1st Time WCer

[Dawgdoc]

Heres what Ive got so far. Am I missing something? Some little odds or ends? I will be cooling an 8800 GTX, and at this point in time Q6600, which will very shortly turn into an E8400 once Im done benching it.

I was all ready to do this about 3-4 months ago, and then someone convinced me to go subzero instead. Well...now Im back to finish up what I planned on starting. I am planning on making a custom WC loop that is as quiet as possible.

CPU : D-tek FuZion, Nozzle Accelerator Kit, Pro Mount Kit for FuZion

GPU : Swifttech MCW60-R + swifttech ramsinks

Rad : PA 120.3 + shroud

Fans : Antech tricool 120mm X 3. Can I use these fans? I got a killer deal at Microcenter $1.99 each and I loaded up with like a dozen If these are going to be too loud or some other issue I can get different ones without much of an issue.

Pump : Swifftech MCP655 vario

Tubing : Clearflex 7/16 ID, 5/8 OD - 20 feet. Yea more than I need, but its cheap and I always like extra pieces in case of mistakes, or future uses. Ill prob actually bump this up to 30-35 ft in case I plan on changing something on the fly later too.

Clamps : What do you think?"worm-clamps" or some of the fancy plastic clamp clips I see at petras, jab-tech, sidewinder, etc...?

Res or T-line or Fill-port: Which one and why? Im not really sure what would be easiest? One of those DD fill-ports looks about as simple and easy as possible and Im leaning towards that.

Do I need any special additives for the liquid? Any special cleaners for the parts when they arrive new?

Mounting the rad......Im having trouble visualizing this. I have one of those HUGE rocketfish cases from BB that I was planning to use, so I was going to mount the rad INSIDE of the case. I know I will need to cut some holes in the top to mod the case, and Im ok with that. If I mount INSIDE the case do I still need a shroud? How will I mount the rad? Just rig it with some screws, or nuts/bolts, or does it come with mounting equipment?

Do I need any different/extra nozzles for the rad?

Also, do I lose any cooling capacity by mounting the rad inside of the case as opposed to outside? How much louder would it be mounted outside of the case?

Its alot of questions I know But Id really appreciate any input and thx in advance!

[ben333]

Welcome to Pro/Forums!



Your selection of parts looks good I myself don't mess with any plastic crap so I'd go down to the hardware store for a handfull of metal hose clamps. Make them tight but don't crush the plastic barbs. Same with hose, I just get it at the hardware store because its cheaper but I guess thats up to you. For my T line, here is my parts list (PVC)

1" PVC T
3x 1" PVC to 1/2 threaded adapters
2x 1/2 threaded barbs
1x 1/2 threaded cap

That setup is cheap and lets you bleed your system fast. Or you can use a T line with a fill port, it takes longer to fill the system but in the end takes up less space and may look better. I don't see how nozzles could help in a radaitor, you just want the water in and out, no special way to do it Mounting the rad inside the case means that the rad is warming your case or your comp is warming the rad with hot air. Outside the case the rad gets fresh air. But the difference is small. I myself like to mount inside because it looks much better and is quieter. To mount it in a case you need to cut out the fan holes and drill the screw holes. I don't think it comes with mounting gear but you can use the same screws you'd use on a CD drive to mount the rad.

Did I miss anything? Good Luck!

[Dawgdoc]

Hey Ben

Thx for the reply and the info.

I guess I dont really fully understand how I will be using either the fill port or the res.

For the fill port, I just put a "T" in the loop somewhere, and off the 1 non-inline nipple I rub a vertical/riser tube which then connects to my fillport? And the fillport is just a cap that unscrews that allows you to drain/fill correct?

And a res.....I guess that has some kind of drainage/fill area (plug or something) and you unhook it, bring it outside the case, and drain/fill?

[billbartuska]

Originally Posted by Dawgdoc

Hey Ben

Thx for the reply and the info.

I guess I dont really fully understand how I will be using either the fill port or the res.

For the fill port, I just put a "T" in the loop somewhere, and off the 1 non-inline nipple I rub a vertical/riser tube which then connects to my fillport? And the fillport is just a cap that unscrews that allows you to drain/fill correct?

And a res.....I guess that has some kind of drainage/fill area (plug or something) and you unhook it, bring it outside the case, and drain/fill?

Something like this.

[ben333]

What bill said except not may people have drains lol.

