Poblem in CHGCAR and WAVECAR

Message

Yeneneh · #1 Post by **Yeneneh** » Mon Jan 23, 2006 1:03 pm

Hi everybody!
I am calculating the band structure for polymers to see how the Spin-Orbital coupling affects the energies of the bands. First I did a self consistent calculation by setting the INCAR file as

PREC = Accurate
NBANDS = 40
ISMEAR = 0
ISYM = 0
EDIFF = 1.e-7
NPAR = 1
LORBIT = 1
VOSKOWN = 1
ISPIN = 2
GGA_COMPAT = .FALSE.
LNONCOLLINEAR= .TRUE.
LSORBIT = .FALSE.
ISTART = 0
ICHARG = 2
LWAVE = .TRUE.
LCHARG = .TRUE

I and then I did a non-self consistent calculation by setting ( LSORBIT=.TRUE. , ISTART=1 and ICHARG=11 , SAXIS= 1 0 0 , NBANDS=80) and for the other INCAR parameters I used the default values. For the POTCAR I am using LDA (LEXCH = CA in POTCAR). When I do a non-selfconsistent calculation most of the time I encounter the following error messages

ERROR: charge density could not be read from file CHGCAR for ICHARG>10
And
ERROR: while reading WAVECAR, plane wave coefficients changed 23962, 11981

I am using vasp.4.6.28 parallel version. Is there any one who can help me on this problem?

Thanks!!

Veronika · #2 Post by **Veronika** » Tue Jan 24, 2006 8:11 am

Thats obious: you changed the number of bands!
(NBANDS = 40 --> NBANDS = 80)

Yeneneh · #3 Post by **Yeneneh** » Tue Jan 24, 2006 8:32 am

Even if I did not change the number bands I am getting these error messeges. However if I do a non-selfconsistent calculation the NBANDS is at least twics as the NBANDS for self consistent calculation unless I took very many NBANDS for self consistent caluculation.

Thanks!!

lmiao · #4 Post by **lmiao** » Tue Jan 24, 2006 4:24 pm

sometimes an incorrect ISMEAR will cause this error, but I am not sure whether others will also. I actually have another question: In order to get an accurate DOS, do I need to use ICHARG=11 (assume that I have calculated CHGCAR from previous scf run) Or can I just use ISTART=1 to start from a correct wavefunction and to obatin DOS? Or both ISTART=1 and ICHARG=11 have to be used together, like the one used by Yeneneh? Reply will be appreciated!

#5 Post by **admin** » Wed Jan 25, 2006 2:00 pm

you have to continue with the converged CHGCAR, because most probably, you will increase/change the k-mesh to get a denser k-grid to calculate the DOS accurately. Then, WAVECAR will not be read correctly because the wavefunction-coefficients are stored k-point wise
concerning the READ error of CHGCAR: please check whether the FFT meshes have changed.

wangzhiyong_520 · #6 Post by **wangzhiyong_520** » Thu Jan 25, 2007 2:03 pm

Now ,I encouter the same problem too, "charge density could not be read from file CHGCAR for ICHARG>10 " so I hope that someone can help me,I'M looking forward to your reply!thanks

#7 Post by **admin** » Mon Jan 29, 2007 2:09 pm

please make sure that
1) the CHGCAR really is in the working directory at runtime
2) the fft meshes of CHGCAR are compatible

gus.hart · #8 Post by **gus.hart** » Wed Dec 19, 2007 7:35 pm

More verbosely, grep the OUTCAR file from the SCF calculation and find the values of NGXF, NGYF, and NGZF. Set these parameters explicitly in the INCAR file that you use when calculating the band structure (the ICHARG=11 run).

yinwei_li · #9 Post by **yinwei_li** » Tue Nov 17, 2009 11:03 am

The main points is in this sentence "plane wave coefficients changed ", I think the ISMEAR you used in scf and noscf process is different, therefore, the plane wave coefficients changed in these two process is not identical. You can find the values of NGXF, NGYF and NGZF in the CHGCAR or OUTCAR of the scf, and then add these three parameters in the INCAR of the nonscf. OK, the problem is resolved.

Good Luck!