Learning Forum

[amkum]

Hello all,

In a few multiphase simulations, the absolute pressure falls below zero and goes negative near the throat. 

I am using the Zwart-Gerber-Belamri cavitation model. Fluid is nitrogen. 

The reference pressure is 0 Pa.

I also checked the pressure limiter and set the minimum absolute pressure to 1000Pa. No change was observed even after this. 

 

Thanks in advance!

Amit

 

 

[Rob]

Turn off node values to check it's a true result and not due to contour smoothing. The multiphase models (including cavitation) don't always use the pressure limiter so if convergence isn't great a nonphysical pressure can be calculated. 

[amkum]

Hi!

Thanks for the reply.

I turned-off the node value still I can see negative pressure near the throat.

It can be also seen in the pressure plot. 

 

 

[Rob]

What gas density did you use? 

[amkum]

Hi,

I have simulated in two different ways.

1st: Imposing the constant fluid properties

2nd: Imposing fluid properties using the RGP table.

Both solutions converged. 

 

Thanks,

Amit

 

 

 

 

[Rob]

And both give a negative pressure? If you take the gas density what pressure would that be at? 

[amkum]

Yes!

Both gives the negative pressure.

The constant fluid properties of Nitrogen was imported from the Fluent Database. I am using the default values in case of constant fluid properties case. (Ref Tem 298.15K).

However, changing the evaporation coefficient for Cavitation model changes it. If I change the evaporation coefficient from 0.25 to 0.7, the  Absolute pressure value changes to positive. However, the mass flow reduces from the 0.22kg/s to 0.16kg/s, which is not desirable. The experimental mass flow rate is 0.28kg/s. Therefore, the mass flow error increases if I use evaporation coefficient of 0.7.

If I reduce the evaporation coefficient from 0.25 to lower values, the absolute pressure near the throat becomes more and more negative.

 

 

 

[Rob]

The default is 50, why did you drop it that far? https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v231/en/flu_th/flu_th_sec_multiphase_cavitation.html  They're factors so may need some tuning. 

 

 

[amkum]

Thanks for reply.

Yes! The default value of Fevp is 50. However, This needs to be reduced to have a better match between experimental data ( mass flow , pressure profile, and vapor fraction).

Incorrect values of evaporation and condensation coefficients also resulted in a divereged solution. 

 

[Rob]

OK, another possible is the rate of phase change for the cavitation. They're generally a pain to converge well, so also look at time step. If the pressure goes negative it means you should have either lower density gas (it expands) or more gas: if the mass exchange is unstable you may need to switch to transient. By unstable I mean bubbles form & collapse quickly: the bulk mass fraction may remain relatively unchanged. 

[amkum]

Thanks for the reply!

I will set up the transient run and keep you posted. 

Thanks a lot for your suggestions!