REAL_OPT: internal ERROR

[murphy]

Dear VASP users,

I am doing the ENCUT test for a PbSe supercell containing 64 atoms.
I wanted to scane ENCUT= 200 300 350 400 450 500 600 700 800.
But after ENCUT= 600, I got the following errors consistently:

WARNING: PSMAXN for non-local potential too small
LDA part: xc-table for Pade appr. of Perdew
POSCAR, INCAR and KPOINTS ok, starting setup
WARNING: small aliasing (wrap around) errors must be expected
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32
REAL_OPT: internal ERROR: -48 -48 -48 32

My INCAR file is:

SYSTEM = PbSe

ENCUT = 600
ENAUG = 750

SIGMA = 0.1
LREAL = Auto
PREC = Medium
EDIFF = 0.0001
EDIFFG = -0.001
IBRION = 2 # atomic minimisation algorithm (2 or 1)
POTIM = 1.0 # "size" of atomic step
NSW = 0 # maximum number of atomic steps
ISIF = 7 # controls what should be optimised
ALGO = Fast

I am using 8x8x8 Gamma-centered k-mesh.
I did found a thread addressing the same problem in this forum.
But I after I changed the position of Pb and Se, the error still exists.
And seemingly I have to use the PAW-PBE potential always in this series calculation,
therefore I think I can not change the pseudopotentials.

Is 500 already enough for ENCUT scan? By the way, the error also happens in PbS, PbTe, SnTe supercells.

I do not have this problem in unit cell calculations, is this error associated with cell size or number of atoms?

Please give me some suggestions, thank you very much!

Sincerely,
Murphy

[SKM]

hi
this thread is old. but i have same issue.
did you solve this problem? and how?

Regards

[martin.schlipf]

I'm assuming this is the same issue as reported here
https://vasp.at/forum/viewtopic.php?f=4&t=5354

Let's keep this discussion in a single thread, please. If it is a distinct case, can you provide a set of input files to reproduce the issue?