<p dir="ltr">Hello again,</p>
<p dir="ltr">I might be wrong, but unless there's an experimental Extrude feature that allows tapered/extrusion, I don't think it is possible. You could write directly to Christophe to check this, and maybe suggest the feature, if it's not already under development.</p>
<p dir="ltr">What I did is this:<br>
* avoid Extrude. Commented the lines in your .GEO.<br>
* use a dilate on a duplicate of the original circle, setting the Geometry. CopyMeshMethod=1, so it is also transfinite<br>
* Added manually the lines, surfaces and volumes, setting them transfinite.</p>
<p dir="ltr">I only did it with one quadrant, on the GUI. But it is probably possible to obtain the vertex and line numbers of the copied/dilated surface using a combination of the array[ ] and probably the Boundary{ } function.</p>
<p dir="ltr">I won't be as elegant as your current script, though.</p>
<p dir="ltr">Good luck</p>
<div class="gmail_quote">On Apr 21, 2016 11:12 AM, <a href="mailto:felix.a.salazar@gmail.com">felix.a.salazar@gmail.com</a> wrote:<br type="attribution"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><p dir="ltr">Dear Florian,</p>
<p dir="ltr">That's a very nice .GEO file. Very organized. The problem is the dilation. When you Extrude the entities, you're assigning those entities mesh information. Then, you modify the geometry, it no longer remains connected. The lines, determined by the position of the vertex, looks correct. But if you show the geometrical surfaces, they are still the ones created with the extrusion, with the exception of the dilated one. Hence, when gmsh reach a surface whose mesh coordinates are known from the geometry, i.e. the one opposite to the extrusion, it finds discrepancies.</p>
<p dir="ltr">I look at the documentation, and to my knowledge, there's no option that could easily help you. Even if you remove the Layers and create an unstructured mesh, just as a test, would cause errors, at least in my gmsh version.</p>
<p dir="ltr">I'm thinking of an alternative. Let me do some test and I'll get back to you.</p>
<p dir="ltr">Sorry for the delay. Time zones I guess...</p>
<div class="gmail_quote">On Apr 21, 2016 4:48 AM, "Florian Hoffmann" <<a href="mailto:florian.hoffmann@inutech.de">florian.hoffmann@inutech.de</a>> wrote:<br type="attribution"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
  
    
  
