[Wien] 'stop - follow, nstep > maxfoll' in aim - it's okay now

[Kevin Jorissen]

The problems with aim have been sorted out, and they were actually my
problems, not aim's.
* The calculations where aim never stopped : I was using an old version of
aim that still contained a bug which has been fixed some time ago.
* The calculation where I got the 'stop-follow; nstep > maxfoll' message : I
have some suspicions about this calculation (not because of core leakage,
though) and there may well be the sort of discontinuities that prevent aim
from doing its job.  Expanding the lm-list will not help here, the
calculation itself will have to be cured (may be unphysical).

Many thanks to Javier and Peter for their help,

Kevin.



----- Original Message -----
From: "Javier Fuhr" <fuhr at crmc2.univ-mrs.fr>
To: <wien at zeus.theochem.tuwien.ac.at>
Sent: Friday, July 11, 2003 5:40 PM
Subject: Re: [Wien] 'stop - follow, nstep > maxfoll' in aim


> Dear Kevin,
>
> Your input file seems okay, I would suggest to increase MAXFOLL and try if
> with that you can solve the problem. If that does not work, it could be a
> problem with some discontinuity of the electron density between the
> muffin-tin and the interstitial region. Sometimes it is necessary to have
a
> high value of GMAX, which control the description of the electron density
on
> the interstitial region. If you continue having problem send me the struct
> and an clmsum files so that I can look at them.
>
> Javier.
>
> On Thursday July 10 2003 15:16, Kevin Jorissen wrote:
> > Hello WIEN-users,
> >
> > I try to calculate the Bader-charge of an atom using the aim-program and
a
> > pretty default input-file case.inaim :
> >
> > SURF
> > 1                          atom in center of surface
> > 20 0.0 3.1415626535     theta, 20 points, from zero to pi/2
> > 40 0.0 6.2831853072       phi, from pi/4 to  3 pi/4  (depends on
> > symmetry!!) 0.07 0.8 4                 step along gradient line, rmin
(when
> > reached it assumes the gradient line ends at the atom), every 4th step
it
> > checks wether gr.path is behing/in front an already found surface
> > 1.6 0.1                    initial R for search, step (a.u)
> > 2 2 2                      nshell
> > IRHO                       "INTEGRATE" rho
> > WEIT                       WEIT (surface weights are available in
> > case.surf), NOWEIT if surface put int by hand
> > 30                         30 radial points outside min(RMIN,RMT)
> > END
> >
> >
> > For some materials this seems to work, i.e., it produces charges that
> > appear reasonable and are only very slightly sensitive to some of the
input
> > options that I varied.
> > In another case (a simple calculation for MnO2), the program stops
rather
> > soon and writes to the screen :
> > follow
> >
> > Judging from case.outputaim, it has obviously not ended its task.  The
last
> > lines read :
> > [...]
> > distance to atom  -1    3.7027
> >
> > :PC   0.313181  0.686819
> >
> >  0.935392 -0.1506E-01 -0.2760E-01 -0.2066E+00  -3 -0.2493E+00
0.1862E+00
> >
> > :POSOUT   0.260640E+01  0.571596E+01  0.508372E+01
> > :STEP    2 r=  2.41944033 iat=   0 ipos=   0 time(sec)=   1.95000 nstep=
> > : 68 POSINIT   0.308557E+01  0.614706E+01  0.324576E+01
> >
> >  We are inside a pseudo-atomo
> >
> > :POSOUT   0.266922E+01  0.573260E+01  0.504469E+01
> > :STEP    3 r=  2.31944033 iat=   0 ipos=   0 time(sec)=   0.62000 nstep=
> > : 26 POSINIT   0.313195E+01  0.606144E+01  0.322298E+01
> > :
> >  :NSTEP gt maxfoll          501         500
> >
> > Of course, I could try to raise maxfoll.  But as the imposed limit of
500
> > worked fine for other cases (using the same case.inaim), I suspect I'm
> > making a mistake here, probably because I don't fully understand all the
> > input options.
> >
> > Can anyone offer me some good advice ?
> >
> > Thanks,
> >
> > Kevin
> > EMAT
> > University of Antwerp
> > Belgium
> > kevin.jorissen at ua.ac.be
> >
> > _______________________________________________
> > Wien mailing list
> > Wien at zeus.theochem.tuwien.ac.at
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
> --
> ----------------------------------------------------------
> Javier Daniel FUHR
> CRMC2-CNRS
> Campus de Luminy, Case 913
> 13288 Marseille Cedex 9 - FRANCE
>
> e-mail: fuhr at crmc2.univ-mrs.fr
>
> >From France x = 0, from Abroad x = 33
> TEL:  x.6.62.92.28.97
> FAX:  x.4.91.41.89.16
> ----------------------------------------------------------
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>