Re: Plans for 2010: examples of undocumented features "group_materi_stokes 0 -yes"

Dear Tochnogers,

I tried simplification of the "cuartometric.dat" file.

------------- start ----------------
echo -yes
number_of_space_dimensions 2
derivatives
materi_velocity
materi_stress
condif_temperature
end_initia

options_mesh   -fixed_in_space -fixed_in_space
options_solver -matrix_superlu

(# if options_convection is set to -no it disables rho * v * grad(v) )
(options_convection -no) 

(# if options_inertia is set to -no it disables rho * dv/dt )
(# used for stationary processes )
(options_inertia -no)

start_define
  left_edge1  geometry_line 0
end_define
start_define
  left_edge2  geometry_line 1
end_define
start_define
  lower_edge  geometry_line 2
end_define
start_define
  upper_edge1 geometry_line 3
end_define
start_define
  upper_edge2 geometry_line 4
end_define
start_define
  upper_edge3 geometry_line 5
end_define
start_define
  right_edge  geometry_line 6
end_define
start_define
  temp_point  geometry_point 0
end_define

left_edge1  0.     0.      0.      0.9    1.e-3
left_edge2  0.     2.      0.      2.4    1.e-3
lower_edge  0.     0.      1.2     0.     1.e-3
upper_edge1 0.     2.4     0.73    2.4    1.e-3
upper_edge2 0.75   2.4     0.94    2.4    1.e-3
upper_edge3 0.95   2.4     1.2     2.4    1.e-3
right_edge  1.2    0.      1.2     2.4    1.e-3
temp_point  0.24   0.891 2.e-3

bounda_unknown    0  -left_edge1  -velx -vely
bounda_time       0  0.
bounda_unknown    1  -left_edge2  -velx -vely
bounda_time       1  0. 
bounda_unknown    2  -lower_edge  -velx -vely
bounda_time       2  0.
bounda_unknown    3  -upper_edge1 -velx -vely
bounda_time       3  0.
bounda_unknown    4  -upper_edge2 -vely
bounda_time       4  0.    0.
                     5.    0.5
		     1.e6  0.5
bounda_unknown    5  -upper_edge3 -velx -vely
bounda_time       5  0.
bounda_unknown    6  -right_edge  -velx -vely
bounda_time       6  0.
bounda_unknown    7  -temp_point  -temp
bounda_time       7  0.    0. 
                     3.    0.
		     5.    1.
		     1.e6  1.

group_type                          0  -materi  -condif
group_materi_viscosity              0 1.82e-05 (N.s/m2 @ 20 C) 
group_materi_density                0  1.205  (kg/m^3 @ 20 C)
group_materi_memory                 0 -updated_without_rotation
group_materi_elasti_compressibility 0 7.65e-06 (m^2/N)
group_materi_stokes                 0 -yes
group_condif_density                0  1.205  (kg/m^3 @ 20 C)
group_condif_capacity               0 1006.0 (J/Kg C )
group_condif_conductivity           0  1.89e-05 (m^2/sec diffusion coeff.)
group_condif_absorption             0 0.0

control_mesh_macro             0 -rectangle 0 21  36
control_mesh_macro_parameters  0 0.6 1.2  1.2 2.4

control_timestep              10  0.01     0.05
                                  0.05     0.5
		                  0.5      1.
		                  1.       5.
		                  2.5     10.
		                  5.      20.
		                 25.     250.
		                100.    1200.
control_timestep_iterations   10  3
control_print                 10  -time_current -post_node_rhside_ratio
control_print_gid             10 -yes

end_data
---------------- end -------------------

>Dear Tochnogers,
>
>Throughout 2010 I will try and list how to implement important and undocumented features in tochnog. If you check the 
differential equation for the convection diffusion equation, it only offers the option of specifying the velocity field,  
group_condif_flow  which is a scalar that represents the fluid velocity.
>
>I was trying to solve a case which involved a non uniform velocity field and had to dig deep into the tochnog code to try and 
implement a solution for making group_condif_flow point to the velocity field. After getting the code to work, I realized that 
although not documented, the original code had the option. 
>
>I am providing you tochnogers with the source code and solution, including some GiD generated movies. The solution 
represents a 2-D fluid flow problem. Tochnog first solve the problem and after some time, I let a gas source start 5 seconds into 
the solution to give time for a velocity field become organized.
>
>The solution was obtained using  options_solver -matrix_superlu. If you are not using a version with superlu, comment out this 
line.
>
>This is the dat file. three more e-mails to follow.
>
>I am trying to send the mpeg movies generated with GiD but I am having troubles posting them due to their size, so I am going 
to post them at the tochnog website
>
>Enjoy!
>
>
>Fernando Lorenzo

Minoru TANAKA
NIT
Department of Architecture
[email protected]

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com