Air-oil mixture

Jiricbeng · October 12, 2020, 08:28

Hi there,

I simulated a hydrodynamic bearing operation being lubricated by oil. Since it is possible that air may be sucked into the bearing, I included the air onto a surface as boundary condition. The trouble is, that the computation runs only with Air at 25C which has constant density. Applying Air ideal gas leads to divergence due to overflow. Playing with timescale does not work as well as improving mesh quality. Are there any ideas for oil - ideal gas mixture?

ghorrocks · October 12, 2020, 20:36

It is normal that when you make the physics more complex that convergence is harder. You have replaced an incompressible fluid with a compressible one, so now you should expect pressure wave effects, compression, adiabatic heating and cooling and all those sort of complex physics.

Without specific details I cannot help, but this simulation sounds highly transient and will require a VERY small time step. Have a look at the CFL numbers for your flow now you are using a compressible fluid, that will guide you to the sort of time step size you need.

Jiricbeng · October 13, 2020, 04:01

Thank you for the reply. Is there any document how to set CFL for multiphase analyses? Must the Courant number be always below 1?

ghorrocks · October 13, 2020, 04:55

No, CFX is an implicit solver and is not restricted to a CFL or Courant number. I generally recommend you use adaptive time stepping, homing in on 3-5 coeff loops per iteration for transient simulations.

If you want to Courant Number (or CFL for compressible flows) then aim for about 1 to 10 - the actual requirement could be higher or lower than that, but that is a starting point.

October 12, 2020, 08:28	Air-oil mixture	#1
Jiricbeng Senior Member Jiri Join Date: Mar 2014 Posts: 221 Rep Power: 13	Hi there, I simulated a hydrodynamic bearing operation being lubricated by oil. Since it is possible that air may be sucked into the bearing, I included the air onto a surface as boundary condition. The trouble is, that the computation runs only with Air at 25C which has constant density. Applying Air ideal gas leads to divergence due to overflow. Playing with timescale does not work as well as improving mesh quality. Are there any ideas for oil - ideal gas mixture?

October 12, 2020, 20:36		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,816 Rep Power: 144	It is normal that when you make the physics more complex that convergence is harder. You have replaced an incompressible fluid with a compressible one, so now you should expect pressure wave effects, compression, adiabatic heating and cooling and all those sort of complex physics. Without specific details I cannot help, but this simulation sounds highly transient and will require a VERY small time step. Have a look at the CFL numbers for your flow now you are using a compressible fluid, that will guide you to the sort of time step size you need. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

October 13, 2020, 04:55		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,816 Rep Power: 144	No, CFX is an implicit solver and is not restricted to a CFL or Courant number. I generally recommend you use adaptive time stepping, homing in on 3-5 coeff loops per iteration for transient simulations. If you want to Courant Number (or CFL for compressible flows) then aim for about 1 to 10 - the actual requirement could be higher or lower than that, but that is a starting point. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Air Liquid Mixture simulation in a hydrocyclone	Navillera	Fluent Multiphase	2	September 7, 2019 23:50
multiphaseEulerFoam convergence problems	Stefanie.S.W.	OpenFOAM Running, Solving & CFD	6	August 28, 2019 04:15
How to derive an Equation of State for a mixture of Liquid-vapor and air	mg.mithun	Main CFD Forum	0	April 12, 2016 11:32
Radiation interface	hinca	CFX	15	January 26, 2014 17:11
air bubble is disappear increasing time using vof	xujjun	CFX	9	June 9, 2009 07:59

October 13, 2020, 04:01		#3
Jiricbeng Senior Member Jiri Join Date: Mar 2014 Posts: 221 Rep Power: 13	Thank you for the reply. Is there any document how to set CFL for multiphase analyses? Must the Courant number be always below 1?