CFD Results to Sim Mechanical

[Anonymous]

So, Simulation 2015 has a new way to get SimCFD loads into SimMech. I upgraded from 2014 to 2015 in the middle of a big project just so I could use this feature. But, it's not going so well.

Here is a solid in Inventor representing the fluid volume of a maifold. This geometry was used in SimCFD.

Here is a shell of the same geometry. This is what I used in SimMech.

I set up and ran a transient flow analysis in SimCFD. Here is the pressure result from one time step of interest.

Now, following the steps in the help article, " Apply Simulation CFD Fluid Boundary Reactions to a Simulation Mechanical Stress Analysis", I ran the FEA.

But, when I plot the applied loads, they look like this:

What am I missing?

[Andrew.Sartorelli · Accepted solution]

Hi Steven,

From the screenshots that you uploaded it looks like you are using plate/shell elements in your analysis. The CFD mapping utility only supports solid elements such as bricks and tetrahedrons. Using plate/shell elements will not produce the desired result, as you have seen from your model.
______________________________________________________________

If my post answers your question, please click the "Accept as Solution" button. This helps everyone find answers more quickly!

[arizawilmer]

is anyway of over passing that limitation?

[AstroJohnPE]

Hi stevenaking,

The images are a little too small for me to clearly see what the values are. Also, without knowing the size of the mesh, I would not be able to convert the pressure from CFD to nodal forces in Sim Mech. Have you done the hand calculation to show that the conversion is wrong? Otherwise, I can imagine how the elements around the joints between intersecting segments are larger and therefore have a larger nodal force. In other words, I cannot say that there is something wrong just from looking at the images.

The obvious way to avoid the limitation is to use brick elements instead of plate elements.

Or since the pressure results are approximately smooth and linear in the -Z direction, you may be able to get suitable results by manually applying a pressure in Sim Mech -- such as a hydrostatic pressure. This becomes a little tricky with plate elements (and your model in particular) since you need to indicate which side of the element has the pressure applied. There is no "inside" and "outside" surface of the plate elements; just a "top" and "bottom" side which you designate (using the Element Definition input).

One final comment/question. Are the CFD results based on the proper gage pressure? Atmospheric pressure occurs somewhere between the first and second nozzle from the left where the pressure result is 0, so the pressure at the left end is below atmospheric pressure?

[Anonymous]

John,

Does the CFD mapping utility support plate elements in simulation mechanical 2016.  Otherwise is there any way to work around this?  I have a large thin walled shell that has plate elements attached to it.  My computer preforms very poorly when I try mesh/solve my tank as a solid.

Thanks,

Mcihael Schroepfer

Polar Tank Trailer