The ABINIT Users Mailing List ( CLOSED )

["Ane Xu" <xzp02@mails.tsinghua.edu.cn>]

Dear XG:
    Thanks a lot for your helpful reply, I have tried some cases, but
still have some questions.
1) I make changes that
    tolmxf 5.0d-4, toldff 5.0d-5, I set ntime=50, in the first 49 steps
it's alright, but after the 50th step it writes strange coordinates into
the GEO file, like following:
/*
 Atom list        Reduced coordinates          Cartesian coordinates
(bohr)
    C(1)   -135.67034 117.03091 146.79584
**********1261.894931165.09655
    C(2)    188.93904-142.77457 -99.03563
3593.35136**********-786.03093
    C(3)    -99.39994-186.22634  40.50058     112.14069**********
321.44702
    C(4)    -99.39994-186.22634 -40.75415
112.14069**********-323.45958
    C(5)   -391.06710-561.48009 386.48280
-870.56994**********3067.45589
    C(6)   -391.06710-561.48009-386.73637
-870.56994********************
    C(7)   -135.67034 117.03091-147.04941
**********1261.89493**********
    C(8)    188.93904-142.77457  98.78206    3593.35136**********
784.01837
    C(9)    -99.95091  24.64285 120.01347    ********** 265.71345
952.52888
    C(10)   195.30226 105.21562 300.76874
1777.230391134.495592387.15627
    C(11)   308.07778 -77.32945-144.61444
4667.83687-833.81083**********
    C(12)   308.07778 -77.32945 144.36087
4667.83687-833.810831145.77055
    C(13)  -118.47131-127.14750 -79.70579
-592.84514**********-632.61289 
*/
    I do not know what happened, do you have any idea on that?
>> /*
>>  fconv : at Broyd/MD step  24, gradients have not converged yet.
>>   max grad (force/stress) = 4.3791E-03 > tolmxf= 5.0000E-05 ha/bohr 
>> (free atoms) */
>>    The grad is too large after 24 steps! Is my tolmxf too large?
maybe
>> result from 5.0D-3 or 5.0D-4 is convictive?
>I suspect that your convergence on the forces is not well controlled, 
>using toldfe 1.0d-6 . You would better choose a stopping criterion for
the SCF convergence >based on the forces, i.e. toldff 5.0d-6 This will
give sufficiently accurate 
>forces to propagate the nuclei, and to achive convergence of atomic
positions with target> residual forces less than 5.0d-5 .
>    I used toldff 5.0d-6, tolmxf 5.0d5

2) I have noted that in the log file it comments:
/*
 pspatm : COMMENT -
  the projectors are not normalized,
  so that the KB energies are not consistent with
  definition in PRB44, 8503 (1991).
  However, this does not influence the results obtained hereafter.
 pspatm: epsatm=   -0.01143560
         --- l  ekb(1:nproj) -->
             0    0.945422
 pspatm: atomic psp has been read  and splines computed 
*/
    I use Hartwigsen-Goedecker-Hutter psp for Carbon, PBE GGA for
calculation. I wonder how will this affect my results, is there some
comparison comments on the psp of carbon used in ABINIT ?


Best wishes

Ane