simulation settings for wind tunnel testing

[Anonymous]

Hi everybody,

Autodesk CFD is quite new for me, so I may not know everything. I'm currently building a model wind tunnel and I want to do the simulation to see how the air moves insde the model wind tunnel (because I'm going to build one in real life). In front of my wind tunnel, there is a fan with fixed RPMs and the air moves from the fan, pass through a section which will "convert" turbulent air flow to laminar air flow, and will eventually pass through the test section (it's an open wind tunnel). Assuming that there is no wind outside the wind tunnel in the beginning, which settings should I apply to the simulation so that the simulated air flow inside the wind tunnel would be nearly as same as the one in real life?

What I did was creating an inner air volume by using "Void fill", then velocity = 0m/s and pressure = 101325 Pa in front of the wind tunnel (using "Boundary conditions") and pressure = 0 Pa at the end. So I don't know if what I did, was correct (enough).

Thanks in avance!

[frederic.gaillard.7]

Hey, 

did you have a cfz file to share with us ? 
Thx 

fred

[Anonymous]

hey @frederic.gaillard.7 

yes, I do have my .cfz file, but for somehow I couldn't share it on this forum (I got an error). I'll try it another time.

Anyway, you can find the screenshot of my wind tunnel.

[frederic.gaillard.7]

Hello @Anonymous , 

You can sahre your cfz file through a compressed folder. 

Thx 

Fred

[Anonymous]

Hi @frederic.gaillard.7 

Here is my cfz file.

[frederic.gaillard.7]

Hello Ly, 

Nice step up,  it's looking good ! 
However there is few thing missing on your simulation before you can run it properly. 

1. Rotating Region Volume
   • You will a cylinder volume who encompass your fan. i joint a example below 
   • 
   • The orange-transparent part is the rotating region that is missing from your design right now
2. Boundary Condition 
   • The good practice for these type of scenario are the following 
     • Intlet : velocity or flow rate BC
     • Outlet : 0 Pressure gage BC
   • Right now you have a pressure and a velocity applied on the same surface. Delete the pressure boundary condition (101325Pa) and keep the 0 Pa at the intlet. 

Correct these drawback before lauching the simulation. 

here you have different ressource that might help you to overcome this problem 

Rotating Region Geometric Considerations

Best practices for rotating regions 

Hope it helps 

Fred

[Anonymous]

Hi @frederic.gaillard.7 

How can I determine the inlet flow rate if I only know the RPM of the fan? And what do you mean exactly about the rotation region (1st point)? Do I have to change the material of the object that surrounds the fan (air) to "Rotation region"?

[frederic.gaillard.7]

Hello Ly, 

''How can I determine the inlet flow rate if I only know the RPM of the fan?''

The easiest answer is to use the fan curve of your system. Do you have access to the fan curve of your ventilator ? 

If you don't have this information you will have to run it with two open boundary (0 pressure gage)

Do I have to change the material of the object that surrounds the fan (air) to "Rotation region"?

You have to create a volume around your fan. This volume must be surround by fluid. Inside this particular volume the air will rotate at 1200 RPM. 

Hope it helps 

Fred

[frederic.gaillard.7]

Do I have to change the material of the object that surrounds the fan (air) to "Rotation region"?

[Anonymous]

Hi @frederic.gaillard.7 

I did what you said about adding a "Rotating region" volume around the fan, but CFD gave me this error: "Setting cannot be applied to <fan> : No neighboring fluid found". So how can I apply "Rotation region" and "air" as materials to the volume arround the fan?

[frederic.gaillard.7]

Hello LY, 

You need 3 parts to achieve this

1. the fan itself

2. the rotating region surrounding the fan

3. and finally a fluid region who is surrounding the rotating Region

Fred

[Anonymous]

Hi @frederic.gaillard.7 

This is what I got.

[frederic.gaillard.7]

can you clear the window, i see the half of it 🙂 

THx

[Anonymous]

@frederic.gaillard.7 

Does this seem to be correct? There is a fan, which is "wrapped" by a "Rotation region" volume and that volume is surrounded by air as fluid. As Boundary Conditions I've set the pressure at the inlet and outlet equals to zero.

[frederic.gaillard.7]

Hello Ly, 

It's hard to tell from your picture if it's set correctly, but it's a step foward if you're able to run it. 

I join two picture. this is how i'm using RR. 

On the second picture you have cylindre part in yellow that represente the rotating region. Inside you have the impeller. 

You can share your last cfz. It's easier for me to look at it 🙂 

THX

Fred

[Anonymous]

@frederic.gaillard.7 

here is my recent .cfz file

[Anonymous]

Hi @frederic.gaillard.7 

I think the Rotating region volume is (only) usefull when you use a turbomechanism that its rotational plane is parallel to the moving direction of a fluid (an impeller in your example), because I tried to do at your way and the results are not quite as realistic as I expected. Or I've done something wrong, which is also a possibility.

[frederic.gaillard.7]

Hello Ly 

I think you understand the concept of a rotating region now 🙂 
Good job ! 

What you have to keep in mind is when you are using this tool, everything inside this region will rotate at a prescribed angular velocity. So if your look closely at your model, a lot of the air of your system will rotate even if they don't i really life experiment. So what you have to do is to shrink the RR volume close to the nose of your impeller : 

A more custom rotating region volume might be more suited for your purpose. This way you will minimize the angular momentum to your fluid.

Hope it help 

Fred

[Anonymous]

Hi @frederic.gaillard.7 

Is this actually quite normal? Because the airspeed in front of and behind the propeller is nearly neglectible in comparison with the airspeed inside the rotating region. 

[frederic.gaillard.7]

Hello Ly, 

Few points here : 

• Is this actually quite normal? Because the airspeed in front of and behind the propeller is nearly neglectible in comparison with the airspeed inside the rotating region. 
  • Based on my experience i will say that is quite normal after 16 iterations, specially if you set a ramp up before reaching the full speed of the impeller. It will take more iterations before the velocity spread over the entire computational domain. 
  • I dont know the entire context of your simulation but it seems appropriate to add a line in your rotating region material editor to keep the rotational speed to his operational speed. 
  • 

• Extend the length of your air volume. You will have a better continuity over your computational domain and your flow will be more realistic. 
• 

*** 
I have to double check my claim about the rotating region geometry. i'm not sure if it is mandatory to proceed the way that i told you. I will come back to you on this matter quickly. 

Regarding this issue can you send me the step file of your impeller i will make some test if your allowed to. 

Thx 

Fred

[Anonymous]

Hi @frederic.gaillard.7 ,

Here is my latest .cfz file. You can run simulations on it