*spin-off* Feasibility of Water Cooling Radiator Setups

Most top end air coolers have more surface area and higher fin density then a 120mm radiator in most all in one water cooling systems.
Can you link to some specific examples of retail water cooling systems that have all the technical details about their size, area and cooling capabilities and compare them to some air cooling systems?
almighty said:
Water temp doesn't vary that much between rads ... You can have several radiators and general performance between them will be a few degrees at best...
How big role does the fan generated air flow play in all this in your opinion?
almighty said:
20-30 fins per inch
That sounds about twice as many fins than my Zalman has
almighty said:
Should easily blast a 175w heat load.
With how much air flow?
 
Not to be a complete snark...well, okay, to be a complete snark, clearly almighty is on to something that all the engineers and designers in the industry simply don't understand or can't imaging. It's innovative, ground-breaking and could really change things in tangible ways! If only the hardware designers at Intel, AMD and IBM had such new ideas!

:)
 
You wouldn't put a massive heat load on a 50^2mm radiator would you?
I could put if I combine it with a suitable fan and have it cool down pretty much anything. Only problem will occur if the air flow gets so strong that the friction from it begins to heat the whole thing up, not cool it :)
 
If you match it up to a suitable heat load then yes.

You wouldn't put a massive heat load on a 50^2mm radiator would you?
Well that's not what you said, and that's also no different to a HS+F. No matter what size HS you use, it'll never get more than 5 degrees above ambient (as long as you match it up to a suitable heat load...). In the case of consoles, the heat load isn't really the variable we control. It'll be whatever hundreds of watts these consoles produce at launch. So the variables we have to work with are cooler sizes and airflow, be that HS area and fan speed, or radiator and fan speed. Whatever cooling system is emplyed, there'll be x amount of watts heat going in and that same x watts needed to be discharged into the air. Unless radiators are intrinsically better at that than aluminium (and surely if they're made of Al, they can't be any better) then there's no way to get better heat discharge. The advantages of water cooling as I see it are a good thermal absorber for removing heat at source, and good transport to get that heat to the point of removal, whereas a heat-sink placed on the mobo needs airflow out of the box.
 
Not to be a complete snark...well, okay, to be a complete snark, clearly almighty is on to something that all the engineers and designers in the industry simply don't understand or can't imaging. It's innovative, ground-breaking and could really change things in tangible ways! If only the hardware designers at Intel, AMD and IBM had such new ideas!

:)

Intel and AMD have started offering water coolers for there CPU's.
 
Do you have any figures on how much better they are compared to their regular coolers? How about size and noise level comparisons?

http://www.legitreviews.com/article/1784/1/

Its not the best but the asetek cooler its based on is not the best one available either.

Corsairs H series are much better coolers

Personaly I think a better fan would greatly improve performance as the standard one will no doubt struggle with the 1.5" thick radiator.

If you look at the air cooler they used its massive compared to the radiator, could be double the fin area!

http://www.legitreviews.com/article/1212/1/

Wouldn't be surprised if they used the same fans that they used in the review as well which are much better then the fan on the Intel cooler.

Put the same fans on both and you might find that the watercooler produces the same temperatures while having a lot less fin area.
 
Last edited by a moderator:
You should go read [H]'s reviews of the Corsair units. They're great, but they cost 2-3x an air cooler that's almost as good. Consumer designs are never about "the best" - they're about "good enough"
 
http://www.legitreviews.com/article/1784/1/

Its not the best but the asetek cooler its based on is not the best one available either.
How can you make it better while maintaining size and noise levels or preferrably shrink it further and make it more quiet and cheaper? Also take into account that MTBF was 3x higher for the air cooler.
Corsairs H series are much better coolers
Aren't they also with much bigger fin area fans that provide significantly better air flow?
If you look at the air cooler they used its massive compared to the radiator, could be double the fin area!
True but as you saw from the tables the intel "inbox" watercooler lost on every single metric they had there.
Wouldn't be surprised if they used the same fans that they used in the review as well which are much better then the fan on the Intel cooler.
By better you mean faster spinning that move more air?
Put the same fans on both and you might find that the watercooler produces the same temperatures while having a lot less fin area.
I'd love to hear about how the physics of that theory of yours work out. Proper formulas were provided earlier in this thread, go ahead and fill them with numbers!