  <div bgcolor="#FFFFFF" text="#000000">
    Dear Felix,<br>
    <br>
    thank you very much for your quick response.<br>
    Please find attached a simplified example highlighting my problem.<br>
    It is a simple cylinder with an internal block structure. The 3D
    geometry is obtained by extruding several surfaces (=blocks for
    hexaeder meshing). Extrusion is done in line 161 of the *.geo file.<br>
    With the variable 'withDilate' in line 186 one can switch on/off the
    dilate command I was trying to obtain the frustum-like outline with.<br>
    I also attached two screenshot showing my problem when meshing: On
    the left hand side there is the cylinder (Extrude only) and on the
    right hand side one can see the conical frustum (Extrude+Dilate) but
    one can also see the problem with the mesh.<br>
    If you could point me to the bits that I would have to change/add in
    order to make this work, that would be really really making my day.<br>
    <br>
    Thank you very much.<br>
    <br>
    Best regards,<br>
    <br>
    Florian<div class="elided-text"><br>
    <br>
    <br>
    <br>
    <br>
    <div>Am 21.04.2016 um 06:18 schrieb Felix
      Salazar:<br>
    </div>
    <blockquote type="cite">
      <p dir="ltr">Hello,</p>
      <p dir="ltr">I did something similar not too long ago. My geometry
        had a basic structure, and for different cases, the angles and
        lengths of the sections varied. Just make sure that Extrude
        ReturnLateralEntities option is on, and check the contents of
        the extruded array.  Start with the B->C extrusion only, and
        make sure that you're using the Layers option in the extrusion,
        to get your structured mesh. Depending on your options, you
        might need to explicitly recombine your elements to get hexas.
        Could we see your .GEO file? Just from the B-B surface to the
        extrusion? You can remove the details of the A-A section, to
        protect IP</p>
      <div class="gmail_quote">On Apr 20, 2016 5:47 PM, "Florian
        Hoffmann" <<a href="mailto:florian.hoffmann@inutech.de" target="_blank">florian.hoffmann@inutech.de</a>>
        wrote:<br type="attribution">
        <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
          <div style="word-wrap:break-word">Dear all,
            <div><br>
            </div>
            <div><br>
            </div>
            <div>I would like to generate a block structured
              hexaeder-only-3D-mesh for a geometry consisting of several
              conical frustums. Something like this:<br>
              <div><br>
              </div>
              <div><br>
              </div>
              <div>  __________ </div>
              <div>  \                 /</div>
              <div>   |               |<br>
                    \             / <br>
                     |           |          </div>
              <div>C_|           | _ C      <br>
                     \          /            ^  extrude B->C<br>
                B _ \____/  _ B    _|_</div>
              <div>A _  |___|   _ A</div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div>Section A-A has a non-circular cross-section, and
                section B-B and all others above have circular
                cross-section. I successfully created A-A and the
                circular cross-section B-B. Both consist of several
                surfaces/blocks (see attached picture). In my *.geo
                script I try to work with variables. There I collect the
                IDs of all surfaces forming the cross-section B-B (= the
                bottom of the first conical frustum) in an array
                (=cross_section_BB_surfaces[]). Nevertheless, when I try
                to Extrude+Dilate these surfaces in order to build the
                first conical frustum (=B-B -> C-C), during meshing I
                get several errors about „Could not find extruded vertex
                …“ . Extrusion-only of B-B to C-C (without Dilate) works
                fine, but this would just give me a cylinder outline and
                it is not what I want.</div>
              <div>Since I need a block structured mesh and also because
                A-A has a non-circular cross-section (see bottom plane
                of geometry in attached picture) I would really like to
                make use of the extrude command for creation of C-C and
                the other frustum above.</div>
              <div><br>
              </div>
              <div>(1) Is there a way to make this work (without having
                to create all frustums including their blocked structure
                by hand) ?</div>
              <div><br>
              </div>
              <div>(2) If not, can anybody think of a solution for how
                to script this in an efficient and handable way ?</div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div>Thanks a lot.</div>
              <div><br>
              </div>
              <div>Cheers Flo</div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><img src="cid:part2.01050306.01080900@inutech.de" height="710" width="1262"></div>
            </div>
            <br>
            <br>
            <div>
              <div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word">
                <div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word">
                  <div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word">
                    <div>
                      <div style="margin:0in 0in 0.0001pt;font-size:11pt;font-family:Calibri,sans-serif"><span style="font-size:x-small;font-family:Arial"><span style="font-size:10pt" lang="DE">Mit
                            freundlichen Grüßen<br>
                            --</span></span><strong><strong><span><span style="font-size:12pt" lang="FR"><br>
                                Dr.-Ing. Florian Hoffmann</span></span></strong></strong><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">inuTech
                          GmbH                   </span><span style="font-family:Helvetica;font-size:12px">Phone
                             : <a href="tel:%2B49-%280%29911-323843-22" value="+4991132384322" target="_blank">+49-(0)911-323843-22</a></span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">Fuerther
                          Strasse 212         Fax        : <a href="tel:%2B49-%280%29911-323843-43" value="+4991132384343" target="_blank">+49-(0)911-323843-43</a></span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">90429
                          Nuernberg               E-Mai l   :<span> </span></span><a href="mailto:sebastian.peetz@inutech.de" style="font-family:Helvetica;font-size:12px" target="_blank"><a href="mailto:florian.hoffmann@inutech.de" target="_blank">florian.hoffmann@inutech.de</a></a><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">Germany
                                                     Internet  : </span><a href="http://www.inutech.de" style="font-family:Helvetica;font-size:12px" target="_blank"><a href="http://www.inutech.de" target="_blank">http://www.inutech.de</a></a><br style="font-family:Helvetica;font-size:12px">
                        <br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">inuTech
                          GmbH</span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">Sitz
                          / Registered Office: Nuernberg</span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">Handelsregister
                          / Companies' Register: AG Nürnberg HRB Nr.
                          19026</span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">Geschäftsführer
                          / Managing Director: Frank Vogel</span><br style="font-family:Helvetica;font-size:12px">
                        <br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">*****************************************************************</span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">DIFFPACK
                          - THE NEW GENERATION OF SIMULATION-SOFTWARE!</span><br style="font-family:Helvetica;font-size:12px">
                        <br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">Go
                          to </span><a href="http://www.diffpack.com" style="font-family:Helvetica;font-size:12px" target="_blank">http://www.diffpack.com</a><span style="font-family:Helvetica;font-size:12px"> to
                          read the details</span><br style="font-family:Helvetica;font-size:12px">
                        <span style="font-family:Helvetica;font-size:12px">*****************************************************************</span></div>
                    </div>
                  </div>
                </div>
              </div>
            </div>
            <br>
          </div>
          <br>
          _______________________________________________<br>
          gmsh mailing list<br>
          <a href="mailto:gmsh@onelab.info" target="_blank">gmsh@onelab.info</a><br>
          <a href="http://onelab.info/mailman/listinfo/gmsh" rel="noreferrer" target="_blank">http://onelab.info/mailman/listinfo/gmsh</a><br>
          <br>
        </blockquote>
      </div>
    </blockquote>
    <br>
    </div><pre cols="72"><font color="#888888">-- 
Mit freundlichen Grüßen
--
Dr.-Ing. Florian Hoffmann
inuTech GmbH            Phone    : <a href="tel:%2B49-%280%29911-323843-22" value="+4991132384322" target="_blank">+49-(0)911-323843-22</a></font><div class="quoted-text">
Fuerther Strasse 212    Fax      : <a href="tel:%2B49-%280%29911-323843-43" value="+4991132384343" target="_blank">+49-(0)911-323843-43</a>
90429 Nuernberg         E-Mai l  : <a href="mailto:florian.hoffmann@inutech.de" target="_blank">florian.hoffmann@inutech.de</a>
Germany                 Internet : <a href="http://www.inutech.de" target="_blank">http://www.inutech.de</a>

inuTech GmbH
Sitz / Registered Office: Nuernberg
Handelsregister / Companies' Register: AG Nürnberg HRB Nr. 19026
Geschäftsführer / Managing Director: Frank Vogel

*****************************************************************
DIFFPACK - THE NEW GENERATION OF SIMULATION-SOFTWARE!

Go to <a href="http://www.diffpack.com" target="_blank">http://www.diffpack.com</a> to read the details
*****************************************************************
</div></pre>
  </div>

</blockquote></div></blockquote></div>