mistermack Posted December 9, 2021 Posted December 9, 2021 I think a paying fusion technology is inevitable. In fact, I'm going to argue that it could be happening right now, given the right investment. Even with the the old Joint European Taurus (JET) that first lit up in 1983. The JET holds the current record for Q, the input / output equation, with 24 MW in and 16 MW out, a net deficit of 8 MW. So how could be made to pay? I think you could make that pay by selling the waste heat in a district heating scheme. The 24 MW input will produce about 24 MW of waste heat. The 16 MW output will probably have produced nearly 50 MW waste heat, in the process of generating the electricity. So you're looking at about 75 MW of waste heat for a net input of 8 MW electricity. In the right location, you could sell that in a district heating scheme, cutting out huge amounts of fossil fuel burning. In Sweden, half of the residential and commercial premises are heated by district heating so it's not like it's a pipe dream. So even with technology that's forty years old, you could be using fusion energy now.
Prometheus Posted December 9, 2021 Posted December 9, 2021 My understanding is that Q is a poor metric of performance as it doesn't take account of the energy put into the entire reactor, just that used to create the plasma. Apparently things like confining that plasma with powerful electromagnets takes an awful lot of energy which are not taken into account with Q. If you use Q_total (i.e all the energy used by the reactor) as a metric then the value for JET is more like 0.01. 1
mistermack Posted December 9, 2021 Author Posted December 9, 2021 On 12/9/2021 at 3:30 PM, Prometheus said: My understanding is Expand Any link for that? I havn't heard it, and it would seem to be a dishonest representation of performance. I'm not denying that it's right, but I would be amazed if it was.
exchemist Posted December 9, 2021 Posted December 9, 2021 (edited) On 12/9/2021 at 3:30 PM, Prometheus said: My understanding is that Q is a poor metric of performance as it doesn't take account of the energy put into the entire reactor, just that used to create the plasma. Apparently things like confining that plasma with powerful electromagnets takes an awful lot of energy which are not taken into account with Q. If you use Q_total (i.e all the energy used by the reactor) as a metric then the value for JET is more like 0.01. Expand Yes. And as far as I am aware, little to no work has yet been done on how to get the energy out of the plasma, once it exceeds break-even. So it seems to me there is a hell of along way to go. On 12/9/2021 at 3:49 PM, mistermack said: Any link for that? I havn't heard it, and it would seem to be a dishonest representation of performance. I'm not denying that it's right, but I would be amazed if it was. Expand Here is one link that goes into it. https://whyy.org/segments/fusion-energy/ The energy you actually input to the plasma is only a fraction of the energy needed to drive all the systems at the facility. And then the energy output from the plasma goes through various lossy processes, including a Carnot efficiency factor (<50%) in the steam turbine or whatever heat engine is used to convert heat to electricity. According to this article, to break even and become a net provider of electricity to the grid, you will need a Q factor of around 10. With laser "inertial confinement" systems it is even worse, apparently. (The article also corroborates @Prometheus's figure of 0.01 (1%) for JET.) Edited December 9, 2021 by exchemist 2
Prometheus Posted December 9, 2021 Posted December 9, 2021 I got the numbers from Sabine Hossenfelder's blog - not verified the numbers myself but she seems like a trustworthy source (a theoretical physicist).
mistermack Posted December 9, 2021 Author Posted December 9, 2021 To be honest, I think those comments are more relevant to the inertial fusion research, than Tokamak designs. The shorter the period that the experiment runs, the more energy is used in the heating up from cold. So the overall Q is depressed. With a system that is constantly running, those losses dwindle to closer to zero. There is a lot of argument over whether heating up energy should be included, and it makes a big difference to the numbers that can be claimed. Getting the energy out is done by heat exchange to steam turbine. It's not bothered with at the moment, with the record duration of a plasma being six and a half minutes, but that doesn't mean that there are fundamental problems there.
