From owner-nwchem-users@emsl.pnl.gov Thu Sep 30 01:09:50 2004 Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1]) by odyssey.emsl.pnl.gov (8.12.10/8.12.10) with ESMTP id i8U89o5J011789 for ; Thu, 30 Sep 2004 01:09:50 -0700 (PDT) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.12.10/8.12.10/Submit) id i8U89o2t011788 for nwchem-users-outgoing; Thu, 30 Sep 2004 01:09:50 -0700 (PDT) Date: Thu, 30 Sep 2004 01:09:27 -0700 From: Paul Ha-Yeon Cheong Subject: Summary: DFT Single Point convergence failure "error on integrated density" To: nwchem-users@emsl.pnl.gov Cc: Paul Ha-Yeon Cheong Message-id: <0C7C17D1-12B8-11D9-85A4-000A95D7EC7E@chem.ucla.edu> MIME-version: 1.0 (Apple Message framework v619) X-Mailer: Apple Mail (2.619) Content-type: multipart/alternative; boundary=Apple-Mail-4-112967103 X-Probable-Spam: no X-Spam-Hits: 0.16 X-Scanned-By: smtp.ucla.edu Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk --Apple-Mail-4-112967103 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; format=flowed Dear fellow NWChem users, The question was: "How does one increase the number of maximum iterations in the SCF cycle of a DFT calculation to fix a convergence problem?" The answer is: "specifying maxiter in the scf directive does not influence the maximum number of iterations of DFT SCF cycles. This can be done instead by adding 'iterations #' in the DFT directive." I am adding the original e-mail inquiry, input, and output at the end of this e-mail. Thanks to Edoardo Apra and Erich Vorpagel, I've finally found a solution to this problem. All the best, Paul. --Original e-mail-- I am having a problem trying to converge a custom DFT single point. The output specifies that there is an error on the integrated density on each SCF iteration, and then dies after 30 iterations. My question is: how do you fix the input file so that the single point converges? - I increased the maxiter to 300, but the job continues to die after 30 iterations. - I specified DIISOFF, but this job fails with the same error message. - I have other jobs with different geometries and those jobs complete successfully, so I know it isn't a problem with this particular input's directives. - I prefer not to increase the grid because I am trying to compare this calculation result to others that I have performed previously. I will summarize the responses for the e-mail list. Thank you for your time. Sincerely yours, Paul. ------------------------NWChem Input File------------------------- basis spherical * library 6-31G* end scf thresh 1.0e-8 maxiter 300 end dft xc HFexch 0.218 slater 0.782 becke88 nonlocal 0.542 perdew91 0.167 vwn_1_rpa 0.129 lyp 0.871 grid ssf euler lebedev 75 11 end task dft freq ------------------------NWChem Output File----------------------------- WARNING: error on integrated density = 0.27D-05 greater than required accuracy of 0.10D-06 Memory utilization after 1st SCF pass: Heap Space remaining (MW): 32.85 32846686 Stack Space remaining (MW): 33.55 33550228 convergence iter energy DeltaE RMS-Dens Diis-err time ---------------- ----- ----------------- --------- --------- --------- ------ d= 0,ls=0.0,diis 1 -976.9229705741 -2.93D+03 1.05D-02 3.32D+00 109.1 WARNING: error on integrated density = 0.22D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 2 -975.1721149639 1.75D+00 4.26D-03 1.59D+01 231.5 WARNING: error on integrated density = 0.22D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 3 -976.8396976825 -1.67D+00 2.31D-03 2.80D+00 353.8 WARNING: error on integrated density = 0.22D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 4 -977.1511589082 -3.11D-01 1.36D-03 6.04D-01 474.2 WARNING: error on integrated density = 0.23D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 5 -977.2436289916 -9.25D-02 8.56D-04 1.03D-01 595.1 WARNING: error on integrated density = 0.23D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 6 -977.2711666219 -2.75D-02 6.18D-04 2.93D-02 716.4 WARNING: error on integrated density = 0.23D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 7 -977.2831326016 -1.20D-02 4.94D-04 1.23D-02 836.6 ... (Truncated) d= 0,ls=0.5,diis 30 -977.2978675695 -2.66D-08 7.19D-07 2.55D-08 3633.9 WARNING: error on integrated