1. 14/09/2010
IMP5 CPOs T. Johnson

2. Outline Overview of IMP5 CPOs Bugs & examples
What’s new 4.08b? (ready to be release after meeting - no official date)
New feature: structures containing arrays of structures
WAVES is test case
Arbitrary variables and unstructured grid available for 2D quantities in WAVES 2

3. IMP5 cpos There are still no cpos for fast particle/instability output
ALPHAS
Fusion reaction model
COREPROF (NOT IMP5)
There are still no cpos for fast particle/instability output
NBI
Beam deposition model
Collision couple the distributions
INPUT
Machine descriptions
Experimental data
Other input +
Self-consistency
NBI: neutral source
inside injector
DISTSOURCE: source for kinetic ion/electron models
Kinetic model (e.g. Fokker-Planck)
Datajoiners: e.g. new cpo merges IC & EC waves
DISTRIBUTION: distribution functions
RF HEATING
Wave model
ANTENNAS: boundary condition
at antenna
WAVES: Wave field
OUTPUT
CORESOURCE: input for ETS (NOT IMP5)

4. IMP5 cpos (my interpretation)
We have KEPLER actors for most models;
although not all of them fill completely the output cpos
ALPHAS
Not available: Maybe ASCOT?
COREPROF (NOT IMP5)
NBI
NEMO / ASCOT
INPUT
Machine descriptions;
JET NBI coming soon
Experimental data
Other input (usually ETS) +
NBI: neutral source
inside injector
DISTSOURCE: source for kinetic ion/electron models
FPSIM
Not available: ACT6
DISTRIBUTION: distribution functions
RF HEATING
TORIQ / EVE / GRAY
ANTENNAS: boundary condition
at antenna
WAVES: Wave field
OUTPUT
CORESOURCE: input for ETS (NOT IMP5)

5. Bugs & examples CPOs for testing
When testing that your code can read CPOs; then you need examples
You may not find what you need (some fields missing / no CPO exists)
You may even be the first person to use these cpos!
If not, then you should either
write test-routine yourself that fill your input cpo
please share your routine/template cpo!
or contact IMP5 Leadership; we’ll ask someone to do it for you – not recommended!
Bugs or bad design
There will be mistakes in the CPOs!
Please let us know when you find them!
If there’s something you don’t like in the cpos; let us know!
We are IMP5 and we have to make this work together! 5

6. What’s new 4.08b?
Next revision is 4.08b; ready to be release after meeting (no official date)
2D spatial grids in WAVES:
4.08a grids (psi, theta)
4.08b grid in arbitrary coordinates (dim1, dim2)
4.08c will include new 2D grid for DISTSOURCE and DISTRIBUTION
New feature “arrays of structure”
IMP5 cpo WAVES was test pilot:
In 4.08a datais arrays of frequency index, e.g. …%Eplus(jfreq, ind1,…)
In 4.08b the WAVES are split into an array of coherent waves (e.g. one for each RF antenna or EC beam); each one being a structure with properties Eplus, Epar…; e.g. waves%coherentwave(jfreq)%...%Eplus(ind1,…)
Advantage e.g.: merge waves from ECRH, ICRH into one CPO;
4.08a: copy every field (Eplus, Eminus…) separately
4.08b: copy entire ”frequency”-structures
Modified indexing of modules in waves%antennaslh_setup
NBI: Specification of beam injected species simplified 6

7. Changes: waves (4.08a / 4.08b) 4.08a 4.08b grids changed Array
(name changed to coherentwave)
In wave field split in one structure per frequency
4.08a: …%Eplus(jfreq, ind1,…)
4.08b: frequencies(jfreq)%...%Eplus(ind1,…)
2D grid changed to general from (npsi,ntheta) to general coordinates (ndim1,ndim2)

8. Changes in waves/beamtrace
nbeams remove in 4.08b; replaced by array of frequencies
all arrays of frequencies removed
4.08a
4.08b 8

9. Changed: waves/fullwave/pol_decomp
nmpol obsolete in 4.08b; replaced by size(…) (only one frequency; only one mpol)
all array indexing of frequencies removed
4.08a
4.08b 9

10. Changes: wave/grid(1d/2d)
4.08b – 1D
(npsi)
2D grid changed from (npsi,ntheta) to (ndim1,ndim2);
Thus different psi/rho are needed for 1D and 2D grid.
grid_type added
4.08b - 1D & 2D
(npsi)
4.08b - 2D
(ndim1,ndim2)
Allows for unstructured grids
(ntheta)
frequency vectors absolete in 4.08b
(ndim1,ndim2) 10

11. Changes: waves/antennaLH_setup
Position index of the module in the poloidal and toroidal directions added
Dimension of Amplitude/Phase of the TE10 mode injected changed
4.08a: Array 3D (nantenna_lh,max_nma_phi,max_nma_theta)
4.08b: Matrix (nantenna_lh, max_nmodules) 11

12. Changes: NBI NBI species description simplified;
pointers to the plasma composition removed
4.08a
4.08b
Added: zion=ions charge 12