12 Replies Latest reply on Mar 28, 2011 4:19 PM by Martin Kaib

# Fluid Subdomain Problem in an Assembly

Hello:

I'm attempting to model flow in a simple flange assembly. The flange assembly has a pipe running through its center with a heated gas in it, and the flanges are water-cooled. When I attempt to set up a fluid subdomain within the pipe and select the lid on its end, for some reason, the fluid subdomain is only half of the pipe's volume. See the attached document for pics.

However, if I model a simplified part version of the total assembly, and construct a fluid subdomain, the subdomain takes on the complete volume of the pipe like it is supposed to.

Being new to Flow Simulation, I've never run across this before. Can someone explain why this is happening, and how I can correct this problem?

Thanks.

Marty

• ###### Re: Fluid Subdomain Problem in an Assembly

my guess would be you have a hidden part that is active or your computational domain has a boundary there.

• ###### Re: Fluid Subdomain Problem in an Assembly

Bill:

Thanks. There are no hidden parts. I'm not sure what you mean by a computational boundary though. There are two lids on the pipe at each end only.

Marty

• ###### Re: Fluid Subdomain Problem in an Assembly

NOt too sure then. I took another look at the images. In the one that show the half domain - what is the circular edge inthe half with no fluid domain?

On the domain - go to the top of the tree and have it show the domain. It is a black box that shows the extents of hte flow domain. If that was inthe middle then you would not get fluid onthe other side. How did you get the BC applied to the part with no fluid with out getting the dreaded "BC not laying on the fluid boundary errror msg? Try editing that BC and see if will close with the check mark. In any event this sort of thing is usually a geomtery problem.

• ###### Re: Fluid Subdomain Problem in an Assembly

Bill:

The circular edge that you see, is an extrusion that I created that lays on top of the pipe, so that I could grab the extrusion's surface to create a surface heat source. I didn't want to capture the whole outer surface of the pipe to create the heat source.

I don't know why I don't get the dreaded error for the BC not lying in the fluid region, so there is no error to clear in that respect. When I run the solver, I get errors for vortex crossing, Mach number exceeded, and negative pressure. Some of these errors may be related to the skewed fluid domain.

Initially, I had the pipe set up as two pieces, one for the front, and one for the back. But when I ran into the fluid domain, and it looked like it was cut in half, so I decided to create the pipe in one piece and pass it through the whole assembly to fix the issue. However, it still remains. I'm stumped.

It looks like I have the "show" for the fluid domain turned on, but I don't see it. Please see the attached.

Marty

• ###### Re: Fluid Subdomain Problem in an Assembly

Hello,

I was able to cause the same thing to happen, by changing the computational domain (Z min = -4).  Even though one of the BCs is outside the comp domain, Flow Sim doesn't complain.  When I set Z min to zero, Flow Sim complained that a body was outside the domain, which was true.

Trying resetting the computational domain, and then tell us what happens.  The attached screenshot show how to reset the domain.

Rich.

• ###### Re: Fluid Subdomain Problem in an Assembly

Rich:

Thanks for the feedback.

I didn't change the computational domain's volume yet. Please take a look at the attached. It appears that the assembly is fully captured by the domain's envelope.

Marty

• ###### Re: Fluid Subdomain Problem in an Assembly

Hello Martin,

Interesting!  Would you mind posting your model?

Rich.

• ###### Re: Fluid Subdomain Problem in an Assembly

Rich:

Here's the assembly...

Marty

• ###### Re: Fluid Subdomain Problem in an Assembly

Hello Martin,

I tried to open the assembly, but it needs all the sub parts, so I wasn't able to open it.

Please try again using: File, Pack and go, and then post the output.

Rich.

• ###### Re: Fluid Subdomain Problem in an Assembly

Rich:

Here's the assembly again. I don't think that it grabbed all of the lids for some reason.

Marty

• ###### Re: Fluid Subdomain Problem in an Assembly

Hello Martin,

By using the tool, Check Geometry, and the Create Fluid body assemblys option, Picture-4 shows the blank area in the middle.

You added some volumes to be used as volume heat sources.  They must be disabled in component control as shown in Picture-5

Picture 6 shows the results.

Rich.

• ###### Re: Fluid Subdomain Problem in an Assembly

Rich:

Didn't know that would affect the fluid subdomain. And I didn't realize that I had the Volume Heat Source feature hidden, and not suppressed. Another lesson learned for me.

I am not familiar with the Component Control feature in Flow Simulation, does suppressing the feature in the model yield the same result?

Thank you and everyone else for helping to diagnosis the problem.

Regards,

Marty