input.tgyro Alphabetical List

LOC_BETAE_SCALE

Definition

Amount to scale electron beta

Comments

  • To shut off electromagnetic fluctuations, set LOC_BETAE_SCALE=0.0

Default

1.0


TGYRO_DEN_METHOD*

Definition

Rules for evolving species.

Choices

  • TGYRO_DEN_METHOD*=-1: Use species to enforce quasineutrality (can use this for more than one species simultaneously)

  • TGYRO_DEN_METHOD*=0: Do not evolve species

  • TGYRO_DEN_METHOD*=1: Evolve species by solving transport equation

  • TGYRO_DEN_METHOD*=2: Species evolved with alpha particles from DT reaction as source

Comments

  • Electrons are TGYRO_DEN_METHOD0, the first ion is TGYRO_DEN_METHOD1, etc.

  • To evolve electrons, with first ion used to set quasineutrality, and second ion density fixed, set

    TGYRO_DEN_METHOD0=1
    TGYRO_DEN_METHOD1=-1
    TGYRO_DEN_METHOD2=0
    
  • To evolve the second ion density, and adjust the main ion to enforce quasineutrality, set

    TGYRO_DEN_METHOD0=0
    TGYRO_DEN_METHOD1=-1
    TGYRO_DEN_METHOD2=1
    
  • For a DT main ion plasma (1), with He4 ash (2), an impurity species (3) and a fast ion NBI species, set:

    TGYRO_DEN_METHOD0=1  #electron
    TGYRO_DEN_METHOD1=-1 #main ion DT
    TGYRO_DEN_METHOD2=2  #ash He4
    TGYRO_DEN_METHOD3=0  #impurity
    TGYRO_DEN_METHOD4=0  #fast ion
    

Default

0


TGYRO_DT_METHOD

Definition

Method for handling DT mixture (as separate or hybrid species).

Choices

  • TGYRO_DT_METHOD=1: two separate D and T species (requires two separate profiles)

  • TGYRO_DT_METHOD=2: a single hybrid D-T species with density \(n_{DT}\) and mass 2.5 AMU

Comments

  • When TGYRO_DT_METHOD=1, the fusion cross section is computed as

    \[S = n_{D} n_{T} \left\langle \sigma v\right\rangle\]
  • When TGYRO_DT_METHOD=2, the fusion cross section is computed as

    \[S = \frac{n_{DT}^2}{4} \left\langle \sigma v\right\rangle\]

Default

2


LOC_DX

Definition

Step length, \(\Delta z\), used to compute the finite-difference approximation to the Jacobian.

Comments

  • This can be safely increased for robustness. Root is unchanged.

  • This is dimensionless, with \(\Delta z = \Delta (a/L_x)\), where \(L_x\) is any of the gradient scale lengths.

Default

0.1


LOC_DX_MAX

Definition

Maximum length, \(\Delta z_\mathrm{max}\), of any Newton step.

Comments

  • This should probably always be smaller than unity.

  • This is dimensionless, with \(\Delta z = \Delta (a/L_x)\), where \(L_x\) is any of the gradient scale lengths.

Default

1.0


LOC_ER_FEEDBACK_FLAG

Definition

Flag to signal electric field evolution

Comments

  • LOC_ER_FEEDBACK_FLAG = 0 : Do not evolve electric field

  • LOC_ER_FEEDBACK_FLAG = 1 : Evolve electric field

Default

0


LOC_EVOLVE_GRAD_ONLY_FLAG

Definition

Flag to allow decoupling of profile and corresponding gradient scale length evolution

Choices

  • LOC_EVOLVE_GRAD_ONLY_FLAG=0: Self-consistently evolve profiles and gradients.

  • LOC_EVOLVE_GRAD_ONLY_FLAG=1: Evolve gradient scale lengths while holding density, temperature, and rotation profiles fixed.

Comments

  • By setting this flag to 1, one effectively transforms TGYRO from a transport code into a parallel job management routine which solves for the local flux-matching gradients at each specified radius, while holding local density, temperatures and rotation values fixed. This process allows one to more efficiently calculate local flux-matching solutions and stiffness assessments, while eliminating changes to the gyroBohm normalizations and plasma parameters such as \(T_e/T_i\) and collisionality arising from model performance at other radii.

