NSHOT: The shot number to be simulated.
TINIT : The start time of the simulation. Note that this is in 'TRANSP time' which is equal to 'JET time' - 40s, i.e. time 0 in
TRANSP is 40s into the JET shot.
FTIME : The end time of the simulation. This is also in 'TRANSP time'.
TRDAT interpolates UFILE data onto two fixed time grids covering the period between TINIT and FTIME, these grids are specified using the following
two namelist variables:
TGRID1 : The spacing between time points for scalar input data.
TGRID2 : The spacing between time points for profile input data.
These entries control the time resolution of the input data as TRANSP sees it. The main code will then interpolate to intermediate time points as required.
TRANSP has a heirarchy of different timesteps for the different physical processes: heating & current drive, evolution of the equilibrium, solution of the transport equations etc.
Some of these timesteps are fixed (Heating & Current Drive) whereas others are adjustable (equilibrium and transport equation timesteps) and will adapt to avoid too large a variation
in a particular signal. The following parameters control the variation and initialisation of the different time steps used:
DT_SOURCES : The fixed time step used for all heating sources (ECH, LH, ICRF, Neutral Beams).
DTMING, DTMAXG : Minimum and Maximum allowed time steps for the advance of the MHD equilibrium. Note that all other timesteps are
constrained by DTMAXG and so this must be at least MAX( DTMINT, DTMINB ). DTMAXG is used as the initial value of the equilibrium timestep but the eventual value used
will be determined by macroscopic properties of the equilibrium or input boundary.
DTMINT, DTMAXT, DTINIT : Minimum, Maximum and Initial time step to use for the solution of the heat and particle balance equations. The actual
step size used is controlled by changes in the temperature or density.
DTMINB, DTMAXB, DTINIB : Minimum, Maximum and Initial time step to use for the solution of the Poloidal Field Diffusion Euqation.
NZONES : The number of radial zones used when solving the 1D particle and heat balance equations. This is usually set to 50 but can be changed
depending on memory and time constraints.
NZONES_NB : The number of radial zones used in the Neutral Beam Monte Carlo module (NUBEAM). 20 zones is recommended or 60 zones with smoothing.
NZONES_FP : The number of radial zones used in the fast ion Fokker-Planck model (20 zones recommended with default 'dxfsmoo'
smoothing).
NZONES_FB : The number of radial zones used when evaluating fast ion effects such as beam-beam interactions. Effectively the number of
radial zones used in the fast ion distribution function for these cases. Note that NZONES/NZONE_FB, NZONES_NB/NZONE_FB and NZONES_FP/NZONE_FB
must all be integers.
The following two variables control when TRANSP outputs data:
SEDIT : This sets the time period between successive output of profile data.
STEDIT : This sets the time period between successive output of scalar data.
Obtaining the fast particle distribution involves utilising TRANSPs capability to output ASCII Compressed Files (ACFILES). This allows outputing of all or part
of TRANSP COMMON to a file. The settings for this are as follows:
SELOUT : A list of variable names to be written to the output ASCII file. By default this is all of them (preferred).
SELAVG : A subset of the above variables to be averaged prior to being output. By default none are averaged. For Fast particle
output select 'FBM BMVOL BDENS2 EBA2PL EBA2PP'.
OUTTIM : An ascending list of times when the ACFILES should be written. A maximum of 99 can be written. The output files are
named 'runID'.DATA1 for OUTTIM(1), 'runID'.DATA2 for OUTTIM(2) etc.
AVGTIM : This is the averaging time in seconds used for the variables specified in SELAVG. The variables are averaged
over the period AVGTIM prior to a particular OUTTIM. Note that the times specified for OUTTIM must be separated by at least
AVGTIM.
MTHDAVG : The averaging method to be used. 1 is the default option,
set to 2 to sample variables after each heating source timestep and set to 3 to use a sample period specified by AVGSAMP.
AVGSAMP : If MTHDAVG was set to 3, this variable specifies the sampling period in seconds.
The following two flags control the field and current orientation within the tokamak during the simulation. Because the rotation velocity is defined such that positive is
co-current rotation, these switches should normally only have a weak effect, however they can have an important effect for low field or smaller devices. See the TRANSP
website entry 'Field_Orientation_Controls' for more information.
NLBCCW : TRUE indicates that the B field is counter-clockwise when viewed from above. For JET, set this to FALSE.
NLJCCW : TRUE indicates that the current is counter-clockwise when viewed from above. For JET, set this to FALSE.
Several components of TRANSP have been parallelised and consequently when a parallel job is submitted the user can specify whether each of these components is run in serial or in
parallel. At JET it is usual to only run NUBEAM and TORIC in parallel. Note that if MPI resources are allocated to a run, at least one of the following must be set to 1.
If no MPI resources are allocated, both must be set to 0.
NBI_PSERVE : Set equal to 1 to run the NUBEAM neautral beam and fusion fast ion model in parallel. Set to 0 to run in serial.
NTORIC_PSERVE : Set equal to 1 to run the TORIC full wave ICRF heating module in parallel. Set to 0 to run in serial.
MDS_SERVER : When running TRANSP over the fusionGRID at PPPL the inputs are not written to Ufiles, instead the PPF system is queried
directly from PPPL using MDS+. In order to do this it is necessary to connect to the JET MDS+ server which is 'mdsplus.jet.efda.org'. Note when including this
in the namelist file the single quotes must be included.
MDS_TREE : This argument specifies the name of the MDS+ tree which allows the input data to be read from the PPF system, this should
always be set to 'TRDATA'.
MDS_PATH : This specifies the path in the MDS+ tree where the input data calls are located. This should always be set to '\top.INPUTS'.