I found out that the radiator itself is 37mm thick. Add to that 25mm for the cooling fan and you get to 62mm thickness for the combo. That wouldn't really leave much room inside the console box for other stuff and I've yet to hear about a console that would have used 120mm fans.
 
This page: http://www.hardocp.com/article/2011/08/01/corsair_h80_high_perf_liquid_cpu_cooler_review/4

says it all...the perf per dollar is far more important to any mass-produced consumer item. All the water coolers are >2x the price of the air cooler. Scale that to mass production and it will get worse since the water cooler solution is far more laborious and requires higher-skilled labor to assemble. What's more, if you look at this page:

http://www.hardocp.com/article/2011/08/01/corsair_h80_high_perf_liquid_cpu_cooler_review/3

The air cooler is within 1.5 degrees C on the overclocked, at load, results and 0.8 degrees at stock clocks. All while being quieter than the Corsair unit.

If you can sell a 0.8 degree cooling advantage at 3x the cost with more noise to a console product development director then please be nice to him or her because they're going to get fired soon.
 
Pretty much the only reason why water cooling works that well in a PC is that you can move all the heat energy near the exhaust so you can push it directly out of the case. Most air coolers simply spin around the air within the case while hoping the case fans remove some of it. In a console that won't be true.
 
Pretty much the only reason why water cooling works that well in a PC is that you can move all the heat energy near the exhaust so you can push it directly out of the case. Most air coolers simply spin around the air within the case while hoping the case fans remove some of it. In a console that won't be true.

Ill try and do it all in one.

Improve by either making a better water block, Improving radiator fin design.... There's loads of way to improve performance without physically chaining the overall size of the unit.

The point is that with a decent fan you can have a water cooling unit that's in this case atleast half the physical size and get roughly the same performance.

Air flow rating for fans meens nothing for water cooling, Static pressure is much more important.

Fan lay out can make a such a differenece, My old lay out used to have my fans pushing the hit air from my computer through the Rad and out the top, I moved it so that my fans pushed the colder ambient air through the Rad and into the case. Better Myanmar 10x drop.

You can always have the Rad at the front of the console and have itsucking unfold ambient air instead of the warm air from inside the consoles.

You mension size? You seem to not understand that you can get radiators all different sizes and thicknesses.

A much smaller radiator would work for a console.
 
I don't happen to think that a liquid which can transfer 100 Watt with a flow rate of 2.5 ml/s with just a 10 degree temperature increase is a poor transfer medium.

Then you don't know of many heating/cooling fluids my friend ;-) The only benefit that water has is that it's cheap. And 100 Watts of heat in 2.5 ml/s with a 10deg temperature increase tells you very little about water's relative heat transfer performance. There are many oils and synthetic organic refridgerant fluids that would outperform water by a large margin. Hence why water isn't used in heatpipes ;-)
 
Then you don't know of many heating/cooling fluids my friend ;-) The only benefit that water has is that it's cheap. And 100 Watts of heat in 2.5 ml/s with a 10deg temperature increase tells you very little about water's relative heat transfer performance. There are many oils and synthetic organic refridgerant fluids that would outperform water by a large margin. Hence why water isn't used in heatpipes ;-)

If heat sinks were so super duper all top PC enthusiasts would be using them, Heat sinks are good as a cheap thing.

But if you want pure cooling performance then you move over to water cooling.

No heat sink I've ever used offers any were near the temperature drop that I've got from water cooling.

Its just hard to explain to a group of people that have little to no experience water cooling any thing.

The very few bits of information that is on the internet of water cooled Xbox 360 jasper consoles also show a massive drop in temps.

This thread is about if its feasible and it is, It even a good option from a performance stand point but because there's so little data of water cooling consoles people are not taking it. Instead relying in theory's and formulaes
 
Last edited by a moderator:
Actual thermal performance only matters up to the point it is good enough. Then it's a question of how cheaply the thermal solution can be made and installed.

The fact that these mods were custom-installed is a strike against them in the comparison. Someone took the special effort to install the cooler and to do it well, and probably did a good application of a good thermal compound.
Graphics boards ship in far fewer numbers than a console, but even they have pretty poor cooler installation a lot of the time.

This is why I wanted to know about someone re-seating a standard cooler with good thermal compound.
 
