Google  TTM    

  Home
  Mission
  Staff
  History
  Location
   
  Open Positions
   
  Education
  Lectures
  Theses and Jobs
  Downloads
   
  Research
  Departments
  Projects
  Downloads
   
  Experiments/Simulation
  Test rigs
  Measurement
  techniques
  Numerical codes
  Facilities and
  workshop
   
  Publications
   
  Ebooks

 


LINARS --- controll.txt
 
LINARS creates this file if not present with standard parameters given in the table below.

LINARS ignores lines without a "&" on the left-hand side!
Lines with a "&" LINARS compares the keywords with a internal list and sets the number after the "=" to the variable, if the notation is correct. The order of the keywords given in the description below is not necessary, as LINARS compares each keyword with its own internal list,


"controll.txt" file parameter description
########################
###                                        ###
##   Controll File for flow-solver   ##
###                                        ###
########################

 
& R = 287.0 gas constant
& kap = 1.4 adiapatic index

& TimeIterations = 1000 number of timeiterations
& reload = 1 0...new calculation with values of initialization, see below
     section "Initialization"
1...loads primitive NS values from stored file "primVar.bin"
     if file not fount --> same operation as reload = 0

& moduloReadControl = 10 LINARS reloads each Xth timestep "controll.txt" file
0... LINARS doesn't read the controll.txt file during simulation.
& moduloWriteConsVar = 10 LINARS saves each Xth timestep "primVar.bin" file
0... LINARS doesn't save temporary restart files.

 constTimeStep = 1.0e-6 in the standard "controll.txt" given without a "&" on the left-hand side. This means LINARS uses the steady flow solver. If you want to use the unsteady flow solver set the sign "&" at the left-hand side and the dimensional time for each timestep behind the "=" (e.g.: 1e-6s)! Therefore the CFL number below is ignored.
& cflOnFinest = 10 CFL number on finest grid.
& cflOnCoarser = 10 CFL number on coarser grids, if the multigrid solver is used.
& TVDonFinest = 1 specifies the spacial discretization scheme for the inviscid (EULER) fluxes
0... 1st order accurate interpolation
1... Total Variation Diminishing scheme, 3rd order accurate
2... van Albada lmiter, 2nd order accurate
& TVDonCoarser = 1 If the multigrid solver is usedspacial --> discretization scheme on coarser grids.

########################
##              Initialization             ##
########################

if variable "reload" = 0, the flowfield is initialized with the specified values of the primitive variables.
& Ptot = 1.013e5 total pressure
& Ttot = 293.150000 total temperature
 InVelU = 0.000000 velocity in x-coordinate
 InVelV = 0.000000 velocity in y-coordinate
 InVelW = 0.000000 velocity in z-coordinate

& IsQuasi3D = 1 0... if a fully 3d simulation
1... if imax, jmax or kmax of each grid block is 1
(see "geom.txt")
!!!You can also use a mixed grid topology of blocks with imax>1, jmax>1 and kmax>1 and blocks with imax=1, jmax=1 or kmax=1 (quasi 3d) ---> then use IsQuasi3D = 0

########################
##           Solver Settings            ##
########################
 
& nMultiGrid = 1 number of multigrid levels, if 1 only finest level and hence no multigrid is used!
& nPreIterationsCoarse = 0 number of pre iterations on coarset level (NS and turbulence model), variable is only used if multigrid>1!
& RelaxScheme = 2 0... four step Runge-Kutta integration scheme
1... one point implizit Newton-Raphson intergration scheme
      (not recommended!)
2... fully implizit Newton-Raphson scheme with Alternating
      Direction Implizit solver.
      Recommended for steady and unsteady solver!
& UnderRelaxParam = 0.5 variable is used only for implizit solvers, recommended value "0.5" for steady solver. Use ">0.9" for unsteady solver.
& NewtonIterations = 1 number of Newton-Raphson iterations. Use "1" for steady solver and ">1" for unsteady solver.
  GaussSeidelIterations = 1 only for one point implizit Newton-Raphson intergration scheme, don't change this value!
  ModuloGaussSeidel = 1 only for one point implizit Newton-Raphson intergration scheme, don't change this value!

########################
##    EULER, VISCOUS,.........      ##
########################
 
& UseViscous = 1 0...EULER
1...VISCOUS, laminar calculation
2...TURBULENT -> Spalart Allmars
3...TURBULENT -> Wilcox k-omega
4...TURBULENT -> Menter SST k-omega
5...TURBULENT -> Durbin's k-epsilon-v'2-f
6...TURBULENT -> Turbulence Potential Model
     use without multigrid solver!
7...TURBULENT -> Laminar Kinetic Energy Model

& DynViscosity = 6.0e-05 With "&" on the left-hand side --> constant viscosity in the whole flow domain with specified value.
Without "&" --> dynamic viscosity is calculated by the Sutherland law with specified values in "grid.cpp" and "definitions.h" source files.
& TurbPreIterations = 0 number of pre-iterations with the specified turbulence model, while the Navier-Stokes equations are left unaltered (frozen).
& TurbReload = 1 0... starts turbulence model with standard initialization values
1... loads turbulence values from stored file "primVarXXX.bin"
    if file is present, if not -> same operation as TurbReload = 0

   
########################
##         Transition Modells          ##
########################
 
& UseTransition = 0 0... simulation without a transition model
1... Menter's γ-ReΘ transition model, implementation
      is not yet finished, use only with UseViscous = 4
& TransPreIterations = 0 number of pre-iterations with the specified transition model, while the Navier-Stokes and the turbulence model equations are left unaltered (frozen).
& TransReload = 1 0... starts transition model with standard initialization values
1... loads transition values from stored file "primVarXXX.bin"
   if file is present, if not -> same operation as TransReload = 0

########################
##       Unsteady / Phase Lag       ##
########################
 
& PLIFAType = 0 0...without
1...pitchratio 1:1, add line "& timeStepsEachPeriod = X"      X=pitch/gridVelocity/constTimeStep!!!
     X is an integer value, therefore set first an appropriate
     value of X and recalculate the constTimeStep value! 2...phaselaged boundary conditions

   
########################
##             Grafik Output            ##
########################
 
& moduloOutputGr2 = 10 0... no output file
X... each Xth timestep LINARS writes the output file,
      see 2dOutput.gr2 for a 2d simulation and
      field.gr2 for 3d.
& moduloOutputGrf = 0 0... no output file
X... each Xth timestep LINARS writes the output file,
      see XXX.grf
& moduloOutputTec = 0 0... no output file
X... each Xth timestep LINARS writes the output file,
      see tecplot.plt
& moduloOutputBoundStatistics = 0 0... no output file
X... each Xth timestep LINARS writes the output files for
      inlet outlet and wall, see stxSeq...
& moduloOutputResid = 2 DON'T use 0!!! except you reprogram LINARS
X... each Xth timestep LINARS writes the residuals into the
      res_log.txt file.
   
   
   
   
   
   


  TTM - TU Graz // Inffeldgasse 25A // A-8010 Graz // Phone: +43 (316) 873 7226 // Fax: +43 (316) 873 7239
eMail: TTM@TUGraz.at // Contact informations & Impressum // TUGonline // Graz University of Technology © 2005