*spin-off* Feasibility of Water Cooling Radiator Setups

What if insulation breakdown causes a transformer to fail short? Can it cause a fire? One in a million times things go horribly wrong regardless ...
 
well it would need to be on the bottom and the side if they want to stick with the current console design.

Well thats one of the benefits of the next gens, It's not to late to alter the internal lay out of the case to fit water cooling parts.

PC enthusiasts like me do it for one reason, Overclocking....

Heat and the ability to keep the processors cool is one of the major factors that determine processor clock speed. The extra cooling performance could be used to turn the clock speeds up by an extra 20%, Maybe more!

So while it might cost a bit more per unit they could have a cooler, quieter console that could be clocked to gain an extra 20%+ more raw performance...

Still they'll more then likely air cool anyway...

What if insulation breakdown causes a transform to fail short? Can it cause a fire? One in a million times things go horribly wrong regardless ...

It depends on the fluid, I've known people on PC to have pretty bad leaks while using there machine and nothing has gone poof! Most water cooling fluids these days are designed to be as none conductive as possible so in the event of a leak you'll just turn the machine off and alert Microsoft of a hardware failure.

I've had it drip into my power supply, PC survived fine and worked perfectly after it dried, But believe me seeing coolant dripping out your case and running along you desk was a complete shit yourself moment, Leak was caused by user error :oops:
 
I'm not sure overclocking would bring any significant benefits to the table. The first limit they are likely to run into is the TDP.
 
I'm not sure overclocking would bring any significant benefits to the table. The first limit they are likely to run into is the TDP.

Overclocking always brings benefits, Especially from the GPU side of things as they scale quite nicely, A 20% GPU over clock would near enough net you 20% more performance. A 20% over clock on a console would pretty increase whole system performance by 18-19%

But yes TDP is a pain in the back side but in a case were they're hitting thermal limits before they hit power limits they could really set the clocks high if they go with water cooling.
 
Last edited by a moderator:
You really underestimate the capacity of heatpipes ...


Heat pipes were a nice invention but they're not the future because everyone would be using them and not water cooling, If you go into any decent PC forum and tell them you have a heat pipe based heat sink that out performs a decent water loop you will be laughed off the forum or have loads of people telling you the opposite.

You need lots of massive heat pipes to handle a decent load, And then a decent amount of fins to dissipate the heat coupled with loads of air flow, Massive size and no doubt weighs a bit on heavy side as well.

Even Intel and AMD are shipping pre-made water cooling systems with there CPU's now and not heat pipe based heat sinks.

Heat pipes may be better in theory but in practice they're no match for water cooling.

19052010393.jpg


Those results speak for themselves, Google shows that on air cooling they're not far off being twice as hot under the same load!
 
I bet there are more people using heat pipes than water blocks... I know there are around 120 million console users with heat pipes alone.
 
Shifty have you ever water cooled anything?
My head in hot weather? :p There are lots of things I've never done that I understand because I've had someone explain the principles behind them. I don't need to build a water cooled system to be convinced it's ideal. I just need someone to address the scientific principles that I already know.

I appreciate your enthusiasm, but you're not tackling the technical arguments in a technical fashion. "It works, so people should use it," isn't convincing in itself. I accept the empirical evidence such as your Xbox360 example, but that won't ever be enough to convince engineers to fit watercooling to tens of millions of consoles, especially when nobody else is. These systems are understood and need to be explained. If you can't explain it, then wait for someone else who can rather than just repeating yourself. ;)

For the record, as an outsider with no experience of water cooling, I accept in a new console that a single, closed unit, designed and manufactured for the job, will have no leaks or issues fitting, unlike user parts. I don't think there'll be any technical problems incorporating water cooling. My concerns are cost and efficiency, and a general curiosity to actually understand how it works because at the moment I can't reconcile the claimed results with what I know abuot heat transfer, and that's bugging me!
 
You need lots of massive heat pipes to handle a decent load, And then a decent amount of fins to dissipate the heat coupled with loads of air flow, Massive size and no doubt weighs a bit on heavy side as well.

Heat pipes may be better in theory but in practice they're no match for water cooling.

While its been stated before, Please note that there is NO difference in the amount of fins / airflow needed to remove a certain amount of heat genererated within a box, regardless of what cooling method is used.
The only advantage with water cooling for the wattage we are likely to see in a console is that the heat will efficiently be transported to the radiator so the rest of the system will see lower temperatures than an aircooled setup.

Comparing to desktops with OC are not really valid, since they are usually using 4 or more 12cm fans on the radiators. In that situation watercooling will shine, because the layout of the standard chassis does not support efficient layout of GPU coolers. (i.e. supporting 12cm fans that can be ducted directly to the outside of the chassis.
 
Those results speak for themselves
Spamming the thread twice with a huge picture of a single poorly controlled experiment doesn't make it so, it only speaks of your bias. The data for CPU coolers is more plentiful with more precise benchmarking.
 
Spamming the thread twice with a huge picture of a single poorly controlled experiment doesn't make it so, it only speaks of your bias. The data for CPU coolers is more plentiful with more precise benchmarking.

