VERY BAD NEWS! internal error in subroutine IBZKPT ??

Message

lahaye · #1 Post by **lahaye** » Thu Jun 01, 2006 1:16 pm

Hello,

I have followed the procedure to get a accurate DOS, by running at first
a run with few k-points, and then a second run with increased number
of k-points and with ICHARG=11 in INCAR.

When I start the second run, I get this error message:

Code: Select all

 running on    4 nodes
 distr:  one band on    1 nodes,    4 groups
 vasp.4.6.28 25Jul05 complex 
 POSCAR found :  1 types and   32 ions
 scaLAPACK will be used
 WARNING: for PREC=h ENMAX is automatically increase by 25 %
        this was not the case for versions prior to vasp.4.4
 LDA part: xc-table for Pade appr. of Perdew

 VERY BAD NEWS! internal error in subroutine IBZKPT:
 NKZ>IKPTD      81

 VERY BAD NEWS! internal error in subroutine IBZKPT:
 NKZ>IKPTD      81

 VERY BAD NEWS! internal error in subroutine IBZKPT:
 NKZ>IKPTD      81

 VERY BAD NEWS! internal error in subroutine IBZKPT:
 NKZ>IKPTD      81

I appreciate advice or comments on what I have been doing wrong.

Thanks,
Rob.

#2 Post by **admin** » Thu Jun 01, 2006 3:03 pm

the parameter IKPTD is set to 45 in main.F (in vasp.4.4 and in vasp.4.6), so either choose a smaller number of k-points along each direction or increase IKPTD, (and NKDIMD and NTETD) in main.F and recompile vasp.

Phani Kanth · #3 Post by **Phani Kanth** » Tue Jun 29, 2010 9:29 am

Sir,

Would it be sufficient to increase the NKDIMD and NTETD proportionate to increase in IKPTD, in that case?

Thanks and Regards,
Phani kanth

#4 Post by **admin** » Thu Jul 01, 2010 7:29 am

yes, absolutely, please set these 3 parameters to the values needed for your k-mesh

hzkvictory · #5 Post by **hzkvictory** » Fri Mar 08, 2013 5:01 am

mkpoints.F not main.F