Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

How do Floworks interations relate to time?

Status
Not open for further replies.

phonix

Military
Jan 27, 2007
54
If I had a chamber with a gas inlet and a gas outlet, and there was a pressure difference between the inlet and outlet, such that the gas would flow through the chamber which started at atmospheric pressure and later reached a steady state -- I want to see the gas flow as it first rushes into the chamber. Not after it reaches the final state.

How would I do this? Set goals of some sort? Do iterations map to real time, or to refinement of a final solution?

I ask because I viewed the results as they were being computed and very early on it looked like the gas was flowing right through and you could see it progress through. And this was just a few minutes into a two hour calculation. So does that mean I had already reached my goal after a few minutes? Or does it still need the full two hours to be able to show me the best result of the beginning of the gas flow?


 
Replies continue below

Recommended for you

Have you tried a time dependent analysis with output options specified to save the initial results? I don't think the iterations are related to real time in a steady state analysis.
 
Ah yes. That is it. I am learning a lot every day so even yesterday's questions sound silly to me now.
 
Hi,
to be "exact", what you see in the calculation preview is the value field at each iteration needed to make a "step" converge (in a steady-state analysis, you have only one "step"). The initial field is obviously a "null" one, or a uniform gradient, or some very simple "first-guess" solution. Afterwards, the Navier-Stokes equations will govern progressively the values computed in each cell. Convergence will be achieved when all the "goal" values at an iteration will differ, from the previous iteration, less than the thresholds.
In a transient analysis, if you request, say, 1000 steps in order to capture the evolution of the fluid field at intervals of 1 ms for a total duration of 1 s, then you will have 1000 x "steady-like" solves; if each solve takes, say, 20 iterations, then in total you will have approx 1000 x 20 total iterations (I say "approx" because the number of iterations will not necessarily be constant).
Do never confuse the solve evolution in the preview with the time-dependent evolution of the fluid field!!! Even if it can seem very alike (in some simple cases), it can be extremely misleading!!!

Regards
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor