jmc wrote:
Well the coils used in JET were copper as they were cheaper then superconductors, whereas the coils in a fusion power plant will be super conducting. I realise you still have to put in energy to keep them refrigerated but I believe the numbers have been done and its condsiderably less than the fusion energy you get out.
The biggest limitation is the heating you have to put in to drive current non-inductively (1/3 of out put power) and that's assuming your neutral beams are 3 times as efficient as the state of the art.
No, this isn't the 'lost' thermal energy in the coils, this is the energy stored in the magnetic field and would be the same whether generated by copper, SC, or by the thoughts of God himself; that magnetic field has within it 1GJ. ITER's will have around 50GJ, meaning it has to run for over 100s at 500MW just to pay back the energy cost of forming the magnetic field.
As the plasma collapses in JET, and it takes the magnetic field 'with it' (as plasmas do), the whole of the physical torus physically deflects vertically by a few cm (from what I've been told) and in any case you can hear the collapsing field as a big 'thump' (even in the control room behind the multi-metre concrete walls). It's a lot of magnetic energy!!... it's a quater tonne of TNT equivalent!
When ITER's plasma collapses, the magnetic field alone will be like 10 tonnes of TNT being set off in the chamber.
So, with that little gem of info, does anyone have any doubts over the viability of tokamaks??.... nah! should be fine!
Sorry misread, I though you said GW. And JET does consume about a GW of power in its field coils. It only contains 350 MJ of magnetic energy and I think ITER is closer to 10GJ.
The ITER plasma should produce 500MW of neutron power and so needs to last 20s to "break even", plasma pulses are envisaged to be 15 minutes long.
What you say about the plasma taking the magnetic field out with it, only applies to disruptions 98% of JET pulses are ramped down quietly. Finally the magnetic energy in the toroidal field coils can be thought of as potential energy in a giant induction coil and as such should be reclaimable with high efficiency.
Your right to be concerned about the 10 tonnes of TNT in equivalent stored energy though that might be realeased in an event such as a disruption. Its something people are aware of and disruption mitigation systems are being designed and one will be installed on ITER. Not sure how reliable they are yet though.
"..the energy in the toroidal field magnets at full strength is about 41 GJ."
Anyway, the point being that it only ramps down 'gently' most of the time because it can only run up to saturation once over the course of a few seconds. And that's the tokamak's lot. It can't do no more! It's like going to buy a new car and you say "OK, start her up" and the guy turns over the demo car on the starter and it goes 'chug chug' and the engine doesn't start up and the seller says "there you go.. it turned over - the engine's a good 'un. We're good for your order for one, yeah??!?"
Oh, I calculated it simply by using the field of 5.3T to calculate the energy density and then multiplying it by the plasma volume.
5.3^2/2(4*pi*10^-7) * 850 = 9.5GJ
But maybe the blanket, shielding materials and coolant pipes are so thick that the volume the magnetic field exists over is 4 times larger than the plasma volume.
Tokamaks can do more, there's a number of ways of driving current in the plasma without being reliant on transformer action. Bootstrap current driven by the plasma's own pressure, neutral beam current drive, Lower Hybrid Current Drive, Electron Cyclotron current drive. (Both ECCD and LHCD are form of current drive driven by radio waves)
Lower Hybrid Current drive has proved itself in Tora Supra where pulses of 6 minutes 30 seconds were obtained.
Indeed. There is also a smaller (say 10%) contribution from the poloidal fields.
And you might note that I have given a 'healthy' 20% confidence of achieving these continuous current drive processes, in the finger-in-air 'assessment' I made on the other thread. The highest 'success likelihood' rating I've given any 'non-engineering' issue.
I am more concerned over the instabilities.
I might rather speculate that the current 'H-mode' transport barrier enhancement to power rating is (self-evidently) going to increase the likelihood of instabilities, and so I might also rather speculate that it might be found that a successful tokamak would need to be run in L-mode. Remember - ITER was originally designed to run L-mode and then this 'new physics' came along and people started [prematurely] breaking open the champagne!
What was the state of the edge when Tore Supra ran its long stability run? L- or H- mode?
Just imagine it, though, picture it: 10 tonnes of TNT taking up some 10m^3 of ITER's 800m^3, then you set it off...
Were it to be found that H-mode is a dead end, then the consequence is, of course, that ITER will need to be scaled up [yet again] by an order of magnitude. Eventually, as playfully commented on elsewhere, we'll end up with a minimum tokamak size which is the size of the Sun!
I think the chances of getting current drive working (eventually) are higher than 20%.
The biggest problem I can think of is control of burning plasmas, unlike non-burning plasmas, in a burning plasma you have a wildcard heat source you can't control that in many cases will exhibit positve feedback (if your on the wrong side of the fusion rate coefficient curve then more heating=more fusion= more heating) this wildcard heat source will in turn change the resistivity of the plasma in a choatic and uncontrolled manner which could result in lost equilibrium and a catastrophic disruption.
ELMs will be bad for ITER, but that's because ITER is an obsolete design from the 90s, there are new fancy divertor configurations that can safely spread the heat over a larger area. ELMs are small fry compared to the disaster a disruption could reap on the vessel.
At JET we can control disruptions and avoid them quite reliably. My concern is whether this level of control will be lost when the plasma starts burning.
Tokamaks will make good fusion-fission hybrids, but ignited plasmas may prove to be too unstable to maintain in a tokamak configuration.
jmc wrote:I think the chances of getting current drive working (eventually) are higher than 20%.
