The field amplitudes returned to Transp from Toric have been renormalised ( link ?? ) . Preliminary results appear to indicate that the fields are still within the range anticipated in the QLO renormalisation, but the normalisation factor is further away from unity than was the case previously ( see previous report - Sep 2008 )
Shot 69409 was rerun with Transp version JET_48, which is linked to Toric built from SVN release 729. The four JET A2 antenna were combined, and run with NPHI = 25, 48, NMDTORIC=63. Results are compared with a similar run ( J18 ; JET_47, Toric 678 ) [ Note that a previous comparison with run J16 was difficult to interpret, since the A2 antenna position has been corrected between runs J16 & J17.
Figure 1 E fields, Toric 678 ( blue ) & 729 ( green ) , from the Toric NetCDF output files. The Vn 729 renormalised amplitudes are expected to be larger by a factor of 2 Pi
Figure 2 Field amplitudes ( on axis ) from the .fppmod files written for Transp, where the J17 / svn 789 values have been divided by 2pi, showing that the change in normalisation has been applied to the Transp values, as expected.
Comparison of the on axis field values written to the Toric NetCDF files & the Transp .fppmod files show changes corresponding to the overall normalisation change ( factor of 1/2pi removed from efld_nor in t4_mod_fields ). It is clear from the code that this factor is propagated to eleout_zenorm, which in turn normalises the values written to .fppmod. The .fppmod values are normalised to total antenna power, & so are increased by a factor of sqrt(Power/tpw) = sqrt(2.34/0.15) ~= 4 relative to the NetCDF values. [ Is this factor correct ?? see updated Normalisation calculation .. ]
The total power deposition to electrons & minority ions ( seen by Transp ) is unchanged, with the exception of the innermost point :-
Figure 3 Minority heating - radial distribution at 5 sec
Figure 4 - QLO renormalisation vs Time
Power calculated by the wave code ( PWAVMIN_3 ) has not changed significantly, but the power calculated by Transp before renormalisation ( PQSLMIN_3 ) is much larger. Consequently a larger value of the renormalisation factor ( MQKPRHO_3 ) is needed to bring the power deposited in Transp ( PQLNORM_3 ) back to the original Toric value
The renormalisation factor in ( Transp library ) fppmod is restricted to the range 0.001 and 100., so the new average value of approximately 20 is a little too close to the upper limit - figure 4 shows a maximum of ~60. in only 1 second of plasma.
The renormalisation factor for Spruce was 1. on average, so there is once more a significant difference between the Spruce and new Toric field amplitudes.Also, the Toric calculation of power deposition ( PQSLMIN ) is significantly higher than the Transp QLO calculation, which suggests a disagreement between these two codes - or, that the normalisation to antenna power ( in fppmod ) is wrong ( still ) ??