LWAVE and LCHARG make differences

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
hlzya
Newbie
Newbie
Posts: 12
Joined: Fri Mar 17, 2006 3:12 pm
License Nr.: 641

LWAVE and LCHARG make differences

#1 Post by hlzya » Thu Jan 07, 2016 8:43 am

Dear Admin,

I'm trying to do some calculations on LaVO3 in DFT+U but there are some troubles. When I only change the parameter LWAVE or LCHARG, they can finish but the results will be different.

Here is the result;
LWAVE=T LCHARG=T; 18 0.02610000 -653.421794 -0.225406
LWAVE=T LCHARG=F; 16 0.02999400 -653.420867 -0.229999
LWAVE=F LCHARG=T; 16 0.02999400 -653.420867 -0.229999
LWAVE=T LCHARG=T; 16 0.02999400 -653.420867 -0.229999

It will converge in different steps and different energy. But It shows that the parameter LWAVE or LCHARG only control the form of output (whether CHGCAR and WAVECAR are written or not), I'm confused that how these two parameters change the results? Did I do wrong somewhere? Or maybe I compile VASP wrongly?

Here is my INCAR:
SYSTEM = default

Start parameter for this Run:
ISTART = 0
ICHARG = 2
INIWAV = 1

Electronic Relaxation:
ENCUT = 600
ENAUG = 650
PREC = Accurate
IALGO = 38
NELM = 1000
NELMIN = 4
# NELMDL = -12
EDIFF = 1E-04
# NBANDS = 7
GGA = PE
# VOSKOWN = 1
LREAL = Auto
# WEIMIN = 0

Ionic Relaxation:
EDIFFG = -0.03
NSW = 1000
IBRION = 1
ISIF = 2
POTIM = 0.1
# IOPT = 2
# ISYM = 0

DOS related values:
SIGMA = 0.1
ISMEAR = 0

Spin polarized:
ISPIN = 2
MAGMOM = 47*0 8*-5 8*5 16*0

File writing
LWAVE = .FALSE.
LCHARG = .FALSE.

Calculation of DOS
NPAR = 16

LSDA+U
LDAU = .TRUE.
LDAUTYPE = 2
LDAUL = -1 2 -1
LDAUU = 0 3.0 0
LDAUJ = 0 0 0
LDAUPRINT = 2
LMAXMIX = 4
LASPH = .TRUE.

Thanks a lot for your help

admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

Re: LWAVE and LCHARG make differences

#2 Post by admin » Thu Jan 07, 2016 11:21 am

This has nothing to do with output parameters. Your system is relatively large and exhibits many similar minima.
When you repeat the calculation with fixed output parameters the result will be the same (cf. e.g. line 1 and 4).
Make the convergence criteria more severe.

Post Reply