Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo
Close
From: 张 小东 <zhangxiaodong1028@gm...>  20130428 03:09:19
Attachments:
Message as HTML

Dear phonopy users, I want to calculate the phonon spectrum of Bi2Se3 and got the force_constants using DFPT implemented in VASP. However, when i calculate the eigenvalue of phonon band, and error occurred. The error is "numpy.linalg.linalg.LinAlgError: Eigenvalues did not converge" The band.conf is: ATOM_NAME = Se Bi DIM = 2 2 2 BAND = 0.0 0.0 0.0 0.50 0.50 0.50 0.50 0.50 0.0 0.0 0.0 0.0 0.0 0.50 0.0 EIGENVECTOR = .TRUE. FORCE_CONSTANTS = READ The INCAR is: SYSTEM = Bi2Se3 ## static calculation######### ISMEAR = 0;SIGMA=0.005 ENCUT=400 PREC=Accurate ALG0=Normal ############################## ## structure optimization #### #NSW=300 #IBRION=1 #ADDGRID=.TRUE. #ISIF=3 #EDIFF=1E6 #EDIFFG=0.001 #POTIM=0.3 ############################## ## Phonon #################### IBRION=8 NSW=1 Is anyone has some suggestions? best wishes, Xiaodong Zhang Institute of Modern Physics, Northwest University ps. The error is somewhat related to the numpy. I guess it is the result of the convergence problem of solving dynamical matrix. 
From: Atsushi Togo <atz.togo@gm...>  20130428 03:19:41

Hi, I guess the values of FORCE_CONSTANTS are something unusual. I may suggest: 1) FORCE_CONSTANTS file is incorrectly generated in some reason. You can compare FORCE_CONSTANTS and force constants in OUTCAR to check it. 2) Force constants are not well calculated in VASP. It's difficult to help for me on this. But I see one point something that isn't recommended: NSW=300 should be divided into, e.g., NSW=40 x some number of VASP runs, because FFT mesh should be regenerated according to cutoff energy when lattice parameters change largely. Togo On Sun, Apr 28, 2013 at 12:09 PM, 张 小东 <zhangxiaodong1028@...> wrote: > Dear phonopy users, > > I want to calculate the phonon spectrum of Bi2Se3 and got the > force_constants using DFPT implemented > in VASP. However, when i calculate the eigenvalue of phonon band, and error > occurred. > The error is "numpy.linalg.linalg.LinAlgError: Eigenvalues did not > converge" > > The band.conf is: > ATOM_NAME = Se Bi > DIM = 2 2 2 > BAND = 0.0 0.0 0.0 0.50 0.50 0.50 0.50 0.50 0.0 0.0 0.0 0.0 0.0 0.50 > 0.0 > EIGENVECTOR = .TRUE. > FORCE_CONSTANTS = READ > > The INCAR is: > SYSTEM = Bi2Se3 > > ## static calculation######### > ISMEAR = 0;SIGMA=0.005 > ENCUT=400 > PREC=Accurate > ALG0=Normal > ############################## > > ## structure optimization #### > #NSW=300 > #IBRION=1 > #ADDGRID=.TRUE. > #ISIF=3 > #EDIFF=1E6 > #EDIFFG=0.001 > #POTIM=0.3 > ############################## > > ## Phonon #################### > IBRION=8 > NSW=1 > > Is anyone has some suggestions? > > best wishes, > Xiaodong Zhang > Institute of Modern Physics, Northwest University > > ps. The error is somewhat related to the numpy. I guess it is the result of > the convergence problem of solving dynamical matrix. > > >  > Try New Relic Now & We'll Send You this Cool Shirt > New Relic is the only SaaSbased application performance monitoring service > that delivers powerful full stack analytics. Optimize and monitor your > browser, app, & servers with just a few lines of code. Try New Relic > and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr > _______________________________________________ > Phonopyusers mailing list > Phonopyusers@... > https://lists.sourceforge.net/lists/listinfo/phonopyusers >  Atsushi Togo http://atztogo.github.com/ atz.togo@... 
From: Aldo Humberto Romero <aromero@qr...>  20130428 15:31:11

