Is it okay to have a closed volume with just one pressure inlet and no outlet?

- August 28, 2021 at 1:58 pm
  
  KiranS
  Subscriber
  
  Hi. I have a simple 2D rectangular geometry with one side as pressure inlet and remaining as walls. It is a transient problem, so I'm using a udf to define pressure at inlet. But as I run the calculation, after a few iterations, it run into floating point error. The pressure values run into infinity. Any suggestions to fix this?
  Thank you
- August 30, 2021 at 1:31 pm
  
  Karthik Remella
  Administrator
  
  Hello What problem are you trying to model? Having just one flow boundary might make the problem ill-posed - conservation of mass might be an issue.
  Karthik
- August 30, 2021 at 5:38 pm
  
  DrAmine
  Ansys Employee
  
  Pressure boundary will allow in and outflow so, it's okay.
  What are trying to model?
- August 30, 2021 at 7:06 pm
  
  KiranS
  Subscriber
  
  Hi. Thanks for the reply. I'm trying to model a cryocooler, where the cylinder is connected to a varying pressure, oscillating between 12 bar and 18 bar. Conservation of mass indeed seems to be the issue. My continuity equation is diverging to infinity after just 6-7 iterations. I'm a total noob. I suspected it might be due to high pressure, so I gave 15 bars as operating pressure and just the 3bar oscillation in the udf. TNow the calculation runs without any issues but the final pressure contour shows pressure varying between -3 and +3 bar, instead of 12 and 18 bar. Maybe giving operating pressure is not the right way to do this?
- September 1, 2021 at 6:56 pm
  
  KiranS
  Subscriber
  
  This worked after setting the operating pressure and reducing the time step size.
- September 7, 2021 at 1:03 pm
  
  Karthik Remella
  Administrator
  
  Thanks for letting us know ! Good luck with your model.