Mixture model not give me accurate results?

mariam.sara · July 31, 2015, 12:20

Hello I am attempt to model water and Al2O3 particles flowing inside a tube with a constant heat flux at the tube wall. I applied the mixture model without granular to simulate the problem I chose water liquid as primary phase and Al2O3 particles as the secondary phase I set the diameter of the particles as 45 nm the problem is mixture model not give me temperature contours with values less than expected? what the problem? may somebody guide me to the reason?

mariam

mariam.sara · August 2, 2015, 10:30

anyone reply to my question?

CFDYourself · August 14, 2015, 06:39

Your question isn't clear.

Are you saying that it has cooled more than you expected or less than you expected ?

mariam.sara · August 14, 2015, 07:12

Thanks CFDYourself for the reply. I mean the results of mixture model for temperature are less than the results predicted from experimental work for the same fluid properties, boundary conditions....etc

mariam

CFDYourself · August 14, 2015, 10:13

Alright, what you are describing in this context is model errors.

Discrepancies exist between data and calculations occur even when numerical error in the calculation has been accounted for. I'm assuming you have achieved grid independence and a converged solution.

A lot of CFD processes rely on empirical models and make a lot of assumptions. k-e for example, famously uses empirical constants derived from simple lab test cases, and makes sweeping assumptions such as isotropic flow (boussinesq hypothesis) and wall functions. As a result standard k-e can be massively out. But its very much horses-for-courses, short of full DNS there's no "good" and "bad" turbulence model, it depends what model is appropriate to the flow you are trying to model.

systematic errors are errors resulting because of differences between your experimental rig and the conditions in the model, such as flow conditions, fluid properties etc.

You should be sure of the uncertainty in your experimental results.

Bear in mind also that the two phase model itself is full of assumptions that may or may not count for your case. What is the agreement like when there are no particles in the flow, in both the experiment and the simultion ? this will tell you whether the issue is due to the flow modelling, or due to the multiphase modelling.

mariam.sara · August 14, 2015, 10:52

Hi james
The experimental results I want compare with not mine its from a paper from the literature. No turbulence in my case its a laminar case representing by flow of nano particles of Al2O3 with water into thin tube and there is a constant heat flux at tube wall the geometry is 2D and the mesh quality is good and convergence achieved.

mlooti · May 25, 2016, 09:16

Hi sara

do you manage to solve your problem? I have the same cases.....

mariam.sara · June 3, 2016, 15:47

No my problem not solved yet with mixture model? Do you have any ideas may assist me?

Quote:

Originally Posted by mlooti

Hi sara

do you manage to solve your problem? I have the same cases.....

ykb3950 · August 4, 2016, 02:21

hi.....
please tell me how to apply solid in secondary phase because i am trying water + ash is used in mixture model. water is select for primary phase but ash(not only any solid ) is not select in secondary phase.
please tell how to solve this problem in fluent

hkhosravi · August 12, 2016, 02:19

Hi mariam
I faced with same problem, but my result was acceptable. Which of viscosity and conductivity model were used? Your experimental data related to Kim [2009] paper?

July 31, 2015, 12:20	Mixture model not give me accurate results?	#1
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Hello I am attempt to model water and Al2O3 particles flowing inside a tube with a constant heat flux at the tube wall. I applied the mixture model without granular to simulate the problem I chose water liquid as primary phase and Al2O3 particles as the secondary phase I set the diameter of the particles as 45 nm the problem is mixture model not give me temperature contours with values less than expected? what the problem? may somebody guide me to the reason? mariam

August 4, 2016, 02:21	mixture model	#9
ykb3950 New Member yatish Join Date: Aug 2016 Posts: 3 Rep Power: 10	hi..... please tell me how to apply solid in secondary phase because i am trying water + ash is used in mixture model. water is select for primary phase but ash(not only any solid ) is not select in secondary phase. please tell how to solve this problem in fluent

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Tutorial.the mixture model and evaporation-condensation error	orchiders	FLUENT	3	January 22, 2017 03:09
Superlinear speedup in OpenFOAM 13	msrinath80	OpenFOAM Running, Solving & CFD	18	March 3, 2015 06:36
air and water vapour mixture - multiphase model	Saba	FLUENT	0	February 10, 2009 13:05
mixture model for rotating bowl	garima chaudhary	FLUENT	1	January 30, 2008 19:24
Mixture and Eulerian Model	Hassid S.	FLUENT	1	August 9, 2007 08:24

August 2, 2015, 10:30		#2
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	anyone reply to my question?

August 14, 2015, 06:39		#3
CFDYourself Member Jim Join Date: May 2015 Posts: 47 Rep Power: 11	Your question isn't clear. Are you saying that it has cooled more than you expected or less than you expected ?

August 14, 2015, 07:12		#4
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Thanks CFDYourself for the reply. I mean the results of mixture model for temperature are less than the results predicted from experimental work for the same fluid properties, boundary conditions....etc mariam

August 14, 2015, 10:13		#5
CFDYourself Member Jim Join Date: May 2015 Posts: 47 Rep Power: 11	Alright, what you are describing in this context is model errors. Discrepancies exist between data and calculations occur even when numerical error in the calculation has been accounted for. I'm assuming you have achieved grid independence and a converged solution. A lot of CFD processes rely on empirical models and make a lot of assumptions. k-e for example, famously uses empirical constants derived from simple lab test cases, and makes sweeping assumptions such as isotropic flow (boussinesq hypothesis) and wall functions. As a result standard k-e can be massively out. But its very much horses-for-courses, short of full DNS there's no "good" and "bad" turbulence model, it depends what model is appropriate to the flow you are trying to model. systematic errors are errors resulting because of differences between your experimental rig and the conditions in the model, such as flow conditions, fluid properties etc. You should be sure of the uncertainty in your experimental results. Bear in mind also that the two phase model itself is full of assumptions that may or may not count for your case. What is the agreement like when there are no particles in the flow, in both the experiment and the simultion ? this will tell you whether the issue is due to the flow modelling, or due to the multiphase modelling.

August 14, 2015, 10:52		#6
mariam.sara Senior Member Join Date: Jan 2011 Posts: 339 Rep Power: 16	Hi james The experimental results I want compare with not mine its from a paper from the literature. No turbulence in my case its a laminar case representing by flow of nano particles of Al2O3 with water into thin tube and there is a constant heat flux at tube wall the geometry is 2D and the mesh quality is good and convergence achieved.

May 25, 2016, 09:16		#7
mlooti Member Tom Join Date: Oct 2015 Posts: 67 Rep Power: 11	Hi sara do you manage to solve your problem? I have the same cases.....

August 12, 2016, 02:19		#10
hkhosravi New Member Hamed Kh. Join Date: Dec 2010 Location: Tehran , Iran Posts: 18 Rep Power: 16	Hi mariam I faced with same problem, but my result was acceptable. Which of viscosity and conductivity model were used? Your experimental data related to Kim [2009] paper?