Modelling Fluid Flow Through a Solid Ceramic Foam

[Anonymous]

Hi, 

I am quite new to CFDs. Using X Ray tomography I was able to take a CT Scan of a Reticulate Porous Ceramic. I then used Fusion 360 to repair its mesh and exported it as a .obj file after endlessly failing to do it through a stl or step file. I then imported this .obj onto CFD by creating an external volume which tightly encapsulates the solid, as I want to model fluid flow through the solid, as opposed to around it like most other applications. 

I am trying to model flow of Steam at 1200 degree C through the porous Ceramic foam (Silicon Carbide) and have tried many techniques that I read on multiple forums. There are many different problems I am facing:

1) I cant keep the fluid at 1200 and the solid at 24 degrees. Even if I set these properties at initial conditions, and boundary conditions both, as soon as I run the first iteration, either both fluid and solid are at 24 or 1200 C.

Note: Both Fluid and Solid are set to Variable Properties using the scenario environment.

2) I cant model the flow through the ceramic, velocity magnitude through that region is 0.

3) Static Pressure, Wall Forces, Strain Rate, Shear Stress Values are all logically incorrect. 

I am looking to calculate the temperature gradient, pressure gradient, and whatever mechanical properties CFD can solve for to test the mechanical and thermal performance of this ceramic under the fluid flow. 

I am unable to attach the CFZ file here (It is very large due to the ceramic mesh size). I have uploaded it to my onedrive and the link is as follows: 

https://anu365-my.sharepoint.com/:f:/g/personal/u5922125_anu_edu_au/EtnEQ9p5ARRNrdsdQuASzuMB1p43Rr7k...

I need to urgently produce results, all help is greatly appreciated. 

[frederic.gaillard.7]

Hello @Anonymous , 

Really cool project, i used to work with CT scan with heavy material with high atomic number and the beam hardening aliasing. Really rich physic behind this !
Anyway regarding to your project, after a quick look, i notice that you had recirculation over your outlet mainly caused by a lack of mesh. 

outlet recirculation

Before adding the thermal layer to your analysis, i recommand you to focus only on the flow. Right now your continuity equation (mass conservation principle) is not respected and you can not have reliable result. 

To overcome this particular difficulty i suggest you two things : 

1. add a dense and uniform mesh over your water (vapor) volume 
2. You might need to extend your external volume as well. Your ceramic solid is to close of your 0 pressure boundary condition. i'm pretty sure that this particular disposition is responsible for the recirculation behavior. 

In this thread the problem of recirculation is adressed in more detail.

Furthermore, your solution must be resolve with turbulence. In this case laminar flow doesn't occurs specially with high reynolds number. 

Last thing, as i said earlier, try to achieve a good flow simulation first. Uncheck hydrostatic pressure, change the compressibility to ''incompressible''

If you have other question don't hesitate.  We will tackle the thermal consideration once your flow will be  stabilized and modelize correctly.
Good Luck

Fred

[Anonymous]

Hi frederic.gaillard.7,

First and foremost, thank you very much for your prompt reply and for helping me out with the identifying some problems with the gas flow. I took your recommendations on board, increased the length and mesh density of the external volume which was modelling the gas flow. I have even generated a lot of results for different combinations of gases and ceramic materials under the following conditions:

Inlet:

Mass Flow Rate 1kg/s, Pressire 1.5 atm gage, temperature 1200 degree C

Outlet:

Pressure - 1 atm Gage.

I noticed that most simulations and forum posts online use a Pressure of 0 Pa at the outlet. However, my supervisor said not to use 0 Pa, as that would mean the gas is being released to vacuum or model an unrealistic scenario. Since we have a facility where this can be experimentally tested, he wants me to use an outlet pressure of 1 atm. 

Now, I have two problems:

(1) I am supposed to check the results for air at inlet with 1.5 atm and with 10 atm. I am not sure if this over-constraints the problem in any way. When CFD does simulate it my results are shocking, the max static pressure for 1.5 atm is around 728,705 Pa whereas for 10 atm is  635,584Pa.

(2) I am not sure if my results are correct, and if I can trust them. 

Can you please advise me on this? Again, I have sent the new share file to you via a Private Message. 

Thank you very much and I look forward to hearing from you soon.

Cheers,

Shikhar

[frederic.gaillard.7]

Hello Shikhar, 

About The pressure, more specifically 0 pressure gage, it only means that the discharge is under standard atmospheric condition. If you check closely you several type of pressure : 

type of pressure

Here is some intersting reading about pressure BC. I'm sure it will answer most of your question. 

It will check your result later and give you some news.

Thx 
Fred

[frederic.gaillard.7]

Hello Shikhar, 

Extra comment about your model that you shared. 

There is a few issue related to your boundary condition and your geometry. 

Boundary Condition issue

Boundary Condition

In your design you should have only one pressure BC located at the exit of your model. 

Right now you have 1 at the intlet and 4 on the lateral surface, those should be removed. What they will do is they will introduce fluid recirculation through these surface and screw up your mass conservation principle (continuity equation).  Simply by applying the right pressure at the exit, static pressure will slowly rose all over your computational domain. The rise of the pressure will depend on your mass flow at the enter of your model. 

External Geometry

What are representing the air volume surrounding your ceramic foam ? 

Is it a wind tunnel experiment ? The air volume represent the physical dimension of that tunnel ?

Thx 
Fred