Re: [Wien] LAPW1 error using HDLO's for Sm f states in SmCo5

[Dürrschnabel , Michael]

Dear Mr. Tran, dear Wien2k users,


thanks for the hint. I tested it and it worked fine for the case that you 
neglect spin-orbit coupling. However, in this particular material system 
spin-orbit coupling plays an important role and should be switched on in your 
calculation. If you do so and run e.g.


"runsp_lapw -p -so -orb -i 150 -ec 0.0001 -cc 0.01 -NI"


the scf cycle runs up to LAPWSO. Afterwards, I get in the console:


   FERMI - Error
   cp: cannot stat '.in.tmp': No such file or directory
   >  stop error

The file "uplapw2.error"- the only error file with non-zero file size- contains 
the following output:

   'FERMI' -  # of eigenvalues eq 0, check case.scf1
   **  testerror: Error in Parallel LAPW2

There is no further error output.


Best regards,


Michael Duerrschnabel


Additional info:


case.inso (created by initso)


WFFIL
4  0  0 llmax,ipr,kpot
-10  1.5Emin, Emax
0 0 1   h,k,l (direction of magnetization)
 3   number of atoms with RLO
1 -1.58 0.0010 CONT atom-number, E-param for RLO
2 -4.56 0.0001 STOP atom-number, E-param for RLO
3 -4.56 0.0001 STOP atom-number, E-param for RLO
0 0  number of atoms without SO, atomnumbers



Von: Wien  im Auftrag von 
t...@theochem.tuwien.ac.at 
Gesendet: Freitag, 22. September 2017 21:34:13
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] LAPW1 error using HDLO's for Sm f states in SmCo5

Hi,
the number of lines for the 1st atom is 7 since you added an HDLO:
"0.306  0" --> "0.307  0"

FT

On Friday 2017-09-22 18:20, MD wrote:

>Date: Fri, 22 Sep 2017 18:20:37
>From: MD 
>Reply-To: A Mailing list for WIEN2k users 
>To: wien@zeus.theochem.tuwien.ac.at
>Subject: [Wien] LAPW1 error using HDLO's for Sm f states in SmCo5
>
> Dear Wien2k users,
>
> I tried to run a scf cycle for SmCo5 using HDLO's for the Sm f-states
> using Wien2k v 17.1. LDA as exchange-correlation potential was selected.
> The system is spin-polarized and you need spin-orbit and the Hubbard U
> corrections (here: U=0.3 Ry, J=0, SIC). The Sm f-states lie close to the
> Fermi level. The RMT's of Sm are quite large in this system and, thus,
> it might be of advance to use HDLO's for the Sm f states (my case.in1
> see below). However, LAPW1 stops immediately whether spin-orbit or +U
> was selected or not throwing the following error (the scf cycle was
> started via wien2web):
>
>  'INILPW' - Invalid k-point file on unit   0
>
>
>  'LAPW1' - INILPW aborted unsuccessfully.
>
> Did I make a mistake or is it a bug? The case.in1 file should be
> correct, according to slide 16 of
> http://susi.theochem.tuwien.ac.at/events/ws2017/notes/PB-getting_started.pdf.
> Please let me know if you need more information than added below.
>
> Best regards,
>
> Michael Duerrschnabel
>
>
>
> Additional info:
>
>
> case.in1 file:
>
> WFFIL  EF= 0.5   (WFFIL, WFPRI, ENFIL, SUPWF)
>  8 10   6   ELPA pxq hm (R-MT*K-MAX,MAX L IN
> WF,V-NMT,lib,gridshape,hm/lm)
>   0.306  0  (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global
> APW/LAPW)
>  00.30 0. CONT 1
>  0   -3.20 0.0001 STOP 1
>  10.30 0. CONT 1
>  1   -1.58 0.0010 CONT 1
>  30.30 0.0010 CONT 1 // APW+lo for Sm f states
>  30.30 0.0010 CONT 2 // This should account for the HDLO of the
> Sm f states
>  20.30 0.0010 CONT 1
>   0.304  0  (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global
> APW/LAPW)
>  10.30 0. CONT 1
>  1   -4.56 0.0001 STOP 1
>  20.30 0.0010 CONT 1
>  00.30 0. CONT 1
>   0.304  0  (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global
> APW/LAPW)
>  10.30 0. CONT 1
>  1   -4.56 0.0001 STOP 1
>  20.30 0.0010 CONT 1
>  00.30 0. CONT 1
> K-VECTORS FROM UNIT:4   -9.0   1.5   187   emin / de (emax=Ef+de) /
> nband
>
> ---
>
> case.struct:
>
> SmCo5
>
> H   LATTICE,NONEQUIV.ATOMS:  3 191_P6/mmm
>
> MODE OF CALC=RELA unit=ang
>
>   9.459973  9.459973  7.502405 90.00 90.00120.00
>
> ATOM  -1: X=0. Y=0. Z=0.
>   MULT= 1  ISPLIT= 4
> Sm NPT=  781  R0=0.1000 RMT=2.5000   Z: 62.000
>
> LOCAL ROT MATRIX:1.000 0.000 0.000
>  0.000 1.000 0.000
>  0.000 0.000 1.000
> ATOM  -2: X=0. Y=0.6667 Z=0.
>   MULT= 2  ISPLIT= 4
>   -2: X=0.6667 Y=0. Z=0.
> Co NPT=  781  R0=0.5000 RMT=2.1900   Z: 27.000
>
> LOCAL ROT MATRIX:1.000 0.000 0.000
>  0.000 1.000 0.000
>

