RE: tcl force

From: JC Gumbart (gumbart_at_ks.uiuc.edu)
Date: Sun Oct 18 2009 - 22:45:56 CDT

All I mean is that if you use tclforces, you need to use addenergy in
addition to addforce IF you want to track the energy as well. Addforce
alone contributes nothing to the total energy.

 

 

From: Seungho Choe [mailto:sec53_at_pitt.edu]
Sent: Sunday, October 18, 2009 9:00 PM
To: JC Gumbart
Cc: namd-l_at_ks.uiuc.edu
Subject: Re: namd-l: tcl force

 

What I see in the manual is that "addenergy" is optional for normal runs.
However, if I want to explicitly see the additional energy term, then I need
to add "addenergy" as you mentioned.
Is this what you meant ?

Thanks,
Seungho

JC Gumbart wrote:

>From my reading of the manual, you do have to manually add the energy, in
addition to the force.

 

I think your tcl script is fine, except it's more complicated than
necessary. The force is just F = -k * (z-z0); you don't need to set
conditions on r0.

 

 

From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf
Of Seungho Choe
Sent: Saturday, October 17, 2009 1:08 PM
To: JC Gumbart
Cc: namd-l_at_ks.uiuc.edu
Subject: Re: namd-l: tcl force

 

Hi JC and NAMD users,

I'm trying to compare both methods: tcl force & SMD.
Could you please look at my tcl script again and tell me whether it is
correct ?
I wonder whether I'm missing some factors in the script. Thanks a lot.

Best,
Seungho

JC Gumbart wrote:

Yes, you can. But it will only work in 1D as you've proposed, e.g., not n =
(1,1,1).
 
Regarding your initial question, see the addenergy term in the manual:
http://www.ks.uiuc.edu/Research/namd/2.7b1/ug/node43.html
 
 
-----Original Message-----
From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf
Of Seungho Choe
Sent: Saturday, October 17, 2009 12:21 AM
To: Seungho Choe
Cc: namd-l_at_ks.uiuc.edu
Subject: Re: namd-l: tcl force
 
Hi,
 
I wonder whether I can use SMD with v=0, n= (0,0,1) to restrain the
center of mass of a molecule in z direction ?
It seems that in this case I don't need to worry about the energy term
that I mentioned in the previous my post. Could someone clarify this ?
Apology if this issue was already discussed before.
 
Best,
Seungho
 
 
 
 
Seungho Choe wrote:
  

Dear all,
 
I'm trying to restrain the center of mass (only z position) of a
molecule. Below is my tcl script. What I'm wondering is whether a new
energy term from the added force is applied correctly to the total
energy. Do I need to explicitly add the energy term in the script ? Is
there any way to check this ? Any comments are welcome. Thank you in
advance.
 
Best,
Seungho
 
************************************************************************
set numatoms 63230
 
set atoms {}
for {set i 63207} { $i <= $numatoms} { incr i} {
   lappend atoms $i
}
 
print "atoms:$atoms"
 
foreach atom $atoms {
   addatom $atom
}
 
set groupid [addgroup $atoms]
 
 
# Set spring constant
set k 7.0
set z0 -3
 
proc calcforces {} {
 
global atoms groupid numatoms k z0
loadcoords p
 
#print "$p($groupid)"
 
set z1 [lindex $p($groupid) 2]
 
print "z comp:$z1"
 
set r01 [expr $z1-$z0]
#print "r01:$r01"
 
if {$r01 > 0} {
   set n01 {0 0 -1}
} elseif {$r01 < 0} {
   set n01 {0 0 +1}
}
 
# Calculate the "force"
set force [expr abs($k*$r01)]
 
set forcez [vecscale $force $n01]
 
print "forcez:$forcez"
 
addforce $groupid $forcez
 
}
 
 
    

 
  

 

 

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:23 CST