let the madness begin...

EricFX1984

[H]ard|Gawd
Joined
May 21, 2010
Messages
1,122
well it has begun...

issues so far... not sure how I should bend the evap into the cooler... not sure if a D5 pump will be ok for these low temps...

DSCF1858.jpg


DSCF1859.jpg


DSCF1860.jpg
 
this will all be ran in my storage room which is right next to my office!


this was the first test... I will be fully submersing the evap once I patch the hole in the side or the cooler

DSCF1861.jpg



this is what it looks like so far

DSCF1862.jpg
 
last night I talked to a guy that has been running a D4 pump @ -30c for about 4 years... so I am not worried about it... he did mention that if you plan to do any work at all to the pump you must must must wait for it to warm up or it will break :rofl:
 
started my first test, not hooked to anything yet... do not have a working temp probe ATM...

trying to freeze winter windshield washer fluid

18:55: turned on

19:01 noticed tube is frosty...

19:04: had to shut down small coolant leak
 
patched up the leak in the cooler (res) using hot glue :rolleyes:

starting test (attempting to freeze winter windsheild washer fluid) starting at room temp of 21c

22:27: started chiller

22:29: observed condensation on suction line

22:34: observed ice crystals forming on suction tube

22:40: coolant is very cold to the touch, similar to the last beer in the bottom of the ice cold water in a cooler

22:53: no real changes, still cold
 
ran some chilled coolant through the water block and the block turned to ice in a few seconds!
 
here are some pics...

it only takes a few mins to get the block turned to ice...

DSCF1864.jpg


DSCF1863.jpg
 
room temp @ 16:46 82.3F


coolant temp before start up 58.6F


coolant temp at start up (16:49) 58.6F

----------

coolant temp @ 16:51 57.5F

-----------

" " @ 16:53 54.9F
light condensation on suction line and coolant tubes

--------


" " @ 16:55 50.2F
no change in condensation

---------

" " @ 16:57 48.2F
slightly more condensation

----------

" " @ 16:59 44.2F
heavy condesation

----------

" " @ 17:01 41.9F

----------

" " @ 17:03 39.9F

----------

" " @ 17:05 37.2F

----------

" " @ 17:07 35.4F

----------

" " @ 17:09 33.4F
even more condensation, like its about to rain!

--------

" " @ 17:11 31.3F
light frost on suction tube, none on coolant tubes

--------

" " @ 17:13 30.0F
heavy frost on suction tube

-------

" " @ 17:15 30.0F

-------

" " @ 17:23 19.4F
heavy frost on water block, light frost on coolant tubes

--------

" " @ 17:26 18.2F

--------

" " @ 17:30 16.2F

---------

" " @ 17:33 13.8F
frost around barbs

--------

" " @ 17:43 8.6F
frost on coolant tubes, frost on D5 pump

--------

" " @ 17:58 2.8F
almost on frost on the intake coolant hose?

--------

" " @ 18:02 1.2F

--------

" " @ 18:07 0.0F
WOW! :)

-------
 
18:12 -1.3F

------

18:22 -3.5F

-------

18:32 -5.3F

-------

18:41 -7.1F

-------

18:57 -8.9

-------

19:20 -11.0F

------

19:26 - shut down, Beer time! -11.2F
 
I am going to fire it up today and use a lazer thermometer to get temps on the block face
 
ok it takes around 45 mins to pull to 0F at the water block face... not too bad if you ask me
 
And for us super noobs, how would that be? :p

lol, oh sorry... well the main thing is to keep air from hitting the cold parts... if you go to the hardware store you can buy foam sleeves and basically seal them up and that takes care of most of it...

kneaded eraser will take care of the board
 
Last edited:
Was wondering if you considered dropping the thermostat probe from the AC into the bath to see if it would go/register/allow enough adjustment to moderate the cooling so the AC did not run all the time. Perhaps you are just not there yet in your testing.

Not sure of your goals for the system, you may need to run flat out once you get a heat source going. I was thinking it would help with the frost and condensation issues if you had some control. I am guessing that you would need to swap out the existing thermostat with say one for a fridge or freezer so it works in the temp range you are looking for but I guess it depends on what the operating range ends up after you get a CPU or other heat source introduced to the system.

Anyway, great pics, please keep us updated as you work, looks like big fun.

