Another: New WC for my PC

Hello, everyone, first let me thank you for reading again one more thread about a watercool noob. (And a very big one)

So, by summer ill be investing some money on a WC for my ring, so far ive been reading stuff and looking at parts, i still havent decided everything but i already got a idea!

My idea is: not to buy a kit, but choose from parts. The assembly idea for the WC is something like this:
- A reservoir from Thermaltake a 5.25” Drive Bay Tank.
- A pump at the bottom of case, in the case upper part (The case is possibly going to be a Antec P-180)
- A radiator with 1 120mm fan on the top of case (Its a crazy idea ill explain bellow). Ive seen a rad(Alphacool NexXxoS HC 120-LE) that could handle 720cc inside, i thought it could be good since the most time water stays there the more it will cool, is this ok?
- And a block on the CPU & GPU

The flow will be: from reservoir -> pump -> radiator -> CPU -> GPU -> back to resevoir.

The crazy idea:
So, as ill probably going to get a P-180, the idea is to put the radiator on top of the case and attach it to the top fan hole! So the fan on top will blow air outside taking heat out and cooling the rad. What do you think about this idea? If i fo get that case im surelly not "modding" any hole on it for the rad!

My system specs:
Tbred - B 2600+ @ 11.5 * 192, a7n8x-deluxe, 6800GS, 1gb Ram, 3 hdds, eneremax liberty 500.

Why do i want watercool:
Mostly cause silence and since i moded my case i got that modding feeling and well you know it...

Also the WC is supposed to last for long, it ill be on this system for the next 2 years, and the ill take it to my new system. I understand that it ill probably need to change the CPU & GPU blocks, but i want minimal changes in components now chosen.

So, now i apreciate comments, sugestions and advices on this!
Thank you.
Cheers.

EDIT: i expect to spend some 150€ on this.

150 Euros is only 180 USD. That's a little light for a brand new WC system. For example, a pump from Danger Den (the D5, great pump) will run you 75 dollars. Then with a CPU and GPU block you're looking at say another 40 dollars a piece and you're already up to 155. Then of course you need a reservoir and a radiator which we'll say will cost you another 70 dollars together and you're at 225. Then you still need hose, clamps, ramsinks for your video memory, and perhaps a bit of coolant. Then you throw in shipping and you're around the 300 dollar mark or around 250 euros.

The other thing that I've learned is that you should always go Pump > CPU > GPU > Radiator > Reservoir > Pump. I've heard this from a lot of people and this is how I'll be setting up my system when it gets here this week.

lolol @ 250€ /300€

well im glad im planing ahead, so ill have 4 or 5 months to gather money!

Ive been reading that the pump adds some heat to water and it would be better to make water move into the rad first, to cool it down...

a6s0lu7 said:

Ive been reading that the pump adds some heat to water and it would be better to make water move into the rad first, to cool it down...

Click to expand...

That would have the exact opposite effect and make your rad less effective. Without going into too much detail, you want your rad at the point in your loop where your water is the hottest. Right after all the blocks for example.

Bad Maniac said:

That would have the exact opposite effect and make your rad less effective. Without going into too much detail, you want your rad at the point in your loop where your water is the hottest. Right after all the blocks for example.

Click to expand...

Wasn't it proved it doesn't matter where the rad is? I remember it was...

Flip-Mode said:

Wasn't it proved it doesn't matter where the rad is? I remember it was...

Click to expand...

I think it was proved not that it didn't matter, but that the matter is small, like .05ºc. Meaning, placing components in different places in the loop will only change the temp +/- 0.5ºc. The consensus is to order the components as such that its using the least amount of tubing possible, and not creating any tight bends or kinks.

citronym said:

I think it was proved not that it didn't matter, but that the matter is small, like .05ºc. Meaning, placing components in different places in the loop will only change the temp +/- 0.5ºc. The consensus is to order the components as such that its using the least amount of tubing possible, and not creating any tight bends or kinks.

Click to expand...

Well, for more or less 0.5c i guess ill put it where it fits better with the tubes.

So, what parts do you guys recommend? And do you think i can fit the rad on top of the case??

Thks.
cheers

a6s0lu7 said:

Well, for more or less 0.5c i guess ill put it where it fits better with the tubes.

So, what parts do you guys recommend? And do you think i can fit the rad on top of the case??

Thks.
cheers

Click to expand...

Of course you can, you can fit it anywhere you want, you just need to be a little creative as to mounting.
I was going to get a double rad and place it on top of the case but I found a deal in the classifieds I couldn't resist so I'm going with a single rad. Sorry, my point is that I was going to mount it on the top outside.

citronym said:

I think it was proved not that it didn't matter, but that the matter is small, like .05ºc. Meaning, placing components in different places in the loop will only change the temp +/- 0.5ºc. The consensus is to order the components as such that its using the least amount of tubing possible, and not creating any tight bends or kinks.

Click to expand...

I thought the one exception to this rule was that you need to put the res right before the pump, something about it damaging your pump or something to that effect. Is this incorrect?

Davan said:

I thought the one exception to this rule was that you need to put the res right before the pump, something about it damaging your pump or something to that effect. Is this incorrect?

Click to expand...

Nothing I recall.

Davan said:

I thought the one exception to this rule was that you need to put the res right before the pump, something about it damaging your pump or something to that effect. Is this incorrect?

Click to expand...

It makes it easier to fill the loop, and lessens the chance of running the pump dry.

