• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Watercooling Setup .. Please give me some help and advice ?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Nomortal

New Member
Joined
Jul 29, 2001
here is my system:
1.33 tbird @ 1540mhz
Lian li pc70 case
386 megs pc 150 kingmax
hercules gf2 ultra

i wanna go watercooling .. cuz my case is so loud and im tried of air cooling .So here is what
I plan on getting ..

Aquastealth Aquacoil Radiator
DD Maze 2 or Switftech waterblock ? any advice is appreciated
Eheim1250 or 1048 pump
Im gonna use all 1/2 tubing .. or should i not ?
no resivior .. closed system ..

please help me out with some comments and advice that could make this better .. better equipment to use .. etc .. thanks
 

Flash

Member
Joined
May 21, 2001
It sounds like you're on the right track. Part of the difficulty in purchasing water cooling equipment is that there aren't many head-to-head comparisons showing what's best. For instance, if anyone has compared the DD cooling cube to the Aquacoil cube I'm not aware of it. Seems like a no brainer that one of these hardware sites would put these similar rads to the test (not to mention the other popular radiators), but no such luck. Instead we get endless comparisons of the same gd heatsinks.

Same goes for water blocks. Someone did compare one of the Maze blocks to one of the Swiftech blocks (the DD block won), but both designs have changed several times in the interim.

Given all of that, my recommendation would be to go with one of the DD kits. That'll save you shipping costs and if you have a problem you'll have just one place to go for support.
 

AMDGuy

Member
Joined
Apr 10, 2001
Flash (Jul 29, 2001 11:13 p.m.):
It sounds like you're on the right track. Part of the difficulty in purchasing water cooling equipment is that there aren't many head-to-head comparisons showing what's best. For instance, if anyone has compared the DD cooling cube to the Aquacoil cube I'm not aware of it. Seems like a no brainer that one of these hardware sites would put these similar rads to the test (not to mention the other popular radiators), but no such luck. Instead we get endless comparisons of the same gd heatsinks.

Same goes for water blocks. Someone did compare one of the Maze blocks to one of the Swiftech blocks (the DD block won), but both designs have changed several times in the interim.

Given all of that, my recommendation would be to go with one of the DD kits. That'll save you shipping costs and if you have a problem you'll have just one place to go for support.

Actually this site compared the Aquacoil and one of the Cube type radiators and found a 1C difference in temp. I'd say get the Aquacoil. It'll work fine. I have on in my PC and am quite happy with it. Small size fit perfect. Cooling cubes are too cumbersome for small cases and don't provide that great of an improvement in cooling. I guess water is pretty efficient even with a small radiator.
 

Flash

Member
Joined
May 21, 2001
AMDGuy (Jul 30, 2001 11:52 a.m.):Actually this site compared the Aquacoil and one of the Cube type radiators and found a 1C difference in temp. I'd say get the Aquacoil. It'll work fine. I have on in my PC and am quite happy with it. Small size fit perfect. Cooling cubes are too cumbersome for small cases and don't provide that great of an improvement in cooling. I guess water is pretty efficient even with a small radiator.

I'd take that comparison with a large grain of salt. The reviewer was using a Danner 1.5 pump which he later measured 19.2 gph running through the Aquacoil radiator. The cube has many more passes, and therefore restricts flow more, so I'm sure its performance really suffered due to that puny pump.

Anyways, I've got a CC on order and I'll do some comparisons for y'all between it and the Aquacoil (or the Leufken equivalent which I'm using now), using both a puny Rio 180 pump and a beefier Eheim 1250. I'll also compare the Leufken water block to the Maze2. It should be interesting to see wich single component makes the biggest difference, and what the differences really are.
 

Richard

Senior Member
Joined
Jan 15, 2001
I've gotta agree with AMD guy on this one. You have to keep in mind what these small radiators were originally designed for. Even a 1.5 GHz Tbird is a joke in comparison to what these oil coolers and tranny coolers have to put up with. My motorcycle uses a small oil cooler similar to the Aquacoil that keeps the engine at 150 degrees Farenheit @ 55 MPH. Temps quickly go up at a redlight. (Up to 200 degrees before the fan kicks in, then the fan maintains the temp of the bike at about 190-200 degrees - even on a hot 90 degree day!) That's quite an accomplishment considering I have a 140 HP 900 cc Ninja.

