DICE!! coolers please check my Pot out before i get it made.

[cerbrus2]

Hi guys basically looking for some experianced dice coolers. who can check out a Pot i desighned to see if they think it will work well or not.

The pot will be made of Copper milled out of one peice. Dont worry about the cost I have the lump of it sat here allready lol.

And I was thinking of doing the retaining coller out of S1.

Retaining Collar.

The Pot itself.

Cross section veiw.

 

[kitfit1]

Some dimentions would be a good help here cerbrus2.
But without dimentions, i would say that it might be a good idea to make the bottom of the pot much thicker than it looks in your drawings. With pots, MASS is what it is all about. For benching in general nowdays, it's not worth making a pot that will only benefit Dice. LN2 is where you WILL end up, so make the bottom as thick as you can to hold as low a temp as possible. To give you an idea, i borrowed a pot a couple of months ago that was 5kg of solid copper. Easily held a 990x at 5.5ghz with LN2 with just the occasional top up.

 

[JonJ678]

I have no worries about its performance, as I know little about CO2 / N2 cooling. I do have concerns about manufacture however.

How are you planning on milling it? Aside from being staggeringly wasteful in terms of copper, I don't think its possible to make that on a mill. Thin, parallel walls and a structure deeper than the length of a milling bit will be tricky.

Putting it on a lathe and turning it internally should be possible, though to echo kit above, thicker walls seem like a good idea here too. Heat capacity of the block is important. The design is cylindrical, so perhaps you meant "turn" when you said "mill", and I'm making a fuss about nothing.

I've got a related question, what's the advantage to making pot's entirely out of copper? Mostly brass, with a copper base brazed on, would be far easier to make and I believe perform identically. It would also be cheaper.

 

[cerbrus2]

Some dimentions would be a good help here cerbrus2.
But without dimentions, i would say that it might be a good idea to make the bottom of the pot much thicker than it looks in your drawings. With pots, MASS is what it is all about. For benching in general nowdays, it's not worth making a pot that will only benefit Dice. LN2 is where you WILL end up, so make the bottom as thick as you can to hold as low a temp as possible. To give you an idea, i borrowed a pot a couple of months ago that was 5kg of solid copper. Easily held a 990x at 5.5ghz with LN2 with just the occasional top up.

Sorry basically the bottom of the pot is 35mm Thick. And the walls ar 9mm thick. (without the shamfer for the retaining collar).

Total height is 220mm, and the outer circ is 100mm and this is reduced to 70mm at the bottom. so hopefully should give ample coverage.

I have no worries about its performance, as I know little about CO2 / N2 cooling. I do have concerns about manufacture however.

How are you planning on milling it? Aside from being staggeringly wasteful in terms of copper, I don't think its possible to make that on a mill. Thin, parallel walls and a structure deeper than the length of a milling bit will be tricky.

Putting it on a lathe and turning it internally should be possible, though to echo kit above, thicker walls seem like a good idea here too. Heat capacity of the block is important. The design is cylindrical, so perhaps you meant "turn" when you said "mill", and I'm making a fuss about nothing.

I've got a related question, what's the advantage to making pot's entirely out of copper? Mostly brass, with a copper base brazed on, would be far easier to make and I believe perform identically. It would also be cheaper.

I have a 130mm by 300mm solid copper roundbar. Came from a warehouse I was refurbing. it used to be a Massive Buzz bar for their electrical supply.

Yeah and I meant to put Machined not milled DOH!. Gave the plans to a local machine shop and they said its not a problem to do it. And they are Soing it for free. Well they just want to keep the scrap copper. whitch i dont mined as it didnt cost me anything at all. And at the end of the day I get a solid Copper pot.

Again with the using different materials. One i already have the copper and it will cost me nothing. Rather than having to purchase s1 or bronz to braze onto it. And also. machining it out of one peice garentees it to be solid and pretty much bombe proof. it will probably out live me lol.

And using S1 for the collars means i can get different collars made up for it. to suit what ever CPU i am using it on. AM2, AM3, 1155, 1366. 2011. etc etc.

 

[kitfit1]

I would go with a bottom of at least 70mm. Dice pull down time will be rubbish, as will LN2. But both Dice and LN2 will hold the load much much better and without using much of either during any given benching session.