AnsweredAssumed Answered

Vacuum simulation set up

Question asked by Patrick Hands on Dec 10, 2014
Latest reply on Dec 18, 2014 by Jared Conway

Like • Show 0 Likes0
Comment • 16

Hi,

I asked this questions a few backs and got good feedback on how to set the problem up. I believe I did what I was informed to do but I am unsure.

I have a manifold design for my thesis and I am trying to simulate a vacuum drawing up through up from 16 nozzles to pick up small objects.

I was told to place the design is a small box and apply a 1atm pressure to the walls and leave the base pressure free.

I then applied a outlet volume flow rate to the outlet which will be attached to the vacuum pump.

as you can see in the images I don't think the results are right and I am wondering have I done something wrong in the set up??

There is no movement in the flow trajectories and the velocity is zero?

No one else has this question

Outcomes

16 Replies

Chris Michalski Dec 10, 2014 7:51 AM
Mark Correct
Correct Answer
According to the color chart you have some incredibly high velocity somewhere. What does it look like if you decrease the max velocity? There is likely a leak that is causing high velocity through a small opening which makes the rest of your flow trajectory look like 0m/s.

So you have the top tube as a defined volume out? I'm a bit confused because you have trajectory arrows going both up and down there. So there is major turbulence at that point.

I'm struggling to recall the previous discussions about setting this up to know where the air is supposed to enter the system at... the 16 parts (nozzles) shown wireframe at the bottom? Then through the middle solid part and to the butterfly cavities before exiting at the tube on top?
Can you include a screenshot that includes your feature tree to help know what you've included?
Like • Show 1 Like1
Actions
- Patrick Hands @ Chris Michalski on Dec 10, 2014 7:59 AM
  Mark Correct
  Correct Answer
  Yes the top tube is outlet volume flow at 2 m^3/s..... this is the outlet that is connected to the vacuum pump.
  
  The air should be drawn through this manifold system.... which enters at the nozzles..... into the butterfly area.... through the centre of this butterfly area and through the manifold before leaving the outlet at the top back to the vacuum pump.
  Like • Show 0 Likes0
  Actions
  - Chris Michalski @ Patrick Hands on Dec 10, 2014 8:13 AM
    Mark Correct
    Correct Answer
    If you display just the mesh, does it look like your setup? With such a large volume and small diameter passages I'd be concerned that it is not accurately representing your model. You should be able to look at the mesh (either solid or fluid cells) and easily envision your system based on its recreation in square cells.
    When you reduce the max velocity shown in your trajectory plot what does it look like? What about increasing the # of trajectory lines or the max length?
    Try creating a surface plot based on velocity and choose a few higher values. Where it creates surfaces with high velocity would help indicate where you may have an issue.
    Like • Show 1 Like1
    Actions
Jared Conway Dec 11, 2014 12:03 AM
Mark Correct
Correct Answer
please link your previous post
elaborate on your setup
and build a really simplified version that you can test with and post

i'm not following your setup but based on what i'm understanding, it doesn't make much sense

if you could also include what your goal is, that would be helpful to make sure the setup suggestions match your goals
Like • Show 0 Likes0
Actions
- Patrick Hands @ Jared Conway on Dec 11, 2014 3:57 AM
  Mark Correct
  Correct Answer
  Hi Guys,
  
  First thanks so much for trying help. I am new to this flow simulations and its required in my thesis for my masters.
  
  I have created a new model which only contains what I need.
  
