a2y conversion error using PAW

[Michael.Friedrich]

Hello,

a brief summary of the problem I have is that I produced a KSS file with abinit 7.10.4 using PAW atomic data and the conversion to yambo (v4.0.0) failed because apparently the KSS file is not formated as it is expected from a2y. Former tries with NC pseudopotentials worked perfectly. I used the 'new' JTH - v0.2 PAW atomic data provided at the abinit web site suitable for abinit higher v7.6. Is this a known problem?

I generated a _KSS file in abinit using this input file:

prtvol 1
ixc -001012
iscf 17
#npwkss 1000
nbandkss -1
kssform 3
#irdwfk 1
#kptopt 3
diemac 5
enunit 2

ngkpt 6 6 6
nshiftk 1
shiftk 0 0 0
nband 60
nbdbuf 5
# nsym 1

rprim
0.540032689021 0.000000000000 0.841644042805
-0.270016344510 0.467682027566 0.841644042805
-0.270016344510 -0.467682027566 0.841644042805

acell 3*1.02398565369242E+01

natom 10
ntypat 3
typat 1 2 3 3 3 1 2 3 3 3
znucl 3 41 8

xred
2.83532832601605E-01 2.83532832601605E-01 2.83532832601605E-01
0.00000000000000E+00 0.00000000000000E+00 0.00000000000000E+00
8.67763054245485E-01 6.08931784775957E-01 2.20201886565342E-01
2.20201886565342E-01 8.67763054245485E-01 6.08931784775957E-01
6.08931784775957E-01 2.20201886565342E-01 8.67763054245485E-01
7.83532832601605E-01 7.83532832601605E-01 7.83532832601605E-01
5.00000000000000E-01 5.00000000000000E-01 5.00000000000000E-01
1.08931784775957E-01 3.67763054245485E-01 7.20201886565342E-01
7.20201886565342E-01 1.08931784775957E-01 3.67763054245485E-01
3.67763054245485E-01 7.20201886565342E-01 1.08931784775957E-01

pawecutdg 24
ecut 23
nstep 500
toldfe 1.0d-12
istwfk *1
# Use only symmorphic operations
symmorphi 0

I run into following error within a2y:

\ \ / / / \ | \/ | | _ \ / __ \
\ \/ / / ^ \ | \ / | | |_) | | | | |
\_ _/ / /_\ \ | |\/| | | _ < | | | |
| | / _____ \ | | | | | |_) | | `--" |
|__| /__/ \__\ |__| |__| |______/ \______/



<---> Checking input file ...datao_KSS
<---> DBs path set to :.
<---> KSS Header...abinit version 7.10.4
<---> Pseudo Potential KB form factors...yes
<---> Atom's informations...done
<09s> Symmetries...At line 192 of file a2y_db1.F (unit = 11, file = 'datao_KSS')
Fortran runtime error: I/O past end of record on unformatted file

apparently, the header is read incorrectly. I inserted 2 x read(11) in a2y_db1.F so that it gets the header right but then the energies and wf are formated different than read by a2y...

<---> Checking input file .../Users/michaelf/Work/LiNbO3/pawlda/optic/ags/datao_KSS
<---> DBs path set to :.
<---> KSS Header...abinit version 7.10.4
<---> Pseudo Potential KB form factors...yes
<---> Atom's informations...done
<---> Symmetries...[SI no]......[-I no]...[TR yes]
<---> RL vectors...done
<---> Closing shells against inversion...done
<---> :: ng_vec was increased to close the G-shells
<---> Energies...At line 263 of file a2y_db1.F (unit = 11, file = '/Users/michaelf/Work/LiNbO3/pawlda/optic/ags/datao_KSS')
Fortran runtime error: I/O past end of record on unformatted file

So theres no easy solution for me without going deeper into the codes...

Does anybody have a suggestion?

Thank you
Michael Friedrich

[Michael.Friedrich]

I also tested abinit v7.4 with GBRV PAW atomic data and yambo v3.4.1 with the same unsatisfying results...
Is there something wrong with the abinit run after all?
There's one warning at the stage of the KSS generation in abinit which however I don't consider relevant for my problem:

Calculating and writing out Kohn-Sham electronic Structure file
Using conjugate gradient wavefunctions and energies (kssform=3)

Sorting g-vecs for an output of states on an unique "big" PW basis.

remove_inversion: WARNING -
Removing inversion related symmetrie from initial set
The inversion was not found in the symmetries list.
Program uses the original set of symmetries
Since the number of g's to be written on file
was 0 or too large, it has been set to the max. value.,
computed from the union of the sets of G vectors for the different k-points.
Number of g-vectors written on file is: 4375
Number of bands written on file is: 60

I'm all out of ideas...

[myrta gruning]

Dear Michael

yambo works only with norm conserving pseudopotential.

Best regards,
m