CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   modelling a classroom using Airpak! (https://www.cfd-online.com/Forums/fluent/38319-modelling-classroom-using-airpak.html)

 Amid October 20, 2005 16:30

modelling a classroom using Airpak!

can we post here our questions about Airpak?

i am modelling a classroom (8.4 X 7.2 m) uses DV, the process is domianted by natural convection which is the source of the problem. Since natural convection was heavily dominated in my problem, i would have had no choice but to use Transient to capture the physics of problem (even though the problem itself is steadystate), which brings me to my next quetion ==> is there any ways to achive convergence by playing around with the URF, where tdo i start??

my last question is about results, i am comparing my model to experimental data, my results agree with the measured data at all points except at floor level (0.1 m). all my 12 points at the floor level is off by approx 2-3 C. i know that for a realistic solution we need to use the Low-reynold number near wall areas, and iam guessing airpak uses the Law of Walls coupled with the standard k-e to get results. i have placed the first grid at the distance of 10mm from the wall. i need however to verify that iam not meshing the viscous sublayer, so i need y+. Is there anyways to make airpak give the value for y+??

i was wondering for a typical DV problem for a office what should the the typical the convergence critrea, for somereason i dont think Airpak's pre-define convergence critrea is enough for a problem with natural convection

what should the the typical discrepency percentage between experimental and CFD

thanks

 All times are GMT -4. The time now is 20:01.