Its recommended on pumps that can't prime themselves. So long as you have the input tube a little higher than the pump, it will be fine.

a6s0lu7 said:

The crazy idea:
So, as ill probably going to get a P-180, the idea is to put the radiator on top of the case and attach it to the top fan hole! So the fan on top will blow air outside taking heat out and cooling the rad. What do you think about this idea? If i fo get that case im surelly not "modding" any hole on it for the rad!

Click to expand...

As far as I know, using the heated air from a blow hole at the top of your case to cool a rad would not be the most ideal placement.
The air you would be using to cool the rad will all ready be above room temp.
The ideal placement should be in the front of your case, using room temp air drawn in over the rad to get the maximum cooling.
It depends on the size of the rad and the size of the case if you can physically fit the rad in there with a 120mm fan on it.
Good case ventilation will help exhaust the heat from all the components and the heat from the rad.

Some other folks on this fourm might have some better ideas for the P-180 case, I'm not sure because I've never worked with that case.

And you can save a few bucks if you use a "T" line in place of a res.

RIG RIDER said:

As far as I know, using the heated air from a blow hole at the top of your case to cool a rad would not be the most ideal placement.
The air you would be using to cool the rad will all ready be above room temp.
The ideal placement should be in the front of your case, using room temp air drawn in over the rad to get the maximum cooling.
It depends on the size of the rad and the size of the case if you can physically fit the rad in there with a 120mm fan on it.
Good case ventilation will help exhaust the heat from all the components and the heat from the rad.

Some other folks on this fourm might have some better ideas for the P-180 case, I'm not sure because I've never worked with that case.

And you can save a few bucks if you use a "T" line in place of a res.

Click to expand...

Hi.

That also crossed my mind already, but since i never worked with any watercooling i was looking for some feedback from who did.
I think the rad is the most challenging part for me, cause i want to place it in a nice place without doing any mod on the case.
Another idea is to attach it to the back of the case, but once again there is anoter fan blowing air there.
The front one seems an xcelent idea. i could also put it inside the case in the lower chamber at front so that it wont heat main components... Thks! Now ill just have to see if the rad i chose fits there.

Another question, is the rad i told above (Alphacool NexXxoS HC 120-LE) good or should i pick comething else? My understanding from reading at the rad specs is that it can have 720cc of water inside, so the more time the water stays inside the rad, the cooler it will be. Is this correct?

thnks,
cheers.

a6s0lu7 said:

Hi......<snip>....The front one seems an xcelent idea. i could also put it inside the case in the lower chamber at front so that it wont heat main components... Thks! Now ill just have to see if the rad i chose fits there.

Click to expand...

/\Acctualy that's what I was intending for your setup.

Prices in USD.

Hydor L30/Eheim 1046 = ~50
Maze4 = ~50
Swiftech Apogee = ~50
Black Ice Pro = ~30
Copper T-line (for 3/8" OD cupper tubing) and End Cap= ~1
7 feet of 1/2" clearflex = ~10
11 clamps = ~5
VGA Ramsinks = ~8

Total ~204 USD or about 170 Euros, not including the price of the coolant (You can just use 1 part automotive antifreeze, 9 parts distilled water.)

You dont' really need a reservior, it just makes things easier, but I don't see a point in it unless your pump is submerged instead of inline.

As to the placement, you have to have your T-line before the inlet of the pump in order to prime it, once it is primed all the bubbles will work their way into the T-line eventually through diffusion caused by turbulence. I would also recommend using distilled water initially to prime the pump and once the system is bleeded to add the antifreeze afterwards. You can place the rest of the rest of the components wherever you want in the loop, so long as you get the inlet and outlet barbs correct (I once mixed them up on a Swiftech Storm, whoops.)

Edit: I forgot the price of fans, so add another 6-7 Euros for a good Panaflo or Sanyo Denki 120x38mm.

a6s0lu7 said:

- A radiator with 1 120mm fan on the top of case (Its a crazy idea ill explain bellow). Ive seen a rad(Alphacool NexXxoS HC 120-LE) that could handle 720cc inside, i thought it could be good since the most time water stays there the more it will cool, is this ok?

Click to expand...

Actually, this turns out to be false. Flow rate is king. I believe there's a sticky on this topic that can explain why this is a misconception.

MoreGooder said:

Actually, this turns out to be false. Flow rate is king. I believe there's a sticky on this topic that can explain why this is a misconception.

Click to expand...

Not true. The flow rate through a rad has no effect on temps assuming non laminar flow. The faster the water travels through the loop, the sooner it comes back. No matter how fast the water flows it still spends the same percent of time in the rad. Say, at 1gpm each molecule of the water is in the rad 25% of the time. At 2gpm it's there the same 25%, just twice as often for half the time.
With rads air is king. The air never comes back, so the more air (flow) the more heat removed. That isn't to say flow rate isn't important in certain setups, like with Storm blocks.

I stand corrected. However, the original premise posed was that a large rad will mean that the water will remain in the rad longer and thus be cooled better. THAT in particular is what I was referring to, and is false. On top of that, a large rad will likely have a degradation in flow rate due to more restriction, further degrading it's impact.

You are right though. Air flow through a rad is king. I guess that makes water flow through the rad "queen" then?

Wow! I just re-read that sticky on flow rate. There have been tons of new posts to it. Man, now I'm confused too.

So, if Graystar is correct, it is false to assume that higher flow rate through a radiator is automatically superior to the mass of water in the system.

But, in the end, no matter what, Delta T is what it's all about.

How' bout this imaginative extreme example: You can't cool down a nuclear reactor with a Chevette H/C and the intake from a 747 jet engine.