Tok's cooling Tower....

tokamac · 09-12-2002, 09:05 PM

Ok I know it has been a while since I got this thing going, but I finally got hold of a digital camera... so what to you think of my first attempt at watercooling?
And yes this is milkcrate technology at it's finest

Here is the inside of the top half: 3 heater cores (8"x7.5"x2" each, fin area)
The parallel distribution manifolds are 3/4" ID pvc and all the braided hose you see is 5/8" ID.

Here is another top down view...

Air is vented in through the grated bottom by a 6.5" 235CFM rotron (however I only run it at below audible and have not tested it any higher, so I'm guessing it is pushing maybe ~70 CFM or less... The shroud is made of rubber roofing mat, so the fan is completely suspended and vibrationally isolated...

Click here for a closer view of the fan/shroud....

The pump and electrical wiring are in the very bottom of the fan crate... (I have the computer, Eheim, and fan plugged into a custom electrical box that is then plugged into my switch box that sits under my monitor.)

Click here for the front side of the Fan Crate...

The complete project

The filling/air removal tubes:

edit: Oh yea I forgot this image: The workstation:

I used more rubber roofing mat to seal/center the 13x13" floor tile top, makes a nice beer holder

Well there you go. All I gotta say is that was much more expensive than I first planned, but the quiet, ahhhh...
any comments or suggestions are welcome...

Specs:
1800XP @ 1830Mhz, 2.05V 166FSB
EPOX 8K3A+
Maze3 1/2" barbs
2 GPM (timed bucket method)
Idle 30 C
Load 38 C
Ambient 21 C
H2O (input) 27 C
H20 (output) don't know yet...

WebMasta33 · 09-13-2002, 12:12 AM

That's just huge

Nice job if ya like it. But ya need to put the bezel back on your case!

TonyM · 09-13-2002, 12:16 AM

Got so many twists and turns Im surprised the water actually flows, but good job its quite a unique setup, can you post some numbers please.

tokamac · 09-13-2002, 07:10 AM

WebMasta33,
yea, I plan on hooking the hoses thru the back as soon as I get some high flow quick connectors... I am a function before form kind of guy (see milk crates

), but I just couldn't bring myself to destroy the front making holes for the hoses

Tony,
the the twists and turns theory is most critical/significant only in high velocity/small ID hose applications... Once the water leaves the case of the computer it is in 5/8" ID hose till it gets to the cooling tower at which point it is in 3/4" ID pipe... and on top of that any hard 90 turn is done in 3/4" pipe (no restriction from barbs!) once the water leaves the distribution manifolds the flow is 1/3 what it was entering...

I unhooked the return hose and stuck it into a bucket and ran the system for a bit, then took a gallon jug and filled it up with the water comming out of the radiator array... I did this a few times and got about ~30 secs +/- 2 so this is about 2 gallons per minute... now since it is a "sealed" system the flow is probably a bit higher with the hose connected.

Specs:
1800XP @ 1830Mhz, 2.05V 166FSB
EPOX 8K3A+
Maze3 1/2" barbs

2 GPM (timed bucket method)
Idle = 30 C (XP diode)
Load = 38 C (XP diode)
Ambient = 21 C (exterior thermocouple)
H2O (input/load) = 27 C (2nd exterior thermocouple)
H20 (output) = don't know yet...

ItsSoLARGE · 09-13-2002, 08:08 AM

An engineering marvel

bigben2k · 09-13-2002, 09:20 AM

Nice work. I like the PVC "thing", I'm writing an article on using PVC to make an airtrap...

One Q: it looks like you have 2 sets of hoses coming in and out of your milk crate. What's the deal? Also, you have 2 refill points. Why?

Just curious...

tokamac · 09-13-2002, 12:45 PM

bigben2k,
origionally I only had the one fill tube at the exit of the ratiator array ( clear 3/4" vertical tube ).... however the flow was so fast that it was not effective enough at removing all the air (i.e. immediately

)... so I added the 1 1/2" PVC air trap at the input to the radiator array. Also works better for filling and measuring the temperature of the "hot" side of the system before it enters the radiators. Of note I recomend a larger PVC cleanout than the one I have... the liquid still moves fast enough to transport the "foam" (small bubbles) through the system (perhapse a 3" would be about right...) what do you think?

