Heat-sink trans coolers. Who's got them, and how do you like them?

http://www.coolingzone.com/library.php?read=493


Look nice, cant see effective cooling. Too little surface area, too much fluid volume. The plate style is much more effective: probably 12 8x1 double contact sided pates for a total area of 192 square inches of cooling. That finned tube has the surface area of the can plus the fins and the fins are not in contact with the can but for a thin width. MHO.

Look at link on top and get the slide rule out...

Fin efficiency where L is the fin height andnote that k is the conductivity of the fin material. A is the cross-sectional area that's transferring heat up the fin. Usually this is the fin thickness times the flow length of the heat sink. Moving air produces heat transfer coefficient h (remember the Nusselt number?) and P is the "perimeter" of the fin. The perimeter is basically two times the flow length of the heat sink plus two times the thickness of the fin (negligible for thin fins); think of it as the length of a string wrapped around the fin at a constant height.
So now that you have the fin efficiency, what do you do with it? The way to apply it is to use instead of just (hA) in the log-mean temperature difference (see March column). Check that for a given heat flux, the surface temperature increases when you use the fin efficiency; that way, you'll know you wrote the formula correctly. Refrain, however, from placing too much emphasis on what the actual fin efficiency value is. It's not a design goal, but a calculation tool, and possibly a diagnostic to help you figure out why the heat sink performance isn't quite as high as you'd like.
Fin efficiency calculator (if your calculator-maker supports hyperbolic tangent…)
Fin height L
Heat sink flow length
Fin thickness
Heat transfer coefficient h
Material conductivity k (default = 205, extruded aluminum)
Area A = (fin thickness)*(heat sink flow length)
Perimeter P = 2*(heat sink flow length + fin thickness)