CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Diffusor Simulation (http://www.cfd-online.com/Forums/cfx/101717-diffusor-simulation.html)

 ChrisR May 11, 2012 17:54

Diffusor Simulation

Hey there,

i'm trying to get a good simulation from a (fsae)car diffusor.
But my result seems to be false.

I measure the Force in y with different types of Expressions.

1. force_y@Diffusor
2. areaInt(Pressure)@Diffusor

but it says its just 40 N downforce.

I work with a half car simulation.

I set Side and top to wall. Car and Diffusor to Wall
Specify Inlet(65 km/h) and Outlet(0 pa)

Ground is not moving at this state of the simulation.

Turbulenz modell is SST ( tried it also in k-e)
Reference Pressure is 1 atm

Why are my Forces so low? I should get at least >200N.

It seems like none of the Steamlines in cfx post is going under the car.

http://postimage.org/image/ubqywvcup/
http://postimage.org/image/ubqywvcup/
Greetings

 evcelica May 11, 2012 22:05

Just a couple thoughts:
Perhaps the ground not moving is making a difference. Why didn't you make it move, you just specify wall velocity of 65 km/h?
How do you know what it "should" be?
You did remember to multiply be 2 for symmetry correct?

 ChrisR May 12, 2012 04:57

We did some simulations with a Achmed Body. It was 300 N Downforce with the same Parameters.

When i turn on wall velocity i get this error message

Quote:
 +--------------------------------------------------------------------+ | ERROR #002100080 has occurred in subroutine CHECK_NORMV. | | Message: | | The specified velocity vector on the boundary patch | | | | Boden | | | | has a significant normal component at one or more faces. One of | | these face locations is | | | | (x,y,z) = (-6.77959E-02,-7.27100E-02, 6.78937E-01). | | | | The angle between the specified velocity and the element surface is| | 41.851 degrees at this face. This is considered an error because | | it implies that the mesh is moving. The following are possible | | reasons for the error message: | | 1. There is a setup error; for example, an incorrect axis of | | rotation. | | 2. There may be a meshing problem; for example, the nodes on a | | rotating surface might not lie on the surface of revolution. | | 3. The boundary is curved and the mesh is very coarse. In this | | case, you may modify the tolerance by increasing the | | expert parameter 'tangential vector tolerance wall' | | from its default of 20 degrees. | +--------------------------------------------------------------------+
How can i solve this?

Should i make a better mesh or increase the expert parameter?

Update : After Improving my mesh i still get this failture? Whats wrong with my simulation? Even if my expert parameter is set to 90 deg it dont work

 ghorrocks May 13, 2012 06:53

Are you trying to put the velocity on the car? That obviously will not work. You have to put the velocity on the road surface.

Your question is essential why are my results not accurate, that is a FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

 ChrisR May 13, 2012 11:16

Moving ground now works, there where 2 Surfaces directly at the same spot.

But my results are still really bad. When i start i get a error in the symmetry plane it wouldnt be symmetry. So i have to change my expert parameters. Where is this issue from?

 ghorrocks May 13, 2012 18:33

Posting the error message would help.

 ChrisR May 24, 2012 12:00

+--------------------------------------------------------------------+
| ERROR #002100013 has occurred in subroutine Chk_Splane. |
| Message: |
| The symmetry boundary condition requires that the boundary patch |
| mesh faces form a plane or axis. However, face set 11 in the |
| symmetry boundary patch |
| |
| Sym |
| |
| is not in a strict plane, which means that at least one of its |
| faces is not parallel to the others. To make the solver run |
| you can do one of the following: |
| |
| (1) Make sure that this symmetry boundary patch is in a plane or |
| axis by checking and regenerating the mesh. |
| (2) If the symmetry boundary patch is an axis rather than a |
| plane, change the tolerance of the degeneracy check by |
| increasing the value of the Solver Expert Parameter |
| 'degeneracy check tolerance' (the default value is 1.e-4). |
| (3) Increase the value of the Solver Expert Parameter |
| 'vector parallel tolerance' (the default value is 1 deg.). |
| Note that the accuracy of the symmetry condition may decrease |
| as the tolerance is increased. This is because the tolerance |
| is the number of degrees that a mesh face normal is allowed |
| to deviate from the average normal for the entire face set. |
+--------------------------------------------------------------------+

i really dont get it. My Mesh have 3.000.000 Cells - and a good quality -

 ghorrocks May 24, 2012 18:44

Don't play with the expert parameters. The error message clearly says your symmetry plane is not a plane. So find the root cause error and fix it. This is easy to do by loading it in the CFD-Post and looking at your symmetry plane.

 ChrisR May 28, 2012 13:20

The Symetrie Plane is just a plane in a box where my car is placed. it is just a normal plane.

I dont get it! :mad: It is not the first time with this model( i have 3 different diffusors) i get this error message!

 ghorrocks May 28, 2012 18:43

The error message is very clear. Some bit of your symmetry boundary does not lie on the plane. You have to find it and fix it.

 Cola1988 May 31, 2012 05:21

+--------------------------------------------------------------------+
| *** INSUFFICIENT MEMORY ALLOCATED *** |
| |
| ACTION REQUIRED : Increase the real stack memory size. |
| |
| Details : |
| Requested space : 11918376 words |
| Current allocated space : 2147483646 words |
| Current used space : 2137664714 words |
| Current free space : 9818932 words |
| Number of free areas : 1 |
+--------------------------------------------------------------------+

The Calculation PC have 32GB of Ram and it seems like it isnt enought? is it possible that a calculation computer has not enough ram? In my opinion thd calculation takes only much longer to calculate?

Can I setup the calculation options in a way it doesn't need as mutch ram?
What can i do?

Any guidiance would be great!

 ghorrocks May 31, 2012 06:00

The error message tells you exactly what to do. You need to increase the real stack memory size. Have a look in the CFX documentation for the solver manager for how to do that.

This error has nothing to do with the amount of physical memory in your machine.

 All times are GMT -4. The time now is 10:49.