Re: [Wien] Wien2k v 17.1 bandstructure calculation

[Dürrschnabel , Michael]

Thanks! I was not aware of the post before. Maybe I was searching the mailing 
list using the wrong keywords.


Von: Wien <wien-boun...@zeus.theochem.tuwien.ac.at> im Auftrag von karima 
Physique <physique.kar...@gmail.com>
Gesendet: Samstag, 22. Juli 2017 13:59:09
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] Wien2k v 17.1 bandstructure calculation

In this message you find the correction of this error:

http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16069.html

2017-07-22 13:18 GMT+02:00 Dürrschnabel, Michael 
<duerrschna...@geo.tu-darmstadt.de<mailto:duerrschna...@geo.tu-darmstadt.de>>:

Dear Wien2k users,


I am running Wien2k version 17.1 on a Dual Core Workstation machine with 
operating system Linux Mint v 18.1, latest Intel fortran compiler and MKL 
libraries. The following error seems to be new in version 17 of the Wien2k code.


After compiling the source code (no error), I started a first test k-parallel 
calculation (GaAs.struct from the examples). The SCF cycle was finished 
yielding no problem. However, when trying to calculate the bandstructure via 
w2web the following problem occurred:

After having created the k-list I pressed the button "x lapw1 -band -p" and got 
and "end-of-file error". The lapw1 Fortran code and the respective *.def files 
were ok in my point-of-view. Therefore, I decided to run the command "x lapw1 
-band -p" via console manually and I got no error. Thus, I believe that there 
is a bug somewhere in the perl code of Wien v 17 (file: 
SRC_w2web/htdocs/exec/band.pl<http://band.pl>). If you replace 
"band.pl<http://band.pl>" in version 17 by that from version 16 everything in 
the bandstructure calculation via w2web runs again fine. If you need further 
information to reproduce the error please let me know.


Best regards,


Dr. Michael Duerrschnabel



___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at<mailto:Wien@zeus.theochem.tuwien.ac.at>
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] Wien2k v 17.1 bandstructure calculation

[Dürrschnabel , Michael]

Dear Wien2k users,


I am running Wien2k version 17.1 on a Dual Core Workstation machine with 
operating system Linux Mint v 18.1, latest Intel fortran compiler and MKL 
libraries. The following error seems to be new in version 17 of the Wien2k code.


After compiling the source code (no error), I started a first test k-parallel 
calculation (GaAs.struct from the examples). The SCF cycle was finished 
yielding no problem. However, when trying to calculate the bandstructure via 
w2web the following problem occurred:

After having created the k-list I pressed the button "x lapw1 -band -p" and got 
and "end-of-file error". The lapw1 Fortran code and the respective *.def files 
were ok in my point-of-view. Therefore, I decided to run the command "x lapw1 
-band -p" via console manually and I got no error. Thus, I believe that there 
is a bug somewhere in the perl code of Wien v 17 (file: 
SRC_w2web/htdocs/exec/band.pl). If you replace "band.pl" in version 17 by that 
from version 16 everything in the bandstructure calculation via w2web runs 
again fine. If you need further information to reproduce the error please let 
me know.


Best regards,


Dr. Michael Duerrschnabel


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html