[Pw_forum] Convergence with respect to ecutwfc

GAO Zhe flux_ray12 at 163.com
Tue Jun 28 11:15:14 CEST 2011


Normally, ecutrho = ecutwfc * 4 is as the default setting of PWscf. However, it is suggetsted that ecutrho be in the range of 8~12 times of ecutwfc. Therefore, when you were doing ecutwfc vs total energy convergence test, please set ecutrho, too.

--
GAO Zhe
CMC Lab, MSE, SNU, Seoul, S.Korea


At 2011-06-27 13:25:24,"priyanka goud" <priyankagoud8 at gmail.com> wrote:

Dear PWSCF users,

                                     Iam calculating band structure of an organic crystal.For the same I have performed scf calculatiions
with different cutoff.Results are as follows.
                       

 

 

 

ecutwfc

 

 

 

Total Energy

 

 

 

Difference

 10

-923.6688

 

15

-945.1988

    21.53

20

-953.0299

   7.8311

25

-954.8061

   1.7762

30

-955.0794

   0.2733

35

-955.1120

   0.0326

40

-955.1538

   0.0418

45

-955.2164

   0.0626

50

-955.2699

   0.0535

55

-955.3043

   0.0344

60

-955.3239

   0.0196

65

-955.3352

   0.0113

70

-955.3416

   0.0064

75

-955.3475

   0.0059

80

-955.3571

   0.0096

85

-955.3725

  0.0154

                         
            Kindly suggest me upto which decimal point should I see the total energy for convergence. And here which cutoff can I take for post processing (nscf) calculations.I am attaching the plot ofecutwfc verses Total energy.

 Thanks In advance .     
                
     
B.Priyanka
Research Scholar,
Computational Lab,
IICT,INDIA .


                   
         
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20110628/bbf24ee9/attachment.html>


More information about the users mailing list