Convergence problem for ionic relaxation in a charged system

[umklapp]

I am optimizing a structure in a charged system for a particular magnetic state; However, Even if I try difference relaxation strategy(quasi newton and CG), different subspace diagonalization method(Davison and Residual minimum), symmetry tag(ISYM= -1,0, default), and different initial geometry, I am always getting a bouncing dE(the energy fails to decrease continuously) as follows: Do you have some idea to overcome this ionic convergence problem?
48 F= -.26219840E+03 E0= -.26219840E+03 d E =-.327147E-03 mag= 6.0000
49 F= -.26219893E+03 E0= -.26219893E+03 d E =-.527741E-03 mag= 6.0000
50 F= -.26219919E+03 E0= -.26219919E+03 d E =-.259314E-03 mag= 6.0000
51 F= -.26219809E+03 E0= -.26219809E+03 d E =0.109917E-02 mag= 6.0000
52 F= -.26219562E+03 E0= -.26219562E+03 d E =0.247148E-02 mag= 6.0000
53 F= -.26219434E+03 E0= -.26219434E+03 d E =0.127219E-02 mag= 6.0000
54 F= -.26219575E+03 E0= -.26219575E+03 d E =-.140522E-02 mag= 6.0000
55 F= -.26219585E+03 E0= -.26219585E+03 d E =-.105190E-03 mag= 6.0000
56 F= -.26219623E+03 E0= -.26219623E+03 d E =-.373362E-03 mag= 6.0000
57 F= -.26219663E+03 E0= -.26219663E+03 d E =-.398108E-03 mag= 6.0000
58 F= -.26219668E+03 E0= -.26219668E+03 d E =-.557682E-04 mag= 6.0000
59 F= -.26219637E+03 E0= -.26219637E+03 d E =0.307803E-03 mag= 6.0000
60 F= -.26219559E+03 E0= -.26219559E+03 d E =0.781366E-03 mag= 6.0000

Below is my input file:
PREC = accurate
ISTART = 0
ICHARG = 2
ISPIN = 2
NELECT = 176 #177 for nuetral
IALGO =48
LREAL = Auto
NBAND =140
EDIFF = 1E-5
ISYM =-1

NELMDL = -5
NELMIN = 8
NELM = 40
LMAXMIX = 4

IBRION = 1 # I also tried 2
ENCUT = 530
NSW = 60
EDIFFG = -0.01
ISMEAR = 0
SIGMA = 0.05
ISIF = 0

VOSKOWN =1
MAGMOM =5 -5 -5 -5 5 -5 5 -5 5 -5 5 -5 18*0

LDAU = .TRUE.
LDAUTYPE = 1
LDAUL = 2 0 0
LDAUU = 4.0 0 0
LDAUJ = 0.58 0 0

[support_vasp]

Hi,

We're sorry that we didn’t answer your question. This does not live up to the quality of support that we aim to provide. The team has since expanded. If we can still help with your problem, please ask again in a new post, linking to this one, and we will answer as quickly as possible.

Best wishes,

VASP