Water to water cooling

[elvper]

Ok let's get this straight evaporation will allow for the cooling effect which is in relation to thermodynamics and will result in lower temperature but the rate at which you are planing to do this cooling is dependent to the rate of evaporation, just placing the rads in a exterior tank will evaporate water when warmed but will probably not be at a suitable rate for water cooling the computer since it is a very slow process. It will need some form of evaporator or way to heavily remove water and send it very dry air to be effective. Which also brings the problem or high humidity where if the exterior humidity is too high evaporation will not occur and overheating is a very real issue that can occur

I sorta doubt it would become too hot, as hot water strongly increase the amount of evaporation (and thus cooling).

100% humidity doesn't stop evaporation. But it causes an equal amount of evaporated water to turn back into water.

And humidity also depends on temperature. Higher temperature = more water the air can hold.

[W-L]

- I sorta doubt it would become too hot, as hot water strongly increase the amount of evaporation (and thus cooling).

- As i said would be connected to the outside through a "chimney", so unlimited amount of air where the evaporated water can go to.

 

How hot are you planning to keep your system idling at, once water absorbs the heat it will contain it for a long time before evaporation will begin to occur. 

 

What I was referring to about the humidity is if the humidity in the air is at 100%, water in a container will not evaporate since the air is fully saturated with moisture already so the container with water will only be a heat sink absorbing heat until it reaches it's max capacity unable to evaporate out of the chimney. The only way for the water to really evaporate at that time is for it to boil which will be way too hot for the computer since it can potentially blow a fitting or hose in the closed loop of the computer when it starts to create steam.

[elvper]

 

How hot are you planning to keep your system idling at, once water absorbs the heat it will contain it for a long time before evaporation will begin to occur. 

 

What I was referring to about the humidity is if the humidity in the air is at 100%, water in a container will not evaporate since the air is fully saturated with moisture already so the container with water will only be a heat sink absorbing heat until it reaches it's max capacity unable to evaporate out of the chimney. The only way for the water to really evaporate at that time is for it to boil which will be way too hot for the computer since it can potentially blow a fitting or hose in the closed loop of the computer when it starts to create steam.

 

Wrong:

- 100% humidity doesn't stop evaporation. But it causes an equal amount of evaporated water to turn back into water.

- And humidity also depends on temperature. Higher temperature = more water the air can hold.

Cold air for example could for example (with random numbers) hold 10 molecules of watervapor. But hot air can hold 20 molecules of watervapor. As we're creating heat, we're also heating the air above the surface of the water and thus the amount of water it can hold.

[ManWithBeard1990]

Could work as long as you can keep the container of water topped up and it has enough surface to evaporate water and the local humidity is low enough. Koolance makes liquid-to-liquid heat exchangers btw.

[elvper]

Could work as long as you can keep the container of water topped up and it has enough surface to evaporate water and the local humidity is low enough. Koolance makes liquid-to-liquid heat exchangers btw.

Thanks.

[W-L]

 

Wrong:

- 100% humidity doesn't stop evaporation. But it causes an equal amount of evaporated water to turn back into water.

- And humidity also depends on temperature. Higher temperature = more water the air can hold.

Cold air for example could for example (with random numbers) hold 10 molecules of watervapor. But hot air can hold 20 molecules of watervapor. As we're creating heat, we're also heating the air above the surface of the water and thus the amount of water it can hold.

 

Humidity is in direction relationship to evaporation of water, temperature is a factor that can play on the rate at which increases or decreases the evaporation rate, it's basically a state of equilibrium it is trying to achieve and balance out at which changes the rate of evaporation. At 100% humidity of the air there will still be evaporation but at a minimal rate where boiling or heating will help in fast evaporation. 

 

How temperature plays a part is the relative air temperature the higher the air temperature the higher the holding capacity of water in the air, which is correct regarding the number of molecules of water the air can hold. 

 

 

This is all theoretical and can either work great or not effectively so just going to general information here you can do it but it may require a lot tweaking and configuration. 

[NotSoEpicMods]

I see this as being more hassle than its worth. Pretty sure with any decent watercooling loop you're gonna hit the voltage or ghz barrier before you hit the temp barrier anyways with water to air exchange.

I'm the kind of person that when he does something he wants to do it about as good as freaking possible (without paying tons extra for a tiny bit extra performance).

You may not be paying in terms of money but you will be in terms of convenience for a tiny bit of cooling performance. Go ahead and do it but you'll eventually come to a point where you'll see it's not worth the hassle. Think of the surface area you'd need for the amount of evaporation required to keep your system within safe temps.

Edit:

I sorta doubt it would become too hot, as hot water strongly increase the amount of evaporation (and thus cooling).

100% humidity doesn't stop evaporation. But it causes an equal amount of evaporated water to turn back into water.

And humidity also depends on temperature. Higher temperature = more water the air can hold.

OK hot water does increase evaporation but if you only have enough surface area to dissapate 50watts of energy at 40C water temp what happens to your system if your CPU pumps out 100watts of energy? Obviously your water temp is going to increase to a point where it will equalize the evaporation rate but at that point your CPU will most likely have hit it's thermal limit because your surface area is only allowing 50watts of thermal dissipation at 40C and maybe let's say 75watts of dissipation at 60C(yes I know it doesn't act linearly like this). In short you're gonna need enough surface area to have enough evaporation to match the heat output at a given temperature. It will work you just have to have enough surface area so the water stayes within safe temp for your CPU.

