Search
Search
Menu
Close

Dear All, there are two channels for you to follow our research. 

1. We started a newsletter called Thoughts on Composites Modeling on LinkedIn. Please subscribe to it to get informed about our research and ideas we want to share with the community.

2. We moved our discussion forum to github at wenbinyugroup · Discussions · GitHub as it has more features which can serve you better. 

Wenbin 

595
Close

Prof. Yu's Research Group in the Cloud

Forum

Strength Ratio Recovery After Homogenization - PreVABS Setup

  1. Dana Frye

    Hello All,

    I have seen in the release notes of VABS 4.0, that the ability to "Compute the safety margin of the cross section under a given load" has been emphasized. I am attempting to do this. I see in the PreVABS documentation (please note that this link -> https://wenbinyugroup.github.io/ivabs/prevabs/index.html is the documentation I'm using - please let me know if this is not the latest and greatest), that to run and recover the stress/stain distribution, failure indicies, and strength ratio distribution, the command is "prevabs -i <cross_section_file_name.xml> -d -e -fi -v ". Correct me if I'm wrong, but in order to calculate strength ratio, the laminae (ply) strength limits must be accounted for. This seems to be noted in the preVABS manual change log, which says for v1.4.0, "Added new inputs for strength property of ‘lamina’ type materials". However looking through the PreVABS documentation, I can't find where to incorporate these ply strengths into the input files, and I can't find the syntax to do so. Any help would be greatly appreciated.

    Thanks, 
    Dana

    Reply Report abuse

    Replying to Dana Frye

    1. Wenbin Yu

      Kevin, I guess that PreVABS does not allow this input for you. However, it is very easy to change the file generated by PreVABS to tell VABS that you need to use failure criterion #2.

      Reply Report abuse

      Replying to Wenbin Yu

      1. Kevin Wang

        Sorry if I'm mistaken but I thought only #4 is supported right now?  To be honest, I haven't tried running with #2 so didn't verify this

        Report abuse

      2. Kevin Wang

        I see, ok to clarify, VABS does support failure criterion #2 for orthotropic materials?

        If so, maybe the question then is to Su on when he expects to release PreVABS to support it.

        Report abuse

      3. Kevin Wang

        Hey Su,

        I saw your e-mail comment but can't find it here but I'll reply to this thread.

        So I was able to run with the failure criteria flag set as #2 on my orthographic material, however the results don't seem to make sense.

        Attached is my test section, where it's a tube of laminate and a filled isotropic inside.  Looking through the .fi file, there are unique values generated for all the orthotropic elements, but they are extremely high.  Is PreVABS/VABS taking the min strength ratio out of all 9 strength constants and outputting the lowest one?

        Thanks

        failure-criteria2.png

        Report abuse

    2. Wenbin Yu

      For VABS, all 1-5 are supported. In fact, you can even provide a user-defined failure criterion if you have a specific failure criterion not supported.

      Reply Report abuse

      Replying to Wenbin Yu

    3. Wenbin Yu

      Yes, you are right. VABS does support 1-5 for orthotropic materials.

      Reply Report abuse

      Replying to Wenbin Yu

    4. Wenbin Yu

      If you can send us the input file, we can debug it for you.

      Reply Report abuse

      Replying to Wenbin Yu

    5. Su Tian

      Hello Dana,

      To provide material strength data, add the following two entries in '<material>...</material>':

          <failure_criterion>...</failure_criterion>

          <strength>...</strength>

      For the 'lamina' type materials, right now only the 'tsai-wu' failure criterion is implemented. This failure criterion needs nine strength constants. However, for the 'lamina' type materials, only the following four strength properties must be input directly from users:

          <strength>

              <xt>...</xt>

              <xc>...</xc>

              <yt>...</yt>

              <s>...</s>

          </strength>

      where

      • <xt> - tensile strength in the x1 direction
      • <xc> - compressive strength in the x1 direction
      • <yt> - tensile strength in the x2 direction
      • <s> - shear strength in the x1-x2 plane

      The rest five constants are optional, with default values derived from these four:

      • <yc> - compressive strength in the x2 direction (default = yt)
      • <zt> - tensile strength in the x3 direction (default = yt)
      • <zc> - compressive strength in the x3 direction (default = zt)
      • <t> - shear strength in the x1-x3 plane (default = s)
      • <r> - shear strength in the x2-x3 plane (default = (yt+yc)/4)

      You can also explicitly input these values to overwrite the defaults.

      I will later add other failure criteria for the 'lamina' type materials supported by VABS, such as 'tsai-hill' and 'hashin', and update the online documentation.

      Please let us know if some inputs are inconvenient or some new features are needed.

       

      Besides, if you want both stress/strain and failure index/strength ratio results, you need to run the command with options '-d' and '-fi' separately.

      Reply Report abuse

      Replying to Su Tian

      1. Dana Frye

        Much appreciated, thank you for the information! I'll work these into my model and run it. 

        In terms of any new features, the only thing that comes to mind is what you touched on - that is having all the failure criteria supported by VABS supported by preVABS (Tsai-Hill particularly). 

        Thanks again. 

        Report abuse

      2. Dana Frye

        Is it still the case that -f, -fe, and -fi options are for SwiftComp only, as shown in the figure below?

        preVABS-options.png

        Report abuse

      3. Kevin Wang

        Hey Su,

        Besides the visualization issues in PreVABS, I've successfully been able to test out the failure criterion regarding isotropic materials, however ran into a hiccup when putting in an orthotropic material.

        See below for the material properties:

        <material name="test" type="orthotropic">
            <density>1</density> <!-- kg/m^3 -->
            <elastic>
              <e1>1</e1> <!-- Pa -->
              <e2>1</e2> <!-- Pa -->
              <e3>1</e3> <!-- Pa -->
              <g12>1</g12> <!-- Pa -->
              <g13>1</g13> <!-- Pa -->
              <g23>1</g23> <!-- Pa -->
              <nu12>0.27</nu12>
              <nu13>0.27</nu13>
              <nu23>0.38</nu23>
            </elastic>
        </material>
        <lamina name="la_test">
        <material>test</material>
        <thickness>0.001</thickness> <!-- m -->
        <failure_criterion>4</failure_criterion> <!-- Tsai-Wu, only supported criterion in 1.4 -->
        <strength>
          <xt>1</xt> <!-- f1tu, Pa -->
          <xc>1</xc> <!-- f1cu, Pa -->
          <yt>1</yt> <!-- f2tu, Pa -->
          <s>1</s> <!-- f12su, Pa -->
          <yc>1</yc> <!-- f2cu, Pa -->
          <zt>1</zt> <!-- f3tu, Pa -->
          <zc>1</zc> <!-- f3cu, Pa -->
          <t>1</t> <!-- f13su, Pa -->
          <r>0.5</r> <!-- f23su, Pa --> <!-- (yt+yc)/4 -->
        </strength>
        </lamina>

        And then as far as the error goes, it happens when running -fi -e

        "I/O error: Failure Criterion =               0 is not defined for an anisotropic material!"

        Any tips? thanks

        Report abuse

      4. Su Tian

        Hello Kevin,

        You need to put the '<failure_criterion>' and '<strength>' blocks inside the '<material>', instead of '<lamina>'.

        Report abuse

      5. Dana Frye

        What is the syntax to add strength data to isotropic materials? See question marks below. Thanks!!

          <!-- =========================================================== -->
          <material name="ti6al4v_iso" type="isotropic">
            <density>4429</density> <!-- kg/m^3 -->
            <elastic>
              <e>116.5E+9</e> <!-- Pa -->
              <nu>0.31</nu>
            </elastic>
            <failure_criterion>mises</failure_criterion>
            <strength>
              <??>827E+006</??> <!-- Pa, yield-->
              <??>896E+006</??> <!-- Pa, UTS-->
            </strength>
          </material>
          <!-- =========================================================== -->

        Report abuse

      6. Kevin Wang

        Just mirroring what Dana said in this post (https://cdmhub.org/groups/yugroup/forum/default-section/prevabsvabs#c4457), is there a timeframe for when other failure criteria will be supported? Specifically looking for max strain criterion (2).

        Thanks

        Report abuse

    6. Su Tian

      Dana,

      Now the '-fi' option is available for VABS. Please download the latest version of PreVABS 1.4.2 on cdmHub.

      Reply Report abuse

      Replying to Su Tian

    7. Kevin Wang

      Follow on to Dana's question,

      1) What exactly are we supposed to put between <failure_criterion> and </failure_criterion>?
      2) To run -fi, do we need strength properties for all the materials in the cross section?
      3) If we have isotropic materials in our cross section, how should we handle the strength properties for now since we're limited on failure criterion?
      4) I've done a test run through with a tube and it sort of gave me results (attached), although it seems to have bugged out on all the cells in the top right (still digging into what I did).  There are now 2 new outputs, EMN and SMN, is there a description for what these are?

      Thanks

      fi-test.png

      Reply Report abuse

      Replying to Kevin Wang

      1. Dana Frye

        Hey Kevin, 

        My model also has both orthotropic materials (for the composite blade skin and spar ply materials), and a metallic material. (for the sleeve). I had a similar question to the one you listed above at number 3 so I've given it a bit of thought. See the figure below for a sketch of my configuration. What I'm doing currently, is modeling the isotropic metallics with the orthotropic material type, but having E1=E2=E3, G12=G23=G13, nu12=nu23=nu13. This is similar to how it is done in the preVABS Airfoil (MH-104) example (https://wenbinyugroup.github.io/ivabs/prevabs/examples/example_airfoil.html) for the gelcoat (see materials table in example). My thoughts are that for cross sections that include both composite and metallic material, if your model is done the way described above, and a tsai-hill or tsai-wu failure criteria is used, for the metallic this ends up simplifying to a von-mises like criteria. That is assuming that for the metallic, the <strengths> are xt=yt=zt, s=t=r, etc. At least I believe this is the case, correct me if I'm wrong. 

        blade-w-metallics-vabs.png

        Report abuse

      2. Su Tian

        Hello Kevin,

        1) The failure criterion options are:

          - For isotropic materials: "max principal stress" (1), "max principal strain" (2), "max shear stress" or "tresca" (3), "max shear strain" (4) and "mises" (5).

          - For orthotropic/anisotropic materials: "max stress" (1), "max strain" (2), "tsai-hill" (3), "tsai-wu" (4) and "hasin" (5).

          - For the "lamina" type materials, currently (in v1.4.2) the only supported failure criterion is "tsai-wu" (4).

        You can use either the keyword or the number to indicate the failure criterion.

         

        2) Yes. You need to provide strength properties for all materials used in the cross-section.

        Report abuse

      3. Kevin Wang

        Thanks Su,

        So for clarity, <failure_criterion> and <strength> are all inside the <material> brackets for isotropic/orthotropic/anistropic materials, however lamina doesn't have a <material> tag.

        Can the <failure_criterion> and <strength> be placed inside of <lamina name="...">...</lamina>?

        edit: ignore the question.  After reading through the 1.4 documentation, the syntax has changed so the question seems solved

        Report abuse

© 2024 cdmHUB.