  This is the link to my original post..... Setting up Vacuum Simulation
  
  I have inserted 5 images below which break down exactly what i'm trying to do and achieve.
  Like • Show 0 Likes0
  Actions
  - Patrick Hands @ Patrick Hands on Dec 11, 2014 5:40 AM
    Mark Correct
    Correct Answer
    As soon as I run a simulation this shows up in the simulation window also
    Like • Show 0 Likes0
    Actions
    - Chris Michalski @ Patrick Hands on Dec 11, 2014 7:20 AM
      Mark Correct
      Correct Answer
      That's a very high Mach Number. Have you tried making a fluid body assembly of the system? (Flow Simulation -> Tools -> Check Geometry).
      This can help show if you have problems with the internals of your model (i.e. accidentally blocking flow or leaks between inside and outside).
      The Vortex just means that the flow at the boundary condition is going in both directions which can happen because that tube is short. Gas coming up from inside tumbles because of the bend in the tube. If you increase the length of that stub that would go away (but isn't truly critical because you are interested in values so far away.)
      Like • Show 1 Like1
      Actions
      - Patrick Hands @ Chris Michalski on Dec 11, 2014 8:15 AM
        Mark Correct
        Correct Answer
        I ran the check geometry and it says all is ok and this is the fluid flow available without any leaks??
        
        Like • Show 0 Likes0
        Actions
        Chris Michalski @ Patrick Hands on Dec 11, 2014 8:24 AM
        Mark Correct
        Correct Answer
        The top looks fine. I would section that near the nozzle to make sure everything is as you expect.
        Then look at the mesh in the nozzles. The holes may be so small relative to the cells that it is not accurately representing the openings.
        whoa. In looking at your solve window you've only got 130k cells. You REALLY need to adjust the mesh. For something this complicated you're going to want at least 1M cells to see anything at all. (This is the reason to only look at 1/4 of the model so you can make the mesh very fine without having an unreasonable number of cells)
        Like • Show 1 Like1
        Actions
        Patrick Hands @ Chris Michalski on Dec 11, 2014 12:44 PM
        Mark Correct
        Correct Answer
        I changed the model to a much simplier one with just 4 nozzles as seen below. I up the mesh to 1m cells and ran the simulation. It exceeded the Mach Number after just 1 iteration. surely this is not correct?
        
        Like • Show 0 Likes0
        Actions
        Chris Michalski @ Patrick Hands on Dec 11, 2014 1:14 PM
        Mark Correct
        Correct Answer
        Don't worry if it starts with high mach flow. The calculations will start VERY far away from the final solution as it iterates until all of the conditions are continuous. It basically used that exhaust pressure inside the entire cavity and when it got to the very small nozzle holes that pressure gradient resulted in a high velocity. As it iterates and distributes that pressure drop the mach # there will drop.
        
        I would make a plan that is parallel to your exhaust and passes through one of your nozzles so you can see the velocity in the small holes as well as further upstream. This will let you verify that the high Mach is at the nozzles. If it is elsewhere then you likely have a problem with your setup.
        Like • Show 1 Like1
        Actions
        Jared Conway @ Patrick Hands on Dec 12, 2014 4:35 AM
        Mark Correct
        Correct Answer
        i think your issues are related to the model, have you simplified it? have you built a really simple test assembly to test your method? please do the latter and post that model.
        
        also what is the goal of your analysis?
        Like • Show 0 Likes0
        Actions
        Patrick Hands @ Jared Conway on Dec 12, 2014 4:43 AM
        Mark Correct
        Correct Answer
        Hi Jared,
        
        Is the model above of four nozzles not simple enough or should I just test one nozzle?
        
        the main goal is just to measure the negative pressure at each nozzle to ensure it is somewhat even and to figure out what diameter pipe I need to achieve the required negative pressure at all nozzles
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Patrick Hands on Dec 12, 2014 4:46 AM
        Mark Correct
        Correct Answer
        the single bank is fine but i think you have some modeling that you need to work on. so i'm thinking something with really simple sweeps is best.
        
        what does "measure negative pressure" mean to you in this case? the pressure there will be somewhere around ambient since that is what it is sucking from.
        Like • Show 0 Likes0
        Actions
        Patrick Hands @ Jared Conway on Dec 12, 2014 4:53 AM
        Mark Correct
        Correct Answer
        Ok I will model something more simpler and try that!
        
        I want to know does it create a strong enough vacuum to pick that part up basically?
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Patrick Hands on Dec 18, 2014 3:15 AM
        Mark Correct
        Correct Answer
        hmm, is the flow pressure at the end of the tube something you can use to estimate the strength of the vacuum?
        Like • Show 0 Likes0
        Actions

Vacuum simulation set up

Outcomes

Related Content

Recommended Content