*spin-off* Feasibility of Water Cooling Radiator Setups
- Thread starter almighty
- Start date
At first when I came across this I was all "aww hell no", but the more I think about it it seems kind of reasonable.
If these manufacturers are so hung up on being quiet as we always hear, well then you wont beat that.
If it's fully self contained and can be reliable, I can think it would be pretty cool.
If these manufacturers are so hung up on being quiet as we always hear, well then you wont beat that.
If it's fully self contained and can be reliable, I can think it would be pretty cool.
Prophecy2k
Veteran
That's what I'm thinking makes the difference. Basically the water works as a heat reservoir, absorbing more heat energy. I guess as it heats up, the speed at which heat is lost to ambient increases, offsetting the decrease in heat absorbtion from the chips. I'm sure there's all sorts of numbers that can be crunched to explain what happens.But if so, surely the size of the reservoir is important? This reminds me of the oil-bath PC that Al linked to a while back. That cooled passively, using the whole surface area of the PC to disipate heat from the oil. My ill-informed gut feeling is that you need a decent amount of water for it to work.
Not particularly Shifty. Well, you're partially correct, but he'll be seeing lower operating temperatures of the core with water because of two other things at play:
- His water is only sensible heating/cooling withing a particular (low temperature) range.
- Because he likely has a whopping great powerful pump blasting water around the circuit at a high flow rate.
Basically, a high flow rate around the circuit can negate the fact that sensibly heated/cooled water is a shitty transport medium (but not as bad as air). But to get better performance in a system you need a powerful pump to get the water flows up so high, and as Alstrong succinctly put it, a massive amount of surface area on the air side to remove the heat at the other end. The larger your flow rate, the bigger the pump needs to be, and the more surface area you need in your radiator as your water will spend less time in the radiator section of the circuit for a given volume of the stuff.
It would be impossible for him to run the same system in a console the same way, because there's no way the console power supply could take the load on the pump, nor could the radiator or pump unit itself be contained within the console.
Water cooling in a console is a pipe dream.
Also, if I understood your suggestion of a system filled with fluid shifty? I suppose that such a system could be possible, however at the external fluid/air interface how would you remove the heat without some kind of high surface area radiator? And without something to provide circulation of the fluid, your fluid would simply remain static in the system, and once the overall bulk fluid temperature has risen to the same temperature of the core, without fluid circulation through your radiator, you have too poor heat transfer there to be able to get the heat out of the system. Only way I can see something like that working is if you had some kind of external heat exchanger connected to the shell of the console, and a pump driving the fluid through it. That or a completely self contained thermosyphoning system that takes advantage of natural convection currents to drive the fluid around a loop, but then say hello to colossal consoles and core operating temperatures in the 100-250deg range (as the temperature of the fluid needs to reach a specifically low enough density for it to work).
A few things I would like to add.
- Flow rate, Higher flow doesn't affect temperatures much after around 200lph
- Adding more fluid doesn't lower the amount the water temperature raises during load, It merely extends the amount of time to reach that peak.
- Water temperature raises between 8-10c over ambient temperature during load
- The temperature of the copper base of the water block is around 12-15c over ambient temperature during load
- Fan speed has limited a effect when you reach a certain fan design
- Loop length doesn't affect performance of the loop, It just affects the time until the water reaches peak. A shorter loop has less capacity so will hit water equalibrium faster
Over the years I've stuck thermal sensors on water blocks and on the inlet and outlet points on the radiator, I've run loops with 500-600mm of tube and loops with only 200mm of tube, Various fan and radiators and just about every CPU water from the last 5 years.
- Flow rate, Higher flow doesn't affect temperatures much after around 200lph
- Adding more fluid doesn't lower the amount the water temperature raises during load, It merely extends the amount of time to reach that peak.
- Water temperature raises between 8-10c over ambient temperature during load
- The temperature of the copper base of the water block is around 12-15c over ambient temperature during load
- Fan speed has limited a effect when you reach a certain fan design
- Loop length doesn't affect performance of the loop, It just affects the time until the water reaches peak. A shorter loop has less capacity so will hit water equalibrium faster
Over the years I've stuck thermal sensors on water blocks and on the inlet and outlet points on the radiator, I've run loops with 500-600mm of tube and loops with only 200mm of tube, Various fan and radiators and just about every CPU water from the last 5 years.
