CVEL Webhead

KEYWORDS FOR SIFT5

Keywords, as shown in Table I, have been re-designed to support changes in EMAP5. Users must use these keywords to describe problems. SIFT5 can translate these keywords into input files for EMAP5. A detailed description of these keywords follows:

Table I. Syntax of keywords for SIFT5.

 
keyword
position coordinates
cell dimensions
sub attributes
#
     
unit
    number in m, cm or mm
boundary
x1 y1 z1 

x2 y2 z2

   
celldim
p1 p2 D p axis (x, y or z)
dielectric
x1 y1 z1 

x2 y2 z2

    e'   e
conductor
x1 y1 z1 

x2 y2 z2

D x D y D z  
eplane
    Frequency q1 j1 q2 j2 magnitude
vsource
x1 y1 z1 

x2 y2 z2

  Frequency polarization (x, y, or z) magnitude
isource
x1 y1 z1 

x2 y2 z2

  Frequency polarization (x, y, or z) magnitude
output
x1 y1 z1 

x2 y2 z2

  axis(x, y, or z) filename
default_out
    filename

#

SIFT5 will skip any line that starts with a "#". Users can use "#" to add comments to their .sif files.


unit

EMAP5 uses meter as the measure of distance. However, users can define their own units in .sif files. SIFT5 converts all geometry specified in .sif files into values in terms of meter internally. The keyword "unit" specifies   the length of one standard mesh cell along all three axes. The following specifies a standard cell to be two centimeters,

unit 2.0 cm

Then, all geometry specified by other keywords are relative to the unit specified above.
If one wants to specify the geometry in terms of inch, the following line should be include in the .sif file,

unit 2.54 cm

Then, all geometry defined by other keywords will have a unit of inch.


boundary

The keyword "boundary" specifies a rectangular brick which forms the FEM region. The FEM region should enclose all dielectric slabs. In addition, the boundary should be as close as possible to the dielectric surfaces to reduce computation time and memory requirements. The boundary should normally coincide with the surface of dielectric materials if only one dielectric object is present.  


celldim

By default, the mesh size for the FEM volume is one unit along each axis. However, users can specify a mesh step in an interval along the x, y or z axes by using the keyword "celldim". For example, the following line defines the mesh step to be two units in [0, 10] along the x axis,

celldim 0 10 2 x

If the unit is defined as one centimeter, five segments will be generated in the interval [0 10] along the x axis. Each segment has a length of two centimeters.


dielectric

The keyword "dielectric" defines a rectangular dielectric slab. (xi, yi, zi), i=1, 2 specifies two nodes on opposing corners of the slab. e' and e" refer to the real and imaginary parts of the complex permittivity, respectively. A configuration can include one or more dielectric slabs. However, the FEM region defined by "boundary" should include all dielectric bodies.


conductor

The keyword "conductor" defines a conductor. At present, SIFT5 can only generate mesh files for patch (i.e. two dimensional) conductors. Traces are typical patch conductors. Thin wires can be treated as patch conductors by using an equivalent width [17]. SIFT5 will use (D x, D y, D z) as mesh steps along the x, y and z axes to discretize the patch conductor and to generate edge grids.

SIFT5 will internally classify conductors into three categories: internal conductors, boundary conductors, and external conductors according to the geometric relationship between the conductors and the FEM boundary. For internal and boundary conductors, (D x, D y, D z) specified by the keyword "conductor" will be ignored. Instead, these conductors will be meshed using the (D x, D y, D z) specified by the keyword "celldim" in order to match grid edges along the boundary.


eplane

The keyword "eplane" defines an incident plane wave source. The units of frequency are MHz. (q1, j1) defines the E unit vector while (q2, j2) defines the direction of propagation. The following defines a 300 MHz plane wave that travels along the +z axis. The polarization of the E fields is along the x axis. The magnitude of the E field is one volt/meter.

eplane 300 90 0 0 0 1.0


vsource

The keyword "vsource" defines a "delta-gap" voltage source. The two nodes specified by (x1, y1, z1) and (x2, y2, z2) in the parameter list specify a line which is the source location. The voltage source can be only defined on metal patches, which are not within the FEM region. The source should be located on grid edges defined by keywords "conductor", "unit" or "celldim".


isource

The keyword "isource" defines an impressed current source. The two nodes specified by (x1, y1, z1) and (x2, y2, z2) in the parameter list specify a line which is the source location. The voltage source can be only defined within the FEM region. The source should be located on grid edges defined by keywords "unit" or "celldim".  


output

The keyword "output" defines the region in which fields of interest will be printed out by EMAP5. (x1, y1, z1) and (x2, y2, z2) specify two nodes on opposing corners of a rectangle. All edges within this rectangle and parallel to the specified axis will be printed out to the file specified in the parameter list. For example,

output 0 0 0 10 10 0 x E1.out

will cause EMAP5 to print out the fields along edges, which are parallel to x axis and within the rectangle area specified by two diagonal nodes (0, 0, 0) and (10, 10, 0), to the file E1.out.


default_out

The keyword "default_out" causes EMAP5 to print all surface fields to the file specified in the parameter list, which is required by FAR. The file generated using keyword "output" can not be used with FAR. For example,

default_out E1.out

prints all surface fields to the file E1.out.