density = 0.24D-05 greater than required accuracy of 0.10D-06 Calculation failed to converge ------------------------------ Total DFT energy = -977.297867582605 One electron energy = -5221.496085196038 Coulomb energy = 2423.541842221345 Exchange-Corr. energy = -136.248642564666 Nuclear repulsion energy = 1956.905017956754 Numeric. integr. density = 162.000382458087 ... (Truncated) Multipole analysis of the density --------------------------------- L x y z total alpha beta nuclear - - - - ----- ----- ---- ------- 0 0 0 0 0.000000 -81.000000 -81.000000 162.000000 1 1 0 0 2.812720 1.406360 1.406360 0.000000 1 0 1 0 -0.347632 -0.173816 -0.173816 0.000000 1 0 0 1 -0.334935 -0.167468 -0.167468 0.000000 2 2 0 0 -103.714684 -682.634369 -682.634369 1261.554055 2 1 1 0 8.718120 -0.000276 -0.000276 8.718672 2 1 0 1 1.169522 -1.624691 -1.624691 4.418905 2 0 2 0 -85.537279 -1896.601610 -1896.601610 3707.665942 2 0 1 1 -6.322998 16.326583 16.326583 -38.976165 2 0 0 2 -98.396322 -476.535120 -476.535120 854.673917 ------------------------------------------------------------------------ hess_anal: energy failure 555 hess_anal: energy failure 555 ------------------------------------------------------------------------ ------------------------------------------------------------------------ hess_anal: energy failure 555 ------------------------------------------------------------------------ ------------------------------------------------------------------------ hess_anal: energy failure 555 ------------------------------------------------------------------------ ------------------------------------------------------------------------ current input line : 0: ------------------------------------------------------------------------ ------------------------------------------------------------------------ This type of error is most commonly associatated with calculations not reaching convergence criteria ------------------------------------------------------------------------ _______________________________________ Paul Ha-Yeon Cheong Member of the K. N. Houk Group Department of Chemistry and Biochemistry University of California, Los Angeles 607 Charles E. Young Drive, East Los Angeles, CA 90095 "Companions the creator seeks, not corpses, not herds and believers. Fellow creators the creator seeks--those who write new values on new tablets. Companions the creator seeks, and fellow harvesters; for everything about him is ripe for the harvest." - Thus Spoke Zarathustra, Friedrich Nietzsche --Apple-Mail-4-112967103 Content-Transfer-Encoding: 7bit Content-Type: text/enriched; charset=US-ASCII Dear fellow NWChem users, The question was: "How does one increase the number of maximum iterations in the SCF cycle of a DFT calculation to fix a convergence problem?" The answer is: "specifying maxiter in the scf directive does not influence the maximum number of iterations of DFT SCF cycles. This can be done instead by adding 'iterations #' in the DFT directive." I am adding the original e-mail inquiry, input, and output at the end of this e-mail. Thanks to Edoardo Apra and Erich Vorpagel, I've finally found a solution to this problem. All the best, Paul. --Original e-mail-- I am having a problem trying to converge a custom DFT single point. The output specifies that there is an error on the integrated density on each SCF iteration, and then dies after 30 iterations. My question is: how do you fix the input file so that the single point converges? - I increased the maxiter to 300, but the job continues to die after 30 iterations. - I specified DIISOFF, but this job fails with the same error message. - I have other jobs with different geometries and those jobs complete successfully, so I know it isn't a problem with this particular input's directives. - I prefer not to increase the grid because I am trying to compare this calculation result to others that I have performed previously. I will summarize the responses for the e-mail list. Thank you for your time. Sincerely yours, Paul. ------------------------NWChem Input File------------------------- basis spherical * library 6-31G* end scf thresh 1.0e-8 maxiter 300 end dft xc HFexch 0.218 slater 0.782 becke88 nonlocal 0.542 