Mesh grid is the first parameter needs to be initialized in using Silvaco to simulate the NMOS structure. Mesh grid parameter could give effects to the simulation process of NMOS. As an example, doping process depends on the y-axis grid location and spacing. In Silvaco TCad, mesh grid is also known as the simulation grid. the correct specification of grid is critical in process simulation. The number of nodes in the grid has a direct influence on simulation accuracy and time.  A finer grid should exist in those areas of the simulation structure where ion implantation will occur, where p-n junction will formed or where optical illumination will change photoactive concentration. Usually the finest grid will defined at the surface of the substrate. this region will be used to form the surface active region of the NMOS transistor.

Best regards.

I would like to add to Andrea, that The device simulator solves numerically the semiconductor equations in the regions of the the device. These equations are the Poisson, current and continuity equations. They are partial differential  equations in position and time.  To solve such equations numerically using digital computers the space of the solution and the time must be discretised. Descretising two dimensional area results in a network of points termed nodes.The array of points may also termed grid. It is so that the solution will be given at these selected points. The number of points per certain unit distance is determined by the rate of change of the change of the function with distance. As hinted by Andrea, in the regions of fast changes one has to densify  the grid points other wise the solution may not be accurate or even may be wrong.  It is so as the grid points increases the solution will be accurate. Unfortunately, the calculations will take longer time. It is so that one must increase the grid points till no improvement in the output of the solution.

Also, in order to decrease the number of points one. can use non homogeneous grid with points density higher in the rapid changing region and much less dense in the slow varying regions.

Best regards 

Grid is required not only for MOS device but also for any device that is to be simulated. This defines the structure and density of the device. Simulation accuracy has direct influence on grid. In denser areas fine grids are required such as, inversion channel of MOSFET when we need to calculate the inversion charge or drain current in the channel. In less dense area, we need coarse grid, such as, in the bulk region of the MOS device.

Thank you for the answers.

In some of the examples i have seen that they have initialized the starting value but end point is not defined.

As in this example starting point in x-axis is 0.0 (with spacing of 0.2 ) till 10. But at the last line for x-axis the point is staring from 14 (with spacing of 0.1 ).. but where it will end.  

for x-axis

x.m l=0.0 s=0.2 

x.m l=10 s=0.2

x.m l=12 s=0.05

x.m l=14 s=0.1

for y axis

As in this example starting point in y-axis is 0 (with spacing of 0.2 ) till 0.1. But at the last line for y-axis the point is staring from 0.3 (with spacing of 0.05 ).. but where it will end.

y.m l=0 s=0.2

y.m l=0.1 s=0.2

y.m l=0.2 s=0.02 

y.m l=0.3 s=0.05

how can change the width in 3rd dimension.

Very useful answers provided above.

is it possible to create triangular,cylindrical or different shapes structures?.....if yes then how to do it...

Thank you for the answer.....I read the manual but i was not able to find for the triangular structure.... i am finding difficult to make angle and joining the two sides.....is it possible to make angle between two sides.....  

I got one example for cut plane but its not very clear to start.....using this method 

Is it possible to find all the characteristics curves ............ 

tonyplot3d mos2ex04_2.str -set cutplane.set -cut 2D_Slice.str

set x_start = 0

set x_end = 1.1

set x_delta = 0.1

set y_start = 0

set y_end = 1

set y_delta = 0.1

set data_in = “2D_Slice.str”

set data_out = potential_map.dat

set num_x_steps = 1 + ($x_end - $x_start) / ($x_delta)

set num_y_steps = 1 + ($y_end - $y_start) / ($y_delta)

go atlas

system rm -f $data_out

set y = $y_start

loop steps=$num_y_steps

set y1 = $y

set y = $y1 + $y_delta

set x = $x_start

loop steps=$num_x_steps

set x1 = $x

set x = $x + $x_delta

extract init inf=”$’data_in’”

extract name=”data_point” y.val from curve (depth,impurity =”potential” material =”Silicon” mat.occno=1 x.val=$”x1”) where x.val=$”y1”

system echo -n ‘$”data_point” ‘ >> $data_out

l.end

system echo ‘ ‘ >> $data_out

l.end

quit

Dear Gurmohan Singh 

what is r, a and z in defenition of cylindrical structure?

can any one plz indicate the error in this code ?

