# Computed Pressure Drop is lower than experimental data

 Register Blogs Members List Search Today's Posts Mark Forums Read

 May 16, 2017, 16:27 Computed Pressure Drop is lower than experimental data #1 Member   Ash Kotwal Join Date: Jul 2016 Location: North Dakota, USA Posts: 92 Blog Entries: 1 Rep Power: 9 Hello Everyone, Objective: To simulate Gas-liquid 2 Phase flow in FLUENT based on experimental data, and compare with respect to experimental data for channel with sudden area of expansion. I'm running the FLUENT 2 Phase flow (gas(Nitrogen)-liquid(Water)) Eulerian-Eulerian multi-phase model, RSM (Reynold's Stresses 5 equation model) model for turbulence calculations. it's 2D steady-state analysis. Boundary Conditions: inlet: pressure inlet, Outlet: pressure Outlet, Volume fraction is used for consideration between phases. (Values for boundary conditions are taken from Experimental calculations) Outcome: Computed Pressure Drop is lower than experimental pressure drop across sudden expansion. Why is it showing lower pressure drop compared to experimental value? How can I get exact pressure drop value for expansion channel?

 May 16, 2017, 23:56 #2 Senior Member   Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,679 Rep Power: 66 Are the differences even significant? You are using nearly a dozen models that try to mimic actual reality, if you want to exactly match results then either improve the models or stop using models altogether. You're using a multiphase model, a turbulence model, the properties are modelled, you're modelling it as steady state, etc. The list goes on and on. Of course this is assuming you made no mistakes along the way.

May 17, 2017, 09:41
#3
Member

Ash Kotwal
Join Date: Jul 2016
Location: North Dakota, USA
Posts: 92
Blog Entries: 1
Rep Power: 9
Quote:
 Originally Posted by LuckyTran Are the differences even significant? You are using nearly a dozen models that try to mimic actual reality, if you want to exactly match results then either improve the models or stop using models altogether. You're using a multiphase model, a turbulence model, the properties are modelled, you're modelling it as steady state, etc. The list goes on and on. Of course this is assuming you made no mistakes along the way.

I activated multiphase eulerian eulerian model, and for catching turbulence in pipe, because the Reynolds number for hydraulic diameter is high. For turbulence I decided to use RSM, that's all.

These are I guess basic models which we need to activate. There is no involvement of energy equation.

The model is stable and it's converging, because the inlet mass flow rate is equal to outlet mass flow rate, by speculation based on the surface monitors.

Also, I used k-epsilon per phase model, but still pressure drop is significantly low.

Value, for example, for 5g/s mixture flow rate (in terms of experimental analysis) is 0.5 kpa, while fluent shows 0.12 kpa.

Sent from my iPhone using CFD Online Forum mobile app