Default

0


LOC_HE_FEEDBACK_FLAG

Definition

Flag to signal helium ash evolution

Comments

  • LOC_HE_FEEDBACK_FLAG = 0 : Do not evolve ash`

  • LOC_HE_FEEDBACK_FLAG = 1 : Evolve ash

Default

0


TGYRO_ITERATION_METHOD

Definition

Variation of Newton method used for root-finding.

Choices

  • TGYRO_ITERATION_METHOD=1 : Local residual (standard) method

  • TGYRO_ITERATION_METHOD=4 : Global residual method (serial version)

  • TGYRO_ITERATION_METHOD=5 : Global residual method (parallel version)

  • TGYRO_ITERATION_METHOD=6 : Simple relaxation method (in development)

Comments

  • If TGYRO_ITERATION_METHOD=5, use (n_evolve+1)*n_mpi rather than simply n_mpi in input.tgyro

  • For example, if evolving \(T_e\) and \(T_i\), then n_evolve+1=3, so set 3*n_mpi in input.tgyro

  • TGYRO_ITERATION_METHOD=6 uses a simple relaxation scheme

\[\frac{dz}{z} = -\mathrm{LOC\_RELAX} \, \frac{Q_\mathrm{tot} - Q_\mathrm{target}}{\max(|Q_\mathrm{tot}|,|Q_\mathrm{target}|,1)}\]
  • In TGYRO_ITERATION_METHOD=6, if \(dz/z >\) LOC_DX_MAX, then \(dz/z =\) LOC_DX_MAX

  • In TGYRO_ITERATION_METHOD=6, LOC_RELAX=0.1 and LOC_DX_MAX=0.05 are good starting points

Default

1


LOC_LOCK_PROFILE_FLAG

Definition

Flag to manage calculation of initial profile

Choices

  • LOC_LOCK_PROFILE_FLAG=0: Use profiles computed from gradients

  • LOC_LOCK_PROFILE_FLAG=1: Use exact profiles

Comments

  • If set to 1, use temperature and density profiles as read from input.profiles, not as calculated from gradients.

  • This has an effect only on the zeroth iteration.

  • This can have a large effect when a coarse radial grid is used. In this case, the profile as calculated from the gradients at only a few points will not match the original (higher-resolution) profile.

Default

0


LOC_SCENARIO

Definition

Power-balance scenario.

Choices

  • LOC_SCENARIO=1 : Fixed integrated powers with static exchange.

  • LOC_SCENARIO=2 : Fixed integrated powers with dynamic exchange.

  • LOC_SCENARIO=3 : Thermonuclear source, radiation and exchange with auxiliary heating from data.

Comments

  • In LOC_SCENARIO=3, pow_e and pow_i in input.profiles are assumed to be auxiliary heating profiles. Otherwise, pow_e and pow_i are the usual total powers to ions and electrons.

Default

1


LOC_MA*

Definition

Ion mass in units of hydrogen mass. First ion is LOC_MA1, etc.

Default

2.0


LOC_Z*

Definition

Ion charge. First ion is LOC_Z1, etc.

Default

1.0


TGYRO_MODE

Definition

High-level control of TGYRO operational mode.

Choices

  • TGYRO_MODE=1 : Operate as transport code.

  • TGYRO_MODE=3 : Multi-job control of GYRO or CGYRO.

Comments

Sample input.tgyro for TGYRO_MODE=1 to run as transport code

DIR TGLF1 1
DIR TGLF2 1
DIR TGLF3 1
DIR TGLF4 1

TGYRO_MODE=1

#-----------------------------------------------------
# NEO control
LOC_NEO_METHOD=1
#-----------------------------------------------------

#-----------------------------------------------------
# Profile control
TGYRO_DEN_METHOD0=0
LOC_TE_FEEDBACK_FLAG=1
LOC_TI_FEEDBACK_FLAG=1
#-----------------------------------------------------

