<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p>Hi Sean,</p>
    <p>I'd set both of those to .false., I'll try a version with .true.
      I've attached my flash.par file, if useful I can also send my much
      longer .log file (which includes many failed iterations of
      modifying these parameters). What is the recommended level of
      magnetic resistivity to minimize this?</p>
    <p>Thanks,</p>
    <p>Rukmani<br>
    </p>
    <br>
    <div class="moz-cite-prefix">On 10/09/2017 09:43 AM, Sean M. Couch
      wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:etPan.59db7cf0.4c043f8b.66bc@pa.msu.edu">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      <style>body{font-family:Helvetica,Arial;font-size:13px}</style>
      <div id="bloop_customfont"
        style="font-family:Helvetica,Arial;font-size:13px; color:
        rgba(0,0,0,1.0); margin: 0px; line-height: auto;">
        I’d agree with Dongwook, it looks like numerical instability. In
        absence of the log file showing the details of your setup and
        run, I would suggest you try the following if you haven’t
        already. Set both E_upwind and E_modification to .TRUE. in your
        flash.par. What these flags do is improve the order of
        reconstruction for the electric fields (see Lee & Deane
        2009).  See the attached gif of a field loop moving across a
        jump in refinement with the various flags on or off.  Not
        exactly the same as your issue, perhaps, but might still be
        worth a try. You might also try adding in magnetic resistivity
        to your sim, if all else fails. </div>
      <div id="bloop_customfont"
        style="font-family:Helvetica,Arial;font-size:13px; color:
        rgba(0,0,0,1.0); margin: 0px; line-height: auto;">
        <br>
      </div>
      <div id="bloop_customfont"
        style="font-family:Helvetica,Arial;font-size:13px; color:
        rgba(0,0,0,1.0); margin: 0px; line-height: auto;">
        <img src="cid:part1.DD5AE1D1.D72BD91E@virginia.edu" class=""></div>
      <div id="bloop_sign_1507556450582258944" class="bloop_sign">Cheers,
        <div>Sean</div>
      </div>
      <br>
      <p class="airmail_on">On October 9, 2017 at 12:07:50 AM, Dongwook
        Lee (<a href="mailto:dongwook@flash.uchicago.edu"
          moz-do-not-send="true">dongwook@flash.uchicago.edu</a>) wrote:</p>
      <blockquote type="cite" class="clean_bq"><span>
          <div>
            <div>
              <title></title>
              <div dir="ltr">Dear Rukmani,
                <div><br>
                </div>
                <div>It looks to me that they may come from numerical
                  instability, although it could be an issue related to
                  the use of AMR. What reconstruction scheme and CFL
                  number are you using? Is this 2D or 3D?</div>
                <div><br>
                </div>
                <div>Best,</div>
                <div>Dongwook</div>
              </div>
              <div class="gmail_extra"><br>
                <div class="gmail_quote">On Sun, Oct 8, 2017 at 12:59
                  PM, Rukmani Vijayaraghavan <span dir="ltr">
                    <<a href="mailto:rukmani@virginia.edu"
                      target="_blank" moz-do-not-send="true">rukmani@virginia.edu</a>></span>
                  wrote:<br>
                  <blockquote class="gmail_quote" style="margin:0 0 0
                    .8ex;border-left:1px #ccc solid;padding-left:1ex">
                    Hi everyone,<br>
                    <br>
                    I get a strange, unphysical feature in the magnetic
                    field structure with the USM solver, seen in the
                    attached slices of B_z and the magnetic pressure.
                    These features, as you can see, track the block
                    boundaries and look like a checkerboard of varying
                    magnetic field strength (and sign?) in individual
                    grid cells. I've tried modifying every possible USM
                    solver switch, including the Riemann solver, order,
                    slope limiter, prolongation method, steepening, etc.
                    Using Riemann Solver = HLLC instead of HLLD makes
                    things a little better, since HLLC is more
                    diffusive, but the feature still eventually appears.
                    Any advice on solving this or even minimizing its
                    effects would be helpful!<br>
                    <br>
                    Thanks,<br>
                    <br>
                    Rukmani<span class="HOEnZb"><font color="#888888"><br>
                        <br>
                        --<br>
                        Rukmani Vijayaraghavan<br>
                        NSF Astronomy & Astrophysics Postdoctoral
                        Fellow<br>
                        University of Virginia<br>
                        <a href="mailto:rukmani@virginia.edu"
                          target="_blank" moz-do-not-send="true">rukmani@virginia.edu</a><br>
                        <br>
                      </font></span></blockquote>
                </div>
                <br>
                <br clear="all">
                <div><br>
                </div>
                --<br>
                <div class="gmail_signature"
                  data-smartmail="gmail_signature">
                  <div dir="ltr">
                    <div style="font-size:12.8px"><font size="2"
                        color="#000000" face="arial, helvetica,
                        sans-serif"><br>
                        =========================================</font></div>
                    <div style="font-size:12.8px"><font size="2"
                        color="#000000" face="arial, helvetica,
                        sans-serif">Dongwook Lee, Ph.D., Assistant
                        Professor</font></div>
                    <div style="font-size:12.8px"><font size="2"
                        color="#000000" face="arial, helvetica,
                        sans-serif">Applied Mathematics and Statistics<br>
                      </font></div>
                    <div style="font-size:12.8px"><font size="2"
                        color="#000000" face="arial, helvetica,
                        sans-serif">University of California, Santa Cruz</font></div>
                    <div style="font-size:12.8px"><font size="2"
                        color="#000000" face="arial, helvetica,
                        sans-serif">Baskin Engineering, Room 353C</font></div>
                    <div style="font-size:12.8px"><font size="2"
                        color="#000000" face="arial, helvetica,
                        sans-serif">1156 High Street, Santa Cruz, CA
                        95064</font></div>
                    <div style="font-size:12.8px"><a
                        href="https://users.soe.ucsc.edu/%7Edongwook/"
                        target="_blank" moz-do-not-send="true"><font
                          size="2" color="#000000" face="arial,
                          helvetica, sans-serif">https://users.soe.ucsc.edu/~dongwook/</font></a></div>
                    <div><br>
                    </div>
                  </div>
                </div>
              </div>
            </div>
          </div>
        </span></blockquote>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
Rukmani Vijayaraghavan
NSF Astronomy & Astrophysics Postdoctoral Fellow
University of Virginia
<a class="moz-txt-link-abbreviated" href="mailto:rukmani@virginia.edu">rukmani@virginia.edu</a></pre>
  </body>
</html>