Convergence problem

MULTIBINIT, aTDEP, ANADDB, optics, cut3d, mrg* postprocessors

Moderators: MMNSchmitt, gonze

Locked
jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Convergence problem

Post by jlwindy » Sat Apr 13, 2019 7:08 pm

Dear all

I want to get the ddb.file to compute the elastic constant to piezoelectric constant.Frist,I optimized the lattice structure following the tutorial (telast1).It's no problem.Then when i got the ddb.file,it's hard for potential residual converdence.How to solve it?

Initialisation of the first-order wave-functions :
ireadwf= 0

iter 2DEtotal(Ha) deltaE(Ha) residm vres2
-ETOT 1 1131.2634175629 -2.232E+03 2.113E+01 5.750E+10
ETOT 2 23.806228100867 -1.107E+03 4.019E+00 9.841E+05
ETOT 3 16.518070661107 -7.288E+00 2.853E+00 8.627E+03
ETOT 4 29.787646794204 1.327E+01 6.060E-01 3.502E+04
ETOT 5 9.3783862366090 -2.041E+01 6.131E-01 3.296E+02
ETOT 6 9.1131521157922 -2.652E-01 4.078E-01 9.984E+01
ETOT 7 9.0854361072602 -2.772E-02 4.802E-01 8.693E+01
ETOT 8 9.0392259378423 -4.621E-02 3.092E-01 2.329E+01
ETOT 9 9.0180788406383 -2.115E-02 4.051E-01 1.851E+00
ETOT 10 9.0155698128447 -2.509E-03 2.409E-01 9.119E-01
ETOT 11 9.0141659678579 -1.404E-03 3.383E-01 7.651E-01
ETOT 12 9.0134453474648 -7.206E-04 1.893E-01 5.637E-01
ETOT 13 9.0132744165990 -1.709E-04 2.827E-01 1.105E+00
ETOT 14 9.0128374773890 -4.369E-04 1.726E-01 4.977E-01
ETOT 15 9.0123882142884 -4.493E-04 2.358E-01 5.209E-01
ETOT 16 9.0122123206579 -1.759E-04 1.923E-01 4.561E-01
ETOT 17 9.0123011219598 8.880E-05 1.977E-01 2.580E-01
ETOT 18 9.0120589467151 -2.422E-04 2.153E-01 7.016E-01
ETOT 19 9.0120326638444 -2.628E-05 1.663E-01 6.304E-01
ETOT 20 9.0119752781907 -5.739E-05 2.420E-01 4.499E-01
ETOT 21 9.0119100985589 -6.518E-05 1.390E-01 2.896E-01
ETOT 22 9.0117798619940 -1.302E-04 2.716E-01 3.434E-02
ETOT 23 9.0118042120148 2.435E-05 1.177E-01 6.764E-02
ETOT 24 9.0117897674422 -1.444E-05 3.055E-01 6.331E-02
ETOT 25 9.0118279191624 3.815E-05 1.096E-01 1.142E-02
ETOT 26 9.0118656319323 3.771E-05 3.440E-01 5.934E-02
ETOT 27 9.0117218163637 -1.438E-04 1.231E-01 3.161E-02
ETOT 28 9.0117784352751 5.662E-05 3.874E-01 2.521E-03
ETOT 29 9.0117172720347 -6.116E-05 1.383E-01 1.207E-01
ETOT 30 9.0117742389971 5.697E-05 4.366E-01 2.024E-01
ETOT 31 9.0117162329102 -5.801E-05 1.555E-01 7.630E-02
ETOT 32 9.0117160119794 -2.209E-07 4.929E-01 6.170E-02
ETOT 33 9.0117554953017 3.948E-05 1.752E-01 2.021E-01
ETOT 34 9.0116953349666 -6.016E-05 5.560E-01 7.095E-02
ETOT 35 9.0117228219874 2.749E-05 1.970E-01 1.235E-01
ETOT 36 9.0117210793042 -1.743E-06 6.275E-01 4.562E-02
ETOT 37 9.0116998428234 -2.124E-05 2.218E-01 4.048E-03
ETOT 38 9.0117019640593 2.121E-06 7.085E-01 1.075E-02
ETOT 39 9.0117193225624 1.736E-05 2.499E-01 2.578E-02
ETOT 40 9.0116725873772 -4.674E-05 8.003E-01 5.693E-03
ETOT 41 9.0116934187290 2.083E-05 2.818E-01 6.058E-02
ETOT 42 9.0117301738969 3.676E-05 9.041E-01 1.406E-01
ETOT 43 9.0116858489546 -4.432E-05 3.179E-01 5.197E-02
ETOT 44 9.0116755117788 -1.034E-05 1.021E+00 2.974E-02
ETOT 45 9.0116968093239 2.130E-05 3.587E-01 9.317E-02
ETOT 46 9.0116833776342 -1.343E-05 1.154E+00 7.315E-02
ETOT 47 9.0116800724668 -3.305E-06 4.045E-01 4.780E-02
ETOT 48 9.0116895561819 9.484E-06 1.303E+00 5.391E-02
ETOT 49 9.0116871344662 -2.422E-06 4.565E-01 3.663E-02
ETOT 50 9.0117247610533 3.763E-05 1.472E+00 1.575E-01
ETOT 51 9.0116834744485 -4.129E-05 5.155E-01 7.767E-02
ETOT 52 9.0116714122596 -1.206E-05 1.664E+00 5.937E-02
ETOT 53 9.0116775553465 6.143E-06 5.821E-01 6.331E-02
ETOT 54 9.0116865074124 8.952E-06 1.880E+00 8.302E-02
ETOT 55 9.0116779287400 -8.579E-06 6.572E-01 3.701E-02
ETOT 56 9.0116726050555 -5.324E-06 2.124E+00 2.777E-02
ETOT 57 9.0116645090243 -8.096E-06 7.423E-01 1.876E-02
ETOT 58 9.0116582693299 -6.240E-06 2.400E+00 7.129E-03
ETOT 59 9.0116540383819 -4.231E-06 8.385E-01 7.420E-03
ETOT 60 9.0116494760614 -4.562E-06 2.712E+00 2.649E-03
-open ddk wf file :telast_2o_DS2_1WF16
-open ddk wf file :telast_2o_DS2_1WF17
-open ddk wf file :telast_2o_DS2_1WF18

