NIMROD TEAM MEETING

26-28 January 2000
Santa Fe, New Mexico

MINUTES

WEDNESDAY, JANUARY 26
9:30 AM-12:30 PM

I .Discussion of ongoing applications
 
 

- Neoclassical Tearing Mode (Tom)
 
 

4 closures are in the code. There is Hirschman-Sigmar and 2 approximations to that closure plua rhw hole model.
 
 

APS results were not an NTM but a tearing mode. Looking for an NTM in a new test case. So far have found only numerical instability. This is being analyzed (new normal modes??).
 
 

- Stochastic transport (Eric, Carl)
 
 

Look for stochastic transport in RFPs. Use Eric's kinetic closure to estimate radial thermal diffusivity.
 
 

Eric will put a parallel version of the field line tracing code on the T3E so that everyone can use it,
 
 

- Spheromak modeling (Carl, Alfonso)
 
 

A paper is being prepared on Carl's spheromak results. Flux amplification vs, applied voltage for many different parameters.
 
 

Calculations of SSPX indicate that the triangles are working well. No problems are seen at the boundary between the quads and the triangles. The problem is that the equilibrium from Corsica does not satisfy regularity. Don Pearlstein is working on that problem.
 
 

- FRC (Alfonso)
 
 

Smooth FRC equilibria have been obtained by filtering the initial data. The beginnings of the tilt mode have been seen. One question is why interpolation seems to introcude noise.
 
 

- D3D High-beta disruption (Alfonso)

There are discrepencies in gamma vs. betaN for NIMROD and GATO. This needs to be understoon. NIMROD finds a resistive mode below the ideal MHD marginal point.
 
 

- High-S (Dalton)
 
 

Linear cylindrical cases have gone to high S (10**9). Still need more convergence info for the S=10**9 case. Toroidal calculations are underway.
 
 

- Two-fluid RFP (Rick)
 
 

Trying to isolate the instability that grows out of the origin when 2-fluid efffects are included. Some indications that it may be related to the electron advection. Could be related to regularity??
 
 

Also attempting to put gyoviscous effects into the model.
 
 

- ??
 
 

2:00 PM - 6:00 PM
 
 

II. Development
 
 

- Higher order elements (Carl)
 
 

The previous good convergence results have been confirmed for cases with a degenerate (r=0) point. Still need to implement this ansatz into the pre-conditioners, and for regularity conditions in poloidal representations. Also need to do it for triangles.
 
 

We may have a released version with biquadratic elements by about April.
 
 

- Parallel computations (Steve P.)
 
 

No progress; work continues.
 
 

- Vacuum (Scott K.)
 
 

No progress; work continues. Work on the passive advection of the shape variable will begin soon.
 
 

- Resistive wall (Tom)
 
 

Tom has been concentrating on NTMs. We're leaning toward using Morrel Chance's code for the external vacuum solution.
 
 

- Neo-classical closures (Tom)
 
 

Discussed above.
 
 

- Kinetic closures/particles (Charlson)
 
 

Trying to push particles in the NIMROD fields. Code is written but there are problems with the particle pushing (don't yet get the right orbits). Aim is to have accurate particle pushing by Sherwood meeting.
 
 

- Pre-/post-processing issues (Scott K.)
 
 

Web seems to be working well.
 
 

Rostom has to work to get funding to Nina!!!!!
 
 

Please put your results on the web in html format so we can copy the figures.
 
 

- Computations on tblocks (Alfonso)
 
 

Already reported.
 
 

- Flows
 
 

Higher order elements may help here because we have a theoretical basis for minimizing the div(b) error. Now we can revisit the time-stepping scheme.
 
 

- ??
 
 
 
 

DINNER AT THE SHED
 
 
 
 

THURSDAY, JANUARY 27

9:30 am - 10:30 pm
 
 

III. Status of PSACI (Dalton)
 
 

Dalton reported status as he knows it.
 
