https://groups.google.com/g/cp2k

[Radmir Khisamov]

Hi, everyone!

I am begginer at cp2k calculations. I am trying to optimize the geometries of coordination compounds with Cu, Cl, P, N, S and H atoms. During optimization, after the first SCF cycle, I get the CPASSERT error at build_overlap_matrix_low routine. Can anyone suggest what I'm doing wrong?

Input file part:

&FORCE_EVAL
 METHOD QS
 &DFT
   CHARGE 0
   MULTIPLICITY 1
  &QS
   METHOD GPW ! default
   EPS_DEFAULT 1.0E-12
  &END QS
  &MGRID
   NGRIDS 4
   CUTOFF 600
   REL_CUTOFF 80
  &END MGRID
  &SCF
   SCF_GUESS RESTART
   EPS_SCF 1.0E-06
   MAX_SCF 300
   &OT
    MINIMIZER DIIS
    PRECONDITIONER FULL_SINGLE_INVERSE
   &END
   &OUTER_SCF
    EPS_SCF 1.0E-7
    MAX_SCF 10
   &END
   &PRINT
     &RESTART ON
     &END RESTART
  &END SCF
  &XC
   &XC_FUNCTIONAL
     &MGGA_XC_B97M_V T
     &END
   &END XC_FUNCTIONAL
   &VDW_POTENTIAL
    POTENTIAL_TYPE NON_LOCAL
     &NON_LOCAL
       TYPE RVV10
       KERNEL_FILE_NAME rVV10_kernel_table.dat
     &END NON_LOCAL
   &END VDW_POTENTIAL
  &END XC
 &END DFT
 &SUBSYS

Error:

 *******************************************************************************
 *   ___                                                                       *
 *  /   \                                                                      *
 * [ABORT]                                                                     *
 *  \___/                             CPASSERT failed                          *
 *    |                                                                        *
 *  O/|                                                                        *
 * /| |                                                                        *
 * / \                                                        qs_overlap.F:415 *
 *******************************************************************************


 ===== Routine Calling Stack =====

            8 build_overlap_matrix_low
            7 build_core_hamiltonian_matrix_forces
            6 qs_forces
            5 cp_eval_at
            4 cp_opt_gopt_step
            3 geoopt_lbfgs
            2 cp_geo_opt
            1 CP2K

[Krack Matthias (PSI)]

Hi

 

You have to provide a full input, preferentially as simple and small as possible, which gives that error.

 

M.

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/70f8ff6f-f548-414a-944c-8520004126d0n%40googlegroups.com.

[Radmir Khisamov]

Hi

I get such errors in various calculations. For example, in calculations with hybrid and with meta-GGA functionals. At the same time optimizations of organic molecules with light elements do not give the error. 

пятница, 25 ноября 2022 г. в 15:55:37 UTC+7, Matthias Krack:

[Krack Matthias (PSI)]

The atomic coordinates are missing.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/7308da48-b332-44a4-a412-0ce80da89d24n%40googlegroups.com.

[Radmir Khisamov]

I'm sorry. I thought that the input files contain an error.

пятница, 25 ноября 2022 г. в 17:06:15 UTC+7, Matthias Krack:

[Krack Matthias (PSI)]

Well, 312 atoms with PBE0 is not really a small system.

I suggest to try the following:

• Drop EPS_FILTER_MATRIX or set the default value 0.0 for it
• Reduce EPS_PGF_ORB to something much less, e.g. 1.0E-20
• Restart from a wavefunction converged with PBE

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/1b0209f4-6a2d-44ff-8d48-d61e282a5982n%40googlegroups.com.