*> *
*> Also, I got the error message: fe__k: non-positive Det[1-Wvv*Cvv*Rho]. I
*
*> checked the density and the dielectric constant in the input, which are *
*> fine. I am wondering how should I solve this problem. *
> Can't help you here; you may need to provide the MDL file you are using.
> Of course, be sure to use the correct value for LJSIGMA: if that is wrong
> by a factor of two, many other things will probably fail.
Here is my mdl file for octanol (united atom model).
%VERSION VERSION_STAMP = V0001.000 DATE = 06/02/16 12:15:14
%FLAG TITLE
%FORMAT(20a4)
OCT
%FLAG POINTERS
%FORMAT(10I8)
10 5
%FLAG ATMTYP
%FORMAT(10I8)
1 2 3 3 4
%FLAG ATMNAME
%FORMAT(20a4)
HO OH CA CB CH
%FLAG MASS
%FORMAT(5e16.8)
1.00800000e+00 1.60000000e+01 1.40200000e+01 1.40200000e+01
1.50300000e+01
%FLAG CHG
%FORMAT(5e16.8)
7.92670050e+00 -1.27556100e+01 4.82890950e+00 0.00000000e+00
0.00000000e+00
%FLAG LJEPSILON
%FORMAT(5e16.8)
1.70000001e-02 1.70000000e-01 1.18000000e-01 1.18000000e-01
1.50000000e-01
%FLAG LJSIGMA
%FORMAT(5e16.8)
0.80000000e+00 1.72300000e+00 2.19150000e+00 2.19150000e+00
2.19150000e+00
%FLAG MULTI
%FORMAT(10I8)
1 1 1 6 1
%FLAG COORD
%FORMAT(5e16.8)
-3.94900000e+00 -2.62800000e+00 -1.68100000e+00 -4.27100000e+00
-2.09200000e+00
-9.72000000e-01 -3.54900000e+00 -8.82000000e-01 -9.25000000e-01
-2.11700000e+00
-1.30200000e+00 -5.94000000e-01 -1.27200000e+00 -4.00000000e-02
-4.18000000e-01
1.55000000e-01 -3.15000000e-01 5.80000000e-02 1.02300000e+00
9.28000000e-01
2.54000000e-01 2.45100000e+00 6.27000000e-01 7.10000000e-01
3.33400000e+00
1.87400000e+00 7.59000000e-01 4.73200000e+00 1.48600000e+00
1.24100000e+00
and my input file for 1drism is here:
&PARAMETERS
OUTLIST='x, T', THEORY='DRISM', CLOSURE='KH',
!grid
NR=16384, DR=0.025, ROUT=384, KOUT=0,
MDIIS_NVEC=10, MDIIS_DEL=0.5, TOLERANCE=1.e-10, mdiis_restart = 1000,
!iter
KSAVE=-1, PROGRESS=1, MAXSTEP=20000,
!ElStat
SMEAR=1, ADBCOR=0.5,
!bulk solvent properties
TEMPERATURE=298, DIEPS=10.3,
NSP=1
/
&SPECIES
DENSITY=6.34d0,
MODEL="./OCT.mdl"
/
The error message is: ERROR> fe__k: non-positive Det[1-Wvv*Cvv*Rho]
STOP 1
Thanks very much for your time!
Joy
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jun 02 2016 - 15:00:02 PDT
