Run Information

Run Time: t=0.14-1.0s
ICRF Module: TORIC (New nl Controls)
ICRF Power: 3MW @80MHz (Simulated)
ICRF Duration: 0.14-1.0s
LH Module: LSC
LH Power: 2.7MW @n0=2.75,4.0
LH Duration: 0.15-1.0s
Density Profile: Experimental, fixed @t=0.2s*1.5
Chi Modification: Yes, NKEMOD=0,-1 and TKEMOD=0.15

Noteworthy Switches: NZONES=20, LEVGEO=6(VMEC6),
QLIM0=15.0, NBOOSM=2, NJSMBOOT=2, NJSMLH=4,
NRAYSLH=50,hLH=0.002,DTMAXG=0.002
* Duplicate of 2282 with different density profile*
* Duplicate of 2283 n=n*2.0 *

Comments:

This run is part of a density scan performed on 2283. The density is simply multiplied by 2.0 throughout the entire run. The run crashes because of a problem involving the election conduction. According to Doug:

This run is having problems due to its Te prediction.  Without going into too much detail,
the solution has run away in such a way as to give Te-->0 near the very edge, even though
there is a large LH-driven heating spike in Te at r/a ~ 0.75.

The edge Te-->0 first caused LSC to crash, because overshoot from the spline interpolation
caused Te < 0 at some locations.  We patched this, and it survived a while longer before
crashing in the poloidal field solver.  This last may be due to the very high resistivity
produced due to Te-->0.

Here are some of the features of the electron power balance:

  a) LSC heating very peaked at r/a ~ 0.75
  b) chi(e) is very small in the edge region: 100cm**2/sec -- i.e. almost no conduction.
  c) due to the high density, convective recycling is very peaked at the edge
  d) the Te boundary condition is from data.

These conspire to cause the Te to do strange things which I could discuss further with
you if you want.

However, I suggest trying two things:

  (a) it might help to increase the radial resolution of the run, and
  (b) it might help to raise chi(e) at the edge... set xkemin=1.0e3 (cm2/sec)

Total Current: 		White
Lower Hybrid: 		Red
Ohmic Current: 		Green
Bootstrap Current: 	Blue