The top crate contains the radiators only... the bottom houses the pump, fan, rheostat and electrical connectins. So there is one hose that comes from the computer to the 1 1/2" cleanout, from there to the radiator array, then out the bottom around the side to the pump inside the bottom crate and one out from the pump to the computer....

Tokamac

bigben2k · 09-13-2002, 12:51 PM

Actually, I'm building mine with 1 1/2 parts too, but I've got that bent T inverted into a street elbow, with clear acrylic... It's complicated, just wait for the article!

With a 3 inch, you could pickup a "closet mount flange", and make a refill point that looks sexy...

tokamac · 09-13-2002, 06:12 PM

that is my one regret: not being able to see the fluid level in the 1 1/2" cleanout (bent T). I look forward to your article... (I'll prolly be kicking myself for not thinking of it first

)

I found that the best way to bleed my system was pick up the top half and turn it all kinds of upside down, sideways and everywhich way... so after about an hour I had all the bubbles out...

Tokamac

FRAGN'STIEN · 09-13-2002, 09:33 PM

Quote:

Originally posted by WebMasta33
That's just huge

Nice job if ya like it. But ya need to put the bezel back on your case!

I would leave it just the way it is. It reminds me of a primered 70 Boss Mustang with a blower sticking out where the hood used to be

utabintarbo · 09-14-2002, 01:22 AM

Quote:

Originally posted by WebMasta33
That's just huge

Nice job if ya like it. But ya need to put the bezel back on your case!

Kind of has that "Terminator" look about it!

Bob

tokamac · 09-14-2002, 10:35 PM

bigben2k,
if you want a crude schematic of how the water moves through the hoses click here.

Tokamac

09-12-2002, 09:05 PM	#1
tokamac Cooling Neophyte Join Date: Aug 2002 Location: G-Town NC Posts: 51	Tok's cooling Tower.... Ok I know it has been a while since I got this thing going, but I finally got hold of a digital camera... so what to you think of my first attempt at watercooling? And yes this is milkcrate technology at it's finest Here is the inside of the top half: 3 heater cores (8"x7.5"x2" each, fin area) The parallel distribution manifolds are 3/4" ID pvc and all the braided hose you see is 5/8" ID. Here is another top down view... Air is vented in through the grated bottom by a 6.5" 235CFM rotron (however I only run it at below audible and have not tested it any higher, so I'm guessing it is pushing maybe ~70 CFM or less... The shroud is made of rubber roofing mat, so the fan is completely suspended and vibrationally isolated... Click here for a closer view of the fan/shroud.... The pump and electrical wiring are in the very bottom of the fan crate... (I have the computer, Eheim, and fan plugged into a custom electrical box that is then plugged into my switch box that sits under my monitor.) Click here for the front side of the Fan Crate... The complete project The filling/air removal tubes: edit: Oh yea I forgot this image: The workstation: I used more rubber roofing mat to seal/center the 13x13" floor tile top, makes a nice beer holder Well there you go. All I gotta say is that was much more expensive than I first planned, but the quiet, ahhhh... any comments or suggestions are welcome... Specs: 1800XP @ 1830Mhz, 2.05V 166FSB EPOX 8K3A+ Maze3 1/2" barbs 2 GPM (timed bucket method) Idle 30 C Load 38 C Ambient 21 C H2O (input) 27 C H20 (output) don't know yet... Last edited by tokamac; 09-12-2002 at 09:50 PM.

09-13-2002, 12:12 AM	#2
WebMasta33 Cooling Savant Join Date: Dec 2001 Location: Nebraska Posts: 526	That's just huge Nice job if ya like it. But ya need to put the bezel back on your case! __________________ ========== --Soyo KT333 DRAGON Ultra (Platinum Edition) --AMD 1.4 @ 1.54 (11x) --768mb of PC2100 (@140FSB) --Asus v8200 GeForce3 @ 210/490 --Maxtor 40gb Quiet + 2 IBM 40gb 60GXPs (removable) + 8gb WD --Audigy --Maze 3, DD Gf3 block, 2x BIX, Eheim1250 ===========

