<html>
  <head>
    <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">Hi,<br>
      <br>
      I think you misinterpreted the code in ewald_LRcorrection. It does
      the same things in 3.3.1 and 4.6.7.<br>
      My guess is that the issue you are observing is due to periodic
      boundary conditions. With group cut-off scheme charge groups are
      whole (not split by PBC), so water molecules are usually whole.
      Then the 3DC correction always works correctly when you have
      charge groups with zero net charge. In the Verlet cut-off scheme
      with domain decomposition, or with charged charge groups,
      (partial) charges moving over PBC change the system dipole. To get
      your system to run correctly you need to ensure that no water
      molecule moves over the periodic boundary in along the
      z-dimension. I think grompp will print a warning for this issue.<br>
      <br>
      Cheers,<br>
      <br>
      Berk<br>
      <br>
      On 2016-06-07 10:12, Sheng Bi wrote:<br>
    </div>
    <blockquote cite="mid:5756863B.019A35.28023@mail.hust.edu.cn"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      <meta name="Generator" content="Microsoft Word 15 (filtered
        medium)">
      <style><!--
/* Font Definitions */
@font-face
        {font-family:"Cambria Math";
        panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
        {font-family:DengXian;
        panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
        {font-family:Calibri;
        panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
        {font-family:"\@DengXian";
        panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
        {margin:0in;
        margin-bottom:.0001pt;
        font-size:11.0pt;
        font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
        {mso-style-priority:99;
        color:blue;
        text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
        {mso-style-priority:99;
        color:#954F72;
        text-decoration:underline;}
.MsoChpDefault
        {mso-style-type:export-only;}
@page WordSection1
        {size:8.5in 11.0in;
        margin:1.0in 1.25in 1.0in 1.25in;}
div.WordSection1
        {page:WordSection1;}
--></style>
      <div class="WordSection1">
        <p class="MsoNormal">Sorry, some type-offs in previous question.</p>
        <p class="MsoNormal">In gmx-4.6.7, ewald_LRcorrection()<b>.</b>
          We loop over excluded neighbours only in condition <b>calc_excl_corr=1</b>.
          But when using Verlet cut-off scheme, <b>calc_excl_corr will
            be 0</b>.</p>
        <p class="MsoNormal"><o:p> </o:p></p>
        <p class="MsoNormal"><o:p> </o:p></p>
        <p class="MsoNormal">--<br>
          PhD student<br>
          State Key Laboratory of Coal Combustion,<br>
          School of Energy and Power Engineering,<br>
          Huazhong University of Science and Technology (HUST),<br>
          Room 302 Power Building, <br>
          1037 Luoyu Road, Wuhan, Hubei 430074 China<br>
          Email: <a class="moz-txt-link-abbreviated" href="mailto:chrishengbee@hust.edu.cn">chrishengbee@hust.edu.cn</a><br>
          Phone: (86)27-87548122 <br>
          <a class="moz-txt-link-freetext" href="http://itp.energy.hust.edu.cn">http://itp.energy.hust.edu.cn</a> </p>
        <p class="MsoNormal"><span
            style="font-size:12.0pt;font-family:&quot;Times New
            Roman&quot;,serif"><o:p> </o:p></span></p>
        <div
          style="mso-element:para-border-div;border:none;border-top:solid
          #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
          <p class="MsoNormal" style="border:none;padding:0in"><b>From:
            </b><a moz-do-not-send="true"
              href="mailto:chrishengbee@hust.edu.cn">Sheng Bi</a><br>
            <b>Sent: </b>Tuesday, June 7, 2016 1:28 AM<br>
            <b>To: </b><a moz-do-not-send="true"
              href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br>
            <b>Subject: </b>[gmx-developers] Question about
            ewald_LRcorrection() function</p>
        </div>
        <p class="MsoNormal"><span
            style="font-size:12.0pt;font-family:&quot;Times New
            Roman&quot;,serif"><o:p> </o:p></span></p>
        <p class="MsoNormal">Dear GMX developers:</p>
        <p class="MsoNormal">               I find a abnormal phenomenon
          when using <b>gromacs-4.6.7</b>. I set up a system containing
          two walls(consist of freeze atoms), one wall has positively
