On Fri, 29 Jun 2012 08:39:58 +0100, "N_Cook"
wrote:

The device was FZT949 (yes, revised, 5 or 6 amp rating/size) ,
http://www.diodes.com/datasheets/FZT9489.pdf
SOT223 package.
http://www.diodes.com/_files/product_packages/sot223-3l.gif
Full spec dissipation requires 4 square inches of copper under the
device. Lower dissipation specs with less of a heat sink. The back
is the device is plastic. Most of the heat comes out the large solder
tab. There is a 1.6mm air gap under the plastic case to deal with
glues and to insure that the tab mechanically hits the copper head
sink area. There is almost no value in obtaining a thermal connection
to the epoxy case, so there would be no benefit to using a thermally
conductive epoxy glue.

the glue
cleared away with soldering iron tip , in the manner of epoxy , with no
smell of cyanoacrylate

Epoxy does not "clear away" when hit with a soldering iron. What it
does is crumble and burn leaving a charred mess. The heat conductive
variety will conduct enough heat to the PCB to also char the PCB. I
agree that if it didn't reek when you hit it with the soldering iron,
it's probably not cyanoacrylate adhesive. That leaves hot melt
adhesives and various acrylic glues. If it seemed to melt away at a
very low temperature, it's probably hot melt. If it took some effort,
and it simultaneously melted and charred slightly, it's acrylic.

Please specify what you mean by the "manner of epoxy"? Melt, char,
crumble, volatize, explode, etc?

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558