Firefly and PC GAMESS-related discussion club



Learn how to ask questions correctly


Re: problematic SCF convergence

Richard
bonarlaw@liv.ac.uk


The usual approach would be to add SHIFT=.t. (maybe DAMP=.t. as well) to the $SCF group - see manual. FSHIFT can also be useful in difficult cases - see Dec 19 2009 post. Sometimes temporarily reducing nconv for a while helps get past a problem geometry during optimizations.

Richard

On Sat Mar 26 '11 5:26pm, Masoud Nahali wrote
---------------------------------------------
>Dear Firefly Users

>My system has passed 23 optimization steps and suddenly at 24th step I have encountered a problematic SCF convergence. As I checked there is not any large spin contamination and the spin parameters are SPIN SZ=2.000
>and S-SQUARED=6.289. The system contains Ni,C,O, and H and its multiplicity is 5. Below is my input file parameters and attached please find the SCF iterations. I appreciate your help in advance to find the reason and solve the problem.

> $CONTRL SCFTYP=UHF RUNTYP=OPTIMIZE DFTTYP=B3LYP ECP=read MAXIT=350 MULT=5
> ICHARG=0 COORD=unique d5=.t. NZVAR=366 $END
> $SYSTEM TIMLIM=600000 MWORDS=200 $END
> $SCF DIRSCF=.TRUE. NCONV=5 $END
> $STATPT OPTTOL=0.0001 NSTEP=150 $END
> $zmat dlc=.true. auto=.true. NONVDW(1)=123,4,123,9,123,14,123,1,
>123,17,123,8,124,4,124,9,124,16,124,6,124,15,124,10 $end
>! to activate P2P inteface and DLB:
> $P2P P2P=.T. DLB=.T. $END
>! to speed up Huckel guess:
> $GUESS guess=HUCKEL $END
> $smp call64=.t. mkl64=8 load=0 ssg32=0 msu32=0 mkllvl=3 $end
> $BASIS gbasis=LANL2DZ extfil=.t. $END
>
>
>
>
>Best Wishes

> Masoud
>--------------
>Masoud Nahali, Sharif University of Technology
>


[ Previous ] [ Next ] [ Index ]           Sun Mar 27 '11 2:37am
[ Reply ] [ Edit ] [ Delete ]           This message read 679 times