CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Cavitation Modelling Options (http://www.cfd-online.com/Forums/cfx/105861-cavitation-modelling-options.html)

 Torque_Converter August 10, 2012 21:10

Cavitation Modelling Options

I'm working on properly modelling cavitation and I know I have some options, but some very incomplete data. I was hoping I could get some of your guys expert guidance on which way to go.

I would like to model the oil vapor as an ideal gas, but I do not know the molar mass, in fact I have NO clue. With the other properties I have some idea. I know density, mean particle size, dynamic viscosity, and saturation pressure at 90 C, the temperature of isothermal operation.

I know instead of ideal gas I could program in properties vs. pressure. If I somehow make an educated guess at this, would it be better than attempting to use ideal gas, or not? If so, how would I write CEL code for rho(P) for example? I haven't seen much in the literature.

Also, if anyone has any insights on the vapor properties of an oil with density at 90 C of about 800 kg/m^3 it would be of great help.

Thanks all

P.S. any additional tips on cavitation modelling you can add that I may not have asked about would also be greatly appreciated to build the best model I can.

 ghorrocks August 12, 2012 08:09

I would not bother modelling the vapour as an ideal gas. Its properties are so far away from an ideal gas as to be not worht the effort.

I assume you are currently assuming a fixed density vapour. Before doing anything on variabel density vapours I would do a sensitivity analysis to determine if it makes any difference. Try doubling and halving the vapour density and see if it makes any difference to the results. If it makes no difference then I think you will find variable density is insignificant and so not worth implementing.

 Torque_Converter August 12, 2012 12:08

That sounds like a good strategy to go forward on. I have a feeling that changing the constant density won't do much.

What seems to be lacking is the drop off in torque production at higher RPMs.

The reason I was grasping for some perfect cavitation/vapor modelling is because I have this device that I just can't get. I thought I had nailed every other factor (geometry, boundary conditions, liquid properties, speed/conditions of test). I guess cavitation, while present in the model, is unlikely to resolve my problems by taking a guess at what might be slightly more accurate, or very likely, no more accurate.

 All times are GMT -4. The time now is 20:59.