#-----------------------------------------------------
# Physics control
# (1=static exch, 2=dynamic exch, 3=reactor)
LOC_SCENARIO=2
TGYRO_RMAX=0.75
#-----------------------------------------------------

#-----------------------------------------------------
# Iteration control
TGYRO_ITERATION_METHOD=1
LOC_RESIDUAL_METHOD=3

LOC_DX=0.02

# TGYRO iterations
TGYRO_RELAX_ITERATIONS=4

# Restart (0=new,1=restart)
LOC_RESTART_FLAG=0
#-----------------------------------------------------

Sample input.tgyro for TGYRO_MODE=3 to run 6 instances of CGYRO

DIR c1 384 GAMMA_E=0.0
DIR c2 384 GAMMA_E=0.02
DIR c3 384 GAMMA_E=0.04
DIR c4 384 GAMMA_E=0.06
DIR c5 384 GAMMA_E=0.08
DIR c6 384 GAMMA_E=0.10

TGYRO_MODE=3

Default

1


TGYRO_RELAX_ITERATIONS

Definition

Maximum number of Newton iterations.

Comments

  • Setting TGYRO_RELAX_ITERATIONS = 0 is useful to generate initial fluxes and other profiles on the TGYRO grid.

Default

0


LOC_RELAX

Definition

Parameter \(C_\eta\) controlling shrinkage of relaxation parameter.

DEFAULT = 2.0

Comments

  • No effect if less than or equal to unity.

  • Only controls the behaviour of TGYRO_ITERATION_METHOD = 1.

  • Experience shows that LOC_RELAX=2.0 is a good choice (default).

LOC_RESIDUAL_METHOD

Definition

Formula for residual (error) in TGYRO root finding.

Choices

  • LOC_RESIDUAL_METHOD=2: \(\displaystyle \sum_p \left|f_p-f^\mathrm{target}_p\right|\)

  • LOC_RESIDUAL_METHOD=3: \(\displaystyle \sum_p (f_p-f^\mathrm{target}_p)^2\)

Comments

  • This parameter only affects LOC_ITERATION_METHOD=4,5.

  • It does not affect TGYRO_ITERATION_METHOD = 1 because each residual is minimized independently.

  • The total residual is renormalized in method 3 to make the value comparable to method 2.

Default

2


LOC_RESTART_FLAG

Definition

TGYRO startup control (new or restart).

Choices

  • LOC_RESTART_FLAG=0: New simulation.

  • LOC_RESTART_FLAG=1: Continue running based on last iteration.

Default

0


TGYRO_THERM_FLAG*

Definition

Thermal/nonthermal species toggle. First ion is TGYRO_THERM_FLAG1, etc.

Choices

  • TGYRO_THERM_FLAG* = 1: Thermal

  • TGYRO_THERM_FLAG* = 0: Nonthermal (energetic)

Comments

  • Only the shared thermal ion temperature is evolved in TGYRO

  • Unequal thermal ion temperatures is not supported

Default

1


LOC_TE_FEEDBACK_FLAG

Definition

Flag to signal electron temperature evolution

Comments

  • LOC_TE_FEEDBACK_FLAG = 0 : Do not evolve \(T_e\)

  • LOC_TE_FEEDBACK_FLAG = 1 : Evolve \(T_e\)

Default

1


LOC_TI_FEEDBACK_FLAG

Definition

Flag to signal ion temperature evolution

Comments

  • LOC_TI_FEEDBACK_FLAG = 0 : Do not evolve \(T_i\)

  • LOC_TI_FEEDBACK_FLAG = 1 : Evolve \(T_i\)

Default

1


TGYRO_WRITE_PROFILES_FLAG

Definition

Manage how new input.profiles files are written.

Choices

  • TGYRO_WRITE_PROFILES_FLAG=-1: Write new input.profiles.* at every iteration.

  • TGYRO_WRITE_PROFILES_FLAG=0: No new files are written.

  • TGYRO_WRITE_PROFILES_FLAG=1: Write input.profiles.new at final iteration

Comments

  • If TGYRO_WRITE_PROFILES_FLAG=-1, then input.profiles.n is written at the nth iteration.

Default

0