[Amps] Advice needed for SS amp

Roger (K8RI) k8ri at rogerhalstead.com
Thu Jun 6 00:48:43 EDT 2013


On 6/6/2013 12:13 AM, Bill Turner wrote:
> ORIGINAL MESSAGE:          (may be snipped)
>
> On Wed, 05 Jun 2013 20:09:34 -0400, K8RI wrote:
>
>> Water cooling is nice, but you are probably going to need either DI or
>> distilled water at those voltages and water cooling can be a royal PITA
>> when things go wrong.
>
> REPLY:
>
> I realize the pitfalls, but with DI or distilled water, I was thinking of
> completely submerging the transistors and heatsink. Sounds a bit loony but
> it might work. The drawback of course is you don't have an unlimited supply
> like if you were using an outside hose connection and cooling only the
> heatsink. .

It's been done with transistors and tubes, although I think the tubes 
were in the 300 to 500 volt range and iy was just to see if it could be 
done.  This is a bit different from where only the tube envelope was 
submerged.  It was a long time ago, but I'll bet there are some on here 
who remember the particulars.

With transistors (particularly computers) the entire circuit has been 
submerged in coolants of at least several types and I have heard of 
several QRO approaches using submersion.

Although at room temp water is the most efficient coolant it might 
present problems with solder.  There must be alloys that are less 
susceptible to the corrosive effects of DI water and I would think 
complete submersion would work and these circuits would be less prone to 
changes in capacitance due to the presence of the water.  I was lucky to 
survive Chem 111 and 112 although I did get good grades.

Silicone 200 fluid of a low viscosity should be effective as well, 
although less effective at heat removal than water it is still quite 
effective.

Whether just the immersion of the transistor case or the entire circuit 
both sound feasible to me.

I like the idea of complete submersion as it removes the problem of 
level control.  with submersion position, or movement  is no longer a 
problem.  1 gram or CC of water increases 1 degree in temperature for 
each calorie of heat removed so it takes relatively little flow to 
handle 1 KW of heat which is the dissipation of an amp putting out the 
legal limit at 60% efficiency.

If refrigerated it can greatly increase the transistor's power handling 
capacity. With computers they just submerge the entire CPU, or CPUs 
directly in the refrigerant.

Good Luck!

es 73

Roger (K8RI)


>
> One way to find out.
>
> 73, Bill W6WRT
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>




More information about the Amps mailing list