[CSC 435] Getting Diffusion to Scale Properly

[Andrew J. Pounds]

Guys -- to get the Jacobi tiling method I sent you yesterday to scale 
properly (the room equilibrates in ~55 simulated sec) you have to be 
very careful about constants and dimensions.

First, here are components from the CSC 330 project...

room_dimension = 5                      ! 5 Meters
speed_of_gas_molecules = 250.0D0        ! Based on 100 g/mol gas at RT
time_between_blocks = (room_dimension / speed_of_gas_molecules) / msize
distance_between_blocks = room_dimension / msize
area = distance_between_blocks**2
timestep = time_between_blocks


Now -- the scalar value is what you have to multiply your "volume 
change" by in the tiling code.  I used the following...

scalar = diffusion_coefficient * timestep / distance_between_blocks**2 * 4

Which produced the following results...


diffusion

   Box dimension =    5.0000000      meters.
   Box Partition =           50  steps.
   Time Step     =   3.99999990E-04  seconds.
   Diffusion C   =   0.17500000

  Simulation required    51.636589      seconds to reach equilibrium.

   Run took           51  minutes and    43.970215      seconds of 
processor time.


   Run took            2  minutes and           10  seconds of wall 
clock time.



-- 
Andrew J. Pounds, Ph.D.  (pounds_aj at mercer.edu)
Professor of Chemistry and Computer Science
Mercer University,  Macon, GA 31207   (478) 301-5627
http://faculty.mercer.edu/pounds_aj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://theochem.mercer.edu/pipermail/csc435/attachments/20160502/544e7840/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: plot.png
Type: image/png
Size: 5688 bytes
Desc: not available
URL: <http://theochem.mercer.edu/pipermail/csc435/attachments/20160502/544e7840/attachment.png>