>Wouldn't heat rise in water too, Peter? You see convection currents in
a pan
>of water that's heating up from below on the stove.
>
>Dave G0OIL
>
The other problem is that heat-insulating bubbles tend to form on the
surface. It needs some kind of pump to keep that boundary layer
continuously "scrubbed" with cooler water.
(Exactly the same applies to air cooling. If you are relying only on
natural convection, there will always be a stagnant, insulating boundary
layer next to the surface. This is why fans work best when blasting
turbulent air directly onto the cooled surface, because that breaks up
the boundary layer. Sucking hot air away from the surface doesn't do
that, so the cooling is far less effective.)
Some SSamp builders have had good success with water pumps, heat
exchangers and other water cooling equipment marketed for computer
overclockers. Only the transistor cooling plate needs to be different
for an RF amp and the size of the water reservoir can be surprisingly
small. I have seen prototypes and this technique can certainly make a
very compact and lightweight 500W DXpedition amp.
73 from Ian GM3SEK
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|