I agree with Togo, before you pass the error to PHONOPY, you have to be sure that your calculation has converged successfully. Check if structure and forces are Ok, etc... as TogoSan said, you have to be sure about convergence... if your dynamical is not well converged, then you can find weird problems.. regards aldo. > Hi, > > I guess the values of FORCE_CONSTANTS are something unusual. I may > suggest: > > 1) FORCE_CONSTANTS file is incorrectly generated in some reason. > You can compare FORCE_CONSTANTS and force constants in OUTCAR to check it. > > 2) Force constants are not well calculated in VASP. > It's difficult to help for me on this. But I see one point something > that isn't recommended: NSW=300 should be divided into, e.g., NSW=40 x > some number of VASP runs, because FFT mesh should be regenerated > according to cutoff energy when lattice parameters change largely. > > Togo > > > On Sun, Apr 28, 2013 at 12:09 PM, 张 小东 <zhangxiaodong1028@...> > wrote: >> Dear phonopy users, >> >> I want to calculate the phonon spectrum of Bi2Se3 and got the >> force_constants using DFPT implemented >> in VASP. However, when i calculate the eigenvalue of phonon band, and >> error >> occurred. >> The error is "numpy.linalg.linalg.LinAlgError: Eigenvalues did not >> converge" >> >> The band.conf is: >> ATOM_NAME = Se Bi >> DIM = 2 2 2 >> BAND = 0.0 0.0 0.0 0.50 0.50 0.50 0.50 0.50 0.0 0.0 0.0 0.0 0.0 >> 0.50 >> 0.0 >> EIGENVECTOR = .TRUE. >> FORCE_CONSTANTS = READ >> >> The INCAR is: >> SYSTEM = Bi2Se3 >> >> ## static calculation######### >> ISMEAR = 0;SIGMA=0.005 >> ENCUT=400 >> PREC=Accurate >> ALG0=Normal >> ############################## >> >> ## structure optimization #### >> #NSW=300 >> #IBRION=1 >> #ADDGRID=.TRUE. >> #ISIF=3 >> #EDIFF=1E6 >> #EDIFFG=0.001 >> #POTIM=0.3 >> ############################## >> >> ## Phonon #################### >> IBRION=8 >> NSW=1 >> >> Is anyone has some suggestions? >> >> best wishes, >> Xiaodong Zhang >> Institute of Modern Physics, Northwest University >> >> ps. The error is somewhat related to the numpy. I guess it is the result >> of >> the convergence problem of solving dynamical matrix. >> >> >>  >> Try New Relic Now & We'll Send You this Cool Shirt >> New Relic is the only SaaSbased application performance monitoring >> service >> that delivers powerful full stack analytics. Optimize and monitor your >> browser, app, & servers with just a few lines of code. Try New Relic >> and get this awesome Nerd Life shirt! >> http://p.sf.net/sfu/newrelic_d2d_apr >> _______________________________________________ >> Phonopyusers mailing list >> Phonopyusers@... >> https://lists.sourceforge.net/lists/listinfo/phonopyusers >> > > > >  > Atsushi Togo > http://atztogo.github.com/ > atz.togo@... > >  > Try New Relic Now & We'll Send You this Cool Shirt > New Relic is the only SaaSbased application performance monitoring > service > that delivers powerful full stack analytics. Optimize and monitor your > browser, app, & servers with just a few lines of code. Try New Relic > and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr > _______________________________________________ > Phonopyusers mailing list > Phonopyusers@... > https://lists.sourceforge.net/lists/listinfo/phonopyusers >  Prof. Aldo Humberto Romero CINVESTAVUnidad Queretaro Libramiento Norponiente 2000 CP 76230, Queretaro, QRO, Mexico tel: 442 211 9909 fax: 442 211 9938 email: aromero@... aldorome@... www: qro.cinvestav.mx/~aromero 