scprqt: WARNING -
nstep= 60 was not enough SCF cycles to converge;
potential residual= 2.649E-03 exceeds tolvrs= 1.000E-10



the in.file

ndtset 3

# Set 1 : Initial self-consistent run

kptopt1 1
tolvrs1 1.0d-18 #need excellent convergence of GS quantities for RF runs

optdriver1 0

# Set 2 : Calculate the ddk wf's - needed for piezoelectric tensor and
# Born effective charges in dataset 3

getwfk2 -1
iscf2 -3 #this option is needed for ddk
kptopt2 2 #use time-reversal symmetry only for k points
nqpt2 1 #one wave vector will be specified
qpt2 0 0 0 #need to specify gamma point
rfelfd2 2 #set for ddk wf's only
rfdir2 1 1 1 #full set of directions needed
tolwfr2 1.0d-20 #only wf convergence can be monitored here
optdriver2 1
# Set 3 : response-function calculations for all needed perturbations

getddk3 -1
getwfk3 -2
kptopt3 2 #use time-reversal symmetry only for k points
nqpt3 1
qpt3 0 0 0
rfphon3 1 #do atomic displacement perturbation
rfatpol3 1 5 #do for all atoms
rfstrs3 3 #do strain perturbation
rfdir3 1 1 1 #the full set of directions is needed
tolvrs3 1.0d-10 #need reasonable convergence of 1st-order quantities
optdriver3 1
#Common input data


# acell COPY RELAXED RESULT FROM PREVIOUS CALCULATION
# Here is a default value, for automatic testing : suppress it and fill the previous line
acell 7.6897611419E+00 7.6897611419E+00 7.9699927278E+00



#Definition of the atom types and atoms
npsp 4
znucl 82 8 40 22
ntypat 3
ntypalch 1


mixalch 0.52 0.48
natom 5

typat 1 2 2 2 3
#Starting approximation for atomic positions in REDUCED coordinates
#based on ideal tetrahedral bond angles

# xred COPY RELAXED RESULT FROM PREVIOUS CALCULATION
# Here is a set of default values, for automatic testing : suppress it and fill the previous line
xred 0.0000000000E+00 0.0000000000E+00 -8.8515205616E-02
5.0000000000E-01 0.0000000000E+00 5.3440017247E-01
0.0000000000E+00 5.0000000000E-01 5.3440017247E-01
5.0000000000E-01 5.0000000000E-01 1.2291109060E-02
5.0000000000E-01 5.0000000000E-01 4.6742375162E-01




#Definition of the plane wave basis set
ecut 30 # Maximum kinetic energy cutoff (Hartree)
ecutsm 0.5 # Smoothing energy needed for lattice paramete
# optimization. This will be retained for
# consistency throughout.
pawecutdg 36
#Definition of the k-point grid
kptopt 1 # Use symmetry and treat only inequivalent points
ngkpt 8 8 8 # 4x4x4 Monkhorst-Pack grid
nshiftk 1 # Use one copy of grid only (default)
shiftk 0.0 0.0 0.5 # This choice of origin for the k point grid
# preserves the hexagonal symmetry of the grid,
# which would be broken by the default choice.

#Definition of the self-consistency procedure
diemac 4.0 # Model dielectric preconditioner
nstep 60 # Maxiumum number of SCF iterations
# enforce calculation of forces at each SCF step
optforces 1



Best wishes
jlwindy

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Convergence problem

Post by mverstra » Sun Apr 14, 2019 5:43 pm

Hello,

you are converging, but badly and slowly. You can add more iterations, more empty bands, and you can try to use metallic occupations (occopt 7 for example) with a very small smearing width. This makes the convergence easier, even for insulators.

Your post is not clear about which dataset is converging badly. It looks like the ddE perturbations at q=0 (it reads the ddk files).
Matthieu Verstraete
University of Liege, Belgium

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Sun Apr 14, 2019 7:16 pm

Dear mverstra

Thanks for your reply.In my calculation,the two sets are difficult convergence.Does convergence have much effect on the final calculation of the elastic constant?

Best wishes
jlwindy

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Convergence problem

Post by mverstra » Sun Apr 14, 2019 10:44 pm

which two datasets?

In general the elastic constants and all DFPT is more sensitive to ecut and nkpt than the ground state is. You have to re-check in detail. The ease of convergence depends more on having a delicate system which is magnetic or close to a metal-insulator transition, or very anisotropic in electronic structure (slabs, layers...)
Matthieu Verstraete
University of Liege, Belgium

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Mon Apr 15, 2019 8:13 am

Dear mvertra
I changed my in.file.It can converge quickly.but a new error appeared.
---- first-order wavefunction calculations are completed ----


==> Compute Derivative Database <==

--- !WARNING
src_file: m_crystal_io.F90
src_line: 192
message: |
Alchemical crystals are not fully supported by the netcdf format
Important parameters (e.g. znucl, symbols) are not written with the correct value
...


--- !ERROR
src_file: m_dynmat.F90
src_line: 5979
mpi_rank: 0
message: |
zhpev returned: 14

I can't understand the eeror.How to solve it?
my new in.file
ndtset 3

# Set 1 : Initial self-consistent run

kptopt1 1
tolvrs1 1.0d-18 #need excellent convergence of GS quantities for RF runs

optdriver1 0

# Set 2 : Calculate the ddk wf's - needed for piezoelectric tensor and
# Born effective charges in dataset 3

getwfk2 -1
iscf2 -3 #this option is needed for ddk
kptopt2 2 #use time-reversal symmetry only for k points
nqpt2 1 #one wave vector will be specified
qpt2 0 0 0 #need to specify gamma point
rfelfd2 2 #set for ddk wf's only
rfdir2 1 1 1 #full set of directions needed
tolwfr2 1.0d-20 #only wf convergence can be monitored here
optdriver2 1
# Set 3 : response-function calculations for all needed perturbations

getddk3 -1
getwfk3 -2
kptopt3 2 #use time-reversal symmetry only for k points
nqpt3 1
qpt3 0 0 0
rfphon3 1 #do atomic displacement perturbation
rfatpol3 1 5 #do for all atoms
rfstrs3 3 #do strain perturbation
rfdir3 1 1 1 #the full set of directions is needed
tolvrs3 1.0d-10 #need reasonable convergence of 1st-order quantities
optdriver3 1
#Common input data


# acell COPY RELAXED RESULT FROM PREVIOUS CALCULATION
# Here is a default value, for automatic testing : suppress it and fill the previous line
acell 7.6903479933E+00 7.6903479933E+00 7.9648543133E+00
strtarget 0.00000000E+00 0.00000000E+00 1.00000000E-05
0.00000000E+00 0.00000000E+00 0.00000000E+00


#Definition of the atom types and atoms
npsp 4
znucl 82 8 40 22
ntypat 3
ntypalch 1


mixalch 0.52 0.48
natom 5

typat 1 2 2 2 3
#Starting approximation for atomic positions in REDUCED coordinates
#based on ideal tetrahedral bond angles

# xred COPY RELAXED RESULT FROM PREVIOUS CALCULATION
# Here is a set of default values, for automatic testing : suppress it and fill the previous line
xred 0.0000000000E+00 0.0000000000E+00 -8.7992160930E-02
5.0000000000E-01 0.0000000000E+00 5.3409116380E-01
0.0000000000E+00 5.0000000000E-01 5.3409116380E-01
5.0000000000E-01 5.0000000000E-01 1.1997404040E-02
5.0000000000E-01 5.0000000000E-01 4.6781242930E-01



