VASP 4.6 Error in IBZKPT: KPT>NKDIM

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
dlonie

VASP 4.6 Error in IBZKPT: KPT>NKDIM

#1 Post by dlonie » Wed Jun 17, 2009 11:23 pm

Hi,

I'm using 4.6 with the following KPOINTS file:
Automatic generation
0
Auto
50

When I run my optimization, I get this error:

VERY BAD NEWS! internal error in subroutine IBZKPT:
NKPT>NKDIM 10300

What is the best way to remedy this situation? Modify the KPOINTS? Increase NKDIM and recompile?

Thanks,
Dave
Last edited by dlonie on Wed Jun 17, 2009 11:23 pm, edited 1 time in total.

admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

VASP 4.6 Error in IBZKPT: KPT>NKDIM

#2 Post by admin » Mon Jun 22, 2009 9:51 am

please note that the automatic k-mesh generation sheme will generate a
(LxMxN)/(#of symmetry operations of your lattice) k-mesh, with approximately
L=50*|b1| +0.5,
M=50*|b2| +0.5
N=50*|b3| +0.5
(b are the lengths of the recipocal lattice vectors, written eg in OUTCAR)
--> for a small unit cell (large b) with little symmetry (almost no symmetry operations) the generated k-mesh might ecxeed the upper limit of NKDIM (10000, at the moment), as has been in your case.
please check:
1) is such a large k-mesh necessary?
2) if yes, please increase the parameter NKDIM in main.F, please search for the the block following the line
-----hard limits for k-point generation package
and re-compile vasp.
Please note that you will have to change IKPTD and might have to change
NTETD (if you use Tetrahedron BZ integration) as well.
Last edited by admin on Mon Jun 22, 2009 9:51 am, edited 1 time in total.

Post Reply