Copper or Aluminum

[BigMofo]

Copper...Copper!!!!

 

[ratbag19]

OK, so copper it is. But now I have another question: Someone told me that my RAM and chipset will get hot also, which is not surprising. Should I worry about it?

Also: Anyone that knows me also knows I always over-engineer things I build. Is it a good idea to slap a thin peice of copper on either side of my RAM and run water through it also? And how about my chipset?

 

[brodo]

there was an article on www.overclockers.com or somewhere I saw, where someone attempted to water cool his RAM. built a custom copper plate and everything. said it did not make a difference for overclocking.

I personally don't think you need to worry about it. But if you must, you could cool it using those heat spreaders and/or heatsinks and perhaps a fan on top as well.

Then again, some other RAM might get hotter than others - but I dunno anything about those stats. (anyone else?)

 

[athlonnerd]

yes, iot was on overclockers, the guy pelt cooled his ram, jet literally did not get 1 mhz increse, there is just no reason to cool a capacitave device. right?

 

[Tecumseh]

i tried google, i heard you can get diamond plates, or something like that, i would like to get ahold of a 2x2 or 2x3 peice.... dunno how possible that is...

Do a Google search on bort. Bort is sometimes called
black diamond. It's gone most of the way toward becoming
diamond from the coal end of the spectrum. For some
military applications CVD diamod is used. Heat spreaders
are made of this stuff. It's black like bort.

 

[frostmeister]

Just throwing a spanner in the works here - and it's made of cusil I saw a post on the overclockers guides about using a cusil coldplate. Not stirkingly better than copper - but enough to make it a feasible alternative for those really wanting to max things out. Oh, melting copper? yeah, if you can get a ceramic crucible, and enough heat, that should do. While you're melting the copper, you could bung in 28% silver (standard cusil alloy) and then pour it into your mould. What are you going to make the mould out of? Well, get some fine sand or clay, an epoxy binder (very little epoxy for this) and then make your mould out of say, a wax prototype. The sand / clay fills the wax, then you melt out the wax, and then have one mould - before you pour in the metal, run a blowtorch over the mould on low air flow, so the big yellow flame you get coats the mould in carbon, providing a smooth finish. Saw it done on an open university programme.

 

[Mr.DLucey]

I don't have any idea. I'd use google to find out.

I think most industrial diamonds are made in tiny sizes because they are used by embedding them in metal to make grinding and cutting tools. I don't think there is actually any practical values to diamonds in watercooling.

I suppose you could conceivably cast a baseplate for a WB with diamond filled copper. I think all the copper/diamond interfaces would limit the usefulness of something like this. Depending on thermal expansion characteristics, the diamonds might be rattling around in little air (or vacuum) pockets. That would be very bad for the thermal conductivity of the baseplate. [/QUOTE

If you do this keep your hours on lapping the bottom for us. You using wet dry paper? LOL

 

[Tecumseh]

Sorry, your cusil fantasy is just that. The guy who originally did the
study based his "estimate" for cusil's thermal conductivity on its
electrical conductivity. The theory was wrong and there is an
article somewhere on the front page that covers this.

 

[Since87]

If you do this keep your hours on lapping the bottom for us. You using wet dry paper? LOL

LOL

That would be a bit of a problem woudn't it?

 

[McWarren]

I've been told that although copper is a better heat conductor, aluminium can dissipate it quicker.

 

[Tecumseh]

Sorry, McWarren, but that is another Internet cooling fantasy.
Look up the old posts by Aesik.

 

[athlonnerd]

I've been told that although copper is a better heat conductor, aluminium can dissipate it quicker.

that makes absolutely no sense, aluminum is bettar a radiating heat, but radiation in this case is neglectable.

doesnt silver have a lowwer melting point, if so itd be real ease to do the mold thing.

frostmeister, yuo relly seem to know what your talkking about, thats a first ont ehse forums , besides the select few, disincluding myself.

 

[frostmeister]

Ahh - Firstly, thanks for the compliments athlonnerd - a bit overstated this time around though, as the cusil alloy figures I put in my last post were the wrong way round - the actual alloy is 28% copper, 72% silver. Makes it a bit more expensive.
Apologies for my cusil theory as well, it was a bit confusing, as I read the article (under peltier cooling/cold plate materials and thicknesses) as saying that the graph on the wesgo metals site was incorrect, and that the graph in the article had been re-calculated, so yeah, cusil is a poorer heat conductor. Why put up a graph based on dodgy data though?
I did see on the wesgo metals website an alloy called silfuz which I'm going to have to find more out about, as it beats silver's heat conductivity by a fair whack. And yes, silver does have a lower melting point and would be easier to mould. It'd also be an easier way to make waterblocks and heatsinks if you could mould them instead of having to faff about with mills, drills, lathes, cutters and ton's of other expensive metalwork equipment... Although thats half the fun isn't it?

 

[Mr.DLucey]

How about giving us the web address for wesgo metals so we can help you look for silfuz.

 

[Mr.DLucey]

I should stop being so lasy and look it up my self if I really want to know, Right?

http://www.wesgometals.com/index.html

 

[Mr.DLucey]

I didn't find anything their. Did you E mail them frostmeister for more information?

 

[frostmeister]

Hmm, no, I haven't e-mailed anyone about this silfuz yet; I'm going to do a search tommorow and see if I can find out any more on it. The wesgo metals website has several pages of alloys, listed in alphabetical order, or alloy number and so on. I'll have a look and find out some more info over the next few days. (I'ts late here and I have work tommorow.