor with 1,200 MWth to have more electricity output, especially since PB11 would enable direct conversion. Unless they are not planning to drive a steam turbine with the heat and just do direct conversion (which still seems to be a bit low, IMHO). Or they are not planning to do direct conversion anymore?
Anyone understand how they get to that number?

Giorgio
Posts: 3061
Joined: Wed Oct 07, 2009 6:15 pm
Location: China, Italy

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Giorgio »

Skipjack wrote:
Tue Apr 06, 2021 4:01 am
So the latest news for TAE is that Da Vinci, their first machine to demonstrate PB11 fusion is now set to begin operation in 2028 (see TAE thread for source).
Copernicus is supposed to demonstrate "viability of net energy" by 2024/25

Prospective length for the commercial reactor vessel: 80 meters
Power out: 1,200 MWth and 350 MWe which seems oddly pessimistic at less than 30% conversion efficiency. I would expect a reactor with 1,200 MWth to have more electricity output, especially since PB11 would enable direct conversion. Unless they are not planning to drive a steam turbine with the heat and just do direct conversion (which still seems to be a bit low, IMHO). Or they are not planning to do direct conversion anymore?
Anyone understand how they get to that number?
A 30% net electric output is coherent with the expected efficiencies of a direct conversion system we was discussing in the "Z-Pinch Renaissance" thread few months ago.
A good chunk of the electric power will need to be used for the following cycle, and most of the heat will be low value heat that is not suitable to generate steam for running a turbine, so it will probably just be discarded.

If they can actually build a working p-B11 aneutronic fusion reactor with those manageable heat fluxes I would not have any complain at all even if the efficiency is only 30%. For a first generation machine it will be already an amazing feat, and improvements will naturally come with following generations once the system is proven to work.
A society of dogmas is a dead society.

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

Re: 2021 state of the horse race: ITER vs. dark horses

Post by Skipjack »

Giorgio wrote:
Tue Apr 06, 2021 7:49 am
A 30% net electric output is coherent with the expected efficiencies of a direct conversion system we was discussing in the "Z-Pinch Renaissance" thread few months ago.
A good chunk of the electric power will need to be used for the following cycle, and most of the heat will be low value heat that is not suitable to generate steam for running a turbine, so it will probably just be discarded.

If they can actually build a working p-B11 aneutronic fusion reactor with those manageable heat fluxes I would not have any complain at all even if the efficiency is only 30%. For a first generation machine it will be already an amazing feat, and improvements will naturally come with following generations once the system is proven to work.
How can 1,200 MWth be "low value heat"?

Post Reply