[Pw_forum] from expint : error # 1, continued fraction failed

Jonathan DuBois jonathan.dubois at gmail.com
Mon Oct 17 18:21:27 CEST 2011


I've been attempting to use hse with the espresso 4.3 and pw.x keeps
crashing with
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     from expint : error #         1
     continued fraction failed
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

any ideas why this might be happening and how I can avoid it?

input file is below.

&CONTROL
    calculation = "scf"
    restart_mode = 'restart'
    prefix = 'Ca_graphene'
    pseudo_dir = './'
    verbosity = high
    outdir = './'
    wf_collect = .true.
/

&SYSTEM
    input_dft = 'HSE'
    ibrav = 0
    ntyp= 2
    nat= 33
    nbnd = 138
    occupations = 'smearing'
    smearing = 'gaussian'
    nosym = .false.
    degauss = 0.001,
    ecutwfc = 160.0
/

&ELECTRONS
    diagonalization = 'david'
    conv_thr = 1.0d-8
    startingwfc = 'file'
/

CELL_PARAMETERS hexagonal
    16.18932808     -9.346904673        0.0000000000
    16.18932808      9.346904673        0.0000000000
    0.00000000       0.000000000        28.345889829


sample output at failure:
     total cpu time spent up to now is  19953.73 secs

     total energy              =    -440.37890917 Ry
     Harris-Foulkes estimate   =    -440.37891169 Ry
     estimated scf accuracy    <       0.00000727 Ry

     iteration #  4     ecut=   160.00 Ry     beta=0.70
     Davidson diagonalization with overlap
     ethr =  5.27E-09,  avg # of iterations =  2.0

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     from expint : error #         1
     continued fraction failed
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%



More information about the users mailing list