water chilling

[SteenkyBastage]

well, I delayed quite some time in my refrigerant cooling plans... but today after 4 hours of literally tearing up a mini-fridge, I have started the ball rolling again.

I plan to let the compressor sit for half a day or so, but tomorrow I plan on hooking just the compressor up and verifying we didn't kill it when we ripped it out.

ahhh... to all who have thought to try this, but haven't got up the nerve... IT'S A ROYAL PITA!!! (that's pain in the anus ) If I knew it would be so much of a hassle to get the entire compressor and lines out of this thing... I never would have tried it.

About 2 hours into it, my friend and I were thinking that it wasn't worth the hassle, but since we had already spent so much time, we didn't want to give up.

So, soon I will get to find out if all the effort was worth it... I certainly hope so.

I plan on leaving my radiator in the loop (after the cpu waterblock) and also chilling the reservoir. Now I just gotta get some good insulation goin on....

 

[Avatar28]

I know what you mean about starting something and finding it a lot worse than thought. I've deteremined to set out and design and build my own case. Man, I never thought that just DESIGNING the thing would be so difficult.

 

[SteenkyBastage]

hey avatar28, welcome to the forums!

yeah, I bit off more than I expected doing this. The main problem is that I thought there was only one straight copper line going to/from the coldplate... but in fact the WHOLE fridge walls were covered with winding copper tubeing... there was literally NOTHING left of the fridge intact other than the compressor and the door

hope you have better luck than me with your case!

 

[SteenkyBastage]

well, we didn't kill it when we ripped it out... that's a plus. I'd have been pretty peeved if we wasted half a day for nothing... Now I get to protect from condensation and find a big enough tank to put the coldplate in...

 

[Tiger]

I plan on leaving my radiator in the loop (after the cpu waterblock) and also chilling the reservoir. Now I just gotta get some good insulation goin on....

That's not a good plan. I'll try explain the logic. You have your evaporator in the coolant and heat is transfered from the coolant into the refrigerant and out to air. Most likely the the coolant is going to be sub-0. You pass it to the water block via insulated line so it doesn't pick up heat from the air and then to the radiator. The only thing that is going to happen in the radiator is that is going to heat the coolant some more. The heat transfered from the core to the coolant raises the temp of the coolant by about 0.2C(ie the coolant is 0.2C hotter when it comes out of the water block than it was when it went in). The ideal is to minimise the heating of the coolant to the proc only. Insulate to minimise any other heat sources.

 

[SteenkyBastage]

yeah, if it doesn't work with the radiator, i'll remove it. I am just not positive it can handle the heat load w/o the radiator.

if it can, that will be nice to have one less 120mm fan, but I'm not optomistic.

I didn't know that the water temps only raised that much. I was under the impression that it was a fairly decent raise in temps. I wasnt expecting subzero temps, just cooler temps.

oh well...I'll experiment and see what works.

 

[Tiger]

Trust me I work with this stuff. Look at my sig.

 

[SteenkyBastage]

yes, but I'll also be using this on two xp's, one northbridge and two video cards. the flow is split in half, so the water is flowing much slower thru the blocks than the radiator/pump.

if the coolant only rises .2C thru all that, then I'll ditch the radiator.

I dont want subzero temps. What I had hoped for is to have the water flowing in a few deg C below ambient, and being warmed up to or slightly above ambient. For my intended purposes, the radiator would help.

now, whether it works out that way or not... we'll see.

Thnx for the input. Yes, if the temps dont change much, or if they're way below ambient, I will remove the radiator. I was looking at this chiller to suppliment my system, hadn't thought of it being able to cool the whole thing by itself.

 

[racecar12]

Trust me I work with this stuff. Look at my sig.

Very good temps. What kind of chiller are you using?

 

[Tiger]

Very good temps. What kind of chiller are you using?

I use a domestic fridge powered with a 110W compressor (R12)and the evaporator immersed in a 20L tub of coolant(40%ethylene glycol) with a Max-jet 750 submerged pump. The flow rate is about 1.2L/min.

 

[racecar12]

I use a domestic fridge powered with a 110W compressor (R12)and the evaporator immersed in a 20L tub of coolant(40%ethylene glycol) with a Max-jet 750 submerged pump. The flow rate is about 1.2L/min.

Sounds like one extreme cooling machine> Got Pics?

Just think....What kind of core temps could I achieve using this to cool a 220watt pelter?

 

[SteenkyBastage]

also, tiger... I dont think my fridge is on the same level as yours. mine is 1/20 HP, using r134a.

110v and 4.5A. Something must be wrong cause I dont think that should be a 500w compressor, right?

 

[knight1976]

Yea yea post pics!!!!

 

[SteenkyBastage]

when trying to bend the evaporator to fit in a 5gal bucket (it was just a tiny bit too big) I managed to spring a leak...

So, all that work in getting the thing out of the fridge was a waste... I can honestly say that I wont be attempting to pull another fridge apart to try this.

If my dual systems temps are too high for un-aided watercooling, I may go to a company who works with this and have them refill the system or get a new one (that isn't inside a fridge!).

 

[knight1976]

That sux man you should have just got a bigger bucket!!!! lol

 

[SteenkyBastage]

LOL, yeah.

I looked and looked all over home depo for a larger watertight container... that was the biggest one I could find.

and man, It was soooooooo close to fitting, all I had to do was slightly bend the evaporator. I think the leak happened at a weld, not on the evaporator itself. In bending the evaporator, I must have loosened up the weld on accident.

oh well... live and learn.

 

[Tiger]

Sounds like one extreme cooling machine> Got Pics?

Just think....What kind of core temps could I achieve using this to cool a 220watt pelter?

I did a write-up on it which you'll find here . There are quite a few pics etc.

 

[Tiger]

also, tiger... I dont think my fridge is on the same level as yours. mine is 1/20 HP, using r134a.

110v and 4.5A. Something must be wrong cause I dont think that should be a 500w compressor, right?

Compressors are rated by work output and it is not as simple as P=IV.
Its important to remember that fridge design and end temps are a result of total system design of which the compressor is a single component. If you want to read more on the subject there is a lot of detail in the Project X thread.