[billbartuska]

Originally Posted by ben333

What bill said except not may people have drains lol.

I don't either.

[jaydee]

I always used the fill port for the drain. Just tip comp over. Drain is handy though if you don't want to unhook the comp to drain.

[ben333]

dawgdoc, are you a member of overclockers.com Forums?

[Dawgdoc]

Originally Posted by ben333

dawgdoc, are you a member of overclockers.com Forums?

hehe....did you just figure that out?

Yea...thats how I came across this place.

Basically, go to ANY online PC/OC forum, and if you see someone named "Dawgdoc" thats me. Ive never seen anyone else use this username ever.

[derelict]

I'm in the same boat as dawg, I'm a bit confused on the fill port. If I understand loops correctly, that fill line in the drawing could be replaced by (or basically is) a res right? And a lot of res' have a cap for filling/bleeding, is that right?

The way I envision the loop working is Pump (out) -> blocks -> rad -> pump (in), is that not correct, or not the best way?

[derelict]

Nobody had adressed the question yet, but yes you do want some additives for the water to keep it from going bad. Someone else will have to make the recommendations. I don't think you need to do anything special about new blocks/parts arriving as long as you don't wipe em on your bum , I clean with alcohol when I leave prints and such.

[billbartuska]

Originally Posted by derelict

...that fill line in the drawing could be replaced by (or basically is) a res right?
... Pump (out) -> blocks -> rad -> pump (in),

Yes, a rez and a T-line perform basically the same function. With a T-line it's easier to add water but a rez works better for bleeding. Since both of these operations are done infrequently, I prefer the easier to set up, neater T-line.

As to component order, it would be best to have the rad complete it's heat removal just before the coolant enters the block(s) so the blocks will "see" the coolest water possible, as the pump also adds heat to the coolant.
So, Pump > Rad > Block(s) > Rez or T-Line > is preferred.
But, since the coolant temp before and after the rad will only change by about 1/2*c, it may be, that in a particular setup, the added restriction of the added tubing length to accomplish the preferred component order will outweigh the benefit of using that preferred component order. Adding 3" of tubing just to get the preferred order may be counter productive, and ugly too.

Coolant:
Distilled (not deionized) water 90-95%
Automotive antifreeze (100%, ethylene glycol type) 5-10%
Iodine: A few drops.
5% antifreeze if the components are all copper/brass.
10% antifreeze if aluminum is present in the loop.

[derelict]

Yeah that makes sense temperature wise, but does having the rad before blocks slow down or weaken the pressure at all (or significantly)? I don't have enough experience with this to know which is the lesser of the two evils. Thanks for your input everyone.

[derelict]

Originally Posted by billbartuska

As to component order, it would be best to have the rad complete it's heat removal just before the coolant enters the block(s) so the blocks will "see" the coolest water possible, as the pump also adds heat to the coolant.
So, Pump > Rad > Block(s) > Rez or T-Line > is preferred.

Yeah that makes sense. Is there any noticeable drop in flow rate / head when going through the rad? I'm not sure which is the lesser of the two evils; higher water temps or decreased head. Thanks for your input bill and others.

[billbartuska]

Originally Posted by derelict

Yeah that makes sense temperature wise, but does having the rad before blocks slow down or weaken the pressure at all (or significantly)? I don't have enough experience with this to know which is the lesser of the two evils. Thanks for your input everyone.

Originally Posted by derelict

Yeah that makes sense. Is there any noticeable drop in flow rate / head when going through the rad? I'm not sure which is the lesser of the two evils; higher water temps or decreased head. Thanks for your input bill and others.

Hmmmm....... There is no evil.

Yes, the pressure is lower every time the water flows through a restriction (rad, block, tubing, etc.), but the order of the restrictions doesn't matter because the flow rate is the same everywhere in the loop.. Perhaps a picture shows it better.

[Dawgdoc]

Ok guys, project is completed! Still needs ALOT of tweaking, but all of the basic constuction/assembly is done.

Full system is Q6600, P5K dlx, 8800 GTS G92, 2 X 1 gb mushkin PC8500, Corsair 520hx, 2 X 320gb 7200.10 in RAID 0.
Cooling is: D-tek FuZion with washer and quad accelerator nozzle, MCW60 + DD Fatboys + swifttech ramsinks, PA 120.3 + EK G3/8 barbs, D5 vario, Micro-Res + DD Fatboy barbs, Tygon 7/16ID 3/8OD

I have some temp issues though.....