exchemist Posted December 9, 2021 Posted December 9, 2021 (edited) On 12/9/2021 at 4:38 PM, mistermack said: To be honest, I think those comments are more relevant to the inertial fusion research, than Tokamak designs. The shorter the period that the experiment runs, the more energy is used in the heating up from cold. So the overall Q is depressed. With a system that is constantly running, those losses dwindle to closer to zero. There is a lot of argument over whether heating up energy should be included, and it makes a big difference to the numbers that can be claimed. Getting the energy out is done by heat exchange to steam turbine. It's not bothered with at the moment, with the record duration of a plasma being six and a half minutes, but that doesn't mean that there are fundamental problems there. Expand No, the link I provided was mainly focused on ITER and confirms the number for JET. So it's about tokamaks. But it does seem that the numbers for inertial confinement systems are even worse. Tell me, how does the heat exchange from this plasma to a steam boiler work? Has anyone produced a design yet for this? I don't think I've seen any steam pipes anywhere, wrapped around the torus, in any of the tokamak designs, but I don't pretend to have followed it all very closely. The torus seems to be festooned in magnets and cables in every picture I've seen. Edited December 9, 2021 by exchemist
swansont Posted December 9, 2021 Posted December 9, 2021 On 12/9/2021 at 3:16 PM, mistermack said: I think a paying fusion technology is inevitable. In fact, I'm going to argue that it could be happening right now, given the right investment. Even with the the old Joint European Taurus (JET) that first lit up in 1983. The JET holds the current record for Q, the input / output equation, with 24 MW in and 16 MW out, a net deficit of 8 MW. So how could be made to pay? I think you could make that pay by selling the waste heat in a district heating scheme. The 24 MW input will produce about 24 MW of waste heat. The 16 MW output will probably have produced nearly 50 MW waste heat, in the process of generating the electricity. Expand I'm not following the math here How does 16 MW output produce 50 MW of waste heat? I don't think you are doing a proper accounting of the energy here. https://www.euro-fusion.org/faq/top-twenty-faq/how-much-power-is-needed-to-start-the-reactor-and-to-keep-it-working/ The power required to keep a reactor working is an interesting question. Energy input is required to keep the plasma hot, because most of the energy produced by fusion is carried away by the neutrons. However 20% is carried by the helium nuclei, which remain within the plasma, so it is possible to reach a point called ignition, at which the production of hot helium is enough to sustain the plasma and the external energy sources can be turned off. It is not clear yet however whether that will be the optimum operating regime in a power plant – being slightly below ignition may give better control of the reactor (while still producing plenty of hot neutrons). So the 16 MW output is mostly radiated away as neutrons - largely unrecoverable, and also a problem* There's no electricity produced anywhere in this process. They haven't gotten to the point of a self-sustaining reaction, where you can turn off the input thermal energy, because (from the numbers above) you'd need to generate ~120 MW for that to happen. And you need to generate more if you want to start siphoning some off to generate electricity. *fusion is sometimes touted as being clean, radiation-wise. The fuel itself will not be more radioactive, which is a leg up on fission, but these neutrons will activate the containment vessel. On 12/9/2021 at 3:30 PM, Prometheus said: My understanding is that Q is a poor metric of performance as it doesn't take account of the energy put into the entire reactor, just that used to create the plasma. Apparently things like confining that plasma with powerful electromagnets takes an awful lot of energy which are not taken into account with Q. If you use Q_total (i.e all the energy used by the reactor) as a metric then the value for JET is more like 0.01. Expand Yes. This is another issue not accounted for in the math. Q=1 would be a PR milestone, but not an indication that fusion power is around the corner.
mistermack Posted December 9, 2021 Author Posted December 9, 2021 I'm very surprised regarding the various claimed different values of Q. I would personally have reported the power gain as MW in of electricity vs MW out of electricity. Maybe ignoring the start up energy, but giving a figure for the plant in continuing running condition. It would have to involve an estimate for electricity out, as you couldn't get the whole shebang generating in five minutes of running. On 12/9/2021 at 6:13 PM, swansont said: How does 16 MW output produce 50 MW of waste heat? Expand I have to admit that I assumed that the 16 MW output figure was electrical energy. Throwing in an estimate of the overall efficiency of the generation process, I was guessing that you would need 50 MW of heat to generate 16 MW of electrical power. If they really do mis-report the power gain as has been reported, then that's going to be nowhere near reality. The whole thing seems to be rather cloudy, and unclear. If you are putting 24 MW in, then you are going to get at least 24 MW of heat out. So the 16 MW figure doesn't seem to correspond to heat or electricity. You would need to do some study, to find out exactly what they are claiming.
