Calculate Nusselt Number: Free Convection Air Vertical Channel Different Temps

[engineer23]

What expression should be used to calculate Nusselt number for free convection (of air) in a vertical channel with walls at different temperatures?

I have the Bar-Cohen and Rohsenow equations, but I am unsure which situation this corresponds to (and thus which constants to use).
Symmetric isothermal plates (wouldn't this imply Ts,1 = Ts,2 which is not the case?)
Symmetric isoflux
Isothermal/adiabatic (I am leaning towards this)
Isoflux/adiabatic

 

[minger]

Wow...I think we're looking at the same book, and I'm surprised this isn't in here. They have good relations for a fully enclosed cavity, which I'm sure you've seen in the next section. They also have a case for flow between concentric cylinders, but nothing about asymmetric heating.

This led me to literature, and I was more surprised on how little I find even in journal papers. The closest I've found so far is FORCED convection between asymmetrically heated plates. I'm going to keep looking for you though.

boy there are a LOT of different types of uniform heating though isn't there?

 

[engineer23]

Maybe I can use the enclosure equations as an approximation.

 

[minger]

That's what I was originally thinking, however the computed Nusselt number can very quite a bit if you assume L/H = 100:1, or 1000:1, etc. This of course affects the convection coefficient. Also, the enclosed space assumes that there will be a complete convection current. In your case of "infinte" parallel plates, I would think that there would be none. There will probably be some swirl, but I wouldn't think that the flow would come "down", so long as both wall temps are higher than the incoming temp.