go atlas

#Meshing

#scale factor applied to all specified grid spacing

mesh space.mult=0.8

#MESH AUTO

x.mesh loc=0.0 spac=0.2

x.mesh loc=0.6 spac=0.02

x.mesh loc=0.9 spac=0.02

x.mesh loc=1.5 spac=0.2

x.mesh loc=2.1 spac=0.02

x.mesh loc=2.4 spac=0.02

x.mesh loc=3 spac=0.2

y.mesh loc=-0.55 spac=0.01

y.mesh loc=-0.01 spac=0.009

y.mesh loc=0.00 spac=0.005

y.mesh loc=0.3 spac=0.02

y.mesh loc=2.0 spac=0.5

#eliminate mesh from unnecessary areas

##Eliminates every second mesh point in a rectangular grid specified

eliminate rows x.min=0.00 x.max=0.85 y.min=-0.55 y.max=0

eliminate rows x.min=2.15 x.max=3.00 y.min=-0.55 y.max=0

eliminate rows x.min=0.00 x.max=3.00 y.min=0.30 y.max=2

eliminate columns x.min=0.00 x.max=0.85 y.min=-0.55 y.max=0

eliminate columns x.min=2.15 x.max=3.00 y.min=-0.55 y.max=0

eliminate columns x.min=0.00 x.max=3.00 y.min=0.30 y.max=2

#Regions

region num=1 y.min=0 y.max=2 x.min=0.00 x.max=3.00 silicon

region num=2 y.min=-0.55 y.max=0. x.min=0.85 x.max=2.15 oxide

#region num=3 y.min=-0.55 y.max=0 x.min=0.00 x.max=0.85 ignore

#region num=4 y.min=-0.55 y.max=0 x.min=2.15 x.max=3.00 air

#Electrodes

#Conductor names are only associated with electrodes

electrode name=float_gate num=1 y.min=-0.25 y.max=-0.01 x.min=0.85 x.max=2.15

electrode name=ctrl_gate num=2 y.min=-0.55 y.max=-0.29 x.min=0.85 x.max=2.15

electrode name=src num=3 y.min=0 y.max=0.00 x.min=0.00 x.max=0.60

electrode name=drain num=4 y.min=0 y.max=0.00 x.min=2.40 x.max=3.00

electrode name=subs num=5 bot

#Multiple electrode statements may have the same electrode name.

#Nodes that are associated with the same electrode name are treated as being

#electrically connected.

#Doping

doping uniform conc=1e14 p.type reg=1 outfile=mydop.dop

doping uniform conc=2e17 n.type y.min=0 y.max=0.2 x.min=0.00 x.max=0.90 reg=1

doping uniform conc=2e17 n.type y.min=0 y.max=0.2 x.min=2.10 x.max=3.00 reg=1

#Contacts

#For contacts directly onto semiconductor, the FLOATING parameter cannot be used.

#This type of floating electrode is best simulated by specifying current boundary conditions on the CONTACT statement.

#For example: CONTACT NAME=drain CURRENT

contact name=float_gate n.polysilicon floating

contact name=ctrl_gate n.polysilicon current

#CONTACT name=subs COMMON=subs1 FACTOR=0.5

#CONTACT name=ctrl_gate COMMON=subs1 FACTOR=0.5

#regrid the mesh (used after MESH, REGION, MATERIAL, ELECTRODE, and DOPING statements)

#REGRID LOG DOPING RATIO=6 OUTF=grid1 DOPF=dopxx SMOOTH=4

regrid log ratio=2 region =1 doping x.min=0.00 x.max=3 y.min=0.00 y.max=0.2 SMOOTH.K=4 MAX.LEVEL=2

save outf=project_kk.str master

tonyplot project_kk.str

#Define some Qss

#MODELS CONMOB FLDMOB SRH FERMIDIRAC

#models model macros lke program erase E.FGCGTUN HEI E.SC.FGCGTUN FG.RATIO N.CONCAN P.CONCAN

#interface qf=3e10

interface qf=3e10 y.min=-0.01 y.max=0 x.min=0.85 x.max=2.15

models srh cvt hei fnord print nearflg

impact selb

method newton

solve init

solve vdrain=0.5

log outf=kk01.log

solve vstep=0.5 vfinal=25 name=ctrl_gate

# plot idvg

tonyplot kk01.log

Very helpful answers but what about spacing x.mesh loc=0.0 spac=0.2. How can we decide spacing to define a structure

Thank you very much Girish

Thankyou Girish Wadhwa Sir for wonderful videos. Kindly make any video for mixed-mode simulations in SILVACO TCAD tool.