Panel | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||
|
ModelInfo
Tell T3P which mesh file to load and what boundary conditions are used for the different side sets in the mesh file (default: Electric)
Code Block |
---|
Wiki Markup |
{toc:style=disc|indent=20px} h3. ModelInfo Tell T3P which mesh file to load and what boundary conditions are used for the different side sets in the mesh file (default: Electric) {code} ModelInfo: { File: coarse.ncdf BoundaryCondition: { Electric: 2 Magnetic: 3 4 Absorbing: 5 6 } } {code} h3. MeshPartitioning To specify the method to partition the mesh {code} |
MeshPartitioning
To specify the method to partition the mesh
Code Block |
---|
MeshPartitioning: { Method: PARMETIS //the other option is ZOLTAN Zoltan: { //if the main method is ZOLTAN, this container will provide further zoltan specific options Method: RCB Dimension: 1 Partition Direction: Z } } {code} h3. Normal finite element parameters |
Normal finite element parameters
Code Block |
---|
FiniteElement: { Order: 2 Order: 2 CurvedSurfaces: on } h3. P-window for short-range wakefield *set the // global order of basis orderfunctions to(can be 0 1...6, 2 is recommended) CurvedSurfaces: on } |
P-window for short-range wakefield
- set the basis order to be 0 (p=0).
...
Code Block
...
FiniteElement: {
...
Order: 0 //p=0 outside of the window CurvedSurfaces: on }
...
- set
...
- an
...
- automatic
...
- moving
...
- window
...
- that
...
- following
...
- with
...
- the
...
- beam
...
Code Block
...
PRegion: { Type: AutomaticMovingWindow
...
Order: 2 //inside the window, p=2 (basis function order) Back: 0.
...
01 //back pudding is 0.
...
01m Front: 0.1 //front pudding is
...
0.1m
...
...
...
StructureEnd: 1.0 //the maximal z. }
Moving-window with mesh refinement for short-range wakefield
- set the basis order to be 0 (p=0).
Code Block FiniteElement: {
...
Order:
...
0
...
...
//p=0 outside of
...
the
...
window CurvedSurfaces: on
...
}
- set an automatic moving window that following with the beam
Code Block MeshRefinement: { Order: 2
...
...
...
...
//inside the window,
...
p=2 (basis function order) Back: 0.01 //
...
back
...
pudding
...
is 0.01m Front: 0.1 //front pudding is 0.1m Subdivision: 1
...
//
...
subdivide each element inside window
...
once
...
StructureEnd:
...
1.0 //
...
the
...
maximal
...
z.
...
}
Gaussian beam going through a cavity
- The first step is to provide beam information:
Code Block LoadingInfo: { Bunch: { Type: Gaussian Sigma: 2e-3
...
...
...
//Sigma (RMS) size of the bunch
...
...
...
...
...
Nsigmas:
...
5
...
...
...
...
...
...
...
...
...
...
...
//beam occupies the location from -5 sigma to +5 sigma, total of 10 sigmas Charge: 1. //charge } SymmetryFactor: 4 //factor by which to reduce the charge to account for symmetry conditions (monopole on axis: use 4, dipole at X (or Y) offset: use 2 in connection with proper electric boundary conditions in one plane) StartPoint: 0. 0. 0. //StartPoint is the position where the beam enters the structure (typically at low Z values) Direction: 0. 0. 1. //
...
Direction along which the bunch
...
will
...
move,
...
at the
...
speed of light (should be the direction of the normal of the face with BoundaryID)
...
...
BoundaryID:
...
5
...
...
...
...
...
//The
...
boundary
...
ID
...
(sidelist
...
number
...
from Cubit), specifies the
...
boundary
...
through
...
which the bunch enters the structure (should be a flat surface, containing StartPoint) }
- Optional: Force analytical BeamBoundaryLoading (can be used if the beampipe is cylindrical). Not required. Default is OFF.
