[Gohu8]

Hello everyone,
I'm studying the trajectory of vortex cores in the wake of a delta wing. But I have a problem. In the experiments I have read in wind tunnel tests, for the same configuration, the vortex cores are going down whereas in my fluent simulation, they are going up.

In the wind tunnel tests, the delta wing has an angle of attack of 17 degrees and the air is coming parallel to the ground (so horizontally).

In FLUENT, as I can't put an angle of attack directly on the wing, the wing stays horizontal but the wind is coming from below at an angle of 17 degrees:

https://ibb.co/c71Xjd




Here you can see the incoming flow (velocity inlet at 17 degrees angle of attack) and the vortex going upwards.

I have also put the gravity as:
X = 0 m/s2
Y = 0 m/s2
Z = -9.81 m/s2

according to the reference axis.
A previous thesis with the same data who did a fluent simulation found vortices going downwards....

I don't really know where is the problem, as I think gravity overcomes the incoming flow who is going upwards, and so at the end the vortices go downwards

Thanks for any help

[Gert-Jan]

You are plotting vectors on the inlet. You should plot streamlines or vectors on a mid plane.

[ghorrocks]

Hugo, I think you have a misunderstanding here. Gravity does not pull the vorticies down. The vorticies are air and they are surrounded by air of equivalent density - so gravity has no effect on it. The vorticies shed off the wing will just go to where the prevailing wind blows it, and that is what your image is showing.

So your image is showing exactly what I would expect.

[Gohu8]

Originally Posted by ghorrocks

Hugo, I think you have a misunderstanding here. Gravity does not pull the vorticies down. The vorticies are air and they are surrounded by air of equivalent density - so gravity has no effect on it. The vorticies shed off the wing will just go to where the prevailing wind blows it, and that is what your image is showing.

So your image is showing exactly what I would expect.

I was thinking like you at the beginning. Even if there is the same angle of attack, is it the same case between those two cases:

- The delta wing is at an angle of attack of 17 degrees to the horizontal, the wind coming horizontally like in this experiment:



- In the CFD simulation, the wing is horizontal but the wind is coming from below at an angle of 17 degrees.




In the wind tunnel experiment, they found that the vortices are going downwards. And the purpose of my CFD simulation is to find the same tendency as in reality (not exactly the same position but at least not going in the opposite direction !)

[ghorrocks]

It is not gravity which pulls the vorticies downwards, it is momentum imparted due to the angle of the wing. If it had the same angle of attack but a negative angle the vorticies would go upwards.

[Gohu8]

Ok thank you for your answer. Then I decided to create a new coordinate frame. I did it. But when I use the probe function to have point coordinates, it displays coordinates in the old coordinate frame and not in the one I created...And I didn't find where we can change the coordinate frame....

Thanks in advance

[Gohu8]

Can someone answer please ?...


Thanks

[Gert-Jan]

Why do you need a new coordinate frame in Post?

But back to your initial question.
I see a rectangular windtunnel with a delta wing at an angle of 17°.
In your simulation, you have a circular windtunnel with a horizontal delta wing (where is it? I can't see it) and an incoming wind at an angle of 17°.

Why this difference? I see no reason to not model the rectangular windtunnel. Do you have a good reason for this?
With the setup you created, the wind will by far not flow straight through your circular windtunnel. You will obtain a secondary flow which will interfere with your vortices. And there will be a wake over the bottom of your circular windtunnel.

So going back to my first post: How do the streamlines, starting at the inlet, look like in 3D? Also, make a plane through the center and plot the velocity and the vectors. What do you see?

[Gohu8]

Originally Posted by Gert-Jan

Why do you need a new coordinate frame in Post?

But back to your initial question.
I see a rectangular windtunnel with a delta wing at an angle of 17°.
In your simulation, you have a circular windtunnel with a horizontal delta wing (where is it? I can't see it) and an incoming wind at an angle of 17°.

Why this difference? I see no reason to not model the rectangular windtunnel. Do you have a good reason for this?
With the setup you created, the wind will by far not flow straight through your circular windtunnel. You will obtain a secondary flow which will interfere with your vortices. And there will be a wake over the bottom of your circular windtunnel.

So going back to my first post: How do the streamlines, starting at the inlet, look like in 3D? Also, make a plane through the center and plot the velocity and the vectors. What do you see?

This problem is now solved and it explains why I want a new coordinate frame... I created a 3d plane sloped by 17° under the wing that represents the floor. And I want to have the coordinates of the vortex core relative to this plane (and not to the trailing edge of the wing which is the initial coordinate frame).

So please have you any idea how to solve this problem ?

[ghorrocks]

In File/Export you have the option of choosing which coordinate frame you want the results in.

[Gert-Jan]

You can create a coordinate frame.
The probe doesn't work with this frame, as you noticed.
But you can export the data using this frame. So if you are able to create a streamline, an iso-surface, or something else which locates your core, you can export it.


But what is wrong with some simple mathematics? It is just a rotation over 17° and maybe a translation......

[Gohu8]

Originally Posted by ghorrocks

In File/Export you have the option of choosing which coordinate frame you want the results in.

Thanks for the answer but how do I define the coordinates of the centre of each vortex ? This is the only data that I am interested in, and when I click on export, they ask me which variable I want to export, the classic ones (velocity, density...)