View Full Version : OverClockers, any new ideas/technology to share?

19-01-2008, 07:08 PM
Most of the threads in this section are well old, and given the speed and volume of new technology in the IT world, I wondering if I've missed anything in regards to cooling technology.


The MG Pony
19-01-2008, 09:31 PM
Cascades are becoming populer for heavy OCing and benching

AutoCascades are geting stronger for the 24/7 ussage and for heavy OCing, average temp is -100C for both type systems

For a standerd single stage average temps are -40 to -50C at 300 to 150Ws

Liquid chillers are as usualy populer add on to a water cooling loop with the temps ranging from just 10c lowe then ambient to -30C

Other then that surf the net mate get tons of stuff

24-06-2008, 04:36 AM
I'm no professional, but i do know some cases where Stirling coolers are making a run trying to replace cascade systems.

An example in development currently is a portable cooler box with about 25L of volume capable of cooling to -92C using a 100W model FPSC (free-piston stirling cooler), designed to replace a small cascade system. (it can also be plugged into a car outlet for a few days w/out draining the battery)

There have been recent successful military electronics cooling applications as well. I'm not sure about one application (secret military stuff) but a stirling cooling system was used to cool electronics operating at 250-300C, where some kind of messy liquid cooling was being used before. (apparently the military was skeptical on that one as they said 200C but when successful they said they really needed a design for 300C, also successful.)

I was also told that using a Stirling cooler has the capability to allow a computer processor to operate 5x better.

These are just some examples i've obtained from talking to Global Cooling, a spin-off of Sunpower Inc (inventor of FPSC).

10-11-2008, 05:20 AM
I think possibly the quickest way is to use a fruit juice machine with glycol and ur standard water jacket cpu block. This system should get a constant low temp. The major drawback is condensation, but i am confident that a thick layer of circuit board laquer with some other type of space filling foam around cpu and block should work. I have yet to test, but if i ever get around to it this is the first place ill let know.

12-11-2008, 02:16 AM
If you're looking to try something a little different, you could try using a kleemenko cycle. less parts than a normal autocascade. If have a little r14 somewhere then i know a gas mix that will work well. i have schematics etc if you're intersted

20-12-2008, 10:55 PM
By Kleemenko Cycle you mean a phase separator less autocascade. As far as performance goes they are very good, even if they involve some extra fun during the building process. The main drawback however with the lack of separation is not only do you need a higher flow to maintain a similar phase separated system, but you cannot go as cold without dealing with oil problems. Thus you end up adding an oil separator, and this can cause entrapment of some of your "sponge" refrigerant like r123, r11, or more recently r600/a.

17-12-2009, 02:13 AM
I was wondering if we can have this moved to the overclockers section. This will be a really good place for it as there will be people looking for this thread as the chip grows more and more popular. Not to mentioon I will be posting some pretty heavy overclocks with this chip as well.

28-12-2009, 05:04 AM
I just want to bump this thread and keep it current, I'm a noob to such technology and also new to this forum, I actually have a phase change prometeia I got off ebay 4 years ago or so and never used other than to test and it worked fine the few times I just randomly turned it on to check it out...however today I dug it out to consider using it on my rig...I got it to -51 degrees no problem...till it started beeping loudly at me and the digital readout was flashing some sort of error symbol...it is a mach 1 unit that was modded with a switch to turn it on and off manually I believe. Anyway I shut it off restarted it within 10 seconds or so and it got down to -45 and stayed there and so I figured it may be working but wasn't quite right...then I turn it back on half hour or so later and now on several attempts it gets to 12 degrees or so and that is IT...you can get it down to 6 degrees if you wait for several minutes and I hadn't felt the condensor before but now it's so hot to the touch after a couple minutes you can barely hold your hand on it for more than a second or two...any ideas what is ailing the unit and as far as fixing it goes...I have no idea where to even start.

04-07-2010, 05:47 PM
Hi ppl, I have been working on a system I designed over a year ago. It takes over-clocking to a new level. The computer components are fully submerged in a special fluid that is non conductive and non corrosive to electronic components. Its viscocity is almost the same as water but it weighs twice as much. The custom made evap coil made from 1/4 copper sits inside the cistom poly tank. The only things that are not in the fluid is the HDDs and the DVD roms.

The compressor, condensor and fan all sit neatly on-top of the custom tank with a brushed aliminium case a seal all the way round seals the fluis into the tank and it can be transported with no leaks.

The unit is quite heavy but is no bigger than a normal tower.

The fluid at full load with an overclicked Q6600 @ 3.7Ghz is at a constant 20.C.

The compressor is a very small R134a LBP capillary system designed by myself. Only problems I have had so far is that it takes 1 hour for the unit to be ready to used at its overclocked state due to the temps.

Other than this if the unit has been running and the fluid is nice and chilled down to its -10.C its good to go

28-08-2010, 02:10 PM
Hi all,
I recently came across a 1hp Toshiba variable speed compressor, it's 3 phase and runs from 30-120hz, i purchased a VFD of ebay to power it.

It's great i can slow it down as it saves power. However it gets bad harmonic vibrations below 55 hz.