My Community

Hi,

I am having problem converging a HSE + SOC calculations. The run crashes after 4-5 cycles of scf calculations with error message which goes as

entering main loop
N E dE d eps ncg rms rms(c)
DAV: 1 0.700209582175E+03 0.70021E+03 -0.38061E+04 27648 0.890E+02
DAV: 2 0.118785843848E+03 -0.58142E+03 -0.52628E+03 32352 0.185E+02
DAV: 3 0.707432166307E+02 -0.48043E+02 -0.47188E+02 46368 0.626E+01
DAV: 4 0.692918935198E+02 -0.14513E+01 -0.14509E+01 33408 0.827E+00
DAV: 5 -0.116920007498E+03 -0.18621E+03 -0.25374E+01 48384 0.205E+01 0.268E+01
DAV: 6 -0.123284078268E+03 -0.63641E+01 -0.24578E+02 34848 0.730E+01 0.486E+01
WARNING: Sub-Space-Matrix is not hermitian in DAV 43
-1.259850098239467E-004

Please note that I get the same error with a noncolinear spin calculation without SOC. I can provide the details of the input file.

Any help will be greatly appreciated. I will be happy to share any more details such as INCAR/OUTCAR etc.

Best regards,
Tanmoy Das.

Which version of vasp do you use? Older versions do not allow combinations like HSE+SOC,
or non-coll+SOC.
If you are using heavy elements, do you use correct LMAXMIX?