When I was a young experimenter, I thought the wattage rating on resistors
meant they wouldn't get hot at the stated power dissipation level.  As time
went on, I found that the wattage rating meant something closer to "the
device won't be damaged" or maybe "the temperature rise won't put the part
out of spec", seeing as how temperature change alters characteristics.

Another eye-opener occurred when I was repairing data terminals, some having
a defective rectifier bridge in their power supply.  These rectifiers were a
serious-looking block of metal, will epoxy fill, and four tabs sticking
out.  I looked up the rating on these things out of curiosity, and here's
what it said:

Max current (attached to a heat sink): 25A
Max current (free air) 2A

At which point, "the repair guy was enlightened".  And yeah, the defective
ones, tended to have loose bolts or had little or no thermal grease under
them.

Thermal physics is interesting to think about, especially as it works a lot
like ohm's law.  Mark's right, that the size of the thing isn't as important
as it looks.  Generating heat makes something's temperature rise, and it
rises until whatever's carrying the heat away (goes up with temperature)
balances the heat being generated.  It can escape through the air, through
the leads to the board (then to the air), or onto a heat sink (and then to
the air)...but it has to escape somehow.

There were these little metal clips you could put over a small (to-5, to-39,
etc.) transistor to increase the surface area.  Many times it was enough to
keep them at a reasonable temperature.  Maybe something like that for your
diode would be enough.

Then again, when the voltage regulating part gets too hot, that's about when
I begin to think, "Maybe I should find a better way to do this..."
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist