6 Replies Latest reply on Sep 5, 2012 8:32 PM by Eric Ryan

    How to fix problem of zero fluid volume?

    Brando Dimapasoc

      Hello all. I'm currently trying to model (roughly) the effect a stent would have on blood flow into an aneurysm, and I ran into a problem with the flow simulation where it says I have zero volume. The model is attached.

       

      When I run a fluid analysis on the assembly without the stent, everything works just fine and I get a non-zero fluid volume when I check the assembly's geometry, as shown in the first picture. However, when I put the stent into the assembly and checked the geometry, the fluid volume shows up as zero (second picture). Any idea what could be going on? At first, I thought it was because the stent's position was under-defined, but after adding a lock mate to fix the stent's position, the result is the same--zero fluid volume.What could the stent be doing that is making the model not work?

       

      Any help is greatly appreciated! Thank you.

       

       

      Pictures:

      http://i47.tinypic.com/2gts1zr.jpg

       

      http://i46.tinypic.com/14c905h.jpg

        • Re: How to fix problem of zero fluid volume?
          Eric Ryan

          Brando,

           

          When I tried to get the fluid volume, there is an error message that the boolean operation failed.  It seems Flow Sim is having trouble with the complex geometry of the stent.   You could try making a simple version of the stent without the diamond shape hole patterns.   You could also try one that is not made using the Flex features, but made with a sweep or extrusions.  Then, see if the fluid volume check works.  This might give you some idea of what geometry features are causing the problem.  You might want to send this one in to Solidworks as well.  They might have some insight as well.

           

          Good Luck,

           

          Eric

          • Re: How to fix problem of zero fluid volume?
            Chris Michalski

            I'm stuck in the land of Flow 2007 so I can't open the file.  Based on Eric's comment about the Boolean Operation Failed you likely have some minute contact point (i.e. tangent curves between the two parts or one of the diamond corners is exactly on the surface), something that it can't resolve. 

            Can you make the stent smaller so that it doesn't actually touch the walls of the other part to test this possibility?

            Or as Eric said, simplify it by eliminating the diamond holes.  Then you can try adding a Porous Media condition to the stent to allow flow into the aneurysm based on the open area.

            • Re: How to fix problem of zero fluid volume?
              Rich Bayless

              1)  You could try running an external study, with just the stent, suppress the rest of the model.  This will help show whether Flow is happy with the stent model. 

               

              2) Run 'Check Geometry' on the part(s), and also check off 'create solid body assembly' and 'create fluid body assembly' options.

               

              3) To help find a gap, try using the 'create lids' tool, to see if the tool can find any openings.

              • Re: How to fix problem of zero fluid volume?
                Brando Dimapasoc

                Thanks for the replies so far, everyone! I've tried some of your suggestions and have a couple of updates. (In order to avoid cluttering the post with huge pictures, I'll just link to the images.)

                 

                Eric and Chris:

                 

                You suggested to make a simplified and smaller-scaled models of the stent in order to determine what geometry was causing the zero-fluid volume problem. In order to test the geometry, I first tried to put the stent into a previous aneurysm model that I know already works with a different stent design. I don't have a screenshot of the model with the "problematic stent" in it, but I did get a nonzero fluid volume, and the flow simulation worked! Here is the model used though, with the original stent in it.

                http://i46.tinypic.com/2130iz7.jpg

                 

                I also tried scaling down the problematic stent in the model from my original post so that it didn't touch any walls of the model at all. Here is the result: http://i45.tinypic.com/oa4vnq.jpg As you can see, still a zero fluid volume.

                 

                *An important question I have though is: Does checking/unchecking "Exclude cavities without flow conditions" have an effect on this? I think I'm a bit unclear on how this option relates to the model as a whole. Can anyone please explain? When I uncheck this option, I do get a nonzero fluid volume! Here is a picture: http://i48.tinypic.com/55qouo.jpg

                 

                However, this doesn't really fix the main root of my problem, which is that when I try to add a boundary condition to my lids so that I can run a flow simulation, SW says that the face does not lie on a boundary between solid and fluid regions, and it doesn't have this problem with the problematic stent absent from the model. Trying to troubleshoot this problem led me to the zero-fluid-volume problem that this thread is about.

                 

                 

                Rich: How do I use the lid tool to find gaps? All I know how to do is use it to create lids at model openings. Also, the stent does have a thickness, though it is very small. As stated above, the stent does indeed work in another model.

                 

                 

                I'm very confused now Thanks in advance for any more help

                  • Re: How to fix problem of zero fluid volume?
                    Eric Ryan

                    Brando,

                     

                    I think one problem might be with your Lids and another is with the Stent part interfering with the Bent Aneurysm part. 

                     

                    It looks like you used the Lid command in Flow Sim to make lids on the Bent Aneurysm part.  I have never had very good luck using that command.  I always model my own lids as solid bodies.    I took your Bent Aneurysm model and suppressed the Lid features.  Then, I sketched on the circular surface, offset the edge, and made an extrusion.  To make the extrusion a separate solid body, I un-checked the "merge results" box.   I did this on both ends of the part where your original lids were.  Here are a couple of screen shots of how this looks.

                     

                    LID-1.jpg

                    LIDS-2.jpg

                    Making the Lids solid bodies lets you change their color, make them transparent, etc. a little more easily.  This is how you had to make Lids before they had the Lids command.

                     

                    To see if the new Lids made a difference, I first scaled the Stent part like you did, and moved it to make sure it was not interfering with the Bent Aneurysm part.  I ran the fluid volume check and it worked, showing a solid and fluid volume.  I was also able to create a fluid body assembly. 

                     

                    Since the new Lids seemed to work, I then tried it with the original size Stent part.  The fluid volume check failed.  To confirm this, I also moved the stent part so that I was sure it was interfering with the Bent Aneurysm part.  The fluid volume check again failed.  So it seems the Stent needs to positioned or scaled to make sure it is not interfering with the Bent Aneurysm part.  This may be hard to confirm.  I tried the Interference Detection command and could not get anything to show up.  I may have to play with this a bit more to find out why.  I also tried sectioning the assembly, but that does not seem to work very well either.

                     

                    For now, you might try making new Lids and adjusting the position of the Stent part to make sure it is not visually interfering with the Bent Aneurysm part.  It might help to adjust the scale or shape of the Stent to help it fit inside the aneurysm.  This may at least get you a first pass through the flow simulation. 

                     

                    Let me know what you find and if you have any questions on what I wrote here.

                     

                    Cheers,

                     

                    Eric