Manfred,
I have build several solid state amplifers, including a 1KW HF amp that I am
currently using on a daily basis. I use Artic Silver. I found that it is
really important to first of all have a flat surface between the copper heat
spreader and the aluminum heatsink. I have always had these surfaces milled.
The milling cost more than the heatsink, but it is worth it. Then put a very
thin layer of Artic Silver between the copper heat spreader and the heatsink,
and underneath the MOSFETS. More is not better, use a line layer. I use a
roller to spread the Artic Silver.
Don
---- Manfred Mornhinweg <manfred@ludens.cl> wrote:
> Paul,
>
> > I'm getting around to my 2 meter kilowatt SSPA. My question is which of
> > the multitude of heatsink compounds do you recommend? I have read some of
> > the comparisons mostly dealing with CPU cooling. But, would like some
> > input from piratical experience. The silver formulations seem to do a bit
> > better.
>
> When I looked for data on these compounds, the best rated one I could
> find was Arctic Silver 5. It's about twice as heat conductive as the
> best ceramic-based compounds, according to its data sheet. But I haven't
> yet measured whether those claims are true.
>
> Between the different ceramic based compounds, there are some large
> differences too. They are not all equal! I found differences up to a
> ratio of ten between them! Check the specs and compare before you buy.
>
> > In addition, I am assuming that I want a layer of the stuff between the
> > copper heat spreader and the heatsink?
>
> Yes. Assuming the spreader is large, and lies very flat against the
> heatsink, a decent grade ceramic compound is good enough there - but a
> better compound is always better! Between the transistors and the
> spreader instead, really the best is soldering! Solder is about 8 times
> better than the best silver particle compound. If it's not practical to
> solder the transistors to the spreader (as is often the case), use the
> best compound you can find. You need only a tiny quantity there.
>
> And if your transistors are expensive, do the math about their cooling
> BEFORE you build that amplifier! If they are cheap, or if you have a
> free supply of them, you can use the empiric method and just see if they
> survive... ;-)
>
> Manfred
>
> ========================
> Visit my hobby homepage!
> http://ludens.cl
> ========================
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|