NIMROD TEAM MEETING
MARCH 25-16, 2000
UCLA KNUDSON HALL
ROOM 1-130N
MINUTES
SATURDAY, MARCH 25
1 PM - 4 PM CODE DEVELOPMENT++
- Higher order elements (Carl)
 
 
 
 

Bicubic elements have been implmented on rblocks. Greatly improved

convergence properties were reported. Still need to have the linear

solves optimized, and also ned to be implemented on tblocks.=20

(Matrix-vector multiplies have been optimized, but new line Jacobi

preconditioning has not.) CFL definition has not yet been updated.=20

Degenerate block works, but has not been implemented for R=3D0 yet. Beta

version may be available by April.
 
 
 
 

- Parallel computations (Steve P.?)
 
 
 
 

AZTEC link works in 2.3.2. Waiting for the next version (version 3) to

continue development.
 
 
 
 

- Vacuum (Scott K.)
 
 
 
 

Nonlinear vacuum implementation is being debugged. The concentration

variable is being succesfully advected. SI for 3-D resistivity has not

yet been coded.=20
 
 
 
 

- Resistive wall (Tom)
 
 
 
 

Nothing to say. Nothing added to code. Morrel's code may be a problem

if it doesn't have the n=3D0 mode.
 
 
 
 

- Neo-classical closures (Tom)
 
 
 
 

Many closures are in the code, and none sem to simultaneously give

poloidal flow damping and perturbed bootstrap current. We shouldn't be

discouraged by this. We're learning things.
 
 
 
 

- Kinetic closures/particles (Charlson)
 
 
 
 

Particles (ions) have been pushed in the fields on a real NIMROD grid.=20

No interactions have been included.
 
 
 
 

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

Rblocks have been extended into the vacuum. Fluxgrid is undergoing

major construction. One problem is that efit reconstructions don't

have good force balance at the end.
 
 
 
 

- Computations on tblocks (Alfonso)
 
 
 
 

Tblocks work and the application (SSPX) willshow (tomorrow) that they

do.
 
 
 
 

- Nimplot/tecplot (Carl, Scott)
 
 
 
 

Interface between xdraw binary and tecplot file has been written. This

has been incorporated in nimplot. This is in:
 
 
 
 

~u2230/code/newnimplot.tar.gz
 
 
 
 

on mcurie.
 
 
 
 

Scott K. will be adding hdf file output.
 
 
 
 

Scott P. and Charlson would like to make a DX animation of a coherent

nonlinear simulation. Looking for input from the team.
 
 
 
 

- ??
 
 
 
 

4 PM - 5 PLANNING
 
 
 
 

Develop goals and plans for next 6 months.
 
 
 
 

Here's the latest
 
 
 
 
 
 

GOAL RESP. PERSON DATE COMMENTS
 
 
 
 

Res. wall BC. Implemented Tom. G 06/00 (slab only) Still on

track/n=3D0 problem?
 
 

Lin. high-beta disr. Pyhsics Alfonso 04/00 linear is working;

nonlinear is running now/Ming to revisit GATO problem
 
 

Linear vac. validated w/GA Scott K. 04/00 Ming to get case to SK for

Sherwood/still on track
 
 

Nonlinear vac. Implemented Scott K. 04/00 Very close; SI resistivity

needs to be implemented
 
 

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

calculated/now needs to be debugged
 
 

AZTEC into current ver Steve P 08/00 awaiting version 3
 
 

OO/C++ structure Ahmet 11/99 conceptually implemented (in Ahmet's

head)/progress related to POOMA's advance/POOMA uses finite

differences; can't be reconciled with NIMROD's FE formulation/this

appears to be DEAD!
 
 

"Fully functional" high ord. eles. Carl S. 04/00 (beta version)

Version 3 by APS (11/00)??
 
 

2-fluid multi-hel. RFP Rick N. 06/00 Numerical problems/still

having problems/needs higher order elements?? (can use beta version)
 
 

Compete SSPX case Alfonso 11/00 Problems w/triangles; eq. is

finally good, now can start the case
 
 

Flow and high-S in ET J.-N. 09/00 Need beta version with high-order

elements; ready for comparison with FAR
 
 

Demo. of nonlinear high-S Dalton 11/00 (linear up to S=3D10**9;

S=3D10**5 nonlinear done; still need to push to higher S)
 
 

Formulate diamag. Corrections Chris H./Jim C. 04/00/giving talk

tomorrow
 
 
 
 
 
 

5PM =F1 6PM OTHER TEAM ISSUES
 
 
 
 

- The view from Germantown (Rostom)
 
 
 
 

Theory budget looks good for next year.
 
 
 
 

1/3 of theory budget will be competed in July.
 
 
 
 

M3D is becoming a "team"
 
 
 
 

- Popova SOW
 
 
 
 

DS will put list together.
 
 
 
 

- The Hazeltine Report
 
 
 
 

DS will tke care of it!
 
 
 
 

- Other ??
 
 
 
 

- Date and place of next team meeting
 
 
 
 

DS will look into having the meeting in Las Vegas in mid-August
 
 
 
 

6 PM ADJOURN
 
 
 
 

Team bonding??
 
 
 
 
 
 

SUNDAY, MARCH 26
 
 
 
 

9:00 AM - 1 PM: RESULTS AND DISCUSSIONS
 
 
 
 

- Neoclassical Tearing Mode (Tom)
 
 
 
 

The trials and tribulations of neoclassical theory and its

implementation in NIMROD were discussed in detail.
 
 
 
 

The "flow damping form" may be promising. we don't know if it's giving

the right answer.
 
 
 
 

Suggestions:
 
 
 
 

1. Sources
 
 
 
 

2. Check with delta-P=3D0
 
 
 
 

3. Convergence
 
 
 
 

4. Try different parameters parameters
 
 
 
 

5. NEAR benchmarks
 
 
 
 

- Spheromak modeling (Carl, Alfonso)
 
 
 
 

Regularity problem for the SSPX equilibrium appear to have been fixed.=20

Linear studies to begin. (Alfonso)
 
 
 
 

Bruce Cohen (LLNL) is giving a NIMROD related paper at Sherwood. This

the first time a non-team member has given a paper using NIMROD, and we

consider this to be a significant development.
 
 
 
 

- FRC (Alfonso)
 
 
 
 

Work in progress; no report.
 
 
 
 

- D3D High-beta disruption (Alfonso)
 
 
 
 

"Generic" shot 87009 was reported. Linear work has progressed.=20

Nonlinear calculations with self-similar heating profiles are

underway.
 
 
 
 

- High-S (Dalton)
 
 
 
 

Lower q(0) back to 0.98 and look at the n=3D1 and n=3D1.
 
 
 
 

New results were reported.
 
 
 
 

- Two-fluid RFP (Rick)
 
 
 
 

Extensive stability analysis indicates that the 2-fluid calculations

should be stable. We still don't understand this problem. Needs

higher order elements for solution?
 
 
 
 

- Diamagnetic corrections (Chris)
 
 
 
 

Chris told us how to put diamagentic (FLR??) effects into the resistive

MHD model.
 
 
 
 

If you put in diamagnetic effects, you also need to put in some sort of

damping (ion Landau damping?) to get rid of the radially propagating

mode.
 
 
 
 

1 PM ADJOURN
 
 
 
 

Amen!