Last edited by a moderator:
A sincere thank-you for taking the time to write out these clear explanations. Suddenly, cooling makes a lot more sense. I feel like I should remember this stuff from school!
I remember looking inside a dual core Mac G5 and seeing warning signs about the toxicity or corrosiveness (can't quite remember which, might have been both) of the liquid in the cooling system. Would this be because they were using a liquid with improved heat capacity and/or thermal conductivity, or would it be a simpler explanation, such as they'd added anti-freeze as a precaution?
That's because apple were idiots and used salt water as fluid which lead to corrosion of the rubber seals and eventually leaks! A well known thing among PC water coolers.
Most fluids have things added not so much to help with thermal performance but to help combat corrosion of the different metals, Also fancy UV color dyes as well.
My fluid is UV blue
The best fluid is just plain distilled water with an anti-corrosive additive or a silver kill coil.
I doubt very much that plain distilled water is the best fluid, it might be best for the systems in use, but it wouldn't cut it if they want to ship it all over the world as freezing during shipping would be a major concern. I can just imagine pallet loads of consoles dripping on the floor of best buy as their cooling systems all froze open during shipping.
Testing on pretty much every water cooling kit shows distilled water with an additive is the best performing cooling.
This is not me being dis-respectful but it seems like a lot of the people who are questioning me have little to no water cooling experience at all.
That's the bit I don't quite understand. If the water only gets a few degrees above ambient, it's not going to be efficiently transfering heat to the air. Hence surely it'll need a very large radiator? Or high air volume. I'm sure someone can provide the maths to show relation to heat disippation, water temperature and surface area. I can imagine a system like PS3 with the radiator on the bottom of a big fan and a peristatic pump moving water around (I'm pretty sure when I used them in biochem that we worked with very watery fluids, but perhaps the amount of effort to squeeze the pipe increases) geared off the fan motor, but I don't know if that's actually a feasible design.
... did you read my post? AT the very least they would need to add something to act as an antifreeze. And I doubt very much you tested every possible solution and clearly provied that plain distilled water (with an additive, which makes it not plain distilled water...) is the best performing cooling. And even if you had it would only prove that systems that were designed to use water... worked best with water. But anyway feel free to keep going hard with that experience is better than science attitude.
I've been through most fluids in the last 10 years and there's more then enough articles using using strict testing conditions that show that water is one of the best fluids to use, Fessor 1 is one of the better fluids that are designed for the purpose of water cooling. And some of these pre-made fluids are desinged to not freeze.
Some fluids have horrible performance.
Have you got much water cooling experience?
In that case answer this, how well does water with a healthy dose (50/50) of antifreeze in it work? That's what would be required to ship water in the northern hemisphere in the winter.
(note that in a car: water + antifreeze works as a better coolant than water alone so it wouldn't necessarily be a problem)
(note that in a car: water + antifreeze works as a better coolant than water alone so it wouldn't necessarily be a problem)
Wouldn't provide any extra cooling performance as the heat that would be dumped into the loop would be much lower then the heat generated by an engine so the extra cooling ability that anti-freeze has would not have chance to be utilised.
What would be of benefit is the anti-corrosive additives and the lack of freezing.
I have know people in forums to play with anti-freeze in loops but there loop was only 15-20% anti-freeze and not a 50/50 split.
But from a pure performance point of view you would be hard pressed to notice a difference.
DuckThor Evil
Legend
That's the bit I don't quite understand. If the water only gets a few degrees above ambient, it's not going to be efficiently transfering heat to the air. Hence surely it'll need a very large radiator? Or high air volume. I'm sure someone can provide the maths to show relation to heat disippation, water temperature and surface area. I can imagine a system like PS3 with the radiator on the bottom of a big fan and a peristatic pump moving water around (I'm pretty sure when I used them in biochem that we worked with very watery fluids, but perhaps the amount of effort to squeeze the pipe increases) geared off the fan motor, but I don't know if that's actually a feasible design.
