SGRCON error

Message

gabral · #1 Post by **gabral** » Fri Feb 10, 2006 12:55 am

The VASP fails due to error:
"VERY BAD NEWS! internal error in subroutine SGRCON:
Found some non-integer element in rotation matrix 2"

This error seems relating to the input structure, But the used POSCAR is just the output (CONTCAR) of previous ionic relaxation calculations of VASP.

What happens here? Anyone can help me?

#2 Post by **admin** » Fri Feb 10, 2006 2:35 pm

please check if your convergence parameters are accurate enough
to obtain reasonable output geometries.

gabral · #3 Post by **gabral** » Sat Feb 11, 2006 2:50 am

My INCAR file as follow:
general:
System = UO2LDAs+U
ISTART = 1 ; ICHARG = 1
# ISYM = 0
ALGO = fast

# NELM = 40
NELMIN = 5
NELMDL = -3
LMAXMIX = 6
# AMIX = 0.2
# BMIX = 0.001
# BMIX_MAG = 0.0001
EDIFFG = -0.001
NSW = 30
# MAXMIX = 30

IBRION = 1
POTIM = 0.5
ISIF = 4
NFREE = 4

NBANDS = 120
# NSIM = 4
# GGA = PE
VOSKOWN = 1
ISPIN = 2
LORBIT = 11
MAGMOM = 0 0 0 0 0 0 0 0 2.8 2.8 -2.8 -2.8
LDAU = .TRUE.
LDAUTYPE = 2
LDAUL = -1 3 3
LDAUU = 0 4.5 4.5
LDAUJ = 0 0.51 0.51
ISMEAR = 0
EDIFF = 1E-5
PREC = Accurate
-----------

and the stdout is as
.....
RMM: 4 -0.109734208798E+03 -0.22270E-07 -0.15176E-08 5215 0.186E-04 0.447E-04
RMM: 5 -0.109734208831E+03 -0.33382E-07 -0.81597E-09 4694 0.127E-04
27 F= -.10973421E+03 E0= -.10973414E+03 d E =-.216649E-06 mag= 0.0000
BRION: g(F)= 0.173E-06 g(S)= 0.285E-07 retain N= 4 mean eig= 2.11
eig: 6.879 0.768 0.768 0.014
reached required accuracy - stopping structural energy minimisation
writing wavefunctions
----
output structure is:
U4O8:
1.00000000000000
4.5850828164967714 0.0000208883368965 -0.0000587557530739
0.0000061515277688 3.5942765861793209 0.0000092180743007
0.0000245124099229 0.0000221535030436 6.3039428829777195
8 2 2
Direct
0.2178455101597675 0.2500017672976527 0.4197880729961659
0.2821533148297019 0.7499987289421063 0.9197876541964458
0.7821540590674243 0.7500001146596649 0.5802113876325371
0.7178469905189431 0.2500034977140801 0.0802116747402537
0.9897405983686803 0.7500000197359459 0.2411243440544729
0.5102586956101343 0.2499985924500308 0.7411230126610723
0.0102594167749661 0.2500000706414963 0.7588758277999381
0.4897412415492236 0.7500014374164214 0.2588768889330395
0.2345947784868536 0.2499986190815056 0.0853166072464893
0.2654037333944865 0.7499973152653351 0.5853186940809162
0.7345963485385999 0.2500008655152598 0.4146833142438296
0.7654053127011892 0.7499989712804948 0.9146825214148409
-------------

I see no unphysical fact relating to this geometry. But the error occurs when I try to make a static calculation with this geometry.

#4 Post by **admin** » Mon Feb 13, 2006 2:39 pm

if the Bravais matrix really is ok, please run with increased SYMPREC
(SYMPREC=1e-6 or even more precise works, the default precision of 1e-5 is too low).

#5 Post by **admin** » Mon Feb 13, 2006 2:54 pm

if the Bravais matrix really is ok, please run with increased SYMPREC
(SYMPREC=1e-6 or even more precise).

DFTSIC · #6 Post by **DFTSIC** » Mon Mar 08, 2010 5:09 pm

Dear VASP masters
I am having same problem and I increased SYMPREC to 1e-6, but still I get same error message as:

VERY BAD NEWS! internal error in subroutine SGRCON:
Found some non-integer element in rotation matrix 2

And here is my INCAR file:

# SCF input for VASP
# Note that VASP uses the FIRST occurence of a keyword
# Extra input added by hand
SYMPREC = 1e-6
# End of extra input

SYSTEM = (Si7 C8)3 (P1) ~431 minimizedSi3/2 0 0 1 surface (VASP)-3rd Si removed-SINGLE POINT-900eV-K661-LOW

NPAR = 1
PREC = Low
ENCUT = 900
IBRION = -1
NSW = 0
ALGO = Normal (blocked Davidson)
NELM = 60
NELMIN = 2
EDIFF = 1.0e-05
EDIFFG = -0.02
VOSKOWN = 1
NBLOCK = 1
ISPIN = 1
INIWAV = 1
ISTART = 0
ICHARG = 2
LWAVE = .TRUE.
LCHARG = .TRUE.
ADDGRID = .FALSE.
# Work Function Options
LVTOT = .TRUE.
IDIPOL = 3
DIPOL = 0.0 0.0 0.5
#
LVTOT = .TRUE.
LELF = .TRUE.
ISMEAR = 1
SIGMA = 0.2
LREAL = Auto
RWIGS = 1.11 0.77

Could you please tell me how can I solve this problem?
Thank you