The biggest problem I can think of is control of burning plasmas, unlike non-burning plasmas, in a burning plasma you have a wildcard heat source you can't control that in many cases will exhibit positve feedback (if your on the wrong side of the fusion rate coefficient curve then more heating=more fusion= more heating) this wildcard heat source will in turn change the resistivity of the plasma in a choatic and uncontrolled manner which could result in lost equilibrium and a catastrophic disruption.
ELMs will be bad for ITER, but that's because ITER is an obsolete design from the 90s, there are new fancy divertor configurations that can safely spread the heat over a larger area. ELMs are small fry compared to the disaster a disruption could reap on the vessel.
At JET we can control disruptions and avoid them quite reliably. My concern is whether this level of control will be lost when the plasma starts burning.
Tokamaks will make good fusion-fission hybrids, but ignited plasmas may prove to be too unstable to maintain in a tokamak configuration.
I sure hope we get Polywell or some other version before ITER gives fusion a bad name.
Engineering is the art of making what you want from what you can get at a profit.
This is somewhat off topic, but I'll throw it in here.
I was under the impression that ITER was expected to reach a Q of about 2. Then I learned that the target is a Q of ~ 10. Concidering the slow pace of Tokamac developmentment and the logrithmic cost escalations with size, why did they aim for this excess power when so many questions about ignition and burning and instabilities remain. Did they hope to skip a generation (despite the fact that a smaller machine could potentially only cost a decade, while ITER is costing well over 2 decades)? Is the confidence of the gains uncertain enough that they felt this excess margin was needed ?
In some ways, it is like Bussard's desire to jump from WB7 to a full up prototype reactor without exploring various geometries and intermediate sizes that could develope the engeenering and science onto a firmer footing. Great if it works, frustrating and possibly time consuming if it fails entirely or partially. Of course, the diffrence is the time scales and budgets involved.
Bussard DID want to explore at least TWO different geometries in small scale before going to the proto-type. Remember, his WB7 was actually the first of his "other geometries", not the robust WB6. I think he would have termed that WB6.1. Oh well! Seems the naming convention diverged after the Valencia paper.
D Tibbets wrote:This is somewhat off topic, but I'll throw it in here.
I was under the impression that ITER was expected to reach a Q of about 2. Then I learned that the target is a Q of ~ 10. Concidering the slow pace of Tokamac developmentment and the logrithmic cost escalations with size, why did they aim for this excess power when so many questions about ignition and burning and instabilities remain. Did they hope to skip a generation (despite the fact that a smaller machine could potentially only cost a decade, while ITER is costing well over 2 decades)? Is the confidence of the gains uncertain enough that they felt this excess margin was needed ?
In some ways, it is like Bussard's desire to jump from WB7 to a full up prototype reactor without exploring various geometries and intermediate sizes that could develope the engeenering and science onto a firmer footing. Great if it works, frustrating and possibly time consuming if it fails entirely or partially. Of course, the diffrence is the time scales and budgets involved.
Dan Tibbets
You should have seen my arguments on that very question re:Polywell on NASA Spaceflight. I may have been a little timid. But I still things have to be taken in steps.
Continuous operation at the following powers:
100 mW - Proof of principle no significant thermal or neutron loads. Short duration operation: 10 seconds a shot - 10 to 20 times a day. 3T 1 m bore SC magnets. Can be used to map out pB11 operating regimes including the resonance region. 100 KV 25A grid supply - with a capacitor bank sufficient for startup. Explore POPS. Possible port for direct conversion experiments.
10 KW to 100KW - 9T 1.5 m bore SC MgB11 magnets. Raise operating density. Include POPS if experiments at lower power show promise or higher power is needed to definitely identify some parameters. Higher density in the reaction space. Can be operated at up to 200 KV - the peak of the pB11 fusion cross section. Include heat shielding of SC magnets. Possibly neutron shielding as well. One port for alpha direct conversion experiments.
100 MW - The largest field 3 m dia SC coils money can buy. The rest according to results of previous experiments.
Engineering is the art of making what you want from what you can get at a profit.
Also keep in mind that if the device can be made smaller rather than larger for each step the magnet field goes up at no extra cost. Half the size. Double the field strength. Power per unit volume would go up by 16 and volume would be 1/8th thus halving the size doubles the power. Roughly.
It is all a matter of stuffing the most current in the least space. And there will be constraints giving us a minimum size. It should be as small as we can make it.
Let me add that boron injection for the 100 mW experimental machine can be done with a blob of boron embedded in the coil casing and injected with a variable power laser beam.
Engineering is the art of making what you want from what you can get at a profit.
There are several boron hydrides that would allow both the boron and the excess hydrogen to be introduced as the same molecule in gaseous form if desired.
That limits your control of the mixture ratio, though. You want a large excess of hydrogen, but the hydrogen burns at the same rate as the boron, and might actually escape more slowly, at least on a particle lifetime basis.
I suppose using two separate feed gases might do it, depending on the ratio between fusion rate and loss rate...
93143 wrote:That limits your control of the mixture ratio, though. You want a large excess of hydrogen, but the hydrogen burns at the same rate as the boron, and might actually escape more slowly, at least on a particle lifetime basis.
I suppose using two separate feed gases might do it, depending on the ratio between fusion rate and loss rate...
For one thing the hazardous material restrictions get expensive for the boranes.
Keep it simple. Keep it cheap. Don't invite weekly inspections.
Engineering is the art of making what you want from what you can get at a profit.