Why is the radiator nearly always the highest point in a loop?

I'm about to build my first watercooling rig,
and I'm wondering if it would be a mistake to do things my way.

Like, why is the radiator nearly always stuck up against the roof of the machine?
Is that purely for convenience, or is there a practical reason for it?
Personally I was thinking about sticking the reservoir in as the highest place of the loop,
because to me it makes sense to keep everything flooded when you power down.
If anything above the reservoir drains out when the pump is powered off,
wont you constantly have to deal with air getting trapped?

Like, why is the radiator nearly always stuck up against the roof of the machine?
Is that purely for convenience, or is there a practical reason for it?

Click to expand...

convenience as a lot of cases have more room there then they do up top.

Being a closed loop, air shouldn't get anywhere after you bleed the loop.

Welcome!

As long as the loop is bled properly you shouldn't have any issues with air being trapped.

As stated above, if it's bled properly, air doesn't get trapped up top.

Your thinking is sound though. The higher the res is, the better. Just like in a car, the radiator cap and overflow tank are the highest point in the system to allow air to bleed properly. Unfortunately, the design of most cases prohibits this happening in PC loops.

The "If its bled, air cant be trapped" argument kind of stumps me though. Even if the system is closed, gravity will still pull the water in a loop down to the lowest point once the flow becomes stagnant, right? Or does the water somehow lock itself due to the reservoir back pressure (until you break the seal, in which case it turns into a fountain)?

So if the case permits, It would be wise to have the reservoir be the top of the loop? I am planning on using a Phantek Enthoo Primo case, which has oceans of space. So if I use a nice bay reservoir and run the radiator from the bottom of the case, it should run best like that?

The way I have it laid out now, is like this.

I would move the 480 radiator to the top, but right now I'm a little worried I'll have to throttle my machine every time it starts up, to get the air gargled out of the radiator.

I would move the 480 radiator to the top, but right now I'm a little worried I'll have to throttle my machine every time it starts up, to get the air gargled out of the radiator.

Click to expand...

you wont...

the res at the top and the pump and rad at the bottom is the way to go if you can, res and fill port as high as you can.

Once the loop is sealed it will pressurize itself. The air will escape and stay in the reservoir. Don't as me how as I am not a mad scientist but some members here are so I'll wait for them to chime in and break down the formula. You'll be ok with a radiator up top. Once the system bleeds the air from that top radiator, it will naturally be removed.

Skydiving without a parachute, gravity is a Bi^@&. But in water-cooling, gravity is your friend!

Think of it this way...

The water both enters and exits the reservoir through the bottom. For the air to end up in the radiator at the top of the loop, it would need to move to the bottom of the reservoir, and flow through the tubing into the top radiator, so the air and water could swap places. But the force of gravity on the water is greater, as water is heavier for a given volume, so any trapped air stays in the top of the reservoir, once it is bled from the radiators into the reservoir. (OK, yes... we must thank Bernoulli too, as here we're relying on his principle as well.)

Hopefully this will result in one of those "Aha!" lightbulb turning on in your brain, moments for you.

Also, most airflow recommendations are to have the top and back fans blowing out which would include the rad in the roof.

circuitwizardry said:

Skydiving without a parachute, gravity is a Bi^@&. But in water-cooling, gravity is your friend!

Think of it this way...

The water both enters and exits the reservoir through the bottom. For the air to end up in the radiator at the top of the loop, it would need to move to the bottom of the reservoir, and flow through the tubing into the top radiator, so the air and water could swap places. But the force of gravity on the water is greater, as water is heavier for a given volume, so any trapped air stays in the top of the reservoir, once it is bled from the radiators into the reservoir. (OK, yes... we must thank Bernoulli too, as here we're relying on his principle as well.)

Hopefully this will result in one of those "Aha!" lightbulb turning on in your brain, moments for you.

Click to expand...

Ha! My light finally came on (damn filament bulb!). When you mentioned both entry and exit portholes are submersed, it clicked. I kept envisioning a roof radiator like an upside down bottle, but those wont drain if the hole is underwater. I'll tell you, that's a load off my mind. I'll change my plans to stick the rad up top. Having the rad on the bottom would have blown hot air into the case, which cant be good.

Thanks for enlightening me guys!

Good thread! I just assumed the radiator was optimally placed at the top because heat rises. Some early liquid cooled motorcycles didn't have a water pump and mounted the radiator higher than the motor to ensure the hot water went to the rad and the cooler water dropped to the motor where needed. My thinking in PC cooling was "Why fight the laws of thermodynamics?"

Its a forced induction (pump) system so heat rising would actually not play a role here. Its really because there is the most room up there in most cases.

Plus this!!! ->

Also, most airflow recommendations are to have the top and back fans blowing out which would include the rad in the roof.

Click to expand...