Code Block |
---|
Loading: { Type: BeamBoundaryLoading Analytical: on // Specify the right-handed coordinate system with its Z-axis along the beamline ( CrossProduct(X, Y) = Z = Direction specified above) Origin: 0.0 0.0 0.0 XDirection: 1.0 0.0 0.0 //this is the direction of the beam offset, if any YDirection: 0.0 1.0 0.0 Beampipe radius: 0.04 Beam offset: 0//and offset of the beam in x-direction of the local 2D coordinate system (value needs to be consistent with StartPoint above) //An analytical solution will be used. // For arbitrary beampipe //offset in x-direction of the local Poisson2D:2D {}coordinate system (value needs to be consistent with StartPoint //this will provide a numerical solutionspecified above) } {code} h3. Time |
Time Integration Parameters
Code Block |
---|
Integration Parameters {code} TimeStepping: { MaximumTime: 10.e-10 //the maximal time to step DT: 2e-12 //delta T } {code} h3. Wakefield Monitor <font color="red">need more expalantion< |
Wakefield Monitor
Code Block |
---|
/font> {code} Monitor: { Type: WakeField Name: wake // Weiland method InID: 5 OutID: 6 Start contour: 0.05 End contour: 0.10(not for protruding structures, beam pipe radius must be the same on left and right side) Name: wake Start structurecontour: 0.0 End structure: 0.1505 // z-position at which the beampipe-cavity transition starts End Smaxcontour: 0.310 // z-position at which the beampipe-cavity transition ends Smax: 0.3 // } {code} h3. Point Monitor To record the field values// atthe longitudinal wake potential will be recorded from s=0 to s=Smax } |
Point Monitor
To record the field values at specified location
Code Block |
---|
Monitorspecified location {code} Monitor: { Type: Point //point monitor Name: monA //an output file called monA.out will be generated //an output file called monA.out will be generated //it contains: t Hx Hy Hz Ex Ey Ez Coordinate: 0.00002, 0.02, 0.1495 //the location } |
Power Monitor
Code Block |
---|
Monitor: { Type: Power ReferenceNumber: 4 //which reference surface to monitor //it containsName: tmymon2 Hx Hy Hz ExTimeStart: Ey0 Ez Coordinate: 0.00002, 0.02, 0.1495 //thewhen power locationmonitor starts } {code} h3. Power Monitor {code} Monitor: { Type: PowerTimeEnd: 30.0e-9 //when it ends ReferenceNumberTimeStep: 4 0.125e-11 //whichhow often referenceit surfacerecords topower monitordensity } |
Volume Monitor
Code Block |
---|
Monitor: { Type: Volume Name: vol Name: mymon2 TimeStart: 0 10.e-9 //when powervolume monitor starts TimeEnd: 30500.0ee-9 //when it ends TimeStep: 050.125e-11e-9 //how often it records powervolume densityfields } |
After T3P finished runs, users should run acdtool to generate mode files for each records of the volume fields using the following command:
acdtool postprocess volmontomode t3pinput <jobname>
The mode files generated can be viewed using paraview.
CheckPoint
request T3P code to checkpointing itself every certain timesteps so that one can restart T3P.
Code Block |
---|
CheckPoint: {
Action: restart //default should be restart. If there is no data available, it will have fresh start.
Ntimesteps: 100 //every 100 times steps, code will checkpoint itself
Directory: CHECKPOINT //the default directory to store checkpointing data
}
|
LinearSolver
The options for linear solvers in the implicit timestepping.
Code Block |
---|
{code} h3. Volume Monitor {code} Monitor: { Type: Volume Name: vol TimeStart: 10.e-9 //when volume monitor starts TimeEnd: 500.e-9 //when it ends TimeStep: 50.e-9 //how often it records volume fields } {code} h3. CheckPoint request T3P code to checkpointing itself every certain timesteps so that one can restart T3P. {code} CheckPoint: { Action: restart //default should be restart. If there is no data available, it will have fresh start. Ntimesteps: 100 //every 100 times steps, code will checkpoint itself Directory: CHECKPOINT //the default directory to store checkpointing data } {code} After T3P finished runs, users should run acdtool to generate mode files for each records of the volume fields using the following command: {{acdtool postprocess volmontomode t3pinput <jobname>}} The mode files generated can be viewed using paraview. h3. LinearSolver The options for linear solvers in the implicit timestepping. {code} LinearSolver: { Solver: CG //other options include MUMPS (direct solver, faster //other options include MUMPSfor less than 32 CPUs) if it is compiled in Preconditioner: CHOLESKY //other options include DIAGONAL PrintFrequency: 50 //if you want print solver convergence history QuietMode: 1 //Set it to 1 if you do not want to print anything Tolerance: 1e-10 //relative tolerance MaxIterations: 3000 //maxima number of iterations before CG quits } |
Load a TEM waveguide mode on a coax port
Code Block |
---|
{code} h3. Load a TEM waveguide mode on a coax port {code} Loading: { Type: PortModeLoading //loading type Port: { ReferenceNumber: 3 //port is at reference surface 3 Origin: 0.0 0.0 -0.011 XDirection: 1.0 0.0 0.0 YDirection: 0.0 1.0 0.0 ESolver: { Type: Analytic Mode: { WaveguideType: Coax ModeType: TEM A: 0.0011 B: 0.0033 } } } Excitation: { Power: 1. Pulse: { Type: Monochromatic Frequency: 10.5e9 Rise periods: 150 Fall periods: 150 T0: 0. TMax: 100.e-9 } } } {code} |