NIMROD TEAM MEETING

1-2 August 2001
514 Engineering Research Building

(1500 Engineering Drive,Madison)

MINUTES

Attendees:


1. Scott Parker

2. Scott Kruger

3. Valerie Izzo

4. Dalton Schnack (tentative)

5. Tom Gianakon

6. Charlson Kim

7. Simon Woodruff

8. Rick Nebel

9. Alfonso Tarditi

10. Jim Reynolds

11. Giovanni Cone

12. Fatima Ebrahimi

13. Ming Chu

14. Jim Callen

15. John Wright

16. Eric Held

17. Hao Tian

18. Sangeeta Mahajan

19. Dylan Brennan

20. Ahmet Aydemir (possible?)

21. Carl Sovinec


8:30 AM-12:00 PM, Aug. 1

  1. Introduction of non-Team attendees


Dalton was a little late!!!


II. Development status

A. Matrix-free 3D solves (Carl)

B. 3D density evolution (Carl)


Full 3-D density solution option is now available.


C. 3D resistivity/moving vacuum (Scott K.)


Status: Packing at plasma/vacuum interface goes unstable with packed grid, only with 3-D solves. This affects the advection of the resistivity only, so it affects only cases that don't require the vacuum to moive. Must be a bug. Should be fixed in "a few months".


To be completed by APS.


D. Particle effects (Charlson)


Coupling particles to the NIMROD grid. Working on parralelization. Grid is kept on a single processor. Nobody thins this is going to be a problem.



Particles can be pushed. 200,000/processor. Working on parallelization. Almost complete. Will have parallel implementation by APS. Will also have results on kink mode stabilization by "energy scaling method".


E. Electron closures

  1. CEL (Eric)


Will program temperature equation and parallel viscous stress tensor, with slab test cases, by APS


  1. Particles (Scott P.)


One jot particle pressure is implemented, can use Vlasov method to cut down noise.



F. Advection (Carl)


Problem in advection may come from phase errors introduced by finite elements at high-k (high-k modes actually don't propagate). The solution is to introduce a more dissipative algorithm to damp oout the high-K modes and avoid the build-up (not instability) of high-k noise.


This is Carl's next code development project.


G. Two-fluid issues (Rick)


Very little agreement between NIMROD and M3D on 2-fluid results, but really aren't yet solving the same equations.


Two development paths:



2. Reduce model to eliminate whistler waves (perp. Force balance) ala M3D.


H. Thermal conduction convergence (Tom)


Thermal diffusion at large anisotropy works very well and converges well.


  1. Graphics (Scotts K. & P.?)


Long discussion about graphics, MDSplus, etc. No consensus about what graphics we want. We should make a list of what kind of graphics capability we want.


Maybe we should say that we want our present xdraw basic capability from whoever supplies our graphics (Klasky or Schissel/IDL).


If we go with AVSexpress, we MUST be supplied the licenses by PPPL/OFES.


REQUEST: All team members, please send a graphics wish list to the team mailing list. Dalton will work with these lists to formulate NIMROD graphics requirements and send them to Scott Klasky and Dave Schissel.


J. New diagnostics through NIMFL (Tom)


A much more flexible version of nimfl was described. This is very useful!


K. Resistive wall (Tom G.)


Module for resistive wall exists for NIMROD. GRIN may give the toroidal matrix now (M3D is using it).


Scott will contact Alex Pletzer. Tom will contacvt Morrell Chance.


III. Applications

A. Non-Team usage

- SSPX (Simon Woodruff)


Summary of Bruce Cohen's and Carl's recent work on NIMROD applications to the SSPX experiment. Plan is to use NIMROD as a numerical experiment to diagnose the same way as SSPX.


- DIII-D (Dylan Brennan)


Studies about determining the onset of tearing modes. Many good suggestions about how to proceed in then future.



Plans for HIT-SI simulation. New grid decomposition was developed and implemented by the user. Complete nonlinear simulation with n=0-5 has been done.


B. High-beta disruption (Dalton)


Blah, blah, blah??..


C. NTM (Tom)


3/2 in 86144 driven by 1/1 crash, but driven island within resistive MHD is n ot large enough to trigger the neoclassical tearing mode.


D. Island control via current drive (Tom)


Paper has been accepted.


E. Merging spheromaks (Ming)


Ming will use NIMROD for study of proposed merging spheromak experiment. Needs pressure and Ohmic heating. Carl will help.


F. RFP toroidal geometry effects (Carl)


Carl described the plans for him and his students in this area.


G. Plasma thruster (Alfonso)


Modeling of magnetic nozzle and exhaust region in the VASIMR plasma thruster.




VI. Next meeting date.


Saturday afternoon and Sunday morning of APS in Long beach. Dalton will make arrangements and send out announcement.



NO °-VIEW FROM GERMANTOWN°¶: NO DOE REP. THIS MEETING