Re: [ESPResSo-users] Are the conical pores in ESPResSo allowed?

[Stefan Kesselheim]

Hi Farnoosh, hi users,

On Mar 14, 2013, at 7:59 PM, Farnoosh Farahpoor <address@hidden> wrote:

> Dear all
> 
> As I have seen in the source code of ESPResSo (interaction_data.h) PORE 
> constraint has two different variable for right and left radius and so it 
> seems that in principal it could be possible to have pores with conical shape 
> in ESPResSo but for initialization (in constraint_tcl.c and lb_boundaries.c) 
> it is supposed that these two radii are equal.
> 
> Is there any technical problem in force calculation or hydrodynamic 
> interaction which force us to just using cylindrical pores?
> 
> Isn't it possible to remove the limit of equality of radii by just a small 
> change in constraint_tcl.c and lb_boundaries.c?

This is a misunderstanding: You can either specify one radius using "constraint 
pore …. radius $radius" or two different radii "constraint pore radii $r1 $r2". 
In principle this should work, and also for LB boundaries. 
You should however use a more recent version of Espresso (the new 3.1.2 works 
for sure, but I believe also 3.1 is ok), because it was only 95% working in an 
older version.
If you have trouble: The dielectric test case shows how to use the constraint 
with conical pore (and decorates it with charges). If you uncomment the last 
two lines it will start vmd and you can look at how the pore looks.
I think I never tried it with lbboundary, but it should work as well.
If you have trouble, please let me know.
Cheers
Stefan