Fluent simulation of Hydrocyclone.. need help pls...

veronicayeo · October 19, 2009, 08:33

Hi guys,

I am trying to predict my hydrocyclone (HC) model using fluent and i have bumped into several problems. Here, i'm trying to model water liquid rotating through the HC, so i have one inlet and two outlets (overflow and underflow). The HC inlet has a pressure of 60psig and mass flow of 0.988kg/s, overflow having a pressure of 30psig and underflow of 45psig.

The problem is the inlet pressure. I've set the operating pressure to 0Pa to analyse just the gauge pressure. But, when i set the inlet pressure to 60psig (414000Pa), the mass flow inlet gives me approximately 4.5kg/s, which is more than i need. Furthermore, the velocity when initialized is somehow incorrect as i predicted using the simple (m=rho*velocity*cross-sectional area).

Aside from that, i specify the outlets as 200000Pa and 300000Pa, then try to run first using RSM model with second order upwind, as stated in most journals. However the residual has a diverging nature, even though the solutions have been relaxed (decreased around 0.2-0.3).

Can someone please provide any suggestion to me? Thx in advance.

esozer · October 19, 2009, 23:29

You are over-specifying the boundary conditions. I think you should remove the inlet pressure specification and only enforce the flow rate. Since you are fixing the pressure at the outlets, inlet pressure should follow from the solution.

veronicayeo · October 19, 2009, 23:54

Oh..k i get it. So i've changed my inlet to msas flow inlet. However i have another question about the operating pressure. Default is 101325, which i presume is the atmospheric condition. the coordinates probably indicates where this pressure would be. So if i set it to 0Pa, does it mean i would just be looking at gauge pressure? If i set it to my inlet, will fluent calculate my inlet based on gauge?

esozer · October 20, 2009, 00:07

Well, I don't know about fluent specifics. But in an incompressible simulation, the pressure values in your simulation doesn't mean anything. All you care about is the pressure variations. You can add or subtract any pressure value from the whole domain. So I guess:

with reference pressure set to 0, you will have absolute pressure results

with reference pressure set to 1 atm, your results would be gauge pressure.

The flow fields should be exactly the same regardless of this choice. Just be consistent with your BC's and initial conditions.

October 19, 2009, 08:33	Fluent simulation of Hydrocyclone.. need help pls...	#1
veronicayeo New Member Join Date: Jun 2009 Posts: 9 Rep Power: 16	Hi guys, I am trying to predict my hydrocyclone (HC) model using fluent and i have bumped into several problems. Here, i'm trying to model water liquid rotating through the HC, so i have one inlet and two outlets (overflow and underflow). The HC inlet has a pressure of 60psig and mass flow of 0.988kg/s, overflow having a pressure of 30psig and underflow of 45psig. The problem is the inlet pressure. I've set the operating pressure to 0Pa to analyse just the gauge pressure. But, when i set the inlet pressure to 60psig (414000Pa), the mass flow inlet gives me approximately 4.5kg/s, which is more than i need. Furthermore, the velocity when initialized is somehow incorrect as i predicted using the simple (m=rhovelocitycross-sectional area). Aside from that, i specify the outlets as 200000Pa and 300000Pa, then try to run first using RSM model with second order upwind, as stated in most journals. However the residual has a diverging nature, even though the solutions have been relaxed (decreased around 0.2-0.3). Can someone please provide any suggestion to me? Thx in advance.

October 19, 2009, 23:29		#2
esozer New Member Emre Join Date: Oct 2009 Location: Ann Arbor, MI Posts: 12 Rep Power: 16	You are over-specifying the boundary conditions. I think you should remove the inlet pressure specification and only enforce the flow rate. Since you are fixing the pressure at the outlets, inlet pressure should follow from the solution. __________________ Free CFD developer www.freecfd.com

October 20, 2009, 00:07		#4
esozer New Member Emre Join Date: Oct 2009 Location: Ann Arbor, MI Posts: 12 Rep Power: 16	Well, I don't know about fluent specifics. But in an incompressible simulation, the pressure values in your simulation doesn't mean anything. All you care about is the pressure variations. You can add or subtract any pressure value from the whole domain. So I guess: with reference pressure set to 0, you will have absolute pressure results with reference pressure set to 1 atm, your results would be gauge pressure. The flow fields should be exactly the same regardless of this choice. Just be consistent with your BC's and initial conditions. __________________ Free CFD developer www.freecfd.com

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
IC engine simulation tutorial by FLUENT	yash	FLUENT	4	April 6, 2015 19:44
IC engine simulation tutorial by FLUENT	yash	Main CFD Forum	1	April 21, 2009 05:05
Simulation of savonius rotor in Fluent	Kirit	FLUENT	0	October 20, 2008 12:05
Warning while fluent simulation	Madhukar Rapaka	FLUENT	8	June 21, 2006 04:18
acoustics simulation with FLUENT	sxf	FLUENT	0	April 15, 2003 12:41

October 19, 2009, 23:54		#3
veronicayeo New Member Join Date: Jun 2009 Posts: 9 Rep Power: 16	Oh..k i get it. So i've changed my inlet to msas flow inlet. However i have another question about the operating pressure. Default is 101325, which i presume is the atmospheric condition. the coordinates probably indicates where this pressure would be. So if i set it to 0Pa, does it mean i would just be looking at gauge pressure? If i set it to my inlet, will fluent calculate my inlet based on gauge?