A large enough radiator could in theory keep a Tbird at near ambient through simple radiant cooling. (Note: We're talking big. Like 12" * 12" stacked plate oil cooler.) Granted the surface area requirement is pretty significant for this kind of cooling, because these rads are designed to function through convection cooling. (Traveling at 55 MPH or at least a fan blowing on them)


I'll try and make sense of all this now. Even a "small" radiator like the Aquacoil is significant when paired with a 120 mm fan.
 

Flash

Member
Joined
May 21, 2001
Yeah, maybe it's just me, but the physics involved is way to complex to be accurately described by these kinds of analogies and unscientific tests. But I *will* find out and then I'll feel better, even if the result is a difference of only 1 degree.
 

Richard

Senior Member
Joined
Jan 15, 2001
Just calculate the BTU load of a TBird processor, and then figure out how many BTUs the coolers are rated for. Then you'll have your scientific answer. I assure you, the oil coolers are far in excess of the BTU load of a TBrid CPU.

Flash (Jul 30, 2001 06:01 p.m.):
Yeah, maybe it's just me, but the physics involved is way to complex to be accurately described by these kinds of analogies and unscientific tests. But I *will* find out and then I'll feel better, even if the result is a difference of only 1 degree.
 

LordSmack

Registered
Joined
May 3, 2001
I've got the Swiftech MCW 462 an it rocks! One thing Flash didn't mention is that the maze2.2 only won by 1*c on the average. the maze 2.2 uses 3/8" hose barbs while the Mcw462 uses/used1/4" quick connect poly tubing. you just pushed it INTO the fitting and you were off no clamps no leaking. so while the hosebarb gave like 1/4" diameter for flow rate the quick connects gave only 3/16"! and it only lost by a degree! I put 3/8" hose barbs on my MCW462. but they are correct because these sites don't review a lot of this stuff you get stuck relying on people you don't know like me hehehe. I would also agree go with a pre modded kit but I would only buy one fro Swiftech or Dangerden. they both have top of the line high high quality parts and if you e-mail or call either one you will know why their the best. If you want any help I would be glad to give you pointers and weblinks to everything you will need. I'm also open minded so I usually try not to give a biased opinion. well most of the time any way
 

SteenkyBastage

Member
Joined
Jul 9, 2001
Location
Land of OZ
fyi, i am using a danner 700 gph pump and i do NOT have to clamp the hose to the barbs (dd maze 2), as the hose fits quite snugly around the barbs, combined with the fact that the pumps you should use dont really put out any pressure.
 

Flash

Member
Joined
May 21, 2001
Richard999 (Jul 30, 2001 06:15 p.m.):
Just calculate the BTU load of a TBird processor, and then figure out how many BTUs the coolers are rated for. Then you'll have your scientific answer. I assure you, the oil coolers are far in excess of the BTU load of a TBrid CPU.

Not trying to be argumentative, but doesn't the btu calculation depend on things like the flow rate of the coolant, the amount/temperature of the air flowing through the cooler, and the temperature differential of the air and the coolant? In other words, I guess I'm asking how they measure the btu rating.
 

Flash

Member
Joined
May 21, 2001
LordSmack (Jul 30, 2001 07:00 p.m.):
I've got the Swiftech MCW 462 an it rocks! One thing Flash didn't mention is that the maze2.2 only won by 1*c on the average. the maze 2.2 uses 3/8" hose barbs while the Mcw462 uses/used1/4" quick connect poly tubing. you just pushed it INTO the fitting and you were off no clamps no leaking.

See, that's another thing. The Swiftech blocks are now using 3/8" fittings, but has anyone gone back to see how the performance stacks up *now*? Nope. BTW, you actually get more flow through the Swiftech connectors because, as Lordsmack (great name!) said, the connector fits outside the hose. Obviously normal 3/8" barbs fit inside and reduce the hose diameter.