#Definition of the plane wave basis set
ecut 30 # Maximum kinetic energy cutoff (Hartree)
ecutsm 0.5 # Smoothing energy needed for lattice paramete
nband 40 # optimization. This will be retained for
occopt 7 # consistency throughout.
pawecutdg 36
#Definition of the k-point grid
kptopt 1 # Use symmetry and treat only inequivalent points
ngkpt 8 8 8 # 4x4x4 Monkhorst-Pack grid
nshiftk 1 # Use one copy of grid only (default)
shiftk 0.0 0.0 0.5 # This choice of origin for the k point grid
# preserves the hexagonal symmetry of the grid,
# which would be broken by the default choice.

#Definition of the self-consistency procedure
diemac 4.0
nstep 120 # Maxiumum number of SCF iterations
# enforce calculation of forces at each SCF step
optforces 1

I want to get the elastic constant under pressure.So i add the pressure parameters.

Best wishes
jlwindy

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Convergence problem

Post by mverstra » Mon Apr 15, 2019 10:02 am

hello again,

the input file is a good start, but it is still not clear to me exactly what you are calculating and especially when (which dataset, at which step) it is stopping. You are using PAW? I am not sure if all of the alchemical stuff is implemented, and in particular for DFPT. Is it crashing during the DDE, the phonons, or the stress response calculations?

The crash is a lapack call to zhpev - you should check online what 14 corresponds to - this is probably not a bug in the code but a combination of inputs which does not work and which is not filtered properly with an error message at startup.

Could you try with norm conserving psp?

As another step, bring down the kpt and ecut, and see if you can run a "test" calculation which is not converged but at least follows the whole workflow. You are mixing alchemy+PAW+stress and I am not sure all of that works together. You also don't mention if you are using GGA, which would be another layer of possible incompatibilities.
Matthieu Verstraete
University of Liege, Belgium

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Mon Apr 15, 2019 11:13 am

Dear mverstra
Thank you very much. I konw the renson of the error.I don't have storage space for caculating.I want to get the ddb.file to compute the elastic constanr and piezoelectric constant.There are two warning in my out.file
--------------------------------------------------------------------------------

dfpt_looppert : total number of electrons, from k and k+q
fully or partially occupied states are 4.400000E+01 and 4.400000E+01.
Initialisation of the first-order wave-functions :
ireadwf= 0

