The ABINIT Users Mailing List ( CLOSED )

[ilukacevic@fizika.unios.hr]

Dear all,

I have a problem with a preliminary convergence test on ecut and ngkpt by 
monitoring how the pressure (at a constant acell) changes. I change ecut from 
25-41 Ha, while raising ngkpt from 3*8 to 3*12. For ngkpt 3*8 and 3*10 all 
the calculations finish normally, but for 3*12 the calculation stops at a 
certain ecut. The .log file reports

-P-0000  fxphas : BUG -
-P-0000   The eigenvector number   1 has zero norm.
-P-0000
-P-0000  leave_new : decision taken to exit ...

just before printing the phonon wavelength (reduced coordinates), norm and 
energies in Ha.

After checking at 
http://www.abinit.org/package/robodoc_4.6.1/Src_4wfs/fxphas_F90.html
I found out that this bug reports when the sum sbb+saa<tol8, though by 
following up the fxphas function code, I could not figure out the meaning of 
these three variables (I guess they are real and imaginary parts of 
eigenvector and tol8 is a numerical zero tolerance - but not sure).

Could anyone help me with an explanation about this error and tell me, if it 
is possible, with which input variables could I improve the calculation.

I am using abinit4.6.5, with TM psp. The input file follows bellow.

Thank You in advance


 ndtset 36   udtset 9  4
 ecut:?    25    ecut+?   2

 #DATASET 1 : SC run
  prtden?1 1
  kptopt?1 1
  tolvrs?1 1.0d-19
  nstep?1  80

#DATASET 2 : NSC run  
  iscf?2  -2
  nqpt?2   0
  getwfk?2   -1
  getden?2   -1
  tolwfr?2   1.0d-20
  nstep?2 400

#DATASET 3 : 
 getden?3   -2
 getwfk?3   -1
  iscf?3   -2
 nqpt?3   1
 qpt?3        0.50d0   0.50d0   0.0d0
 rfdir?3   1 1 1
 tolwfr?3   1.0d-20
 nstep?3  400

#DATASET 4 : phonon at 0.0  0  0.0
    getwfk?4  -2
    getwfq?4  -1
    nqpt?4   1
   qpt?4   0.50d0   .50d0   0.0d0
   rfatpol?4   1  2
   rfdir?4   1 1 1
   rfphon?4   1
   tolvrs?4   5.0d-3   
   nstep?4    25

#Common data****************
ixc 1
kptopt 3
nshiftk  1
shiftk  0.0  0.0  0.0
#   0.5  0.0  0.0  0.0  0.5  0.0  0.0  0.0  0.5
ngkpt 12 12 12

acell  9.5 9.5 9.5
angdeg     90.     90.     90.
#ecut  35
occopt  4
tsmear 0.01
ntime 30
spgroup   225
natom  2
nband 12
ntypat   2
typat  1     2
xred   .00d0          .00d0        .00d0
       .50d0         .500d0        .500d0
znucl     48.      52.
nstep 40
brvltt -1

timopt 2