[yanyanh]

Hi all,

I'm currently doing a car exterior CFD simulation (StarCCM+) and would like to extract the air from one location and reinsert it into another location, kinda like a portal. The idea is to find the best place to flush the waste air of the engine bay.

My question here is, how to define the boundaries for this case. I did some trials with a pressure outlet at one site and couple it to a mass flow inlet at the other side by matching pressures and mass flow. In a simple model it worked, but in the full scale car model I'm not able to get convergence and the two mass flow does not match up.

Does anybody has an idea how I can do this "teleporting" of air (two-way since the mass flow need to match the pressure drop)?

Thank you so much for your help!
yanyanh

[me3840]

I can see a couple ways to do this:

1. Dumb and simple method is to just make a pipe between the two points (even if it wildly goes through other geometry) and set the walls to be slipping.

2. Create similar coordinate systems on each interface side. Create a macro that will use data mappers to iteratively update velocity and pressure back and forth. Run the job with the macro.

3. Write a similar procedure to 2 using user code. Best solution, most effort required.

[yanyanh]

Originally Posted by me3840

I can see a couple ways to do this:

1. Dumb and simple method is to just make a pipe between the two points (even if it wildly goes through other geometry) and set the walls to be slipping.

2. Create similar coordinate systems on each interface side. Create a macro that will use data mappers to iteratively update velocity and pressure back and forth. Run the job with the macro.

3. Write a similar procedure to 2 using user code. Best solution, most effort required.

Thank you very much for the reply.

1. I did this draw-a-pipe way already to compare with the interface/boundary method for "validation". I haven't set the walls of the pipe to non-slip, that I will correct to get closer to the portal method.

2. and 3. What kind of interface are you talking about? Since these two portal boundaries are neither close to each other for in-place interface, nor have the same area for a periodic interface. I'm using a mass flow inlet to adjust to the different surface areas.

Thank you for the help!