[QE-users] FFT grid problems with ph.x

[Jakob Kraus]

Dear QE community,


my phonon calculation for AlN complains about


"Error in routine scale_sym_ops (4): incompatible FFT grid"


, whereas the scf calculation I performed beforehand went through 
without any problems.


I've consulted the user guide and tried increasing the cutoffs for 
wavefunction and density, to no avail.

Additionally, I've tried playing around with the lattice constants and 
setting the FFT grid by hand, to numbers

that are divisible by 12, which did not help either.


Do you have any idea that could help me? I am pretty much stuck at this 
point.

Below is the original espresso.pwi input file for the scf calculation 
that is performed before the phonon calculation.


Yours sincerely,


Jakob Kraus (Technische Universität Bergakademie Freiberg)



&CONTROL
    calculation      = 'scf'
    title            = 'AlN_vib'
    verbosity        = 'high'
    restart_mode     = 'from_scratch'
    tstress          = .true.
    tprnfor          = .true.
    outdir           = './'
    prefix           = 'espresso'
    pseudo_dir       = '/home/kraus/theos_pp/pz.0.3.1/PSEUDOPOTENTIALS_NC'
/
&SYSTEM
    ibrav            = 4
    ecutwfc          = 190
    occupations      = 'fixed'
    nspin            = 1
    ntyp             = 2
    nat              = 4
    celldm(1)        = 5.877048275900463
    celldm(3)        = 1.6012861736334407
/
&ELECTRONS
    electron_maxstep = 300
    conv_thr         = 1e-10
    mixing_beta      = 0.7
/
&IONS
/
&CELL
/

ATOMIC_SPECIES
Al 26.9815385 Al.pz-n-nc.UPF
N 14.007 N.pz-nc.UPF

K_POINTS automatic
5 5 5  0 0 0

ATOMIC_POSITIONS angstrom
Al 1.5550015550 0.8977787708 0.0000000000
Al -0.0000015550 1.7955602350 2.4900000000
N 1.5550015550 0.8977787708 1.9028580000
N -0.0000015550 1.7955602350 4.3928580000