perdew91 0.167 vwn_1_rpa 0.129 lyp 0.871 grid ssf euler lebedev 75 11 end task dft freq ------------------------NWChem Output File----------------------------- WARNING: error on integrated density = 0.27D-05 greater than required accuracy of 0.10D-06 Memory utilization after 1st SCF pass: Heap Space remaining (MW): 32.85 32846686 Stack Space remaining (MW): 33.55 33550228 convergence iter energy DeltaE RMS-Dens Diis-err time ---------------- ----- ----------------- --------- --------- --------- ------ d= 0,ls=0.0,diis 1 -976.9229705741 -2.93D+03 1.05D-02 3.32D+00 109.1 WARNING: error on integrated density = 0.22D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 2 -975.1721149639 1.75D+00 4.26D-03 1.59D+01 231.5 WARNING: error on integrated density = 0.22D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 3 -976.8396976825 -1.67D+00 2.31D-03 2.80D+00 353.8 WARNING: error on integrated density = 0.22D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 4 -977.1511589082 -3.11D-01 1.36D-03 6.04D-01 474.2 WARNING: error on integrated density = 0.23D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 5 -977.2436289916 -9.25D-02 8.56D-04 1.03D-01 595.1 WARNING: error on integrated density = 0.23D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 6 -977.2711666219 -2.75D-02 6.18D-04 2.93D-02 716.4 WARNING: error on integrated density = 0.23D-05 greater than required accuracy of 0.10D-06 d= 0,ls=0.5,diis 7 -977.2831326016 -1.20D-02 4.94D-04 1.23D-02 836.6 ... (Truncated) d= 0,ls=0.5,diis 30 -977.2978675695 -2.66D-08 7.19D-07 2.55D-08 3633.9 WARNING: error on integrated density = 0.24D-05 greater than required accuracy of 0.10D-06 Calculation failed to converge ------------------------------ Total DFT energy = -977.297867582605 One electron energy = -5221.496085196038 Coulomb energy = 2423.541842221345 Exchange-Corr. energy = -136.248642564666 Nuclear repulsion energy = 1956.905017956754 Numeric. integr. density = 162.000382458087 ... (Truncated) Multipole analysis of the density --------------------------------- L x y z total alpha beta nuclear - - - - ----- ----- ---- ------- 0 0 0 0 0.000000 -81.000000 -81.000000 162.000000 1 1 0 0 2.812720 1.406360 1.406360 0.000000 1 0 1 0 -0.347632 -0.173816 -0.173816 0.000000 1 0 0 1 -0.334935 -0.167468 -0.167468 0.000000 2 2 0 0 -103.714684 -682.634369 -682.634369 1261.554055 2 1 1 0 8.718120 -0.000276 -0.000276 8.718672 2 1 0 1 1.169522 -1.624691 -1.624691 4.418905 2 0 2 0 -85.537279 -1896.601610 -1896.601610 3707.665942 2 0 1 1 -6.322998 16.326583 16.326583 -38.976165 2 0 0 2 -98.396322 -476.535120 -476.535120 854.673917 ------------------------------------------------------------------------ hess_anal: energy failure 555 hess_anal: energy failure 555 ------------------------------------------------------------------------ ------------------------------------------------------------------------ hess_anal: energy failure 555 ------------------------------------------------------------------------ ------------------------------------------------------------------------ hess_anal: energy failure 555 ------------------------------------------------------------------------ ------------------------------------------------------------------------ current input line : 0: ------------------------------------------------------------------------ ------------------------------------------------------------------------ This type of error is most commonly associatated with calculations not reaching convergence criteria ------------------------------------------------------------------------ Helvetica3939,3939,3939_______________________________________ Paul Ha-Yeon Cheong Member of the K. N. Houk Group Department of Chemistry and Biochemistry University of California, Los Angeles 607 Charles E. Young Drive, East Los Angeles, CA 90095Futura3939,3939,3939 Frutiger 45 Light3939,3939,3939 "Companions the creator seeks, not corpses, not herds and believers. Fellow creators the creator seeks--those who write new values on new tablets. Companions the creator seeks, and fellow harvesters; for everything about him is ripe for the harvest." - Thus Spoke Zarathustra, Friedrich Nietzsche --Apple-Mail-4-112967103--