exchemist Posted December 9, 2021 Posted December 9, 2021 (edited) On 12/9/2021 at 7:09 PM, mistermack said: I'm very surprised regarding the various claimed different values of Q. I would personally have reported the power gain as MW in of electricity vs MW out of electricity. Maybe ignoring the start up energy, but giving a figure for the plant in continuing running condition. It would have to involve an estimate for electricity out, as you couldn't get the whole shebang generating in five minutes of running. I have to admit that I assumed that the 16 MW output figure was electrical energy. Throwing in an estimate of the overall efficiency of the generation process, I was guessing that you would need 50 MW of heat to generate 16 MW of electrical power. If they really do mis-report the power gain as has been reported, then that's going to be nowhere near reality. The whole thing seems to be rather cloudy, and unclear. If you are putting 24 MW in, then you are going to get at least 24 MW of heat out. So the 16 MW figure doesn't seem to correspond to heat or electricity. You would need to do some study, to find out exactly what they are claiming. Expand I suppose that, to be fair, an initial Q factor of 10 to break even would start dropping as the power of the thing increased. I would expect the power absorbed by all the ancillary systems to become less, as a proportion of the total, as the power output increases. But I still miss any serious discussion of how a practical power extraction and generation system would be designed. Would it be by intercepting the neutrons in some moderator shell construction, surrounding the torus, that gets hot, and raising steam from that? Or would it be by intercepting radiant heat emitted by the plasma itself ? If the latter, how, given that the torus is surrounded by (I think?) superconducting magnets that have to kept very cold? Do you have any idea? It sounds to me like a difficult engineering challenge. Edited December 9, 2021 by exchemist
studiot Posted December 9, 2021 Posted December 9, 2021 (edited) On 12/9/2021 at 7:09 PM, mistermack said: I was guessing that you would need 50 MW of heat to generate 16 MW of electrical power Expand But then it wouldn't all be waste heat would it ? Edited December 9, 2021 by studiot
swansont Posted December 9, 2021 Posted December 9, 2021 On 12/9/2021 at 7:09 PM, mistermack said: The whole thing seems to be rather cloudy, and unclear. If you are putting 24 MW in, then you are going to get at least 24 MW of heat out. Expand That would be dissipated from the system, yes. On 12/9/2021 at 7:09 PM, mistermack said: So the 16 MW figure doesn't seem to correspond to heat or electricity. You would need to do some study, to find out exactly what they are claiming. Expand It’s the energy generated from fusion. From the link I found, you’d retain about 20% of that 16 MW. The neutron losses are presumably because tritium is used
mistermack Posted December 9, 2021 Author Posted December 9, 2021 On 12/9/2021 at 10:57 PM, swansont said: That would be dissipated from the system, yes. It’s the energy generated from fusion. From the link I found, you’d retain about 20% of that 16 MW. The neutron losses are presumably because tritium is used Expand So according to that, you're putting in 24 MW electrical, and the total heat output is about 40 MW, which is a poor return in heat, and certainly not economic. You could do better with ground source heat pumps. On 12/9/2021 at 7:29 PM, exchemist said: But I still miss any serious discussion of how a practical power extraction and generation system would be designed. Would it be by intercepting the neutrons in some moderator shell construction, surrounding the torus, that gets hot, and raising steam from that? Or would it be by intercepting radiant heat emitted by the plasma itself ? If the latter, how, given that the torus is surrounded by (I think?) superconducting magnets that have to kept very cold? Do you have any idea? It sounds to me like a difficult engineering challenge. Expand Wikipedia says this about energy extraction : In the case of neutrons carrying most of the practical energy, as is the case in the D-T fuel, this neutron energy is normally captured in a "blanket" of lithium that produces more tritium that is used to fuel the reactor. Due to various exothermic and endothermic reactions, the blanket may have a power gain factor MR. MR is typically on the order of 1.1 to 1.3, meaning it produces a small amount of energy as well. The net result, the total amount of energy released to the environment and thus available for energy production, is referred to as PR, the net power output of the reactor.[9] The blanket is then cooled and the cooling fluid used in a heat exchanger driving conventional steam turbines and generators.
exchemist Posted December 10, 2021 Posted December 10, 2021 On 12/9/2021 at 11:18 PM, mistermack said: So according to that, you're putting in 24 MW electrical, and the total heat output is about 40 MW, which is a poor return in heat, and certainly not economic. You could do better with ground source heat pumps. Wikipedia says this about energy extraction : In the case of neutrons carrying most of the practical energy, as is the case in the D-T fuel, this neutron energy is normally captured in a "blanket" of lithium that produces more tritium that is used to fuel the reactor. Due to various exothermic and endothermic reactions, the blanket may have a power gain factor MR. MR is typically on the order of 1.1 to 1.3, meaning it produces a small amount of energy as well. The net result, the total amount of energy released to the environment and thus available for energy production, is referred to as PR, the net power output of the reactor.[9] The blanket is then cooled and the cooling fluid used in a heat exchanger driving conventional steam turbines and generators. Expand Thanks. That's informative. What they call the Breeding Blanket seems to be the crucial component. That has led me to this paper,https://nucleus-new.iaea.org/sites/fusionportal/Shared Documents/FEC 2016/fec2016-preprints/preprint0228.pdf according to which ITER will test a design for this blanket, which will then have to be scaled up and proved in something called DEMO, which is a reactor to be built only if and when ITER succeeds in achieving stable fusion. It seems to me it is going to take several further decades of development work before we have commercial fusion power. If we ever get there.
