[huangxianbei]

hi everyone
I'm doing a simulation on the rotating channel in which the water is driven by pressure gradient. I think it's reasonably to use momentum source to implement the pressure gradient for the dimension of pressure gradient is just the same as the momentum source, kg m^-2 s^-2.

The channel is rectangular and the rotating axis is set at the center of the left face where x=0, the x direction is the streamwise ,and the z axis is the rotating axis, y is spanwise. z and x direction are set with 2 pairs of periodic interfaces to make the flow fully developed. However, the results shows no asymmetry aross the y direction and the speed in the x direction is far away from expected, is this method wrong or in other words, no suitable in this problem?

[ghorrocks]

Please post an image of what you are modelling and an image of what you are seeing in the results.

Accuracy is an FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[huangxianbei]

Originally Posted by ghorrocks

Please post an image of what you are modelling and an image of what you are seeing in the results.

Accuracy is an FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[IMG]C:\Users\dodo\Desktop\QQ截图20140115104055.png[/IMG]
the calculation domain is as the image shows[IMG]C:\Users\dodo\Desktop\QQ截图20140115104337.png[/IMG]
here is the result, it's interesting to find that the slop of the calculation curve is the same as the slop in theory, but the y axis U is about 20 time larger than the theory

[ghorrocks]

The attached images did not work, please try again.

[huangxianbei]

Originally Posted by ghorrocks

The attached images did not work, please try again.

Hi, ghorrocks
I tried to submit my image, while only the URL can be used. Is it impossible to use local file in my computer?

[ghorrocks]

No, you cannot see files on your local computer over the web. Not unless your PC is a web server that is

You will have to upload it to the forum.

[huangxianbei]

Originally Posted by ghorrocks

No, you cannot see files on your local computer over the web. Not unless your PC is a web server that is

You will have to upload it to the forum.

the non-dimensional velocity U is done by u/ut， u is the dimensional velocity and ut is the friction velocity which is a constant, in the tunnel flow theory, the slope of the U curve is about -2w, y is the non-dimensional variable which implies the y direction,y=-1 and +1 refers to suction side and pressure side of the channel.

The second picture is the calculation domain of the tunnel, the flow is in the x direction and the whole tunnel rotates with z axis which is set at the center of the left side of x direction surface. Both the x and z direction were set with a pair of translational periodic interfaces.

[ghorrocks]

I have no idea what the diagram means, what the graph is of and what your flow is. Please put some description on your images so we can understand it.

[huangxianbei]

here is the results of DNS, and the condition I chose is Ro=0.5, it's obvious that the U is far away from the DNS results

[ghorrocks]

You still have not explain what the graph is and what the diagram means. I just see a box.

[huangxianbei]

Originally Posted by ghorrocks

I have no idea what the diagram means, what the graph is of and what your flow is. Please put some description on your images so we can understand it.

Hi ,ghorrocks

Thank you for your advise. I tried my best to explain the problem now, if there is any misunderstanding, I'll go on to explain more details