[Pw_forum] failure in trying to compute U

["José C. Conesa"]

Hi all,
I am trying to compute U for Mo in a molybdenum bronze, using 
Cococcioni's method. The calculation crashes. A file CRASH results with 
this content:
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     task #         9
       reading namelist system
  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Standard output shows, after lines
...
      Max number of k-points (npk) =  40000
      Max angular momentum in pseudopotentials (lmaxx) =  3
      Waiting for input...
      Reading input from standard input


these lines (repeated as many times as parallel processes involved):
  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
      Error in routine  read_namelists (17):
       reading namelist system
  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
      stopping ...

My input file is

&control
    title = 'a0.05'
    calculation = 'scf'
    restart_mode='from_scratch'
    verbosity = 'high',
    prefix='namo6o17',
    etot_conv_thr = 2.0D-6
/
&system
    ibrav=14,
    celldm(1)= 5.5325, celldm(2)=0.9974, celldm(3)=2.3467,
    celldm(4)=0.0, celldm(5)=0.0009076, celldm(6)=-0.49868442,
    nat= 24, ntyp= 4,
    ecutwfc=30.0, ecutrho=125.0,
    nbnd= 150, nspin=2,
    starting_magnetization(1)= 0.1,
    starting_magnetization(2)=-0.5,
    starting_magnetization(3)= 0.0,
    starting_magnetization(4)= 0.0,
    occupations='smearing', smearing='mp', degauss=0.01,
    lda_plus_u = .true.,
    Hubbard_U(1)=1.d-8, Hubbard_U(2)=1.d-8, # (values used in the 
previous run)
    Hubbard_alpha(1)= 0.0, # this is alpha 0 applied on atom 1 (Mo)
    Hubbard_alpha(2)= 0.05, # this is alpha, only applied on atom 2 
(also Mo)
/
&electrons
     diagonalization='cg'
     startingpot = 'namo6o17'  #
     startingwfc = 'namo6o17' # (the corresponding files from the 
previous run are copied in the main working directory)
     diago_thr_init = 1.59E-10  # (from the previous run)
     mixing_mode = 'plain'
     mixing_beta = 0.7
     conv_thr = 1.0d-9,
/
ATOMIC_SPECIES
  ...(pseudopotentials for all species)
ATOMIC_POSITIONS crystal
....(names and coordinates of 24 atoms)
K_POINTS automatic
4 4 4 0 0 0

¿Any  hint on the cause of the problem?
Thanks,

-- 
José C. Conesa
Instituto de Catálisis y Petroleoquímica, CSIC
Marie Curie 2, Cantoblanco
28049 Madrid - Spain
Tel. +34-915854766