submerged gainclone...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Funny that someone will think nothing of dropping thousands of dollars on wires but get the vapors when considering a few hundred bucks for a cooling fluid that is inert, has great dielectric properties, doesn't evaporate significantly, is non-corrosive, and has a long track record in high-reliability applications.
 
Don't forget that it can be used as blood plasma replacement, and will reoxygenate blood.

Oh, yeah, and my favorite application: Liquid Cooled Cray Supercomputers. Countless Crays used gallons and gallons of the stuff to cool their massively parallel scalar and vector systems.

I guess on a $48k tube amp, $300-$600 worth of fluorinert would be no big deal. (Hopefully that's what they're using)
 
demogorgon said:


deionized water dont conduct.. there is nothing in it to conduct, and eat up the metal?
my computer waterblocks are holding on just fine after two years of using the stuff..
for a tast it'l work just fine.
it did when i submerged my old computer in the stuff to see if it was anything to shout hourray for.. not that it was anyways, didnt come over 2ghz on my xp1700 that does 2.3 on air. but back to topic..


When water is actually deionized, yes, it is effectively non-conductive as it's resistance is in the megaohms. However, as soon as it comes in contact with metal, it will leech ions, as water is autoionizing, and will become conductive again.

Your watercooling system is not a valid example of the properties of deionized water. Being that it's in a closed loop and I presume you don't change the water very frequently, corrosion won't be immediately noticable. The water will leech what ions it needs to reach equilibrium and will cease to corrode the metal. But make no mistake, deionized water is NOT what you want to be immersing electronics in. Given a bit of time, it will become conductive and you will fry your hardware. The only way to prevent this is to continually replace the water with fresh deionized water and then you WILL have appreciable corrosion.

Just stick to mineral oil or Fluorinert.
 
I work in the high voltage industry and we use transformer oil (not sure of the technical/scientific name) to insulate things upto 330kv and i can commonly get the 50kv stuff for free :) . If your doing it just for the hell of it you could have the oil route to a small radiator and let natural convection circulate and cool the oil for you.

Not sure how well it would work on the small scale but hey its an idea. Would a small radiator out of a car heater be sufficient do you think?
 
elementx said:
i found this site and decided to post it, really interesting stuff!! they submerged a computer mobo in liquid nitrogen, and flrorinert... and it worked, well, sort of...

http://www.octools.com/index.cgi?caller=articles/submersion/submersion.html

it shows that it can be done, but for how long?

A better question: What's the point?

Chip amps run pretty cool anyway. I you wanted to pack a class A amp into minimal space, then you might go to all this trouble. But it is just silly for chips that only rise a couple degrees above ambient when bolted to a minimal heatsink. I suppose you could call it art...

I_F
 
Tweeker said:
Probably more benefit to a class A amp also, but one point I could see if there was a fair amount of circulation is that it would help in keeping all components at the same temperature.

That isn't really a good thing to do if parts are heating up. In electronics you want minimal heat, period. Heat reduces reliability. That's why hot parts are put on heatsinks. If the output transistors are getting hot, you want to get rid of the heat, you don't want to use it to heat up the rest of the amp.

I_F
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.