[Wien] Error while running +/- 0.5e calculation

[Gavin Abo]

The default energy convergence criteria value for runsp_lapw should be 
the same as for run_lapw, which is 0.0001 Ry as given on page 60 in the 
WIEN2k 14.2 usersguide [ 
http://www.wien2k.at/reg_user/textbooks/usersguide.pdf ]. However, if 
you are concerned about what it is, you can specify what value you want 
to use with the -ec flag.

If you do not use the -i flag, the program will only run to 40 
iterations by default.  The program has stopped because you have reached 
the default maximum number of iterations (cycle 40). Usually, you just 
have to continue the calculation and run more iterations with the -NI 
flag or can run a calculation from the beginning and specify more 
iterations with the -i flag.  If you search the mailing list archive, 
you should find that it has been mentioned at least a few times, for 
example:

http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg02833.html

If the calculation does not converge with more iterations (is 
diverging), this might be caused by a ill-defined problem or numerical 
difficulty to converge.  Changing the mixer algorithm or greed value 
might help with that.  There should be discussions in the mailing list 
archive about that, and one of them is:

http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg02638.html

If you run the same calculation in the terminal or w2web, you should get 
the same result with the same WIEN2k version.  You can check case.inm, 
but it might be that you are using a default MSEC1 mixer in WIEN2k 8.x, 
but a different default mixer like MSR1 in WIEN2k 14.2.

On 1/5/2015 7:24 AM, Abed Reg wrote:
> Thanks for all
>
> I have repeated the calculation with the runsp_lapw command (Note: the 
> first calculation is done with the w2web interface) but i have a 
> problem of convergence as the following, a problem that i didn't find 
> with the w2web interface:
>
> in cycle 40    ETEST: .0277655000000000   CTEST: .5085582
> hup: Command not found.
> Invalid null command.
> FORTRAN STOP  LAPW0 END
> Invalid null command.
> FORTRAN STOP  LAPW1 END
> Invalid null command.
> FORTRAN STOP  LAPW1 END
> Invalid null command.
> FORTRAN STOP  LAPW2 END
> Invalid null command.
> FORTRAN STOP  LAPW2 END
> Invalid null command.
> FORTRAN STOP  CORE  END
> Invalid null command.
> FORTRAN STOP  CORE  END
> Invalid null command.
> FORTRAN STOP  MIXER END
>
> >   energy in SCF NOT CONVERGED
>
> First, i want to know the default energy convergence criteria value 
> when we use the commande 'runsp.lapw"
>
> Second, what causes this problem?
>
> I'll be very grateful
>
> -- 
> Mr: A.Reggad
> Laboratoire de Génie Physique
> Université Ibn Khaldoun - Tiaret
> Algerie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20150105/e8ffae2e/attachment.html>