Ambient is 24c

CPU idle is 44/44/42/42 (Q6600 9 X 400 for 3.6Ghz), and load with P95 small FFT is 68/68/63/63 at 1.46 Vcore in BIOS.
Thats the same as my TRUE 120 on my other case which I felt was a sweatbox.

GPU idle is 40C (8800 GTS G92) and load is 45C. Clocks are 800/2000/1100.

These temps are WAY too high IMH
I have bled the system, but Im sure there could be more air bubbles somewhere. Ive turned it every which way you could possibly think of and shaken it so much my shoulders are sore LOL.....

I have new fans on the way (Yate Loon 12DH-12 from Petras) that reviewed very well on the XS fan review to replace the Antec tricools Im currently using.

I think that my Radiator mount/placement could be an issue. From top to bottom here is how it is mounted on the top of my case on the inside. RAD -> Neoprene Gasket -> Fans attached with #6X1/2 inch screws -> Fans directly attached to top of case with fan case screws. Do I need a shroud to move the fans further away from the radiator? The fans are on top of the rad in a pull configuration.....would I get better results moving them to the bottom in a push config?

Also, the hole I cutout for the rad at the top of the case is less than perfect. Cosmetically it looks good since the radgrill is covering it, but I made a simple rectangle that was about 1/8inch smaller than the screw holes on the PA120.3 This leaves approximately 1/4 inch (or so..??) of fan on the periphery which is covered by the aluminum top. Is this significant?

GPU temps are ok, but I think thats likely because its a G92 and not a G80 and produces significantly less heat. I still think it should be cooler though.

Pics to follow later today! I still need a TON of wire management but didnt make sense to start since Im going to be pulling out all of the fans later this week, and wanted to settle up with temps first.

Thx!!

[Dawgdoc]

pics as promised

[ben333]

Lookin good!

Excuse me if its already been said, but I would tip the case upside down and kind of lightly shake it to release air trapped in the radaitor, etc and then run the system until your sure you can not see any bubbles or foam in the lines.

[derelict]

Grats on the system build, I also think your temps are a bit too high, maybe bad contact with block?

[Dawgdoc]

Actually, it looks like a rats nest

But Im not even going to attempt to tackle any wire managment/cleanup until I get everything else under control.

I have had the case in every possible direction you could imagine trying to get trapped water out, but I will try it again

Thx for the info thus far!

[jaydee]

How tight are your mounting screws? Looks like the springs are barely compressed?

[Dawgdoc]

Originally Posted by jaydee

How tight are your mounting screws? Looks like the springs are barely compressed?

They are bottomed out. As tight as it can possibly go.

Its the D-tek promount kit.

[jaydee]

Originally Posted by Dawgdoc

They are bottomed out. As tight as it can possibly go.

Its the D-tek promount kit.

Hummm, maybe try a remount with fresh thermal compound. Temps do seem high for the CPU.

[Dawgdoc]

Did some more work on the loop today.

Remounted the CPU.

Repositioned the res up to the right halfway in one of the 5.25 bays.

Changed the fans on the rad to suck air IN the case from the top, repositioned the 2 120mm case fans to blow air out of the case, and added an additional 120mm fan on the floor of the case to blow air out of the bottom.

So far, after approx 45 mins on P95 small FFT Im at 60/60/58/58 and on the previous setup I was 68/68/63/63. Ambient is still 24c, and my CPU Idles are at 38/37/35/34. Yesterday idles were.....ummm Ill have to check notes but I think 44/44/42/42.

I have not had a chance to run any 3Ds yet to test GPU temps, but I idle at 35/36 now and yesterday was idling at 40/41.

Not sure which change I made helped, and not sure I care!



Still have alot more tweaking. Need to add the radshroud to improve airflow. Need to replace these crapazz Antec tricools with some D12SH-12 Yate Loons from Petras, and still have some air I need to bleed since I emptied the system out tonight and refilled it.

So far not too shabby....I still think the temps are too high on the CPU.

QUESTION: How do you use the bleed screw on the PA 120.3? I turned the case on its front end so the bleed screw is the highest point on the rad. I slowly started to open it up, and little tiny water droplets come out. No noticable air. Am I doing something incorrectly?

[ben333]

I'd open the bleed screw and hold the rad so that the bleeding hole is the highest point and gently kind of shake the radiator to be sure there is no trapped air. Then fill it to the brim and put the screw back.