Steady state fluid flow

[Ralphrail]

I have a heated body inside an enclosure with forced airflow for cooling. When I do a fluid flow analysis, I have been unable to get a result using steady state. If I run as a transient state analysis I have no problem. Can I use the results of a transient state fluid flow analysis for the heat convection analysis? Is it just a matter of using the ufv file instead of the sfv file?

[John_Holtz]

Hi Ralph,

The short answer to your question is yes: you can use the results of the transient analysis in place of the steady state analysis.

If your fluid flow load curve in the transient analysis ends with a constant time period, then a "steady state" solution should occur. However, fluid flow rarely has a steady state solution due to vortices, turbulence, and so forth. So you may notice your fluid flow results "dance around" but show some signs of repeating. If you see the repeating pattern, you may want to run the heat transfer analysis two or three times using a different time step from the fluid analysis, and then mentally average the heat transfer results.

[Joey.X]

John already mentioned some good points; here are a little more comments.

Steady state analysis eliminates the time items in governing momentum equations, thus the result is close to "average" of the physical oscillation result. However, steady analysis is less convergence stable comparing to unsteady analysis since the time item is always a stabilized term in numerical process.

For your question, *.ufv could be used for thermal convection analysis, this implies .ufv could be imported to thermal and run thermal analysis only, It has to be noted that the accuracy depends the model properties with following major factors.
(a) If the flow has unsteady nature, i.e., the flow has no steady state solution at all. This case happens in typical case of "vortex shedding" flow, such as flow through cylinder under certain Reynolds number. For this case, the unsteady analysis is the only option, and the time step has to be small.
(b) Check if "turbulence" is check on in Analysis parameters/local curve/Turbulence if the flow is turbulence flow (user needs to manually check the ranger of Reynolds number). Without checking on turbulence is a common mistake for simulation convergence issue.
(c) Depending on user wants “average" or "time clip" fluid flow result, this corresponds to steady and unsteady analyses respectively. In some stable flows, the average and "time clip " result will be close if the time is longer enough in unsteady analysis, but not always depending on model physical properties. Here are some model tips to achieve converged average result from steady state analysis
- Make sure “turbulence” is turned on for turbulence flow
- Turn on “detect stagnation due to oscillation” from analysis parameter/solution/first >> /Segregated Options/Detect stagnation due to oscillation/Continue to next step
- Increase solution convergence criteria from Analysis parameter/Load curve/Velocity Norm and Pressure norm, both of the defaults are 1e-4
- Some other modeling considerations such as mesh quality, proper boundary conditions.