iter 2DEtotal(Ha) deltaE(Ha) residm vres2
-ETOT 1 1789.8240498592 -1.569E+03 3.553E+01 4.733E+05
ETOT 2 91.464520221009 -1.698E+03 1.955E+01 1.866E+05
ETOT 3 30.760712338420 -6.070E+01 1.246E+00 1.615E+04
ETOT 4 45.900152727585 1.514E+01 2.548E-01 7.524E+04
ETOT 5 11.508799211870 -3.439E+01 7.910E-02 3.279E+03
ETOT 6 9.4807610290486 -2.028E+00 2.116E-02 3.474E+02
ETOT 7 9.1972250918816 -2.835E-01 6.338E-03 1.725E+02
ETOT 8 9.0504298966962 -1.468E-01 7.490E-04 9.894E+01
ETOT 9 8.9845181125379 -6.591E-02 3.176E-04 9.587E+01
ETOT 10 8.9379551818492 -4.656E-02 3.654E-04 2.809E+01
ETOT 11 8.9113719292504 -2.658E-02 2.336E-04 3.055E+01
ETOT 12 8.9030510825323 -8.321E-03 2.245E-04 6.137E-01
ETOT 13 8.9039183943014 8.673E-04 3.601E-05 5.949E-01
ETOT 14 8.8978813524882 -6.037E-03 9.867E-05 2.144E+00
ETOT 15 8.9021913080173 4.310E-03 3.904E-05 8.850E-01
ETOT 16 8.8998657716761 -2.326E-03 1.172E-05 3.554E+00
ETOT 17 8.8995973542492 -2.684E-04 4.928E-06 5.087E-02
ETOT 18 8.9031016765143 3.504E-03 9.558E-06 1.996E-01
ETOT 19 8.8974816349050 -5.620E-03 1.893E-05 6.579E-02
ETOT 20 8.9022616340781 4.780E-03 1.407E-05 9.561E-01
ETOT 21 8.8974845177486 -4.777E-03 9.329E-06 9.342E-02
ETOT 22 8.8965778199918 -9.067E-04 4.882E-06 5.262E-01
ETOT 23 8.8963465412339 -2.313E-04 3.929E-07 4.316E-01
ETOT 24 8.8966128627162 2.663E-04 1.195E-05 2.128E-01
ETOT 25 8.8971340101152 5.211E-04 2.213E-05 4.310E+00
ETOT 26 8.9113493180071 1.422E-02 5.746E-05 1.679E+01
ETOT 27 8.9053508158122 -5.999E-03 3.066E-05 1.520E+01
ETOT 28 8.8977235442458 -7.627E-03 1.753E-05 5.539E+00
ETOT 29 8.9009107961360 3.187E-03 6.059E-05 6.379E-01
ETOT 30 8.8962992949901 -4.612E-03 2.343E-05 1.795E-01
ETOT 31 8.8971253380978 8.260E-04 4.281E-05 2.280E+00
ETOT 32 8.8976903988314 5.651E-04 2.167E-05 3.839E-01
ETOT 33 8.8979567901324 2.664E-04 4.623E-06 6.471E-01
ETOT 34 8.8983695525708 4.128E-04 4.720E-06 2.364E-01
ETOT 35 8.8974020105120 -9.675E-04 1.989E-05 2.667E-01
ETOT 36 8.8969315333466 -4.705E-04 2.997E-06 1.341E-01
ETOT 37 8.8975113297452 5.798E-04 2.884E-05 4.035E+00
ETOT 38 8.8956458348025 -1.865E-03 1.224E-05 5.806E-01
ETOT 39 8.8954161878942 -2.296E-04 4.417E-07 2.403E-01
ETOT 40 8.8956362217527 2.200E-04 3.530E-07 4.151E-01
ETOT 41 8.8985613930681 2.925E-03 1.197E-05 2.255E-01
ETOT 42 8.8961927193841 -2.369E-03 3.876E-06 6.612E-03
ETOT 43 8.8960772551463 -1.155E-04 1.353E-06 8.415E-03
ETOT 44 8.8981128089440 2.036E-03 8.668E-06 2.089E+00
ETOT 45 8.9009186885246 2.806E-03 8.718E-06 3.887E+00
ETOT 46 8.8963454605669 -4.573E-03 3.851E-05 5.905E-01
ETOT 47 8.8967366838283 3.912E-04 2.446E-06 1.092E+00
ETOT 48 8.8966980572991 -3.863E-05 2.898E-07 8.092E-01
ETOT 49 8.8962815696623 -4.165E-04 2.567E-07 1.162E+00
ETOT 50 8.8965229077103 2.413E-04 6.176E-08 9.922E-01
ETOT 51 8.8964273373240 -9.557E-05 7.439E-07 1.324E+00
ETOT 52 8.8978215258689 1.394E-03 4.451E-06 7.519E+00
ETOT 53 8.8956398776468 -2.182E-03 1.278E-05 1.462E-01
ETOT 54 8.8967545671571 1.115E-03 2.597E-06 3.122E+00
ETOT 55 8.8958436714270 -9.109E-04 9.701E-07 1.429E+00
ETOT 56 8.8957871312282 -5.654E-05 6.678E-07 6.805E-01
ETOT 57 8.8955468673043 -2.403E-04 9.938E-07 1.537E+00
ETOT 58 8.8950862176507 -4.606E-04 1.577E-06 5.960E-01
ETOT 59 8.8955148439268 4.286E-04 2.061E-06 1.641E-01
ETOT 60 8.8956857142798 1.709E-04 9.275E-07 9.904E-02
ETOT 61 8.8956253571127 -6.036E-05 9.662E-07 1.197E+00
