Flow between concentric cylinders

Jagan Adithya Elango
Jagan Adithya Elango Altair Community Member
edited February 2023 in Community Q&A

imageimage

Hello experts, I'm trying to model flow in an airgap. I have specified MRF of 600 rad/s at the inner cylindrical surfaces but it doesn't translates on my results. I'm getting 0 velocity. Seeking you help in figuring it out.

 

Tagged:

Answers

  • acupro
    acupro
    Altair Employee
    edited February 2023

    The .inp seems to make sense.  Can you attach the HyperWorks CFD file, along with the .Log file from the run?

  • Jagan Adithya Elango
    Jagan Adithya Elango Altair Community Member
    edited February 2023
    acupro_21778 said:

    The .inp seems to make sense.  Can you attach the HyperWorks CFD file, along with the .Log file from the run?

    I used moving wall instead of reference frame with cylindrical velocity type of 26.7m/s in tangential direction. It seems to be working. image

  • acupro
    acupro
    Altair Employee
    edited February 2023
    Jagan_21383 said:

    I used moving wall instead of reference frame with cylindrical velocity type of 26.7m/s in tangential direction. It seems to be working. image

    I've also found that manually editing the input file to have the inner wall set to type = wall instead of type = auto_wall then allows the reference frame application to work.  (You'll then need to run from command-line or Compute Console after manually editing.)

  • Jagan Adithya Elango
    Jagan Adithya Elango Altair Community Member
    edited February 2023

    Thank you for your response.

  • ydigit
    ydigit
    Altair Employee
    edited February 2023

    It seems to be very thin gap. How much distance do you have between those two cylinders? 
    How many elements do you have in the gap?

    I would recommend min 3 BL on each side and one element in the core gap. Could you take a screenshot with mesh on the small gap? 

     

  • Jagan Adithya Elango
    Jagan Adithya Elango Altair Community Member
    edited February 2023

    Its thickness is 0.7mm. At first I specified boundary layer similar to what you said but then I reduced it to 4 elements since the velocity profile is going to be linear between the inner rotating surface and stationary outer surface right? Would love to know your opinion about this.

    image

  • ydigit
    ydigit
    Altair Employee
    edited February 2023
    Jagan_21383 said:

    Its thickness is 0.7mm. At first I specified boundary layer similar to what you said but then I reduced it to 4 elements since the velocity profile is going to be linear between the inner rotating surface and stationary outer surface right? Would love to know your opinion about this.

    image

    The issue I see is that that the elements at interface would have zero velocity or velocity of rotating cylinder. That leaves only 2 nodes that are really part of the gap. 

    I would still attempt 3 BL on each side and if not 3, then at least 2 BL on each side. 

     

Categories