swansont Posted December 10, 2021 Posted December 10, 2021 On 12/9/2021 at 11:18 PM, mistermack said: So according to that, you're putting in 24 MW electrical, and the total heat output is about 40 MW, which is a poor return in heat, and certainly not economic. You could do better with ground source heat pumps. Expand The total output is 40 MW, but only about ~27 MW could be used; ~13 MW is carries away by neutrons And using any of that 27 MW would likely be counterproductive, since it would tend to cool the plasma down, requiring more input.
mistermack Posted December 16, 2021 Author Posted December 16, 2021 On 12/10/2021 at 11:28 AM, swansont said: The total output is 40 MW, but only about ~27 MW could be used; ~13 MW is carries away by neutrons And using any of that 27 MW would likely be counterproductive, since it would tend to cool the plasma down, requiring more input. Expand I don't think that the energy carried away by the neutrons cannot be used. What they are saying is that the neutrons carry heat away from the plasma, requiring more input as you say, to maintain the plasma. The aim is to achieve a self-sutaining plasma, that doesn't need electrical input, so if heat moves from the plasma outwards, that tends to cool it, so you either need to be generating a surplus of heat, or be putting heat in from the grid. Obviously, generating a surplus of heat, above what is needed to maintain the plasma, is the goal. But the heat carried away by neutrons isn't waste heat, it's available for steam generation. It's just inconvenient, as it works against the plama being able to maintain itself without input. Eventually, the situation will be achieved, where the fuel input can supply all of the heat needed to maintain the plasma, in spite of the heat leaking away via the neutrons, and that would be a useable fusion power station. What I originally was suggesting was that you could make fusion economic before you reached that surplus state, by selling the waste heat in district heating schemes. But it looks like there is a long way to go before they get to that stage. At the moment, with the technology at it's current state, it appears that bigger is much better, when it comes to maintaining a plasma. That's why the ITER project is so big. Maybe, when more is learned about running a continuous plasma, they will be able to do it on a smaller scale, and plants could be sited near to cities, where the waste heat could be used. After all, if the risk of explosion or radiation leak is nil, then I don't think the public would be averse to having a plant nearby.
swansont Posted December 16, 2021 Posted December 16, 2021 On 12/16/2021 at 1:51 PM, mistermack said: I don't think that the energy carried away by the neutrons cannot be used Expand How do you capture it?
mistermack Posted December 16, 2021 Author Posted December 16, 2021 On 12/16/2021 at 2:14 PM, swansont said: How do you capture it? Expand As far as I can make out, that's what the lithium blanket does. It converts the kinetic energy to heat, and also uses it to breed more tritium. The heat is transferred via heat exchangers to steam generators for electrical generation, and then, if I had my way, to district heating schemes.
swansont Posted December 16, 2021 Posted December 16, 2021 On 12/16/2021 at 2:19 PM, mistermack said: As far as I can make out, that's what the lithium blanket does. It converts the kinetic energy to heat, and also uses it to breed more tritium. Expand That thermal energy is not captured by the plasma. Neutron absorption by Li-7 is endothermic. On 12/16/2021 at 2:19 PM, mistermack said: The heat is transferred via heat exchangers to steam generators for electrical generation, and then, if I had my way, to district heating schemes. Expand AFAIK no fusion reactors have incorporated these components. Why would they, when we’re so far from break-even?
