[Pw_forum] El-Ph funny error
Miguel Martínez Canales
wmbmacam at lg.ehu.es
Tue Oct 4 18:55:33 CEST 2005
Dear Dr. Gianozzi,
> so it is not a truncation problem. Could you please submit a tets
> job so that the error can be reproduced?
The batch I'm submitting should make it... although I will test it tonight. For
this particular point, the lines you quoted are:
-0.500000000000000 -0.250000000000000 1.00000000000000
-1.92307692307692D-002 1.92307692307692D-002 1.01923076923077
-1.92307692307692D-002 1.92307692307692D-002 1.92307692307692D-002
The job is:
[...]
# self-consistent calculation
cat > al.scf.in << EOF
&control
calculation='scf'
restart_mode='from_scratch',
tstress = .true.
tprnfor = .true.
prefix='al',
pseudo_dir = '$PSEUDO_DIR/',
outdir='$TMP_DIR/'
/
&system
ibrav= 2, celldm(1) =7.603, nat= 1, ntyp= 1,
ecutwfc =40.0,
occupations='smearing', smearing='gauss', degauss=0.04
/
&electrons
conv_thr = 1.0d-8
mixing_beta = 0.7
/
ATOMIC_SPECIES
Al 26.9815 Al.pbe-n-van.UPF
ATOMIC_POSITIONS
Al 0.00 0.00 0.00
K_POINTS {automatic}
26 26 26 1 1 1
EOF
$ECHO " calculo autoconsistente...\c"
$PW_COMMAND < al.scf.in > al.scf.out
$ECHO " hecho"
# non self-consistent calculation
cat > al.nscf.in << EOF
&control
calculation='phonon'
restart_mode='from_scratch',
prefix='al',
pseudo_dir = '$PSEUDO_DIR/',
outdir='$TMP_DIR/'
/
&system
ibrav= 2, celldm(1) =7.603, nat= 1, ntyp= 1,
ecutwfc =40.0,
occupations='smearing', smearing='gauss', degauss=0.04
/
&electrons
conv_thr = 1.0d-8
/
&phonon
xqq(1) = -0.500, xqq(2) = -0.250, xqq(3) = 1.000
/
ATOMIC_SPECIES
Al 26.9815 Al.pbe-n-van.UPF
ATOMIC_POSITIONS
Al 0.00 0.00 0.00
K_POINTS {automatic}
26 26 26 1 1 1
EOF
$ECHO " calculo no autoconsistente para q=-0.500,-0.250,1.000..\c"
$PW_COMMAND < al.nscf.in >> al.nscf.out
$ECHO " hecho"
# phonon calculation
cat > al.ph.in << EOF
phonons of Al at $qpoint
&inputph
tr2_ph=1.0d-12,
prefix='al',
amass(1)=26.9815,
outdir='$TMP_DIR/',
fildyn='dyn.$qpoint'
fildvscf='aldv'
elph=.true.,
trans=.true.,
/
-0.500 -0.250 1.000
EOF
$ECHO " calculo de lambda y gamma para q=$qpoint...\c"
$PH_COMMAND < al.ph.in >> al.ph.$qpoint.out
$ECHO " hecho"
--
----------------------------------------
Miguel Martínez Canales
Dto. Física de la Materia Condensada
UPV/EHU
Facultad de Ciencia y Tecnología
Apdo. 644
48080 Bilbao (Spain)
Fax: +34 94 601 3500
Tlf: +34 94 601 5437
----------------------------------------
2005 International Year of Physics
"But you don't ask with respect. You don't
offer friendship. You don't even think to call
me Godfather."
Don Vito Corleone
More information about the users
mailing list