[ddungntu]

Dear all,

I am trying to extract the time history of pressure on specific area. I know how to use the monitor point to extract the pressure time history, but I do not know how to get the pressure history on a specific region.

Could anyone can please help me to how to get the pressure time history on a specific region in CFX?

Thank you in advance!

Best regrads,
Dac

[urosgrivc]

Go into CFDpost
Create a Chart
Click XY - Transient or sequence
X axis will be the time, and you can put whatever you like on Y axis

[ddungntu]

Originally Posted by urosgrivc

Go into CFDpost
Create a Chart
Click XY - Transient or sequence
X axis will be the time, and you can put whatever you like on Y axis

Thank you for your help.

Actually, I wonder to plot the pressure time history acting on the specific area (not at points), but I do not know how to create the region in CFD-Post in order to be used in Y axis, as you said.
Could you please tell me the way to solve this?

Many thanks!

[urosgrivc]

Under Data Series (second tab for Chart1)
Your data source can be an expression, check the box, this will be Y

let say you have a named selection named "WING"
then for the average pressure ower a WING surface would be

areaAve(Pressure)@WING

This is an expression which will give you a fixed number of [Pa] at a certain timepoint

You could also evaluate force: force_x()@WING
or whatever you like...
look at expressions

[ddungntu]

yes, your way is very useful.
However, I wonder to plot pressure history on area, say a small rectangular or circular within plate, representing pressure gauge diameter attached on plate in the wet drop test (because using a monitor point seems not proper for this).
so to used your way, I think a user surface may be needed, am I right?

[urosgrivc]

Yes You need to somehow define "Location" on which the expression is evaluated,
but when you have that location than the process is the same except
instead of WING, the surface will now be some other locator


and CFDpost is realy powerful here, location can actually be very complex if you want it to be, it can be another expression

Check out the "ISO CLIP" function
You could clip your surface in X,Y,Z directions and use exactly the part of the surface that you want

[ddungntu]

Yes, I got it. Thank you so much.
By the way, may I ask you a question related to meshing on the water surface?
could you tell me how to make the fine mesh for free surface without split fluid domain?

[urosgrivc]

Well, the meshing possibilities will depend on what geometry you have.
I help myself with BIAS function for edge sizing to concentrate the element where these are needed most

[ddungntu]

yes, the BIAS function is very convenient to refine mesh, but it provided only beginning, middle, end line as bias (--- - - -; - - - --- - -; - - - ---), not provided others.

[urosgrivc]

Yes but if you have a structured mesh, then this is pushed from the edges into the domain.
this doesn't work for the tetrahedral mesh of course
And as said every geometry needs its own approach

[ddungntu]

for more illustration, my fluid domain is a cuboid including water and air domain, say LxBxH=1.4mx1.2mx1.5m, the expected free surface is at H=1m, and I wonder how can I make fine mesh at H=1m?

[urosgrivc]

Yes this is a good example of what I have proposed previously,
If the domain is a simple cube or another simple shape
You can use HEX elements and BIAS on the edges to produce a great mesh which is finer at the height vhere you need it

And if you know where the surface will be, you can imprint additional edges on your geometry to have even better control at vhich height the finest elements vill be

This process also minimizes the growth rate of the elements so the quality can be really high

[ddungntu]

Thank you very much for your kind help and patience.

and I would like to ask you one more question.
When I simulated the wet drop test of one body on calm water, I made to way:
- 1st way: assign an initial velocity to the body, anh fluid domain is steady state
- 2nd way: assign an initial velocity on the inlet face and the top of air I defined the opening as outlet face.

Finally, the 2nd way give me the better results which closer to experimental data (pressure and deformation).

I still do not know why the fist way is not good.
Could you correct me on this matter?

[urosgrivc]

I don't completely understand what you are dropping and where

[ddungntu]

I am trying to simulate the dropping object on calm water, commonly water entry problem - slamming.

[urosgrivc]

There are two main options for this:
#1 rigid body motion
#2 immersed solid

But for this, I would recommend a simulation with mesh deformation and rigid body motion. As I have observed in the past I couldn't get any good data from with a simpler immersed solid simulation because of mesh near walls being too bad to capture the details of the flow and good wall forces, these values are needed as the body is moved by this forces.
You will need to be careful, not to stretch the mesh too far, or some kind of remeshing will be needed, but I haven't done remeshing in CFX yet, so don't have this experience.

[ddungntu]

Thank you for your suggestions.

I am using CFX to model 2-way FSI which coupled with the Mechanical. I used the coupling method to do this, and also used the mesh motion option for this simulation.
If possible, may I have a look at your model you did on this type of simulation for more details?

Once again thank you so much, my dear friend.

[ddungntu]

Originally Posted by urosgrivc

Yes You need to somehow define "Location" on which the expression is evaluated,
but when you have that location than the process is the same except
instead of WING, the surface will now be some other locator


and CFDpost is realy powerful here, location can actually be very complex if you want it to be, it can be another expression

Check out the "ISO CLIP" function
You could clip your surface in X,Y,Z directions and use exactly the part of the surface that you want

Regrading how to specify the region, I have just done by using the user surface format, it works well.