2 Replies Latest reply on Sep 17, 2017 4:33 AM by Tiago Antonio

    XY Plot show incoherent results

    Tiago Antonio


      currently I'm studying the mixing of two fuids in a T-junction that is in the order of magnitude of the micron. As a result the characteristic Reynolds number associated with these flows is very small, so I'm working in the laminar regime. The geometry I'm working is the following:


      The plane you see labeled as "Plane 1" is the section of the mixer outlet where I'm looking for some results.

      After runing the simulation this is the result for the velocity field I get in this section:

      Cut Plot 9.jpg


      Next, what I did was to draw a line(sketch) in this same "Plane 1" parallel to the upper and lower walls that united the left and right walls. By other words, I want to know the velocity field, at a given plane and in my section of study, along the width. With this line I did a XY plot along that line I drew.

      This is what I got:

      Sem Título.jpg

      I have zero velocity in the near walls region. If I select the probe and go to that region, it shows non zero values in this region...

      Can someone tell me what am I missing here?

      Thank you in advance for your help.

        • Re: XY Plot show incoherent results
          Mark Keown

          Can you show us the mesh in this section.

            • Re: XY Plot show incoherent results
              Tiago Antonio


              The mesh is the following:

              Mesh 3.jpg

              I understand that maybe the mesh should be more refined in the near wall region, but the mesh as it is is consuming almost All RAM from my computer. We're talking 8GB of RAM. And on the other hand, I can't use a coarser mesh in other regions because  there are concentration gradients at the center that I need to capture well. Using point parameters, I managed to pick a couple of points and replicated what the XY plot is trying to do. This is what I came up with:

              Sem Título2.jpg


              If you don't consider the scale, wich is obviously diferent, you can see that this graph is exactly the same, but with no zero velocities in the near wall region. Naturally this go round method would do the trick, but using XY plots is much much faster and I have plenty of mixers to simulate, and plenty of diferent flows each. So it's clear that If I can use a method that reduces the time I spend treating results, it would be very nice.