[gmx-users] Re: fudgeQQ (again ...) and fudgeLJ

[Qiao Baofu]

Hi,

I have just tested the fudgeQQ and fudgeLJ, and found one problem:
I use the ffoplsaa force field, and add the atoms I used.(in the
ffolpsaanb.itp file, there is no [pairtype] section, and I only add the atom
informations:
              ; name  bond_type    mass    charge   ptype
sigma      epsilon
               opls_966   CT     6     12.01100     0      A    3.40000e-01
4.5770e-01)
 I write my .itp file by hand, in which in the [ pairs] section, I write the
pairs and the function, but don't give the values of C6 and C12, that is,
[ pairs ]
; ai  aj   fu    c6    c12
   1    5   1
   1    6   1
...........

In the ffoplsaa.itp, I used
; nbfunc     comb-rule    gen-pairs    fudgeLJ   fudgeQQ
    1              2                yes         0.5      0.8333

I used different values of fudgeLJ and fudgeQQ in the 4 tests, see the
following
         fudgeLJ        fudgeQQ     result (from .log file):
LJ-14            Coulomb-14
1.         0.5             0.5
1.48e+02        -8.896e+03
2          1                1
1.96e+02        -1.793e+04
3.        0.25            0.25
9.58e+01        -4.43e+03
4.          0.5            0.8333
1.96e+02       -1.498e+04

The problem is that the LJ-14 doesn't change with the scaling factor of
fudgeLJ. Why?

Does it mean that I must give the C6 and C12 in the .itp file? If yes,
should the C6 and C12 scaled by the scaling factor (0.5) and "gen-pairs =
no; fudgeLJ=1" ?

Another question is that in the .log file, what is the  "Disper. corr."?
What is the difference between it and the 1-4 interacion?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20061209/0f52ba54/attachment.html>