You might want to check the extreme cooling section as a guy has found a new product for protecting parts from condensation etc. I had seen preliminary info about this product a while ago and it seemed almost ideal if a bit pricy at $30, but enough to do a lot of boards. http://hardforum.com/showthread.php?t=1526416
 
Last edited:
looks like next friday or saturday I will be insulating my board and installing this chiller unto my PII 965, I can already do 4.0GHz on air if I run my Kaze Ultra (130+ cfm) instead of my Delta (90 cfm)

I will also be switching over to a tech station to make my life easier!

so I am hoping for a solid 4.5GHz machine
 
well i have to go to St Louis tonight and Kentucky with the 'rents so it might be monday or tuesday before I can get anything done on this beast
 
ok everything is cold at idle (less than 0c) but almost instantly the CPU hits around 18c on load (4.2GHz 1.425v Phenom II 965 Black) is this normal or do I need to tear into the block and reseat?

however the coolant in the res stays around -7c



4ghzidle.jpg



4ghz100.jpg



42ghz100.jpg
 
Last edited:
Nice man very nice! Sweet temps. You got Antarctica in there. I do no think you need to worry about heat XD
 
Last edited:
Nice man very nice! Sweet temps. You got Antarctica in there. I do no think you need to worry about heat XD


thats the thing its still hitting 18c under load... that is not impressive at all really

I have to wait for the temps to come back up, my guess is that the block did not make perfect contact
 
ok it takes around 45 mins to pull to 0F at the water block face... not too bad if you ask me

I've never done this and know basically zero about it, but was wondering if this implies that you must have the cooling system running and pre-cooled before turning on a system that's already maxed-OCed for use with the system?

Just curious, but I got the impression you don't just turn everything on all at the same time with a single button...

(like a freezer always runs... you don't just turn it on/off when food's in/not in it... you wouldn't expect a bag of ice to stay a perfect bag of ice if you threw it in an off freezer and just first turned it on when you first put the bag of ice in? Hope this makes sense...)
 
I've never done this and know basically zero about it, but was wondering if this implies that you must have the cooling system running and pre-cooled before turning on a system that's already maxed-OCed for use with the system?

Just curious, but I got the impression you don't just turn everything on all at the same time with a single button...

(like a freezer always runs... you don't just turn it on/off when food's in/not in it... you wouldn't expect a bag of ice to stay a perfect bag of ice if you threw it in an off freezer and just first turned it on when you first put the bag of ice in? Hope this makes sense...)

you are right, I have to precool the loop.
 
I think this might be the issue. Correct me if I am wrong but this is really uneven and pretty thick on one side.

DSCF2031.jpg



DSCF2032.jpg
 
what past are you using it locks like it cracking because it to cold.
Now about your temps there are about right where they should be-7c to 18 is a 25c split. if you want colder you need to get your water colder -7c is only about 20f if you can get you water to say 0F you temps will be 8c at load. You could add a 2end pump and that will lower it a little because the colder you get the water the thicker it gets.

to give you and idea how to know your temps take my rig.
right now my water temp is 71F or 21c my load temps are 43c. if my water temps are -7c my temps would be 28c colder or about 15c load temps. so as you can see your about right there where it should be.
 
This is pretty awesome.

Can you give more details about what you used as the chiller?

It looks like th ecooling element from an AC unit ripped out of its case and placed in a cooler. is that correct? How many BTU was the AC rated for? I didn't realize a standard AC unit would be sufficient to get the fluid down that cold.
 
Zarathustra[H];1036045462 said:
This is pretty awesome.

Can you give more details about what you used as the chiller?

It looks like th ecooling element from an AC unit ripped out of its case and placed in a cooler. is that correct? How many BTU was the AC rated for? I didn't realize a standard AC unit would be sufficient to get the fluid down that cold.

well this unit is a 5,000 BTU setup, and I just took the evaporator and placed it in a cooler full of winter windshield washer fluid... well basically :D
 
Buy yourself an open air tech station. With that type of setup your conventional closed case will do you no good except used as an awkward and expensive pedestal.
Stdv2_640w.jpg

I'd also say you might need to research what TIMs do best in below 0 temperature conditions.
 
I think this might be the issue. Correct me if I am wrong but this is really uneven and pretty thick on one side.

DSCF2031.jpg



DSCF2032.jpg

Looks to me like:
A. Too much TIM
B. TIM has frozen/not taken to low temperatures very well
C. Not seated properly (lets hope the Ghetto bracket helps :) )

TIM like AS5 starts to crystalize below 0c and freezes at about -20c. I would try using Arctic Silver's Ceramique TIM, can go very low (-150c or so)

Good luck! :)
 
Last edited:
What TIM are you using? And yes for god's sake get you a open station, but not the one he linked, get the TOP DECK.
 
Back
Top