I'm sure someone with a more scientific know how can answear in more deph but what you have to think is that the radiator itself is a huge chunk of metal that absorbs heat and the end tanks do indeed get quite warm to the touch.
From my own finding over the years.
With an ambient of 20c.
- Water temp would be around 25-26c after the unit has been on for a while.
- Water temp rises by 3-4c on top of that during load
- Difference in water temp before it enters and exits the radiator is only 3-4c
- CPU block base raises anything up to 40c during load.
My radiator is a Thermochil PA120.3 and with extreme fans ( 150cfm+ ) can handle up to 400w of heat load.
I have around 600mls of coolant in my loop but does a CPU and two GPU's.
Most important object in water cooling is the block performance.
One simple question about this situation... Just how exactly is a water cooler setup going to be maintained on the consumer item? Remember, it has to last at least 5 years and quite possibly 10 years.
That single fact there makes all the rest of the discussion moot.
*Part of the standard maintenance of a H20 loops is a full flush of the liquid, with an optional cleaning cycle flush, then replacement of liquid. This should be done at least once a year, with some enthusiasts preferring to do this 6 months or less.
That single fact there makes all the rest of the discussion moot.
*Part of the standard maintenance of a H20 loops is a full flush of the liquid, with an optional cleaning cycle flush, then replacement of liquid. This should be done at least once a year, with some enthusiasts preferring to do this 6 months or less.
With modern fluids they dont need to be changed.
I have persoanlly run loops for years without changing the fluid with no problems. I know others that have done the same as well. Having to change the fluid etc.etc.. is overblown by the internet.
Corsair and Antec offer products that come factory sealed, cleaned and filled that require no user maintinece at all. Its units like those that will be used by consoles.
So, what was modernized with Water over the past 2 years?
I have personally run water cooling setups for years as well, anywhere from the beginning of time with the infamous Iwaki pumps down to current Swiftech and Liang pumps with various radiators from custom made bonneville to PA Thermochill 120.2 to the various 4x120 radiators. So I too know a thing or two from first hand working knowledge in the field.
Even the best setup with proper additives (Petra's NUKE compound for instance) have needed to be changed within 2 years. The water liquid and tubes became extremely cloudy. Maybe it was the tubing, but it is the default tubing to be used, Tygon lab quality.
Even if that was just a purely cosmetic feel to it, there remains the fact that the water levels have dropped in those systems. The loops were still functional, but given another 2 years the levels would have been too low to be effective.
I have personally run water cooling setups for years as well, anywhere from the beginning of time with the infamous Iwaki pumps down to current Swiftech and Liang pumps with various radiators from custom made bonneville to PA Thermochill 120.2 to the various 4x120 radiators. So I too know a thing or two from first hand working knowledge in the field.
Even the best setup with proper additives (Petra's NUKE compound for instance) have needed to be changed within 2 years. The water liquid and tubes became extremely cloudy. Maybe it was the tubing, but it is the default tubing to be used, Tygon lab quality.
Even if that was just a purely cosmetic feel to it, there remains the fact that the water levels have dropped in those systems. The loops were still functional, but given another 2 years the levels would have been too low to be effective.
Cloudy tubing could be caused by bad fluid, I use the same for years and the only problems wrong with it are that its dyed slightly blue because of the fluid.
My one Laing DDC is nearly 5 years old now and is still as strong as it was the day I bought it.
If your water levels drooped then you were not air tight or didn't bleed the air out effectively and its just bled itself out over time and thus reduced the water levels.
In this day and age there is no way that a reliable well performing all in water cooling system can't be done.
Lets do the math for a moment ... lets say we design for 10 degrees of water heating for a 100 Watt load, what is the flow rate we need? Lets round down the specific heat of water to 4.
100/10/4 = 2.5 grams per second ... or in other words 2.5 ml per second, you really don't need that much flow to carry the heat. Flow rates are much more about keeping C/W down in the block (more turbulence) than it's about the transport.
