Hi, Have thoroughly researched this subject in this forum, so I hope we can begin at that point. Does anyone have more recent ideas on how this problem occurs? Regarding the theory that the solver puts a -ve pressure when cavitation is present, can anyone cite a reference? This is not an issue of understanding gage pressure versus absolute pressure as has been suggested in past discussions. Have tried all of the suggestions made on previous threads. I know exactly what those poor people were up against! Suggestions greatly appreciated. Thanks, Doug

Hi Doug,

I will give you some information which iam familiar with.

In Fluent Example for cavitation.

The fluid moves from a high pressure zone to the low presure outlet defined as a boundary condition.It needs to pass through a orifice with a small diameter.

The high pressure of the liquid will be changed to drastic low pressure with a very high velocity increasing the dynamic pressure giving rise to the transfer of masses and energy between phases.

the cavitation model is initiated once the pressure falls below a certain level.the values of the flow variables will be obtained from the center of the cell based on the partial differential equations solved.

It need not be a negative pressure to cause cavitation i presume.

Hope these are helpful.

Regards

Rajit

I am the newcomer in the context, if the cavitation model is not activated the domain where the cavitation is going to accure will give the pressure less than the cavitation pressure.

If you set the cavitation on then domain result should show the cavitation happening at the domain, but in any case the pressure should not fall to absolute pressure its not physical. But certainly it can go below vapor pressure of the fluid and if vapor pressure is lower than gauge then gauge pressure is -ve.

I worked on the similar problems while coading for Flowmaster, there presenting cavitation is a trick by putting virtual -ve pressure, this is only to fool the solver, so that it wont check the convergence and mass balance for that domain, I hope Fluent solver also does the same.

but it seems a bug in Fluent as it doesnt convert back the pressure values and reports only -ve pressure value

Gentlemen, Thank you both for your time in replying to a student with limited CFD experience. I believe that you are both correct in that there is cavitation present in the flow. When I used the cavitation model (which I had avoided due to lack of familiarity) I got results that are much more logical. Convergance is more of a problem with the cavitation model, however. Do you have any recommendations for appropriate solver settings that might aid convergence? Model is time steady 3D, with k-epsilon turbulence model, water is the fluid. Again, thank you!

Regards,

Doug

[shahid nadeem]

Dear Friends,
I Would Like to share my experience regarding negative pressure.
i am solving a Perforated Pipe Distributor, the flow enters at atmospheric pressure and exits through the holes to the vaccum vessel maintained at 1 Pa.
The inlet area is 1500 mm and at the outlet there are 16600 number of holes of 10 mm diameter each.

After post processing i am getting a negative absolute pressure of -1358 Pa i am wondering how can there be a negative absolute pressure. WHether it is possible in vaccum conditions or is there any error in solving, if you have any idea please help me.

The Gas is Incompressiblel air at 333
Inlet: 33.33 Kg/s
Outlet 1 Pa