If heat sinks were so super duper all top PC enthusiasts would be using them, Heat sinks are good as a cheap thing.

But if you want pure cooling performance then you move over to water cooling.

No heat sink I've ever used offers any were near the temperature drop that I've got from water cooling.

Its just hard to explain to a group of people that have little to no experience water cooling any thing.

The very few bits of information that is on the internet of water cooled Xbox 360 jasper consoles also show a massive drop in temps.

This thread is about if its feasible and it is, It even a good option from a performance stand point but because there's so.little data of water coking consoles people are not taking it. Instead relying in theory's and formulaes

So the execs of a multi-billion $ company like MS and Sony shouldn't rely on theories and formula, and should instead look to the enthusiast PC building communty, for counsel on how to build their next consoles?

Anyhow, these "top PC enthusiasts" use water cooling in PCs precisely for the reasons posters like myself, AlStrong, Shifty, Mize and others have already explained to death in this very same thread. HOWEVER, this does not specifically make cw circuits a beneficial addition to consoles as fundamentally it seems you have a hard time consciencing that

PC != consoles

And fundamentally, aside from the cooling performance of a water loop over a HS in a console unit there are also many other (sometimes more decisive) factors that console manufacturers need to consider. A cooling water loop might be able to be designed for a console that is performant, and allows the cores to run at cool temperatures, or run at the same temps but clocked higher, however the other considerations implied by such a design may (and I expect very much would) make such a prospect entirely undesirable.

Faux Edit:
Also, I resent your remark about "people with little to no experience water cooling anything", as I've many years experience water cooling 1-2MegaWatt systems ;-) (lol)
 
Actual thermal performance only matters up to the point it is good enough. Then it's a question of how cheaply the thermal solution can be made and installed.

The fact that these mods were custom-installed is a strike against them in the comparison. Someone took the special effort to install the cooler and to do it well, and probably did a good application of a good thermal compound.
Graphics boards ship in far fewer numbers than a console, but even they have pretty poor cooler installation a lot of the time.

Very much agreed.

I would rather have console manufacturers spending more on a well engineered air-cooled solution, than a cheaply constructed and installed cw system which is comparable in implied costs.
 
It was poor cooling that caused RROD on 360 due to heat induced PCB warping.

It that case a cheap air cooler was not enough...
 
Improve by either making a better water block
That only helps with moving the heat from the chip to radiator, not getting the heat out of the console case.
Improving radiator fin design.... There's loads of way to improve performance without physically chaining the overall size of the unit.
Exactly the same can be said about regular air coolers.
The point is that with a decent fan you can have a water cooling unit that's in this case atleast half the physical size and get roughly the same performance.
I've yet to see any proof of that. At least in console context where air cooler can also dump the heat directly outside the box, not like in PC case where it can't.
Air flow rating for fans meens nothing for water cooling, Static pressure is much more important.
They basically come down to the same thing and exactly the same parameters will also effect air coolers the same way.
Fan lay out can make a such a differenece, My old lay out used to have my fans pushing the hit air from my computer through the Rad and out the top, I moved it so that my fans pushed the colder ambient air through the Rad and into the case. Better Myanmar 10x drop.
Ok but by how much did your chipset temperature rise due to that? What about PSU, hard drives and unless GPU is under water as well, that one?
You can always have the Rad at the front of the console and have itsucking unfold ambient air instead of the warm air from inside the consoles.
You can do that with air cooler just as well.
almighty said:
If heat sinks were so super duper all top PC enthusiasts would be using them, Heat sinks are good as a cheap thing.
What water cooling gives is easy distribution of the heat over a LARGE area thanks to being able to transport that heat energy to greater distances than heat pipes. That way you can effectively provide far bigger fin area to the radiator than you could with heatpipes. So basically if you need to cool down 300W+ you probably won't be able to do it with just a cheap air cooler but if you only need to cool down 150W and not more then water makes little sense for it's price.
almighty said:
The very few bits of information that is on the internet of water cooled Xbox 360 jasper consoles also show a massive drop in temps.
Jasper's power usage dropped by over 50W from the original.
It that case a cheap air cooler was not enough...
Did anyone manage to fit some kind of good enough water cooling system into the original XBox to keep it cool enough?
 
Back
Top