wannier: kmesh_get: something wrong, found too many neighbor

Post here questions linked with issue while running the EPW code

Moderator: stiwari

Post Reply
jiachen
Posts: 22
Joined: Sun Sep 11, 2016 3:17 am
Affiliation:

wannier: kmesh_get: something wrong, found too many neighbor

Post by jiachen »

Dear All,

I have a persistent error coming from wannier90. Since I still don't know how to run Wannier90 in stand along mode to get ukk file, I would like to try my luck here. The whole error message is like following:
==============================================================
**WARNING: kmesh has found >12 nearest neighbours**

This is probably caused by an error in your unit cell specification

If you think this is not the problem; please send your *.win file to the
wannier90 developers

The problem may be caused by having accidentally degenerate shells of
kpoints. The solution is then to rerun wannier90 specifying the b-vectors
in each shell. Give devel_flag=kmesh_degen in the *.win file
and create a *.kshell file:

$> cat hexagonal.kshell
$> 1 2
$> 5 6 7 8

Where each line is a new shell (so num_shells in total)
The elements are the bvectors labelled according to the following
list (last column is distance)

| b-vector 1 : ( -0.000312 0.239787 -0.003970 ) 0.239820 |
| b-vector 2 : ( 0.000312 -0.239787 0.003970 ) 0.239820 |
| b-vector 3 : ( -0.000008 0.000296 0.239823 ) 0.239823 |
| b-vector 4 : ( 0.239772 -0.003374 -0.003643 ) 0.239823 |
| b-vector 5 : ( 0.000008 -0.000296 -0.239823 ) 0.239823 |
| b-vector 6 : ( -0.239772 0.003374 0.003643 ) 0.239823 |
| b-vector 7 : ( -0.000320 0.240083 0.235852 ) 0.336551 |
| b-vector 8 : ( 0.000320 -0.240083 -0.235852 ) 0.336551 |
| b-vector 9 : ( 0.239764 -0.003078 0.236179 ) 0.336566 |
| b-vector 10 : ( -0.239764 0.003078 -0.236179 ) 0.336566 |
| b-vector 11 : ( 0.239459 0.236414 -0.007613 ) 0.336586 |
| b-vector 12 : ( -0.239459 -0.236414 0.007613 ) 0.336586 |
| b-vector 13 : ( 0.240084 -0.243161 0.000327 ) 0.341713 |
| b-vector 14 : ( -0.240084 0.243161 -0.000327 ) 0.341713 |
| b-vector 15 : ( 0.239779 -0.003670 -0.243466 ) 0.341736 |
| b-vector 16 : ( -0.239779 0.003670 0.243466 ) 0.341736 |
| b-vector 17 : ( 0.000305 -0.239491 0.243793 ) 0.341747 |
| b-vector 18 : ( -0.000305 0.239491 -0.243793 ) 0.341747 |
| b-vector 19 : ( 0.239451 0.236710 0.232209 ) 0.409010 |
| b-vector 20 : ( -0.239451 -0.236710 -0.232209 ) 0.409010 |
| b-vector 21 : ( -0.240092 0.243457 0.239496 ) 0.417461 |
| b-vector 22 : ( 0.240092 -0.243457 -0.239496 ) 0.417461 |
| b-vector 23 : ( 0.240076 -0.242865 0.240149 ) 0.417482 |
| b-vector 24 : ( -0.240076 0.242865 -0.240149 ) 0.417482 |
| b-vector 25 : ( 0.239467 0.236118 -0.247436 ) 0.417517 |
| b-vector 26 : ( -0.239467 -0.236118 0.247436 ) 0.417517 |

Exiting.......
kmesh_get: something wrong, found too many nearest neighbours
=========================================================

I tried to specify lattice by
=============================================================
!wdata(1) = 'Begin Unit_Cell_cart'
!wdata(2) = 'bohr'
!wdata(3) = '12.380246740125001 0.016137977625 0.000382622625'
!wdata(4) = '0.173942418375 12.379175226000001 -0.015272309250000001'
!wdata(5) = '0.19095813 0.205171151625 12.37713387075'
!wdata(6) = 'End Unit_Cell_cart'
=======================================================
But it seem to be ignored by Wannier90. I ran out of idea here...

Any suggestion is very welcome, Thank you!

Cheers

carla.verdi
Posts: 155
Joined: Thu Jan 14, 2016 10:52 am
Affiliation:

Re: wannier: kmesh_get: something wrong, found too many neig

Post by carla.verdi »

Dear Jiachen,

My suggestion would be to first try and run a wannierization with wannier90.x to determine for sure if it is a problem of wannier90. If that's the case like it seems to be, you can post your problem on the dedicated forum: http://www.wannier.org/forum.html

Cheers,
Carla

Post Reply