[CP2K-user] [CP2K:14782] Re: High cutoff for Force Converage

[Hedley Dong]

Thanks for your advice, Lucas. I have tried decrease EPS_SCF to 3.0E-7, and 
the force did not change compared to EPS_SCF=5.0E-6 at the same cutoff. I 
will try a lower EPS_SCF and take a look of XC_GRID.
 
在2021年2月16日星期二 UTC+8 下午12:08:05<Lucas Lodeiro> 写道：

> Hi Hedley,
>
> Typically forces are more difficult to converge. Forces are more sensitive 
> to convergence parameters, and are affected by "noise" in the derivation. 
> Have you tried to decrease the EPS_SCF? Maybe 5E-6 is not sufficient to 
> ensure a free noise force... 1E-7 or 1E-8 could be useful to smooth the 
> force profile respect to cutoff. And the XC_GRID subsection could be useful.
>
> Regards - Lucas Lodeiro
>
> El lun, 15 feb 2021 a las 23:57, Hedley Dong (<dm... at gmail.com>) 
> escribió:
>
>> [image: 1.PNG]
>>
>> 在2021年2月16日星期二 UTC+8 上午10:44:48<Hedley Dong> 写道：
>>
>>> &GLOBAL
>>> PROJECT DPGEN
>>> &END GLOBAL
>>> &FORCE_EVAL
>>> METHOD QS
>>> STRESS_TENSOR ANALYTICAL
>>> &DFT
>>> BASIS_SET_FILE_NAME BASIS_MOLOPT
>>> POTENTIAL_FILE_NAME GTH_POTENTIALS
>>> CHARGE 0
>>> UKS T
>>> MULTIPLICITY 1
>>> &MGRID
>>> CUTOFF 350
>>> REL_CUTOFF 50
>>> NGRIDS 4
>>> &END MGRID
>>> &QS
>>> EPS_DEFAULT 1.0E-12
>>> &END QS
>>> &SCF
>>> SCF_GUESS ATOMIC
>>> EPS_SCF 5.0E-6
>>> MAX_SCF 300
>>> &OT
>>> PRECONDITIONER FULL_SINGLE_INVERSE
>>> MINIMIZER DIIS
>>> &END OT
>>> &OUTER_SCF ON
>>> MAX_SCF 4
>>> &END OUTER_SCF
>>> &END SCF
>>> &XC
>>> &XC_FUNCTIONAL PBE
>>> &END XC_FUNCTIONAL
>>> &vdw_POTENTIAL
>>> DISPERSION_FUNCTIONAL PAIR_POTENTIAL
>>> &PAIR_POTENTIAL
>>> TYPE DFTD3
>>> PARAMETER_FILE_NAME dftd3.dat
>>> REFERENCE_FUNCTIONAL PBE
>>> R_CUTOFF 15.
>>> &END PAIR_POTENTIAL
>>> &END vdw_POTENTIAL
>>> &END XC
>>> &END DFT
>>> &SUBSYS
>>> &CELL
>>> A 7.552 0.    0.   
>>> B 0.    7.552 0.   
>>> C  0.     0.    18.972
>>> &END CELL
>>> &COORD
>>> @include coord.xyz
>>> &END COORD
>>> &KIND O
>>> BASIS_SET DZVP-MOLOPT-SR-GTH
>>> POTENTIAL GTH-PBE-q6
>>> &END
>>> &KIND Ti
>>> POTENTIAL GTH-PBE-q12
>>> BASIS_SET DZVP-MOLOPT-SR-GTH
>>> &DFT_PLUS_U
>>> EPS_U_RAMPING 1.0E-3
>>> U_RAMPING [eV] 1
>>> L 2
>>> U_MINUS_J [eV] 5
>>> &END DFT_PLUS_U
>>> &END
>>> &END SUBSYS
>>> &PRINT
>>> &FORCES ON
>>> &END FORCES
>>> &STRESS_TENSOR ON
>>> &END STRESS_TENSOR
>>> &END PRINT
>>> &END FORCE_EVAL
>>>
>>> -- 
>> 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 cp... at googlegroups.com.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/cp2k/8485d638-b1d6-4586-b761-fe0193384ac4n%40googlegroups.com 
>> <https://groups.google.com/d/msgid/cp2k/8485d638-b1d6-4586-b761-fe0193384ac4n%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210215/5a341ef9/attachment.htm>