No this thread is just people not accepting results that they don't agree with.

When you get a spare 10-15mins try google searching water cooled 360 temps and you'll see just how much of a temperature drop you get.

Or maybe read some articles of water cooling PC..

And I'm biased? I'm the only one who is actually providing any kind of pictures and performance figures.

People saying this idea is rubbish compared to heat pipes, Where's the PROOF?

I'm proving my side of the argument bit no one is proving there side from aa practical point of view.

People said it would into a console, I've proved it can.

People say it won't perform very well, I provided rough figures and people dismiss them, There is no point in thread as most the people in here have no experience on the topic at hand.
 
Xbox 360 will fully integrated water cooling using separate parts which would take up more space then an 'all in one' system
What version of XB it was? Something new or the original at >>100W?

Heat pipes were a nice invention but they're not the future because everyone would be using them and not water cooling
Well, looking around at Estonian PC enthusiast forums there is at best one watercooled setup per 25-50 high-end aircooled setups.
If you go into any decent PC forum and tell them you have a heat pipe based heat sink that out performs a decent water loop you will be laughed off the forum or have loads of people telling you the opposite.
But what if I'd add "taking less room, being significantly cheaper, zero maintenance/hazard and significantly cheaper" to the mix?
You need lots of massive heat pipes to handle a decent load, And then a decent amount of fins to dissipate the heat coupled with loads of air flow, Massive size and no doubt weighs a bit on heavy side as well.
Not any more than with watercooling as the point where the heat gets removed are the heat sink fins and they need about as big surface area as water cooling radiator. There is no difference if the heat gets to those fins through water or heatpipe except that generally heatpipe radiator is at higher temperature and thus gets by with less surface are/airflow because hotter air with more heat energy gets removed from it compared to the water cooled system.
 
No this thread is just people not accepting results that they don't agree with.
No, your level of debating is weak. Yes, you've contributed some examples which is good and is a starter, but you aren't presenting fair comparisons or reasoning effectively. If watercooling is so awesome, how come heatpipes are more commonplace despite water cooling having been around for longer? (rhetorical question - it's not a sensible argument to prove the engineering value of a solution based on its popularity).

Your argument is, "water cooling is the future because people like me use it and we get lower temperatures." You then just repeat that instead of progressing the argument. As for, "When you get a spare 10-15mins try google searching water cooled 360 temps and you'll see just how much of a temperature drop you get," if people are expected to Google the answers to their questions, than the forum may as well be closed and the FAQ replaced with - "Want to know something? Don't ask here, but go Google it." ;) Your defending your position, so it's your responsibility to come up with the facts that support it when challenged, and then you should counter with your own questions and look to everyone else to provide facts.
 
People saying this idea is rubbish compared to heat pipes, Where's the PROOF?

While this is dangerously offtopic, but I would suggest that mr Fourier might have tinkered a bit with heat transport.

Ontopic: The key point in a console cooling setup is cost, as witnessed at the beginning of the current generation of consoles where the cooling solution was both loud and slightly underperforming. Still the companies must decided that the money saved per unit was worth a fraction of the users complaining.
 
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.

I was speaking more in principle. On the other hand I think your 10deg rise across the radiator is pretty optimistic. If as in almighty's previous post he gets a 4deg temperature rise across his radiator block on a PC system with likely more fans and suface area for heat transfer in his radiator, when designing a similar system for a console you'll be lucky to even get the same 4deg temperature rise, as you may have much less surface area or a lower air flow and thus worse transfer of heat into the air.
 
As you can see it can EASILY be done with existing console cases and big separate parts, So integrating it into a box that could be designed for water cooling and using a small mass produced all in one unit should not be difficult to achieve.

The argument in my mind is not how "easy" it could be done, but how reliable, cost effective etc it would be and if the performance for such would be worth the additional cost.

Water cooling a 360 as an example of what you propose for next-gen proves nothing, as there are millions of 360s and PS3s out there that manage to run fine on air. They wouldn't be sold otherwise.

What you're proposing for next-gen is a significant additional cost and complexity of cooling solution, which also introduces significant extra opportunities for console failure, for questionable* added cooling performance.

My personal argument is that I don't think the cooling benefit (which depending on the overall console system design may even be negligable), if worth the added manufacturing, shipping, reliability and complexity costs that it would imply.

My gut feeling is that this entire notion is something the console manufacturers have already considered in one way or another, and even by looking at the PC space, given that water cooling circuits only exist in the most enthusiast level PC's, it seems my gut feeling regarding the added costs of such a cooling implementation in a console may indeed be true.
 
I think it should be quite clear to everyone here that water cooling in a console would be significantly easier than in a desktop due to consistency of design and the possibility of using a completely solid setup. After all a console isn't like a car that faces lots of vibration from an engine. That does not imply however that it will be cheaper than just using heatpipes. It does imply that it actually makes more sense in such an environment and will likely by cheaper than in a desktop environment.
 
Back
Top