          charges on atoms, another has negatively charges.  I put an
          water molecule(rigid spc/spce/tip3p model) between these two
          walls. <b>Whole system seems like below</b>:</p>
        <p class="MsoNormal">              
          ****************************************</p>
        <p class="MsoNormal">               * -         
          H                          
          +                                          *             </p>
        <p class="MsoNormal">               * -             \         
                         +            
          vacuum                              *</p>
        <p class="MsoNormal">              * -            
            O                    
          +                                          *</p>
        <p class="MsoNormal">               * -           
           /                       
          +                                          *</p>
        <p class="MsoNormal">               * -          H           
                         +                                          *</p>
        <p class="MsoNormal">              
          ****************************************</p>
        <p class="MsoNormal">               I use <b>Verlet cut-off
            scheme, PME method, and ewald-geometry=3DC</b>. Temperature
          of whole system is at <b>0K</b>, controlled by <b>v-rescale
            method</b>. I run whole simulation <b>only in OPENMP
            parallel</b>.</p>
        <p class="MsoNormal">               One can have following
          expected behavior of water molecule without MD simulation:
          Since it can be seen as an infinite panel capacitor, <b>there
            is constant electric field between two walls. So the water
            will change it orientation, but won’t change it’s
            position(Since 0K)</b>.</p>
        <p class="MsoNormal">               This phenomenon can be <b>confirmed
            by gromcas-3.3.1</b> (although there is position shift, but
          quite small). Also, if one use higher version of gromacs, such
          as 4.6.7 or higher <b>with ewald-geometry=3D, either no
            problem</b>. However, if one use <b>3DC</b> (and it should
          be), you can see <b>an obvious force on water molecule</b>.</p>
        <p class="MsoNormal">               I hacked into source code,
          and find this might due to <b>ewald_LRcorrection()</b>
          function:</p>
        <p class="MsoNormal">               Firstly, in gmx-4.6.7,
          ewald_LRcorrection()<b>.</b> We <b>loop over excluded
            neighbours only in condition calc_excl_corr=0. But when
            using verlet cut-off scheme, calc_excl_corr will be 1</b>.
          In gmx-3.3.1, we do loop over excluded neighbours without
          additional if() statement.</p>
        <p class="MsoNormal">               Secondly, in gmx-4.6.7, ,
          ewald_LRcorrection(). When do dipole correction on force, we
          loop over <b>excl_load</b> (Exclusion load distribution over
          the threads). In fact, most water model will have nrexel=2.
          So, as a result, we <b>only do dipole correction on O atom,
            but do nothing on H atoms</b> and we will see an obvious
          force on the whole water molecule. I think this is not purpose
          of 3DC<span style="font-family:DengXian" lang="ZH-CN">。</span></p>
        <p class="MsoNormal">               Best regards<span
            style="font-family:DengXian" lang="ZH-CN">!</span></p>
        <p class="MsoNormal">Sheng Bi</p>
        <p class="MsoNormal"><b>               </b></p>
        <p class="MsoNormal">---<br>
          PhD student<br>
          State Key Laboratory of Coal Combustion,<br>
          School of Energy and Power Engineering,<br>
          Huazhong University of Science and Technology (HUST),<br>
          Room 302 Power Building, <br>
          1037 Luoyu Road, Wuhan, Hubei 430074 China<br>
          Email: <a class="moz-txt-link-abbreviated" href="mailto:chrishengbee@hust.edu.cn">chrishengbee@hust.edu.cn</a><br>
          Phone: (86)27-87548122 <br>
          <a class="moz-txt-link-freetext" href="http://itp.energy.hust.edu.cn">http://itp.energy.hust.edu.cn</a> </p>
        <p class="MsoNormal"><span
            style="font-size:12.0pt;font-family:&quot;Times New
            Roman&quot;,serif"><o:p> </o:p></span></p>
        <p class="MsoNormal"><o:p> </o:p></p>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
    </blockquote>
    <br>
  </body>
</html>