WARNING: dimensions on CHGCAR file are different

[aruth]

I am using VASP 5.2.12

I am trying to generate DOSCAR files for graphene with different impurities. I am using the same INCAR and KPOINTS for all of the relaxations, then I copy CONTCAR-> POSCAR and use new universal INCAR and KPOINTS files for all the DOSCAR calculations. For some of the systems, it works fine, but most of them are crashing with the error: WARNING: dimensions on CHGCAR file are different.


Here is an example logfile, I have many which are more or less the same.

Code: Select all

 running on    8 nodes
 distr:  one band on    2 nodes,    4 groups
 vasp.5.2.12 11Nov11 complex                                                    
  
 POSCAR found type information on POSCAR  C 
 POSCAR found :  1 types and      32 ions

 ----------------------------------------------------------------------------- 
|                                                                             |
|  ADVICE TO THIS USER RUNNING 'VASP/VAMP'   (HEAR YOUR MASTER'S VOICE ...):  |
|                                                                             |
|      You have a (more or less) 'large supercell' and for larger cells       |
|      it might be more efficient to use real space projection opertators     |
|      So try LREAL= Auto  in the INCAR   file.                               |
|      Mind: At the moment your POTCAR file does not contain real space       |
|       projectors, and has to be modified,  BUT if you                       |
|      want to do an extremely  accurate calculation you might also keep the  |
|      reciprocal projection scheme          (i.e. LREAL=.FALSE.)             |
|                                                                             |
 ----------------------------------------------------------------------------- 

 LDA part: xc-table for Pade appr. of Perdew
 found WAVECAR, reading the header
  number of k-points has changed, file:     9 present:    48
  trying to continue reading WAVECAR, but it might fail
 WAVECAR: different cutoff or change in lattice found
 WARNING: stress and forces are not correct
 POSCAR, INCAR and KPOINTS ok, starting setup
 WARNING: small aliasing (wrap around) errors must be expected
 FFT: planning ...(           1 )
 reading WAVECAR
 the WAVECAR file was read sucessfully
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 WARNING: dimensions on CHGCAR file are different
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10
 ERROR: charge density could not be read from file CHGCAR for ICHARG>10

To me, this error sounds like the dimensions of the CHGCAR file are different than the dimensions of the POSCAR file. I checked the CHGCAR and POSCAR files, and the dimensions are only different because POSCAR is printed to higher precision than CHGCAR.

POSCAR

Code: Select all

     9.8694725072717144    0.0000000000000000    0.0000000000000000
     0.0000000000000000    8.5536292067119888    0.0000000000000000
     0.0000000000000000    0.0000000000000000    9.9289699486412406

CHGCAR

Code: Select all

     9.869473    0.000000    0.000000
     0.000000    8.553629    0.000000
     0.000000    0.000000    9.928970

Is this difference enough to cause the error, or is something else causing it?

INCAR for relaxation:

Code: Select all

general:
System = G
ICHARG = 2 !0 charge from initial wave functions, 1 read CHGCAR, +10 non self-consistent
ISMEAR = 0 ! default 1: 0 GAUSSIAN, N>0 method of Methfessel-Paxton order N.
EDIFFG = 0.01 !ion structure (using negative values has caused ZBRENT errors)
EDIFF = 0.001 !electron structure
NSW    = 200 !Number of ionic steps
IBRION    = 2
PREC = HIGH !use ENCUT = 1.3*ENMAX in Potcar
ISIF = 4
LOPTICS= .TRUE.
NPAR= 4 !number of cores per band

INCAR for DOS:

Code: Select all

general:
System = G
ISTART = 1 !1 continuation job
ICHARG = 11 !0 charge from initial wave functions, 1 read CHGCAR, +10 non self-consistent
ISMEAR = 0 ! default 1: 0 GAUSSIAN, N>0 method of Methfessel-Paxton order N.
EDIFFG = 0.01 !ion structure
EDIFF = 0.001 !electron structure
NSW    = 0 !Number of ionic steps
IBRION    = 2
ISIF = 4
LOPTICS= .TRUE.
NPAR= 4 !number of cores per band

[aruth]

I eventually gave up trying to generate the DOS this way. Instead of loading the CHGCAR from the previous run, I set it to run an electronic relaxation step. It is probably much slower, but at least it works.

New INCAR

Code: Select all

ICHARG = 2 !0 charge from initial wave functions, 1 read CHGCAR, +10 non self-consistent
ISMEAR = 0 ! default 1: 0 GAUSSIAN, N>0 method of Methfessel-Paxton order N.
NSW    = 0 !Number of ionic steps
NPAR= 4 !number of cores per band

[patrizio_graziosi]

Hi dear,

I had the same issue when trying to run a nscf after a scf to use c2x.

I sorted it when using the same ENCUT, whereas EDIFF and KPOINTS mesh can differ.

Best,
Patrizio