Hi, Emilio
Actually, I have defined the variable FilterCutoff in my input file, but I do not know how to choose the value for it. I just know it should be smaller than the value of MeshCutoff (for example, if I choose 400 Ry for MeshCutoff, then how about the FilterCutoff). So, the same situation for the variable of FitlerTol, how can I choose a value for it. The input and output files are attached in my former email. Any comment or suggestion is appreciated. Best Fei Mao -----原始邮件----- 发件人: "Emilio Artacho" <[email protected]> 发送时间: 2012年11月13日 星期二 收件人: [email protected] 抄送: 主题: Re: [SIESTA-L] How to fix the eggbox-effect? Dear Fei You could also use filtering (see e.g. FilterTol) Emilio On 13 Nov 2012, at 11:11, 毛飞 wrote: Hi, all Thank you for your kind reply. I also have learned the script eggbox_checker.py in the user’s guide, but I do not know how to use it. Can it be compiled and to be generated as a executable? In the guide, it does not talk too much about how to use this script, I do not know. And when I obtain the Fourier components for each species by the script, I have to define the variable EggboxRemove. In addition, you suggested the use of the block of GridCellSampling. So is there any relationships between these two variables? And do I have to use the two variables together? Any comment or suggestion is appreciated. Thank you in advance. Fei Mao Beijing Normal University, China -----原始邮件----- 发件人: "Ankush Rai" <[email protected]> 发送时间: 2012年11月13日 星期二 收件人:[email protected] 抄送: 主题: Re: [SIESTA-L] How to fix the eggbox-effect? Use the attached script to diagnose the eggbox effect. You can also use grid sampling. Add following lines to your .fdf file %block GridCellSampling 0.5 0.5 0.0 0.5 0.0 0.5 0.0 0.5 0.5 %endblock GridCellSampling care should be taken that grid sampling will need to be modified as per the system. I hope it will help. Best regard..... On Mon, Nov 12, 2012 at 8:24 AM, 毛飞 <[email protected]> wrote: Hi, all Have you ever met the eggbox effect problem before when you do the geometry optimization with Siesta, and how do you fix it. When I optimize the geometry of graphene which is absorbed with a Mg （Magnesium） atom, I encounter the problem, In my case, the value of Max force vibrates up and down, which do not go to convergence. I learn to the user’s guide, it is said that it may fix the problem just by defining the variable FilterCutoff, but I do not know how to define the suitable value for that variable (As I know, it should lower than the MeshCutoff. In my case, the MeshCutoff is 360 Ry, so how about the FilterCutoff? I have attached the input file and output file). Any comment or suggestion is appreciated. Thank you. Fei Mao Beijing Normal University, China -- Emilio Artacho CIC nanoGUNE Consolider, and Cavendish Laboratory, University of Cambridge Tolosa Hiribidea 76, E-20018 Donostia - San Sebastián, Spain, [email protected], +34 943 574039, http://theory.nanogune.eu
