Fluids

Fluids

Solution is not converging properly

    • Nikhil26
      Subscriber

      I am doing air flow analysis on a car model with following conditions;


      model - K-epsilon


      Inlet Boundary con- velocity inlet 16.67 m/s


      Outlet - pressure outlet


      The residuals are deflecting too much also during the solution running sometimes the drag and lift co-efficient goes to negative sides. Whats the problem??? Also I am getting negative value of lift-co-effiecint

    • Karthik R
      Administrator

      Hello,


      What is your mesh quality like? For these types of drag predictions, you need to have a good quality mesh. Also, what near-wall treatment are you using? And what are your y+ ranges?


      Regards,


      Karthik

    • Nikhil26
      Subscriber

      I have nearly 0.78 value of average orthogonal mesh quality. standard wall function


      Also I have applied velocity normal to the boundary, so no issue with angle of attack, I am sure on that.


      But i dont know why the drag value is going negative and residuals are not converging? Can it be the problem with computer configurations??


      Using transient state,I am getting positive values but still results are not satisfying. Whats the results behind so much deflections..

    • Nikhil26
      Subscriber

      Also after analysis I am getting negative values of static pressure in pascal. Is it right???

    • Kalyan Goparaju
      Ansys Employee

      Hello, 


      100 iterations is far too less for the residuals to converge. Did you try running longer? 



      Thanks, 


      Kalyan

    • Nikhil26
      Subscriber

      Actually at inlet I have given 16.67m/s as inlet boundary condition. But after results, when I saw direction of velocity vectors they are totally in opposite direction i.e. from outlet to inlet. Is it any issue with co-ordinate system??

    • Kalyan Goparaju
      Ansys Employee

      Hello, 


      Your vector plot does indicate that the flow is from right to left. Can you share an image of your domain and boundary condition setup? 


      Thanks, 


      Kalyan

    • Keyur Kanade
      Ansys Employee

      Why vectors are inside vehicle as well?


      Is it closed body or open? What is min orthogonal quality? What are domain extends? The domain looks small as compared to the body. 


      You also may want to reduce problem size by considering symmetry. 


      Regards,


      Keyur


       


      If this helps, please mark this post as 'Is Solution' to help others.


      Guidelines on the Student Community


      How to access ANSYS help links


       

    • Nikhil26
      Subscriber

      Inlet boundary conditionsOutlet boundary conditionsFluid domain


      Here are the images of my inlet and outlet boundary conditions and domain

    • Nikhil26
      Subscriber

      yes I went for 500 iterations but direction of vectors went opposite

    • Nikhil26
      Subscriber

       Vectors are not inside the vehicle, they are over the vehicle actually!!


      I am not getting exactly why the velocity is going in opposite direction even though my boundary conditions are right.


      Everything is fine in problem setup.


       

    • Nikhil26
      Subscriber

      Can I set outlet pressure equal to 101325 Pa with diffrent courant number?

    • Kalyan Goparaju
      Ansys Employee

      Hello, 


      Is your car pointed in the correction direction? I.e. x-direction going from front to the rear of the car? Can you please resent the vector image plot with co-ordinate axis included? 


      Thanks, 


      Kalyan

    • Nikhil26
      Subscriber


      Sir, This is how axes system for my model looks like


      actually i dont have the vector image now as I have changed conditions later on so..

    • Karthik R
      Administrator

      Hello,


      It seems like you are using the following boundary conditions to solve the problem. The direction of your velocity inlet BC is also consistent with this. These seem incorrect.



      Please flip these boundary conditions. Use Velocity inlet on the front face (near the front side of the car) and pressure outlet near the backside. Also, change the 'Velocity Specification Method' to 'Magnitude, Normal to Boundary'. This should help your simulation.



      Hope this helps.


      Best Regards,


      Karthik

Viewing 14 reply threads
  • You must be logged in to reply to this topic.