Thrust 05 05.001 Zarnstorff Mike 6/08/2009 17:28 Thrust 05 For this to be useful, you must include an additional requirement: your solution must be consistent with burning at plasma Q~20. I.e., the fraction of external current drive must be low, ideally ~10-20% of the total current, and the external heating must be no more than ~18% of the total plasma heating (for Q=20). Also, there must be a requirement on the maximum variations in the plasma reactivity. E.g, clearly, 100% variations must cause additional problems and not be allowed. What is the acceptable variation? 05.002 Zarnstorff Mike 6/08/2009 17:29 Thrust 05 Cross-connects: you should note that Thrust 17 is attempting to accomplish the same goal, but using other techniques (3D shaping). 05.003 Baylor Larry 6/08/2009 17:32 Thrust 05 An important aspect of controlling a burning plasma is controlling the isotopic mix (D/T)as mentioned in your talk. Perhaps this should be included in "Active control" of your 1 pager. 05.004 Reiman Allan 6/08/2009 17:37 Thrust 05 I thought that the presentation was much improved from the draft thrust write-up. The thrust now is about advanced tokamaks, which does make a reasonable thrust. The one sentence summary should be modified accordingly. 05.005 Greenwald Martin 6/08/2009 17:42 Thrust 05 There is an implicit assumption that the only reactor relevant regime is with beta_n well above the no-wall limit. There may be a solution at higher Bt, high q, high beta_p, but with only modest beta_n. This requires spending more on the magnets, but may result in a much more robust configuration. This option should be investigated... 05.006 Boozer Allen 6/08/2009 17:47 Thrust 05 If the other thrusts were carried out, what would be missing that this thrust would develop? Control of pressure, current drive, and rotation must be done with little power. Inventions in this area must be made before they can be tested, and they need to be shown to be credible by theoretical modeling. The question of how good can the performance can be is misleading. A certain level of performance is required; one must show that it is credible that these performance levels can be achieved. I would agree with Ron Parker that of the three legs sensors, algorithms, and actuators the algorithm issue is so simple compared to the other two that putting the three together is misleading. 05.007 Peng Martin 6/08/2009 17:51 Thrust 05 Great thrust and description. Since controllability is a function of proximity to know instabilities and limits, I recommend the inclusion in the thrust research that characterizes the science of controllability starting from parameter regimes relatively far away from these limits, and develop the science while approaching them progressively. 05.008 intrator tom 6/08/2009 17:55 Thrust 05 Active feedback and control is a good idea, maybe the only alternative for unstable regimes. The parable of the jet fighter plane being dynamically unstable but actively controlled is interesting. On the other hand active control of any system is not compatible with failsafe design. As a machine builder with a lot of hardware experience, I try to stay away from scenarios like this. The safety and reliability standard for a jet fighter is very different (casualties will be pilots and the crash target) from a nuclear facility that is possibly near a population center like a city (where there is a catastrophic downside to system failure). Do we really want this solution? In general simple technology is preferred if we have the option. 05.009 Berk Herb 6/08/2009 17:55 Thrust 05 If we seriously want to base our fusion concepts on operating above passive stability limits we need a good base of experiments that demonstrate such control. This can be done in present day tokamak experiments if feedback tools are to be developed to operate above passive limits. Thus if the concept is worthy, there is a need to document the improvements and reliability achieved by working significant above passive stability limits. Demonstrating this may give credibility to an aggressive feedback direction to planned work on ITER and Demo above passive stabilization limits. 05.010 Temkin Richard 6/08/2009 21:59 Thrust 05 Thrust 5 describes the future need for the Control Systems that would be used to operate the auxiliary systems. This is an interesting topic and the presentation by Alan Turnbull was very thoughtful (and entertaining.) But, this is putting the cart before the horse. We need direct advocacy of the auxiliary systems themselves, and advocacy at a high level! The important technologies for heating, current drive, fueling, etc. must be developed well beyond present day capabilities to meet future needs. We need Thrust 5 to cover the need for the enabling technologies and auxiliary systems. They are not covered in the other thrusts! The Theme 2 write-up on the research needs for auxiliary systems, including plasma heating research needs, is excellent. The research needs for auxiliary systems were supposed to be included in one of the thrust areas, specifically as a major part of Thrust 5. Thrust 5 is called “Develop and Demonstrate the Enabling Science and Technology For Controlling and Su 05.011 Temkin Richard 6/08/2009 22:07 Thrust 05 I would not use the word “actuator” to describe the plasma control systems such as heating, fueling, etc. The word “actuator” is ordinarily used to describe “a mechanical device for moving or controlling a mechanism or system.” Using the word “actuator” will really confuse people, especially non-fusion people. I would replace “actuator” with a specific device or devices such as “heating and fueling system.” I would make the replacement everywhere in the Task 5 write-ups, both the one-pager and the six-pager. 05.012 Milora Stan 6/09/2009 09:46 Thrust 05 Amanda made the point that power density scales like beta^2 B^4. But it scales first and foremost like n^2T^2. Since T is essentially the same for ITER and DEMO it practically only scales like n^2. And n will need to be a factor of 2 at least higher for DEMO than ITER(already a challenge) because the power density needs to be 5 higher. We have not really faced the issue of high density control/sustainment in our experimental programs. Given the importance of the issue, the lack of attention to it in any detail in either thrusts 5 and/or 8 is surprising. A proper treatment of the issue would include theory, simulation, experiments and enabling technology. It should build on Thrust 4. 05.013 Temkin Richard 6/09/2009 11:25 Thrust 05 I put in a previous chit describing issues with this thrust. I want to make it clear that my suggestion is that this thrust either be completely rewritten or be revised to a very major extent. 05.014 Wesley John 6/09/2009 15:38 Thrust 05 Disruption avoidance actions (the 'A' in PAM) will comprise a specialized 'control' response that may combine a number of 'standard' magnetic and control and fueling/pumping actions, plus possible specialized actions such as modest impurity injection, etc. not part of 'stationary-state control'. A wide range of outcomes, ranging from full recovery to Thrust 4 conditions to a fast plasma shutdown 'scram' (ideally not as fast as mitigation) can be envisioned. Avoidance scenarios must be tested. Separation from the other Thrust 2 activities is not acceptable. 05.015 Raman Roger 6/09/2009 16:31 Thrust 05 The goal of fusion research is to burn fuel and produce power. Controlling fusion plasma discharges simply does not mean control current and heat the plasma. In an ignited system no external heating is needed. In a high bootstrap current fraction discharge the plasma will generate a substantial part of its own current. The more this self driven current is the more efficient is the reactor. Thus, capability for precise fueling is a powerful profile control tool. Another benefit of core fueling is increased tritium burn-up (hence less amount of it trapped in walls) and reduced burden on the pumping and tritium processing systems. While there are on-going programs for various other means of control (current, ELMs, RWMs, NTMs), there is no program in the US or internationally to develop precision core fueling capability. This is a severely lacking capability needed for a Demo and quite possibly for the AT phase of ITER. So it should not be hidden in the general thrust descriptions under th 05.016 Chan Vincent 6/10/2009 09:54 Thrust 05 The oral presentation came across as a "brute-force" approach to push the plasma against its natural physics tendencies. I think that approach will fail. A much better approach is to develop an appropriate set of control tools to assist the plasma to reach an optimized self-organized state. The optimized state(s) should be identified through research in scientific understanding. This philosophy needs to be stated up front. 05.017 Brower David 6/10/2009 10:35 Thrust 05 Three elements required for control are 1) sensors, 2) actuators and 3) control algorithms. Emphasis in 6 page thrust is largely on 2) and 3). Importance of 1) or needs for 1) are largely neglected.