ETOT 62 8.8956180976902 -7.259E-06 2.277E-07 1.366E+00
ETOT 63 8.8954827467578 -1.354E-04 6.196E-07 1.414E+00
ETOT 64 8.8954403057837 -4.244E-05 6.679E-07 1.481E+00
ETOT 65 8.8959209932500 4.807E-04 2.519E-07 2.873E+00
ETOT 66 8.8960132682910 9.228E-05 3.861E-07 3.066E+00
ETOT 67 8.8954279717764 -5.853E-04 2.521E-06 1.274E+00
ETOT 68 8.8951739884687 -2.540E-04 6.988E-07 6.348E-01
ETOT 69 8.8950647912038 -1.092E-04 2.804E-07 2.225E-01
ETOT 70 8.8949998776561 -6.491E-05 2.849E-07 3.596E-01
ETOT 71 8.8951766871064 1.768E-04 3.049E-07 7.564E-01
ETOT 72 8.8950353178252 -1.414E-04 3.748E-07 1.576E-01
ETOT 73 8.8951196846642 8.437E-05 2.768E-07 7.424E-03
ETOT 74 8.8950873930994 -3.229E-05 2.822E-08 6.084E-02
ETOT 75 8.8954866607282 3.993E-04 1.120E-06 1.317E-03
ETOT 76 8.8954718548425 -1.481E-05 6.556E-08 8.335E-03
ETOT 77 8.8951735779497 -2.983E-04 2.611E-07 5.275E-02
ETOT 78 8.8951472375738 -2.634E-05 1.009E-07 2.326E-04
ETOT 79 8.8949856915092 -1.615E-04 3.364E-07 1.542E-02
ETOT 80 8.8950365838982 5.089E-05 3.546E-08 2.642E-03
ETOT 81 8.8949820051976 -5.458E-05 4.780E-07 1.254E-02
ETOT 82 8.8949915596477 9.554E-06 1.220E-06 1.494E-02
ETOT 83 8.8948987885215 -9.277E-05 2.086E-07 1.312E-01
ETOT 84 8.8950341384302 1.353E-04 5.318E-07 4.266E-01
ETOT 85 8.8949926222381 -4.152E-05 9.804E-08 2.265E-01
ETOT 86 8.8952589360160 2.663E-04 2.009E-06 4.410E-01
ETOT 87 8.8950139956855 -2.449E-04 6.243E-07 3.481E-01
ETOT 88 8.8951107401508 9.674E-05 2.585E-07 1.452E-01
ETOT 89 8.8949708829568 -1.399E-04 5.166E-07 1.255E-01
ETOT 90 8.8951139528122 1.431E-04 2.182E-07 6.219E-01
ETOT 91 8.8950703083146 -4.364E-05 9.111E-08 4.332E-01
ETOT 92 8.8950839086201 1.360E-05 1.816E-07 5.342E-01
ETOT 93 8.8951524754839 6.857E-05 1.294E-07 7.223E-01
ETOT 94 8.8949802101397 -1.723E-04 9.607E-07 1.625E-01
ETOT 95 8.8954263349483 4.461E-04 2.033E-06 2.611E-01
ETOT 96 8.8951140661534 -3.123E-04 5.080E-06 3.274E-02
ETOT 97 8.8991568720776 4.043E-03 2.874E-05 1.268E+00
ETOT 98 8.8951056275655 -4.051E-03 3.035E-05 6.326E-02
ETOT 99 8.8951934227048 8.780E-05 6.156E-07 1.829E-02
ETOT 100 8.8952735704156 8.015E-05 9.516E-08 2.517E-02
ETOT 101 8.8950478877650 -2.257E-04 3.311E-06 4.812E-01
ETOT 102 8.8950228277836 -2.506E-05 1.301E-07 4.023E-01
ETOT 103 8.8950309994125 8.172E-06 4.707E-08 4.129E-01
ETOT 104 8.8950505249604 1.953E-05 4.153E-08 3.178E-01
ETOT 105 8.8951272993809 7.677E-05 6.007E-08 2.788E-01
ETOT 106 8.8951281334807 8.341E-07 1.473E-07 3.936E-01
ETOT 107 8.8951486452261 2.051E-05 3.653E-07 1.580E-01
ETOT 108 8.8951648543234 1.621E-05 4.153E-07 1.339E-01
ETOT 109 8.8950544615118 -1.104E-04 1.002E-07 1.107E-01
ETOT 110 8.8950861934090 3.173E-05 5.077E-07 1.183E-01
ETOT 111 8.8949986273660 -8.757E-05 7.717E-08 2.671E-01
ETOT 112 8.8950333958587 3.477E-05 9.938E-08 5.513E-01
ETOT 113 8.8949491957615 -8.420E-05 9.806E-08 3.943E-01
ETOT 114 8.8949855842379 3.639E-05 4.146E-07 2.589E-01
ETOT 115 8.8949621183683 -2.347E-05 1.719E-07 1.803E-01
ETOT 116 8.8949635918343 1.473E-06 1.140E-07 2.670E-01
ETOT 117 8.8949731507437 9.559E-06 6.696E-07 2.787E-02
ETOT 118 8.8948982936199 -7.486E-05 4.984E-07 1.633E-02
ETOT 119 8.8948390594574 -5.923E-05 5.303E-07 5.244E-02
ETOT 120 8.8948052830883 -3.378E-05 4.795E-07 3.037E-02
-open ddk wf file :telast_2o_DS2_1WF16
-open ddk wf file :telast_2o_DS2_1WF17
-open ddk wf file :telast_2o_DS2_1WF18

