[CP2K-user] [CP2K:18260] CP2K exercise for converging the cutoff

[Dragoon]

Hi Dobromir and Thomas,

Thank you all for the replies. 
As I dug through the threads in this group. I realized many problems of the 
CP2k Howto <https://www.cp2k.org/howto:converging_cutoff> and exercise 
<https://www.cp2k.org/events:2018_summer_school:converging_cutoff> about PW 
cutoff. Please correct me for my misunderstandings. 
1. Based on one previous discussion 
<https://groups.google.com/g/cp2k/c/ySUAYEgwmhc/>, SCF needs to be 
converged with a large enough REL_CUTOFF, instead of a 1 step calculation 
like in the Howto. 
2. EPS_PGF_ORB 1.0E-07 or lower is usually needed for converging SCF.
3. 1e-3 ev / 3e-5 Hartree energy convergence is good enough for most 
system? For this convergence the exercise system need to go to 800 Ry 
cutoff.
4. GAPW sometimes can achieve a much smaller cutoff than GPW method. The 
default parameters in CP2k are good enough for GAPW.

Thank you again,
Zhuoran 




在2022年12月26日星期一 UTC-5 04:12:08<DMT> 写道：

> In many publications a CUTOFF of 400 is normal and acceptable. I haven't 
> seen anyone to have to go beyond 600, so far. Also, a very good value for 
> NGRIDS is 5.
>
> Dobromir
>
> On Monday, December 26, 2022 at 10:54:36 AM UTC+2 tkuehne wrote:
>
>> Dear Zhuoran, 
>>
>> you can find more detailed informations at: 
>> https://www.cp2k.org/howto:converging_cutoff
>> A very rough a priori estimate for the cutoff can be found at: 
>> https://www.cp2k.org/faq:cutoff
>>
>> Best, 
>> Thomas Kühne
>>
>> Am 19.12.2022 um 06:26 schrieb Dragoon <longdr... at gmail.com>:
>>
>> Hi all,
>>
>> As I'm still struggling with converging the plane wave cutoff, I tried 
>> this CP2K exercise for converging the cutoff for a water system (the link 
>> is https://www.cp2k.org/events:2018_summer_school:converging_cutoff ). 
>> This system has 32 H2O, using PBE functional (no XC smoothing) and 
>> MOLOPT-SR basis. REL_CUTOFF is 60 Ry.
>>
>> I ran this exercise, and extended the cutoff to 2000Ry, yet I'm not sure 
>> at which cutoff either the energy or force is converged. I attached the 
>> result below. <Screenshot from 2022-12-19 00-01-47.png>
>>
>> I do notice that the most coarse grid (#4) has the most grid levels 
>> starting from 800 Ry. I guess that means the cutoff is probably already too 
>> high, and that might be related to a slow convergence. Yet I'm not sure how 
>> to improve that, nor further explain this behavior for this simple system.
>> I really appreciate any explanations and helps.
>>
>> Zhuoran
>>
>> -- 
>> 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... at googlegroups.com.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/cp2k/aa6d141f-3e25-4365-8cbd-f482217bedf0n%40googlegroups.com 
>> <https://groups.google.com/d/msgid/cp2k/aa6d141f-3e25-4365-8cbd-f482217bedf0n%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>> <Screenshot from 2022-12-19 00-01-47.png>
>>
>>
>>
>>
>> ==============================
>> Thomas D. Kühne
>> Dynamics of Condensed Matter
>> Chair of Theoretical Chemistry
>> University of Paderborn
>> Warburger Str. 100
>> D-33098 Paderborn
>> Germany
>> thomas... at upb.de
>> +49/(0)5251/60-5726 <+49%205251%20605726>
>>
>>

-- 
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+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/fda69b0e-7b5d-44a6-8f32-28bd5e56a750n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221226/7b883661/attachment-0001.htm>