[Amps] Liquid cooling

[Manfred Mornhinweg]

Roger,

I see that Bill posted some of our private interchange here! No problem
with me. But now I feel that I have to reply to your comments!

>> I just looked up values for solders: Common 60/40 solder conducts
>> heat about one fifth as well as copper,

> This is the reason we use heat transfer compounds like Artic Silver

But Arctic Silver conducts heat about ten times worse than 60/40 solder!

That's not to say anything against Arctic Silver. It still seems to be
the best thermal compound available. But it's a grease with metal and
oxide filling, which just cannot come anywhere close to the thermal
conductivity of 100% metal, like solder is!

>> In a catalog from an aluminium supplier I found some extrusion
>> stock that looks much like an Omega sign, but closed. Like a round
>> tube
> 
> I'd stick with solid copper with a machined water way and a cover
> silver soldered on rather than one with a removable cover.

I will see what I can do. Certainly teh specially machined copper block 
should be better. But if I can buy that aluminium stock at very low 
price, and use it almost without modifications, that's a powerful advantage.

The core issue is that I need to calculate this stuff! If it turns out 
that the cheap aluminium extrusion is all I need, why should I bother 
with machining a copper block? On the other hand, if the aluminium 
extrusion proves to be unable to cool these MOSFETs well enough, while a 
machined copper block will do, then of course that's the deciding factor.

>> m here? That's no issue for me. It would go into a closed container
>> under the desk, holding maybe 20 liters of water.
> 
> Find for casual operating, but what about contesting.

True, for contesting you need to place a coil of copper tube in that 
container, and run some tap water through it. But I'm not into 
contesting. I'm a technical ham, I like building equipment, and my 
operation on air is not intense. I calculated that such a container with 
20 liters of water would warm up only by 15 degrees, during the wildest 
operating sessions I ever have!

> You are looking for ohms per CM^3 or were you just measuring the 
> resistance between two probes?

I filled a 1 meter long hose with water, and measured end-to-end using 
just the multimeter's probes.

> It's rare to find tap, well, or spring
> water with more than a few hundred ohms per cubic CM.

Even that might be good enough. If not, I can still buy some de-ionized 
water. And then watch it eat ions out of my cold plate over time!

  ;-)Anyway, first I have to decide on the more electronic aspects of my 
amp!  I'm still waiting for the samples of a few candidate MOSFETs, 
ordered a month ago. With international shipping and VERY slow customs 
processing, one needs to have patience...

Manfred.


========================
Visit my hobby homepage!
http://ludens.cl
========================