exchemist Posted December 16, 2021 Posted December 16, 2021 (edited) On 12/16/2021 at 2:40 PM, swansont said: That thermal energy is not captured by the plasma. Neutron absorption by Li-7 is endothermic. AFAIK no fusion reactors have incorporated these components. Why would they, when we’re so far from break-even? Expand He's right about the breeder blanket, according to the link I included in my post of 10th Dec. This is proposed, in the DEMO reactor, to run at 300-500C, be cooled by helium and thus provide the heat for steam-raising. So whatever it is made of it must be able to convert the energy of most of the neutrons to heat. It will have some Li in it, for breeding more tritium. ITER has a blanket, though it won't breed and won't generate power: https://www.iter.org/mach/Blanket Edited December 16, 2021 by exchemist
mistermack Posted December 16, 2021 Author Posted December 16, 2021 On 12/16/2021 at 2:40 PM, swansont said: That thermal energy is not captured by the plasma. Neutron absorption by Li-7 is endothermic. Expand Exactly. It doesn't go to keeping the plasma heated, but it's still available externally to create steam for electricity generation, so it's not wasted energy. On 12/16/2021 at 2:40 PM, swansont said: AFAIK no fusion reactors have incorporated these components. Why would they, when we’re so far from break-even? Expand They wouldn't. I made my original suggestion assuming that they were much closer to break even than they actually are. Having said that, the existing plant is getting on a bit now. ( 1983 in the case of JET, although it's been updated a few times ). So something designed today might be a lot closer to break even. Break even seems to have many different meanings in the fusion industry. It's about time they adopted a standardises way of describing where they are at, that gives a more realistic impression than the phrases they use at the moment. Most people, reading the phrase "break even" would get a very false impression of what it means. And the various versions of Q are misleading too. A situation where a reactor can power itself and run for an extended period would be the gold standard for me. Even if all of it's output was being used to run it's own components. Once you get to that stage, you have the makings to go on and create a viable industry. From what I've read, ITER will not be anywhere near that stage. But it's successor, DEMO will be just about there. But at the current rate of progress and investment, it's about 35 years away. Having said that, all of the doomy predictions of climate change are for fifty years after that, so if a successful fusion industry kicks off mid-century, it still has the potential to mitigate events at the end of the century.
swansont Posted December 16, 2021 Posted December 16, 2021 On 12/16/2021 at 4:17 PM, mistermack said: Exactly. It doesn't go to keeping the plasma heated, but it's still available externally to create steam for electricity generation, so it's not wasted energy. Expand If it’s endothermic (Li-7), the Q of the reaction is not available. The energy of neutrons not absorbed is not available. Lithium melts at 180.5 °C What happens if the water line shuts down? Are you going to run the risk of it getting hot enough to generate steam, and then have the lithium melt when a pump fails?
mistermack Posted December 16, 2021 Author Posted December 16, 2021 On 12/16/2021 at 5:01 PM, swansont said: If it’s endothermic (Li-7), the Q of the reaction is not available. The energy of neutrons not absorbed is not available. Expand I'm not with you there. Where do you think the wasted energy is going? On 12/16/2021 at 5:01 PM, swansont said: Lithium melts at 180.5 °C What happens if the water line shuts down? Are you going to run the risk of it getting hot enough to generate steam, and then have the lithium melt when a pump fails? Expand You would have to ask the designers. But the plasma can be shut down very quickly, and the input electrical power can be switched off in fractions of a second, so I'm guessing that it's not an insurmountable problem.
swansont Posted December 16, 2021 Posted December 16, 2021 On 12/16/2021 at 5:08 PM, mistermack said: I'm not with you there. Where do you think the wasted energy is going? Expand Neutron absorption in Li-7 requires ~2.5 MeV to produce tritium. It’s endothermic. That energy is not available for heating anything. On 12/16/2021 at 5:08 PM, mistermack said: You would have to ask the designers. But the plasma can be shut down very quickly, and the input electrical power can be switched off in fractions of a second, so I'm guessing that it's not an insurmountable problem. Expand Wait. “Ask the designers” implies that this system is in place somewhere. I though this was your proposal. What reactor is doing this?
mistermack Posted December 16, 2021 Author Posted December 16, 2021 On 12/16/2021 at 5:14 PM, swansont said: Neutron absorption in Li-7 requires ~2.5 MeV to produce tritium. It’s endothermic. That energy is not available for heating anything. Expand Yes, but that's just a tiny proportion of the overall energy budget. And the tritium will release the energy again when it's used later. Wikipedia has this about the neutrons : "In a production setting, the neutrons would react with lithium in the breeder blanket composed of lithium ceramic pebbles or liquid lithium, yielding tritium. The energy of the neutrons ends up in the lithium, which would then be transferred to drive electrical production." Also this : The reactant neutron is supplied by the D-T fusion reaction shown above, and the one that has the greatest energy yield. The reaction with 6Li is exothermic, providing a small energy gain for the reactor. The reaction with 7Li is endothermic, but does not consume the neutron. Neutron multiplication reactions are required to replace the neutrons lost to absorption by other elements. Leading candidate neutron multiplication materials are beryllium and lead, but the 7Li reaction helps to keep the neutron population high. Natural lithium is mainly 7Li, which has a low tritium production cross section compared to 6Li so most reactor designs use breeder blankets with enriched 6Li. I don't know the fraction of energy used in the endothermic reaction creating more tritium, I would think that it would be just a tiny bit of the overall energy budget, and in any case, it's just in store till the tritium is used.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now