I hope that made sense it's 2am and becoming increasingly hard to think.

[Schoolofmonkey]

Why do just get a mini bar fridge stick you RAD inside a bucket of water in the fridge.

Cut holes in the side of the fridge, seal them up with silicon and away you go, much better than a RAD in a bucket of water at room temps.

[RushTfe32]

Or use this blue big ice cubes people use to go to the beach. Put your rad inside this cooler, with two or three of those, meanwhile you have another 3 in your fridge, and just change them jaja

[ind3x]

Time to do physics!

Yay!

Alright, English isn't native language blablabla, i won't give it a shot translating the terms, but here the calculations are:

If your loop in total would output 125W

That would be 125J every second.

If your container contains 100 litres of water, we can calculate the following:

 

Q=m*c*Delta temperature

meaning

Q=100kg*4180*1degree celsius=41800J

 

41800J/125W=334,4seconds=5½ minute approx.

If you somehow could manage to make a bath with 100 litres of water in it, it would take a bit more than 5½ minute to warm it 1degree.

 

If you used 10 litres, it would take a tenth (33,4 seconds).

This is the speed it would warm up, but you don't warm it evenly, and i don't know the rates of which the heat would spread to the air around your container.

 

Basically, all i can tell you is how quick it would heat up, in this example i've used 125 watts of power, but if you want to overclock that can easily raise a lot.

If you want more explanations, feel free to ask

 

Best bet:

Try it out, and make a log of it

[Peopie]

@evlper I am very confused at your idea of how evaporation works and how it cools. You say that a cup a water will cool itself to below room temperature, and you showed us some article about it, but just how much cooling is it actually doing? It may be below room temperature, but it is probably less than 1 degree below. I don't think that amount of cooling will be strong enough to cool your computer.

 

If you still wanted to try anyways, I would say the amount of surface area that you will need to cool your submerged radiator will be so large that you would need some sort of underwater fan/pump to evenly distribute the heat given off by the radiator.

 

Good luck with whatever you try though, and let us know of your results!

[elvper]

Time to do physics!

Yay!

Alright, English isn't native language blablabla, i won't give it a shot translating the terms, but here the calculations are:

If your loop in total would output 125W

That would be 125J every second.

If your container contains 100 litres of water, we can calculate the following:

Q=m*c*Delta temperature

meaning

Q=100kg*4180*1degree celsius=41800J

41800J/125W=334,4seconds=5½ minute approx.

If you somehow could manage to make a bath with 100 litres of water in it, it would take a bit more than 5½ minute to warm it 1degree.

If you used 10 litres, it would take a tenth (33,4 seconds).

This is the speed it would warm up, but you don't warm it evenly, and i don't know the rates of which the heat would spread to the air around your container.

Basically, all i can tell you is how quick it would heat up, in this example i've used 125 watts of power, but if you want to overclock that can easily raise a lot.

If you want more explanations, feel free to ask

Best bet:

Try it out, and make a log of it

Ok ty

@evlper I am very confused at your idea of how evaporation works and how it cools. You say that a cup a water will cool itself to below room temperature, and you showed us some article about it, but just how much cooling is it actually doing? It may be below room temperature, but it is probably less than 1 degree below. I don't think that amount of cooling will be strong enough to cool your computer.

If you still wanted to try anyways, I would say the amount of surface area that you will need to cool your submerged radiator will be so large that you would need some sort of underwater fan/pump to evenly distribute the heat given off by the radiator.

Good luck with whatever you try though, and let us know of your results!

How much it cools varies a lot, but the hotter it gets the stronger the cooling effect. I cannot give a straight answer to it. But I doubt that a 100L watercontainer would be able to boil bause off placing a radiator from a GPU in it.

And "surface area" can be "increased" multiple times by just adding a fan that blows into the water.

[ind3x]

@elvper

Come on man, just give it a shot, i can't calculate how quickly the heat can dissipate through water -> plastic sides of bucket -> air

But if you're doing a custom waterloop, just add some easy way of adding/removing the extra tubing, and give it a shot!

If it works, then make a log, and tell us about how good it works, temperatures, overclocks etc.

[elvper]

@elvper

Come on man, just give it a shot, i can't calculate how quickly the heat can dissipate through water -> plastic sides of bucket -> air

But if you're doing a custom waterloop, just add some easy way of adding/removing the extra tubing, and give it a shot!

If it works, then make a log, and tell us about how good it works, temperatures, overclocks etc.

It's about 50-50 chance that I'll do this, not certitude. But if this is done right, I expect good results.

[KingDaddyDaniel]

Have you thought about creating a custom radiator that is more like a giant waterblock? As in two sheets of non-corrosive metal that has a high conductivity. Make it as big as a wall and fill it full of water. Connect it as you would a radiator. Place it in the middle of the room or outside.

No need to add the extra step of water unless you're really set on it. In which case, you should just make the custom radiator/waterblock and then put it in a kiddy pool in your room, fashion the chimney through a window, and post pics.

 

Edit: I had one of my architectural engineering buddies draft a visual representation of what this idea would look like.