Hello,
I am attempting to look at the kpoint grid convergence but the energy increased as the grid size got bigger:
etotal1 -1.1016678839E+04
etotal2 -1.1013220577E+04
etotal3 -1.1012798679E+04
etotal4 -1.1012928525E+04
I've attached my input file.
Thanks for the help!
Stacey
Energy increased with more kpoints
Moderator: bguster
-
mdsimulationgroup
- Posts: 13
- Joined: Mon Feb 25, 2019 2:20 pm
Energy increased with more kpoints
- Attachments
-
Cu_111-4L.in- (3.4 KiB) Downloaded 208 times
Re: Energy increased with more kpoints
Dear Stacey,
Sounds like you might still need to increase the number of k-points to get convergence.
You can check by also reporting the change of energy w.r.t. the most converged case (here 6x6x6):
The difference between 6x6x6 and 4x4x4 is 129.84 mHa but is smaller than w.r.t. 2x2x2 (-292.05 mHa).
Best wishes,
Eric
Sounds like you might still need to increase the number of k-points to get convergence.
You can check by also reporting the change of energy w.r.t. the most converged case (here 6x6x6):
Code: Select all
ETOT, Ha Dela-ETOT, mHa
etotal1 -1.1016678839E+04 -3750.31
etotal2 -1.1013220577E+04 -292.05
etotal3 -1.1012798679E+04 129.84
etotal4 -1.1012928525E+04 0The difference between 6x6x6 and 4x4x4 is 129.84 mHa but is smaller than w.r.t. 2x2x2 (-292.05 mHa).
Best wishes,
Eric