[CP2K-user] [CP2K:13320] SCF Convergence stuck with ADMM HF?

hut... at chem.uzh.ch hut... at chem.uzh.ch
Sat May 16 11:34:27 UTC 2020


Hi

I'm sorry, but I cannot find that specific basis set anymore.
I would have to dig into the archives.
Maybe Joost or Manuel still have it somewhere accessible.

regards

Juerg
--------------------------------------------------------------
Juerg Hutter                         Phone : ++41 44 635 4491
Institut für Chemie C                FAX   : ++41 44 635 6838
Universität Zürich                   E-mail: hut... at chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp... at googlegroups.com wrote: -----
To: "cp2k" <cp... at googlegroups.com>
From: "Nicholas Winner" 
Sent by: cp... at googlegroups.com
Date: 05/15/2020 10:40PM
Subject: Re: [CP2K:13320] SCF Convergence stuck with ADMM HF?

Jurg, 

Reducing EPS_SCHWARZ did help quite a bit in the convergence. In your original paper with Manuel on the ADMM, diamond was one of the case studies and it was stated that: 

Since the condition number of the overlap matrix with the FIT3 basis is unfavorable in the case of bulk C (1.7 × 105), an optimized FIT3 (optFIT3) basis has been constructed that served as the ABS for the two wave function fitting methods. optFIT3 was obtained by minimization of the total energy of the PBE 2 × 2 × 2 supercell with respect to the constraint of a well behaved overlap matrix (the final condition number is on the order of 102). This allows for rather loose screening thresholds (10-6)
 
Which is why I believed there was no issue with EPS_SCHWARZ 1e-6. As I mentioned in response to Matt, I cannot find an aux_basis set in the BASIS_ADMM or BASIS_ADMM_MOLOPT files of the name "optFIT3", which is why I turned to cFIT3. Is it possible to get this optimized basis? If not, I will defer to using 1e-7 for the more ill-conditioned problems.

Best,
Nick

On Friday, May 15, 2020 at 4:42:17 AM UTC-7, jgh wrote:Hi 
 
this behaviour usually points towards a lack of overall accuracy 
in the calculation. In your case, I would decrease 
 
EPS_DEFAULT   and  EPS_SCHWARZ 
 
You might also want not to screen HFX on initial P when you start 
an SCF from an inaccurate initial guess. Screened away integrals  
are lost for the whole SCF procedure. 
 
regards 
 
Juerg Hutter 
 
-------------------------------------------------------------- 
Juerg Hutter                         Phone : ++41 44 635 4491 
Institut für Chemie C                FAX   : ++41 44 635 6838 
Universität Zürich                   E-mail: h... at chem.uzh.ch 
Winterthurerstrasse 190 
CH-8057 Zürich, Switzerland 
--------------------------------------------------------------- 
 
-----c... at googlegroups.com wrote: ----- 
To: "cp2k" <c... at googlegroups.com> 
From: "Nicholas Winner"  
Sent by: c... at googlegroups.com 
Date: 05/14/2020 10:14PM 
Subject: [CP2K:13305] SCF Convergence stuck with ADMM HF? 
 
Hello all, 
 
I have tried to perform static PBE0 calculation of diamond using the ADMM. I have done many of this type of calculation before, but not for this system. I have an issue where the reported "convergence" in the SCF loop is unchanged for many steps. I initially tried using the DIIS minimizer and OT method, but switched to 2PNT CG search to try to improve convergence. This helped a tiny amount (convergence got lower, but still stuck).  
 
I'm unclear as to what the problem is. I have attached my input and output files. Note I stopped this calculation prematurely. 
 
Drawing your attention to the first inner SCF loop of the output file, I can see that SCF step 19-50 all report the same convergence of (0.00000035) with no change. What's more, the "total energy" in that loop is listed as -3903.4843979582 for all steps after 31. So it seems that the energy is extremely well converged... better than the EPS_SCF 1e-7 that I listed in the input file. 
 
I'd like to know if there's any reason why this would be having such a hard time converging; diamond is a large band gap material and should have very good convergence with OT, especially with CG. Also, am I misunderstanding hte numbers reported by the SCF steps? It seems to me the total energy is very well converged to within 1e-9 the end of the first SCF loop, but convergence is listed as .00000035. 
 
Any help would be appreciated. 
 
Best, 
Nick   
  --  
 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 c... at googlegroups.com. 
 To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/c970a006-97f7-4f1e-90de-d5947d22afd2%40googlegroups.com. 
  
 
[attachment "cp2k.inp" removed by Jürg Hutter/at/UZH] 
[attachment "cp2k.out.orig" removed by Jürg Hutter/at/UZH] 
  
  -- 
 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/dadc7f10-683b-4578-854c-6087fbe7ccb2%40googlegroups.com.




More information about the CP2K-user mailing list