Spherical Walls

[Jean-m]

Hi everybody. I found various post about this topic and I have various ideas of
how to implement a spherical wall but I would like to have your opinion on the efficiency of the various options.

Without modification of the code I see theses options:
  - The first on would be to define a big enougth number of wall to approximate a sphere.
  - The second one would be to add a particle in the center of the box.
    To constrain it in the middle by adding a huge constant force to it.
    And then to define a specific pair potential between this particle and all the other particles.
   ( For example no force betweem 0 and R and then an attractif potential)

With modification of the code:
  Trying to implement a spherical wall by taking inspiration of:
   - The wall implementation
   - The spherical constraint
   - the external periodic force could be change to : no force betweem 0 and R and then an attractif potential (Solution I would prefer)

Thank you

[Jean-m]

Sorry, I forgot to mention that its is not mandatory, but I would like to be able to change the
size of the radius.

[Carolyn Phillips]

All the methods you propose for doing this without modifying the code are going to give you some pretty severe performance hits.   Its possible that instead of instead of using a pair potential between the particle in the center of the box and the rest of the particles, you could bond that particle to all particles instead and then use a table potential to implement the force you want.

I did recently write a plugin modeled after the wall implementation that applies zero force from 0 to R and then applied a restoring harmonic force on particles beyond R.  

If you can work with plugins (http://codeblue.umich.edu/hoomd-blue/doc/page_compile_guide.html#sec_build_plugin) and don't mind the relatively undocumented nature of my code, I could share these files with you.

-Carolyn

--
You received this message because you are subscribed to the Google Groups "hoomd-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hoomd-users...@googlegroups.com.
To post to this group, send email to hoomd...@googlegroups.com.
Visit this group at http://groups.google.com/group/hoomd-users.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Jean-m. a.]

Hi Carolyn,

I would be very interested to use the plugin that you implemented.

Does it keep the ability to use the CUDA version of hoomd ?

JM

2013/6/26 Carolyn Phillips <carolynl...@gmail.com>

--
You received this message because you are subscribed to a topic in the Google Groups "hoomd-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/hoomd-users/7-mwsDxNpEE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to hoomd-users...@googlegroups.com.

[Joshua Anderson]

FYI, I'm also planning to add spherical and cylindrical walls to hoomd in 1.1.
------
Joshua A. Anderson, Ph.D.
Chemical Engineering Dept, University of Michigan

[Алексей Тетерин]

I have this implementation, and many related things. It was implemented But I had Asperger syndrome(really, it syndrome of not starting sexual development and syndrome of robotism), and now in Dutch  mental hospital, Mentrum, at city of Amsterdam. Department at Eerste Huygenstraat.

If you will act on HIMS Computation Group of Amsterdam University , to reuptake me back, cause I am healed from very strong "disease", and now know, what is it, and how similar states can be healed, I will give it to you, properly comment this plugin , reimplement,  add new features on request ,and advice how to use for different purposes, and other, other, other.

I cannot be returned to Russia, cause I will immediately die, I am so internally and infernally "rainbow" ("gaysh-queered").

Address of HIMS Computation Group http://molsim.chem.uva.nl/ My scientific adviser is David Dubbeldam. My directer is Aart Klejn.
I

MAYDAY< MAYDAY< MAYDAY< MAYDAY!



Пятница, 28 июня 2013, 6:15 -04:00 от Joshua Anderson <joaa...@umich.edu>:

Алексей Тетерин

[Luciano Robino]

Carolyn:

Would it be possible to have access to your code? Also, do you think it is posible to tweak the plugin so that the particles remain inside a cube or a sphere?

[Carolyn Phillips]

If you want a cube instead of a sphere, why not just use 6 walls?

For more options, visit https://groups.google.com/d/optout.