 

10:30 am - 12:30 pm

IV. Develop plans and goals for next 3 months
 
 

For reference, here are the milestones from the November team meeting:
 
 

GOAL RESP. PERSON DATE COMMENTS
 
 
 
 

1. Res. wall BC. implemented Tom. G 06/00 Changed from Scott K.
 
 

We will use Morrel Chance's code to get the vacuum response. This seems to be a reasonable approach. There are 2 issues: 1) coupling the bc. to the NIMROD solution; 2) the Green's function. Tom will start with a slab geometry to test issue 1. We'll move on to 2 when 1 is understood.
 
 

This milestone includes the incorporation of Chance's code into NIMROD .
 
 

2. Lin. high-beta disr. physics Alfonso 04/00 linear by Sherwood
 
 

There are presently results for gamma vs. betaN for both nimrod and GATO. Nimrod results are about 20% larger This is near the marginal stability point. We don't understand these differences but they shouldn't keep us from going ahead with the nonliear calulations.
 
 

3. Linear vac. validated w/GA Scott K. 04/99 Ming to get case to SK for Sherwood
 
 

Need to get toroidal case for linear validation. Should be complete by Feb.
 
 

4. Nonlinear vac. implemented Scott K. 04/00
 
 

Need to 1) track the passive scalar shape function with 3-d advection; and 2) modify the code to have 3-d resistive diffusion. Will do this semi-implicitly.
 
 

5. Push particles in NIMROD fields Scott P. 04/00 drift orbits
 
 

3-D particle push by Sherwood, not parallel (Charlson)
 
 

6. AZTEC into current ver. Steve P. 04/00
 
 

Done in version 2_3_2. Get AZTEC fully integrated into the "quadratic" version of the code. Steve may lose his funding. It is important that Steve get more fundng.
 
 

7. OO/C++ structure Ahmet 11/99 ???
 
 

Ahmet hasn't shown up at the last 2 meetings, nor has he participated in team e-mail discussions. He hasn't delivered on his promise to report on OO/C++ methods in MHD simulation. It's not clear what his role is, or where he's spending the NIMROD money that he got this year.
 
 

8. "Fully functional" high ord. eles. Carl S. 04/00
 
 

On track and on schedule. Still needs to be implemented on triangles.
 
 

9. Neo-class. paper submitted Tom G. 12/99
 
 

A NTM has not yet been found. Destabilizing perturbed bootstrap currents have been sen, but they are not large enough to cause island growth. There is also a numerical instabiliy related in part to electron advection that is not well understood. Work is proceeding.
 
 

10. 2-fluid multi-hel. RFP Rick N. 04/00 Numerical problems
 
 

Still need to get past numerical instability. Comes from 2-fluid R=0. Also may be related to the electron advection.
 
 

11. Compete SSPX case Alfonso 04/00 Problems w/triangles
 
 

Problems with the triangles appear to have been solved (ie., there was no problem). Te problem is that the SSPX equilibrium provided by LLNL does not satisfy the regularity conditions at R = 0. We cannot proceed further until LLNL solves this problem.
 
 

12. Flow and high-S in ET J.-N. 04/00 High-S works well, flows don't
 
 

Flow problems still not solved.
 
 

13. Demo. of nonlinear high-S Dalton 04/00
 
 

S=10**9 case needs to be understood better (better radial and poloidal resolution). Toroidal case at S=10**5 is underway. Hopefully this will be more interesting in the nonlinear regime.
 
 

14. Formulate diamag. corrections Chris H./Jim C. 04/00
 
 

Chis should write a brief paper/report on how we might incorporate FLR effects within the 2-fluid model.
 
 

15. General closure implementation Eric. H. 06/00
 
 

Eric will implement a general implementation of closures (heat flux, etc) into the NIMROD physics kernel.
 
 
 
 
 
 
 
 

2:00 pm - 4:00 pm
 
 

Continue development of plans/goals
 
 

4:00pm - 5:00pm
 
 

V. Discussion of team issues
 
 

FRIDAY, JANUARY 28

9:30am - 11:30 am
 
 

VI. Continue discussion of team issues
 
 

11:30am - 12:00 noon
 
 

VII. Date and location of next team meeting(s)
 
 

VIII. Adjourn