Dear developers,
      I am trying to do symmetry-adapted MLWF using wannier90. But I found some problem with pw2wannier90.x's compute_dmn
      If I concentrate on silicon's 4 valance bands, and set 
1. nbnd=4 in nscf
2. num_bands =4, num_wann=4 in win file. And projection is
    Si:s
    Si:s
then I will get an error when running pw2wannier90.x as below

compute_dmn: Symmetry operator (    1) could not transform Wannier function (    1).
compute_dmn: The error is       1.0000000.

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine compute_dmn (1):
     Error: missing Wannier functions, see the output.
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On the other hand, if I set
1. nbnd=8 in nscf
2. num_bands =8, num_wann=8 in win file. And projection is
    Si:sp3

then the pw2wannier90.x is fine. But if I set the projection to "random", it will also raise "missing wannier function" error.

What is wrong here? Is there something wrong with pw2wannier90.x?

I packed all files into a tar, can be downloaded here https://drive.google.com/file/d/1-EHYoBIEEqq3-Xv90cP_3l8Gm40v30P6/view
You can run "bash test_running.bash" to get a quick result in a few seconds.

best regards


_______________________________________________
Pw_forum mailing list
[email protected]
http://pwscf.org/mailman/listinfo/pw_forum

Reply via email to