scprqt: WARNING -
nstep= 120 was not enough SCF cycles to converge;
potential residual= 3.037E-02 exceeds tolvrs= 1.000E-10

----iterations are completed or convergence reached----

Mean square residual over all n,k,spin= -50.000E-04; max= 47.946E-08

Thirteen components of 2nd-order total energy (hartree) are
1,2,3: 0th-order hamiltonian combined with 1st-order wavefunctions
kin0= 2.02098815E+03 eigvalue= 2.57587994E+02 local= -1.60445400E+03
4,5,6: 1st-order hamiltonian combined with 1st and 0th-order wfs
loc psp = -4.06035864E+03 Hartree= 1.02661093E+03 xc= -2.24645496E+02
note that "loc psp" includes a xc core correction that could be resolved
7,8,9: eventually, occupation + non-local contributions
edocc= 6.61144605E+02 enl0= 1.21265919E+03 enl1= -2.63939420E+03
1-9 gives the relaxation energy (to be shifted if some occ is /=2.0)
erelax= -3.34986146E+03
10,11,12 Non-relaxation contributions : frozen-wavefunctions and Ewald
fr.local= 1.83193057E+03 fr.nonlo= 1.31587589E+03 Ewald= 2.24519569E+02
13,14 Frozen wf xc core corrections (1) and (2)
frxc 1 = -1.59500719E+02 frxc 2 = 1.45930955E+02
Resulting in :
2DEtotal= 0.8894805283E+01 Ha. Also 2DEtotal= 0.242039960918E+03 eV
(2DErelax= -3.3498614645E+03 Ha. 2DEnonrelax= 3.3587562697E+03 Ha)
( non-var. 2DEtotal : 8.8798506378E+00 Ha)


--------------------------------------------------------------------------------
Perturbation wavevector (in red.coord.) 0.000000 0.000000 0.000000
Found 4 symmetries that leave the perturbation invariant.
symkpt : the number of k-points, thanks to the symmetries,
is reduced to 100 .

--------------------------------------------------------------------------------

-inwffil : will read wavefunctions from disk file telast_2o_DS1_WFK
--------------------------------------------------------------------------------

dfpt_looppert : total number of electrons, from k and k+q
fully or partially occupied states are 4.400000E+01 and 4.400000E+01.
Initialisation of the first-order wave-functions :
ireadwf= 0

iter 2DEtotal(Ha) deltaE(Ha) residm vres2
-ETOT 1 37.868567299436 -2.475E+02 1.298E+07 4.602E+04
ETOT 2 56.361279322605 1.849E+01 1.674E+05 6.043E+04
ETOT 3 15.887560020084 -4.047E+01 1.840E+03 1.600E+04
ETOT 4 4.6423531437261 -1.125E+01 8.381E+01 2.149E+02
ETOT 5 4.6517339038258 9.381E-03 1.627E+00 2.299E+02
ETOT 6 4.5043831087517 -1.474E-01 4.570E-02 9.199E-01
ETOT 7 4.5043023256969 -8.078E-05 7.847E-03 4.472E-01
ETOT 8 4.5040585112419 -2.438E-04 2.272E-03 8.681E-02
ETOT 9 4.5040052901952 -5.322E-05 2.364E-04 8.646E-04
ETOT 10 4.5040047768348 -5.134E-07 7.819E-05 1.307E-04
ETOT 11 4.5040047196023 -5.723E-08 4.400E-06 9.622E-06
ETOT 12 4.5040047177673 -1.835E-09 3.121E-06 4.310E-06
ETOT 13 4.5040047145163 -3.251E-09 2.051E-07 3.894E-07
ETOT 14 4.5040047142823 -2.340E-10 1.126E-07 5.370E-09
ETOT 15 4.5040047142797 -2.615E-12 9.365E-09 3.691E-10
ETOT 16 4.5040047142792 -4.547E-13 4.860E-09 1.015E-10
ETOT 17 4.5040047142793 5.684E-14 7.235E-10 1.268E-11

At SCF step 17 vres2 = 1.27E-11 < tolvrs= 1.00E-10 =>converged.
-open ddk wf file :telast_2o_DS2_1WF16
-open ddk wf file :telast_2o_DS2_1WF17
-open ddk wf file :telast_2o_DS2_1WF18

But there is a convergence before and after this warning. Why? Does it affect the final calculation.The datesat 2 and 3 have the warning.