09-13-2002, 07:10 AM	#4
tokamac Cooling Neophyte Join Date: Aug 2002 Location: G-Town NC Posts: 51	WebMasta33, yea, I plan on hooking the hoses thru the back as soon as I get some high flow quick connectors... I am a function before form kind of guy (see milk crates ), but I just couldn't bring myself to destroy the front making holes for the hoses Tony, the the twists and turns theory is most critical/significant only in high velocity/small ID hose applications... Once the water leaves the case of the computer it is in 5/8" ID hose till it gets to the cooling tower at which point it is in 3/4" ID pipe... and on top of that any hard 90 turn is done in 3/4" pipe (no restriction from barbs!) once the water leaves the distribution manifolds the flow is 1/3 what it was entering... I unhooked the return hose and stuck it into a bucket and ran the system for a bit, then took a gallon jug and filled it up with the water comming out of the radiator array... I did this a few times and got about ~30 secs +/- 2 so this is about 2 gallons per minute... now since it is a "sealed" system the flow is probably a bit higher with the hose connected. Specs: 1800XP @ 1830Mhz, 2.05V 166FSB EPOX 8K3A+ Maze3 1/2" barbs 2 GPM (timed bucket method) Idle = 30 C (XP diode) Load = 38 C (XP diode) Ambient = 21 C (exterior thermocouple) H2O (input/load) = 27 C (2nd exterior thermocouple) H20 (output) = don't know yet... Last edited by tokamac; 09-13-2002 at 07:18 AM.

09-13-2002, 08:08 AM	#5
ItsSoLARGE Cooling Savant Join Date: Feb 2001 Location: So Cal Posts: 125	An engineering marvel __________________ Plumber's Goop: The Watercooler's Duct Tape

09-13-2002, 09:20 AM	#6
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Nice work. I like the PVC "thing", I'm writing an article on using PVC to make an airtrap... One Q: it looks like you have 2 sets of hoses coming in and out of your milk crate. What's the deal? Also, you have 2 refill points. Why? Just curious... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-13-2002, 12:16 AM	#3
TonyM Cooling Neophyte Join Date: Mar 2002 Location: Mount Prospect Posts: 78	Got so many twists and turns Im surprised the water actually flows, but good job its quite a unique setup, can you post some numbers please.

09-13-2002, 12:45 PM	#7
tokamac Cooling Neophyte Join Date: Aug 2002 Location: G-Town NC Posts: 51	bigben2k, origionally I only had the one fill tube at the exit of the ratiator array ( clear 3/4" vertical tube ).... however the flow was so fast that it was not effective enough at removing all the air (i.e. immediately )... so I added the 1 1/2" PVC air trap at the input to the radiator array. Also works better for filling and measuring the temperature of the "hot" side of the system before it enters the radiators. Of note I recomend a larger PVC cleanout than the one I have... the liquid still moves fast enough to transport the "foam" (small bubbles) through the system (perhapse a 3" would be about right...) what do you think? The top crate contains the radiators only... the bottom houses the pump, fan, rheostat and electrical connectins. So there is one hose that comes from the computer to the 1 1/2" cleanout, from there to the radiator array, then out the bottom around the side to the pump inside the bottom crate and one out from the pump to the computer.... Tokamac

09-13-2002, 12:51 PM	#8
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Actually, I'm building mine with 1 1/2 parts too, but I've got that bent T inverted into a street elbow, with clear acrylic... It's complicated, just wait for the article! With a 3 inch, you could pickup a "closet mount flange", and make a refill point that looks sexy... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-13-2002, 06:12 PM	#9
tokamac Cooling Neophyte Join Date: Aug 2002 Location: G-Town NC Posts: 51	that is my one regret: not being able to see the fluid level in the 1 1/2" cleanout (bent T). I look forward to your article... (I'll prolly be kicking myself for not thinking of it first ) I found that the best way to bleed my system was pick up the top half and turn it all kinds of upside down, sideways and everywhich way... so after about an hour I had all the bubbles out... Tokamac

09-14-2002, 10:35 PM	#12
tokamac Cooling Neophyte Join Date: Aug 2002 Location: G-Town NC Posts: 51	bigben2k, if you want a crude schematic of how the water moves through the hoses click here. Tokamac

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)