Unstable Solution in Forced Convection

Jagan

I'm solving air-forced convection of motor using k epsilon turbulence model. I'm getting these warnings:

WARNING: Diss.-rate fully specified by two-layer wall treatment.

WARNING: inflow at outflow boundary: Outlet (mass/time: in=-2.920399e-03, out=4.139305e-02)

And my solution tends to be unstable.

I have attached pictures and input file for reference, Please give me an insight about this warnings and suggest if changes need to be made in the modelling. Thank you.

forcedconvair.inp

acupro

In many instances when there is a large expansion between the inlet and outlet, you'll get reverse flow (backflow) at the outlet.  This is likely what's happening here.  There are typically two options:

1. Activate backflow conditions on the outflow boundary - and define the turbulence and temperature conditions of any incoming flow.  Maybe 'average exiting' as the type?

2. Add a contraction beyond your area of interest - maybe beyond your current outflow to get back to maybe a similar exiting area to the inlet area - similar to the image below where the exit is at the top of the image.  You can make the boundaries 'slip' on the outside of that extension/contraction so as not to influence pressure drop too much if that is of interest.