- Ip has a very strong impact on energy confinement, density limit and L-H transition, hence on extraoplation towards DEMO. For example: A=2.8 vs 3.1 being more attractive because it allows for larger Ip.
- on energy confinement time Ip^1-2 seen robustly in scaling laws even when downselecting subsets as shown by Clemente). The Ip impact in H mode is thought to be due to the pedestal pressure scaling with Ip. Using the IMEP workflow (MISHKA for MHD stability and a critical grad(Te)Te as a proxy for ETG turbulence) a strong Pped scaling with Ip is found Ip^~2. if using another scaling, still with MHD-MISHKA, but making other assumption on the pedestal width scaling, Ip increase can even lead to Ped reduction! Therefore it is essential to understand the physics driving the pedestal width! Also the limit being peeling or ballooning is likely essential too as noted Hartmut. Jeronimo noted that we are speaking of Type I ELMy H modes, while DEMO should rely on ELM free regimes. So at this stage the idea is to understand ELMy H mode Ip impact on confinement. Since this is our largest database... then that should allow us to question better the pedestal limitation and Ip impact of other regimes such as QCE.
- follow up on modelling: integrate other pedestal model with turbulent core in the same spirit of IMEP and test the modification impact on a large database. Would be ideal to down slect ~100 multi machine multi parameters ELMy H modes (with very well diagnosed pedestals) as targets for such large scale validation exercise?
- follow up on experiments (or data mining?): test impact on Pped of fuelling (pellet vs gas), of Ip, of shaping, of B, q95, Zeff etc anything that can help understanding the physics of the width.
- on density limit: if P_sep is the limiting factor (as suggested by Giacomin, more on May 14), no more limitation in density when we go towards ITER and DEMO. The limitation is on the transition back to L mode. Therefore mandatory to explore density limits at high power in all tokamaks. Explore role of Zeff/nsep resistive modes.
- follow up on modelling: explore the density limit in L mode using TGLFsat2 in integrated mdoelling since TGLFsat2 captures resistive modes. Check if Psep seen to be a key scaling parameter.
- follow up on experiments: explore density limits at high power, in detached conditions (high nsep, Zeff) etc.
- on L-H transition. Emiliano recalled E. Delabie scaling of the power threshold with Ip vs B (see PSD^2 meeting https://indico.euro-fusion.org/event/3275/). With a lower rms if using Ip. Emiliano is confirming this result if using only AUG data. In view of DEMO, transiting around the min in density, confirming the scaling with respect to the the power going only to ions as proposed by Ryter 2014 on other machines (other than AUG and CMOD) would corroborate using the ion power scaling instead of the total power.
- follow up on modelling: first pedestal formation have been obtained using integrated modelling, even if not 100% accurate quantitatively, could be used to understand physics interplay btw particle source, Er building and turbulence and the consequences on parametric dependences
- follow up on experiments: it is important to explore the impact of detachement on H mode access, of opacity (fuelling with pellet as expected in opaque ITER/DEMO vs gas) on H access, of torque as Emiliano proposed that Ip could matter in presence of torque due to Vfi*Btheta term in Er (so balanced vs unbalanced torque), Te/Ti, resistivity/Zeff, etc
There are minutes attached to this event.
Show them.