Best wishes
jlwindy

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Mon Apr 15, 2019 7:51 pm

I used the ddb.file to compute elastic constant and piezoelectric constant.The warning appear
ddb_piezo : WARNING -
Acoustic sum rule violation met : the eigenvalues of accoustic mode
are too large at Gamma point
Increase cutoff energy or k-points sampling.
The three eigenvalues are: 2.438948E-05 -2.446268E-04 2.438948E-05

Is it due to an unconvergent warning in the previous step?
Looking forward to your reply
jlwindy

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Convergence problem

Post by mverstra » Mon Apr 15, 2019 10:00 pm

jlwindy wrote:I used the ddb.file to compute elastic constant and piezoelectric constant.The warning appear
ddb_piezo : WARNING -
Acoustic sum rule violation met : the eigenvalues of accoustic mode
are too large at Gamma point
Increase cutoff energy or k-points sampling.
The three eigenvalues are: 2.438948E-05 -2.446268E-04 2.438948E-05

Is it due to an unconvergent warning in the previous step?
Looking forward to your reply
jlwindy

you are accumulating too many questions in one thread. Anyhow, this is not directly related. 2.e-4 is quite large for the Gamma point modes - things are not converged and possibly you have a bad pseudopotential.
Matthieu Verstraete
University of Liege, Belgium

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Convergence problem

Post by mverstra » Mon Apr 15, 2019 10:04 pm

jlwindy wrote:Dear mverstra
Thank you very much. I konw the renson of the error.I don't have storage space for caculating.I want to get the ddb.file to compute the elastic constanr and piezoelectric constant.There are two warning in my out.file
--------------------------------------------------------------------------------

dfpt_looppert : total number of electrons, from k and k+q
fully or partially occupied states are 4.400000E+01 and 4.400000E+01.
Initialisation of the first-order wave-functions :
ireadwf= 0

iter 2DEtotal(Ha) deltaE(Ha) residm vres2

ETOT 117 8.8949731507437 9.559E-06 6.696E-07 2.787E-02
ETOT 118 8.8948982936199 -7.486E-05 4.984E-07 1.633E-02
ETOT 119 8.8948390594574 -5.923E-05 5.303E-07 5.244E-02
ETOT 120 8.8948052830883 -3.378E-05 4.795E-07 3.037E-02
-open ddk wf file :telast_2o_DS2_1WF16
-open ddk wf file :telast_2o_DS2_1WF17
-open ddk wf file :telast_2o_DS2_1WF18

scprqt: WARNING -
nstep= 120 was not enough SCF cycles to converge;
potential residual= 3.037E-02 exceeds tolvrs= 1.000E-10

----iterations are completed or convergence reached----

this is the one that is unconverged - look at the last column.

I still do not know which perturbation it is doing here: You need to be more complete and detailed. Please read the nettiquette on the main page about how to make a good post.




Mean square residual over all n,k,spin= -50.000E-04; max= 47.946E-08
.
.
.
dfpt_looppert : total number of electrons, from k and k+q
fully or partially occupied states are 4.400000E+01 and 4.400000E+01.
Initialisation of the first-order wave-functions :
ireadwf= 0

iter 2DEtotal(Ha) deltaE(Ha) residm vres2
-ETOT 1 37.868567299436 -2.475E+02 1.298E+07 4.602E+04

ETOT 16 4.5040047142792 -4.547E-13 4.860E-09 1.015E-10
ETOT 17 4.5040047142793 5.684E-14 7.235E-10 1.268E-11

At SCF step 17 vres2 = 1.27E-11 < tolvrs= 1.00E-10 =>converged.

this one is fine as you can see.
Matthieu Verstraete
University of Liege, Belgium

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Tue Apr 16, 2019 11:06 am

Dear mverstra

Thank you for you reply.

In dateset3,it includes two kinds of perturbation(atomic displacement perturbation and strain perturbation).I choose the NCPP pseudopotentials.Because I use the alchemical pseudoatoms.

Best wishes
jlwindy

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Tue Apr 16, 2019 1:21 pm

The ERROR appeared again.
--- !ERROR
src_file: m_dynmat.F90
src_line: 5979
mpi_rank: 0
message: |
zhpev returned: 14
That successful run seemed like an accident.Can you tell me how to check online what zhpev 14 corresponds to?

Thank you

pomax
Posts: 7
Joined: Wed Jun 05, 2019 5:40 pm

Re: Convergence problem

Post by pomax » Wed Jun 05, 2019 5:45 pm

Hi jlwindy,
I also had problem to converge a DFPT simulation. I had a metal so I needed to use tsmear and nband. I've optimized tsmear with numerous calculation to get a convergence from e-9 to e-13 using nstep=200. Then, I changed the value of nband from 42(I had 29 occupied band) to 72. I've got a convergence of 1e-18 in 80 step. So, I would recommend to add more band and maybe optimize tsmear.

Cordially, Olivier

jlwindy
Posts: 32
Joined: Tue Jan 08, 2019 4:11 pm

Re: Convergence problem

Post by jlwindy » Mon Jun 17, 2019 2:15 pm

dear pomax

Thanks for your reply.It helps me a lot.

jlwindy

Locked