jmke

no clear response on the "cola trick" , still stands for "boozing it up" in my book

__________________

blind_ripper

give me one day without cola and il die , il even kill for it

__________________

Massman

It has something to do with keeping the ln² temp from fluctuating to much

I'm really looking forward to sunday ...

jmke

anything more scientific than "it has something to do with"?

__________________

blind_ripper

well it hard to explain , but witte whase surely exprest by the Ln² .
Ln² has a boiling point off -196°C or so , to get to that point u need to have a surten amount of Ln² .
the stuff in cola (sugger , the stuff that forms gas dont know the name) in contact with Ln² make's it boil very fast at less Ln² .
the controll is 2°C less controll but the heat absorption is mutch bigger , this what we need .
becaus a quad puts out a huge amount off heat after 1.62vcore , normaal Ln² cant keep up and temps start to rise .
the cola trick is a nice add , it whase found by a little fauld @ ZDnet when we benched there .
little chunks of cola fall in to pot , there is where i saw what happed .
so i did the same thing here @ my home and found out that it helps with high load heat out put .
im really the finder of this trick so all credits come to me .
if i get the WR in 05 il put out the trick on XS , im sure some will say that is allready know but if asked vince (kingpin) and no one did this .
he will be testing it soon 2

no u guys know it so hehe

__________________

Massman

Sweet, that's one for the article

jmke

so Coke is an accelerator for evaporating LN2 ?

DId you spill some by accident to find this out?

__________________

blind_ripper

spilled some in the bucked where we refilled the pot with , phase cooling a guy Can of cola in the Ln² it started to foam and some fell in to the bucked froze up and there we had it little pieces whet into the tube there u have it .
second time i did this to see if it really worked and it did

__________________

jmke

most discoveries are made "by accident" indeed

__________________

geoffrey

Indeed, Coke (like in Coca Cola ) really seem to have an impact on LN2 cooling performance. First you'll have take a small amount of coke and deepfreez it, then you'll have to break the frozen Cola appart into little chunks which are then trown inside you ALU/CU cooling tube. Then add LN2 and do the regular LN2 benchmarking stuff...

What would be the cause of this? I don't think it's is because of a gas escaping from the coke-chunks, that way the effect would go away after a while which and on first sight that didn't seem to happen. Maybe it's because Cola is based on water, which on his turn can handle heat quite good and theirefore causes rapid LN2 boiling. Or maybe not, I'm no engineer

In the end it's hard to judge for me if it really worked, because I didn't compare with or without Coke added, though there was not much swing between idle and load which I though was a real problem when benchmarking with LN2. Also, if this trick does work, this would also mean that there wold be a optimal LN2-Cola mix for each kind of tube. If you're bored you know what to do