Simulating convection depending on density model

burgerma61399 · January 16, 2019, 08:35

Hey everybody,
currently i´m working on my Bachelor-these which is a CFD Simulation of an oven with a constant flow in and out.
One step is to look at the influence of the natural convection of the flow field. The reynolds number inside is pretty low without the convection (Re= 60)
The temperature between the flow at the Inlet and the heated wall is very high. (1000 K)
Now for the natural convection i need to set a density model.
I read a lot about the ideal gas, the incompressible ideal gas and the boussinesq model but nothing specific to my case and which one fits the best.
I guess the boussinesq is wrong because of the high difference in temperature. And because the of that i guess that the pressure difference inside the oven will be to significant to just use the incompressible model...
Can anybody correct me or got a solution for this problem?!

LuckyTran · January 16, 2019, 10:59

Nobody can tell you because these are different ways of modeling. You choose what you want.

If you want the results to look like reality, then choose ideal gas since that most closely mimics reality. There are reasons people use the incompressible ideal gas law and Boussinesq and that's why those models exist.

Which one fits best is entirely subjective and depends on the goal. If your goal is to model this case using the Boussinesq model, then you fail that goal by using ideal gas. What is the objective?

burgerma61399 · January 17, 2019, 04:02

First of all, thank you for your reply.
My goal is to model the flow field as real as possible depending on the convection. So your answer that ideal gas model is the best to mimic reality is exactly what i wanted to hear. Thank you!

I´ve been insecure about it because everywhere i looked regarding convection, there was only incompressible and boussinesq model mentioned...

LuckyTran · January 17, 2019, 07:08

Let's just explore an example. In any introductory fluid mechanics book, the first six chapters are devoted to incompressible, inviscid fluid mechanics despite the rarity of incompressible substances or inviscid fluids. If your exposure to fluids is only this much then you would mistakenly believe the entire universe is full of incompressible substances and non-viscous fluids. Look further and you will see more. Still I would not remove these first six or so chapters at the beginning of fluids text, because they establish a nice foundation for learning things.

Of course ideal gas isn't the end either. There are real gas models beyond ideal gas that have been known for more than a century (and Fluent refprop built-in to calculate a lot of these properties much more accurately than ideal gas model). Someone more picky than you would question why you are using ideal gas and not a real gas model. It never ends... You must start somewhere. "Something mediocre is better than nothing perfect."

January 16, 2019, 08:35	Simulating convection depending on density model	#1
burgerma61399 New Member Join Date: Jan 2019 Posts: 4 Rep Power: 7	Hey everybody, currently i´m working on my Bachelor-these which is a CFD Simulation of an oven with a constant flow in and out. One step is to look at the influence of the natural convection of the flow field. The reynolds number inside is pretty low without the convection (Re= 60) The temperature between the flow at the Inlet and the heated wall is very high. (1000 K) Now for the natural convection i need to set a density model. I read a lot about the ideal gas, the incompressible ideal gas and the boussinesq model but nothing specific to my case and which one fits the best. I guess the boussinesq is wrong because of the high difference in temperature. And because the of that i guess that the pressure difference inside the oven will be to significant to just use the incompressible model... Can anybody correct me or got a solution for this problem?!

January 17, 2019, 04:02		#3
burgerma61399 New Member Join Date: Jan 2019 Posts: 4 Rep Power: 7	First of all, thank you for your reply. My goal is to model the flow field as real as possible depending on the convection. So your answer that ideal gas model is the best to mimic reality is exactly what i wanted to hear. Thank you! I´ve been insecure about it because everywhere i looked regarding convection, there was only incompressible and boussinesq model mentioned... Last edited by burgerma61399; January 17, 2019 at 06:23.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Tait Density model (PVT)	jpina	FLUENT	0	February 18, 2016 06:48
Wrong flow in ratating domain problem	Sanyo	CFX	17	August 15, 2015 06:20
How to model Salinity, density, fresh water & sea water	grayback87	FLUENT	3	December 30, 2014 02:00
Overflow Error in Multiphase Modelling with Two Continuous Fluids	ashtonJ	CFX	6	August 11, 2014 14:32
Model foaming process: change the volume / density	pmp77	CFX	0	November 18, 2008 09:53

January 16, 2019, 10:59		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,673 Rep Power: 65	Nobody can tell you because these are different ways of modeling. You choose what you want. If you want the results to look like reality, then choose ideal gas since that most closely mimics reality. There are reasons people use the incompressible ideal gas law and Boussinesq and that's why those models exist. Which one fits best is entirely subjective and depends on the goal. If your goal is to model this case using the Boussinesq model, then you fail that goal by using ideal gas. What is the objective?

January 17, 2019, 07:08		#4
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,673 Rep Power: 65	Let's just explore an example. In any introductory fluid mechanics book, the first six chapters are devoted to incompressible, inviscid fluid mechanics despite the rarity of incompressible substances or inviscid fluids. If your exposure to fluids is only this much then you would mistakenly believe the entire universe is full of incompressible substances and non-viscous fluids. Look further and you will see more. Still I would not remove these first six or so chapters at the beginning of fluids text, because they establish a nice foundation for learning things. Of course ideal gas isn't the end either. There are real gas models beyond ideal gas that have been known for more than a century (and Fluent refprop built-in to calculate a lot of these properties much more accurately than ideal gas model). Someone more picky than you would question why you are using ideal gas and not a real gas model. It never ends... You must start somewhere. "Something mediocre is better than nothing perfect."