CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Manifold intake simulation (

Super Sonic January 6, 2012 12:30

Manifold intake simulation

I am trying to simulated a simple manifold intake in a transiend state simulation .My problem is that I can't find exaclty the way to do this. For example should I also need to improt some kind of .txt file with some values. If somebody can upload some kind of file to help me I will arrpesiate that. I just need the steps on the set up stage.

Thank you very much!

ghorrocks January 8, 2012 20:35

Define your boundary condition as a function of time (either a CEL expression or a 1D interpolation function) and use that as the boundary value.

Super Sonic January 9, 2012 05:00

Manifold intake
Thank you very much for your respond.I wondered if it is possible to send me an example of what you just explain me, because I am very struggle after one point on this software. But with an example I think I can handle it.

Thank you very much again.

ghorrocks January 9, 2012 05:47

I do not have time to upload an example. But the tutorial examples have examples of using CEL expressions so you should look there.

Super Sonic January 15, 2012 05:38

Intake Manifold
Ok no problem.Do you know how I can relpy by my problem to all, so If it's possible form somebody to help me further.Because still I am very struggle with that issue.

Thanks again:

ghorrocks January 15, 2012 05:50

What input data do you have? Massflow versus time at the exhaust valve? Pressure versus time? This is step 1.

Super Sonic January 15, 2012 16:48

I want to apply atmospheric pressure in the inlet.And I have 2 outlets (2 runners) on which I want to apply 0.035 kg/s (on each) mass flow rate. All I need is a full example on the setup stage in the CFX.

Thanks again.

Super Sonic January 15, 2012 16:49

Pressure versus time Ι want to apply

Super Sonic January 15, 2012 16:51

Because for example, I saw some other files ,as exemples which had also some x,y,z coordinates. That is an issue for me because I don't know what kind of coordinates are those.


ghorrocks January 15, 2012 17:56

I have done this type of analysis many times. It is quite straight forward. Are you interested in the transient or steady state flow? If transient are you trying to model the pressure waves?

You define the mass flow as a function of time, and if you have many cylinders with different timing you need to define multiple functions (or link them with a time offset or what ever is appropriate). This function of time will normally be applied as a 1D interpolation function that returns a mass flow rate for a time value. So you then apply the mass flow boundary condition on the valves with the mass flow rate defined as the function.

If you are trying to model the pressure waves in the system I strongly recommend you do some preliminary work first on pressure waves in a simple straight duct so you can see what is important. Shock wave propogation in a straight duct has a simple analytical solution so you should get this simple case working accurately before attempting the 3D complex version.

Super Sonic January 15, 2012 18:24

Ok, maybe it is better for you (and I terribly sorry that I am doing this to you, it's just I am very struggle on this point of level).

First of all lets say that we have a system (call it manifold intake) and I have 1 inlet and 2 outlets. Now on the intake as a boundary condition I apply the atmospheric pressure. At the outlet (were we have two of them) I want to apply the mass flow rate e.i. 'x' value Kg/s. Now, I want to run the simulation in transient state and by the way were a manifold intake works. I want to start with something simple on this field to understand it correctly, so after that I can carry on with something more complicated. I 've all ready have run a transient state but I don't know how to create expression by function of "something" e.i. time. I tried to understand it by the tutorials and many others ways, but still nothing.I know were to create expressions by I don't know how to define them. That's why an example where I can see what kind of commands you gave, will help me once and for all.

Thanks again for everything.

ghorrocks January 15, 2012 18:56

Have another look in the documentation and tutorials on 1D interpolation functions. They are quite straight forward.

Super Sonic January 16, 2012 08:56

1 Attachment(s)
Hi, I saw tha chapter about the 1D interpolation function. Now on my file I greate a user function and I applied it in the inlet(where I gave all the appropriate data).Now when I trying to run it the programm (ANSYS 12) give me the picture bellow as you can see

Attachment 10859

and the setup has question mark and down in the messages seas:
Cannot DefineRun without a valid definition file.

The only thing I did it was to create the function and put it into the inlet boundary:
Function called: pressure
In the boundary : "pressure(t)"
To be as a function of time


ghorrocks January 16, 2012 17:21

You cannot name a function "pressure". That variable name is already reserved.

Can you post either a screen shot from CFX-Pre or the CCL file?

Super Sonic January 16, 2012 19:29

5 Attachment(s)
The txt file has the values where I gave by importing the file (.txt).You can see also some pitctures

Super Sonic January 16, 2012 19:30

3 Attachment(s)
Those are also from the same error

ghorrocks January 16, 2012 19:50

Your images are barely readable, but it seems you have mm as the input argument. It should be a unit of time (maybe seconds?).

Super Sonic January 17, 2012 12:35

5 Attachment(s)
Ok I gave time int the argument units : sec.
But the programm is still denited to run.I appload you some pictures to see again.
When I put in to the inlet boundary condition the funtion : running (I named it), in to a brackets e.i. "running".

Because When I dot that then the programma it's not accepted to run the simulation because it gives me a guestion mark at the main screen of ANSYS on the Setup step (see the photo)
Should I click to the "Suppress physics error" ?


Super Sonic January 17, 2012 12:35

1 Attachment(s)
And also those

ghorrocks January 17, 2012 17:45

Your images are still hardly readable. Do not use the lossy JPG format, use a lossless format like PNG. Also do not reduce the size of the images.

You appear to have defined the argument unit as "sec". The correct unit is "s". Also your result unit is "KPa". The correct unit is "kPa". The variables format is listed in the documentation.

All times are GMT -4. The time now is 12:12.